Analyseit blog
https://analyseit.com/blog
Analyseit statistical analysis and visualization software blog. Provides the latest news, releases, upcoming new features and informative articles so you get the most out of Analyseit and Microsoft Excel.
engb
© 2022 Analyseit Software, Ltd.

https://analyseit.com/blog/2022/7/analyseitv610survivalanalysisandotherimprovements
Analyseit v6.10: Survival Analysis and other improvements
https://analyseit.com/blog/2022/7/analyseitv610survivalanalysisandotherimprovements
<p>We’re happy to announce a couple of major additions to the Analyseit Ultimate edition: KaplanMeier survival curves and Cox’s proportional hazards survival model. There are also improvements for users of other editions.</p>
<p>If you have <a href="https://analyseit.com/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see: <a href="https://analyseit.com/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a> or visit the <a href="https://analyseit.com/support/download">download page</a>. If maintenance on your license has expired, you can renew it to get this update and forthcoming updates, see <a href="https://analyseit.com/store/maintenance/" target="_blank">renew maintenance</a>. If you already have another edition of Analyseit, you can upgrade to the Ultimate edition for just the difference in cost, see <a href="https://analyseit.com/store/upgrade" target="_blank">upgrading Analyseit</a> .</p>
<h2>KaplanMeier survival curves (Ultimate edition)</h2>
<p><a href="https://analyseit.com/docs/userguide/survivalreliability/kaplanmeier" target="_blank">KaplanMeier survival curves</a> describe and visualize timetoevent data. They are often used in survival and reliability analysis to model time to death or time to failure of a component. You can visualize multiple survival functions, with confidence bands, and test for differences between them. In addition you can describe the key quantities such as the area under the curve or the quantiles.</p>
<p>Example of multiple KaplanMeier survival curves:</p>
<p><img src="/assets/article/62dfd69ad018d92918ea41c6/KaplanMeierSurvivalCurves.png"></p>
<h2>Proportional hazards survival model (Ultimate edition)</h2>
<p><a href="https://analyseit.com/docs/userguide/fitmodel/proportionalhazards/proportionalhazardsfit" target="_blank">Proportional hazards</a> model timetoevent data (survival/reliability analysis). They do not rely on estimating a survival function, but instead estimate the hazard ratios of covariates. We have added proportional hazards to the Analyseit Fit Model analysis so you can include multiple categorical/continuous covariates and complex interaction terms. The analysis report include everything you’d expect: parameter estimates and confidence intervals, covariance matrix, likelihood ratio and Wald test of the contribution of each term, hazard ratios for each term, and the baseline survival function and plot.</p>
<p>Example of Cox’s proportional hazard regression using WHAS500 dataset with multiple covariates and interaction terms including estimation of complex hazard ratios:</p>
<p><img src="/assets/article/62dfd69ad018d92918ea41c6/ProportionalHazards.png"></p>
<h2>Logistic regression improved odds ratio estimation (All editions)</h2>
<p>The estimation of oddratios in Logistic regression has also been improved significantly. The new user interface lets you specify the values for covariates when they are involved in interactions, allowing you to see the effect of, for example, Gender at different Ages when the model includes an interaction such as Age x Gender.</p>
<h2>Monoline user interface (All editions)</h2>
<p>The latest versions of Excel now use the new monolinestyle icons on all the ribbon commands for a clean, consistent user interface. We have updated Analyseit to automatically support this look when using the later versions of Excel.</p>
Releases
Statistics
Wed, 27 Jul 2022 08:03:53 GMT

https://analyseit.com/blog/2021/7/analyseitv590supportfortheupdatedclsiep6ed2protocolandinversepredictions
Analyseit v5.90: Support for the updated CLSI EP6Ed2 protocol and inverse predictions
https://analyseit.com/blog/2021/7/analyseitv590supportfortheupdatedclsiep6ed2protocolandinversepredictions
<p>Recently we’ve been busy updating Analyseit to stay aligned with the latest updates to the CLSI protocols, and added a new inverse prediction feature.</p><p>If you have <a href="https://analyseit.com/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="https://analyseit.com/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a> or visit the <a href="https://analyseit.com/support/download" target="">download page</a> . If maintenance on your license has expired you can renew it to get this update and forthcoming updates, see <a href="https://analyseit.com/store/maintenance/" target="_blank">renew maintenance</a>.</p>
<p><b>New CLSI EP6Ed2</b></p>
<p>The CLSI recently released guideline <a href="https://www.clsi.org/standards/products/methodevaluation/documents/ep06/" target="_blank">EP06Ed2 on the Evaluation of Linearity of Quantitative Measurement Procedures</a> , which replaces the EP06A published in 2003.</p>
<p><img src="/assets/article/610134f684c0d99e70f21d12/CLSIEP06guidelinesmall.png"></p>
<p>EP06A relied on fitting a linear (straight line), 2<sup>nd</sup> (parabolic) and 3<sup>rd</sup> (sigmoidal) order polynomials to the data. A method was then determined to be linear or possibly nonlinear based on statistical criteria. The degree of nonlinearity was then calculated as the difference between the linear fit and the best fitting nonlinear model (parabolic or sigmoidal curves). Nonlinearity could then be compared against allowable nonlinearity criteria.</p>
<p>The new CLSI EP6Ed2 protocol no longer requires fitting polynomial models to determine linearity. Instead, the deviation from linearity is calculated as the difference between the mean of each level and a linear fit through the data. That can then be compared against the allowable nonlinearity criteria. Other changes to the protocol include experimental design and there is now more focus on the structure of the variance across the measuring interval.</p>
<p>By default, the settings in Analyseit are still configured for EP6A as that is still in widespread use. However, to perform a new EP6Ed2 analysis follow these steps:</p>
<p>
</p><ol>
<li>Select a cell in the dataset.</li>
<li>On the <b>Analyseit</b> ribbon tab, in the <b>Statistical Analyses</b> group, click <b>Linearity</b>.</li>
<li>In the <b>Y</b> dropdown list, select the measured variable.</li>
<li>In the <b>By</b> dropdown list, select the level variable, and then:</li>
<ul>
<li>If the values are identifiers, select the <b>Identifier</b>, and then in the <b>Assigned values</b> grid, under the <b>Value</b> column enter the values alongside each level.</li>
<li>If the values are dilutions made by diluting a high pool or mixing high and low pools, select the <b>Relationship</b>, and then select <b>Mixture</b>, <b>Dilution</b>, or <b>Addition</b> depending on how the levels were prepared. In the <b>Assigned</b> values grid, under the <b>Value</b> column for the first and/or last level, type the value (intermediate values are automatically calculated using relative values).</li>
<li>If the values are known/expected/assigned values, select <b>Known</b> values.</li>
</ul>
<p>Note: Computation of linearity only requires the relationship between the levels, so you do not need to enter the assigned values if they are unknown.</p>
<li>Optional: To compare the nonlinearity bias against performance requirements:</li>
<ul>
<li>If the allowable nonlinearity bias is a constant or proportional value across the measuring interval, select <b>Across measuring interval</b>, and then: in the <b>Absolute</b> edit box type the bias in measurement units, and/or in the <b>Relative</b> edit box type the bias as a percentage (suffix with % symbol).<br>
<br>
Note: The allowable bias is the greater of the absolute bias and the relative bias for each level. So, with an absolute bias of 5mg/dL and a relative bias of 10%, the allowable bias will be set at 5mg/dL for all values 0 mg/dL up to 50mg/dL and then at 10% of assigned value for values above 50mg/dL.</li>
<li>If the allowable nonlinearity bias varies for each level, select <b>Each level</b> and then in the <b>Allowable nonlinearity</b> grid, under the <b>Absolute</b> / <b>Relative</b> column alongside each level, type the bias in measurement units or the bias as a percentage (suffix with % symbol).</li>
</ul>
<li>On the <b>Fit Model</b> panel, in the <b>Fit</b> dropdown list, select <b>Linear</b>.</li>
<li>In the <b>X</b> dropdown list, select <b>Expected</b> values and in the <b>Y</b> down list, select <b>Mean</b>.</li>
<li>In the <b>Weights</b> drop down list, select:
<ul>
<li><b>None</b>  Fit an ordinary regression. Use when measurement procedure exhibits constant SD over the measuring interval.</li>
<li><b>Var(Y level)</b>  Fit a weighted regression with weight based on the variance at each level. Recommended when number of replicates per level is 4 or more.</li>
<li><b>Var(Y level pooled)</b>  Fit a weighted regression with weight based on the pooled variance over a subinterval of levels. Recommended when number of replicates per level is 2 or 3.</li>
<li><b>VarFn(Y)</b>  Fit a weighted regression with weight based on the variance function at the mean of each level. Recommended when the precision can be modeled by a variance function.</li>
</ul></li>
<li>If the levels are made by a dilution of a high level, select <b>Force through zero</b> check box. Otherwise, if the levels are produced by mixing a high and low level, clear the <b>Force through </b><b>origin</b> check box.</li>
<li>Click <b>Calculate</b>.</li>
</ol><p></p>
<p><b>New Inverse prediction feature</b></p>
<p>This release also includes a new inverse prediction feature for simple linear regression models. Inverse prediction has several uses including estimating the shelflife of a product. In the context of CLSI protocols it is useful for <a href="https://www.clsi.org/standards/products/methodevaluation/documents/ep25/" target="_blank">CLSI EP25A – Evaluation of Stability of In Vitro Diagnostic Reagents</a> . </p>
<p><img src="/assets/article/610134f684c0d99e70f21d12/InversePredictionLinearFit2.png"></p>
<p>To make an inverse prediction use Fit Model to fit a simple regression model, then:</p>
<ol>
<li>Activate the analysis report worksheet.</li>
<li>On the <b>Analyseit</b> ribbon tab, click <b>Predict</b>, and then click <b>X given Y</b>.<br>
The analysis task pane Inverse Predict X given Y panel opens.</li>
<li>In the <b>Predictions</b> list box, under the <b>Y</b> column, type the values for the response variable.</li>
<li>Optional: In <b>the Confidence interval</b> edit box, type the confidence level, and then in the dropdown list select the confidence bounds. Then in the <b>Method</b> dropdown list, select <b>Simultaneous</b> for intervals that ensure you achieve the confidence level simultaneously for all predictions or <b>Individual</b> for intervals that only ensure confidence for each individual prediction. </li>
<li>Optional: On the <b>Fit</b> panel, in the <b>Predicted</b> values dropdown list, select the style to plot the predicted values on the scatter plot.</li>
<li>Click <b>Recalculate</b>.</li>
</ol><p></p><p></p><p></p><p><br></p><p></p><p><a href="https://analyseit.com/blog/2014/6/recentimprovementsinanalyseit376andourfirstvideotutorial"></a></p><p></p>
Method validation
Releases
Statistics
Using Analyseit
Wed, 28 Jul 2021 12:33:37 GMT

https://analyseit.com/blog/2020/6/analyseit550to565recentimprovements
Analyseit 5.50 to 5.65: Recent improvements
https://analyseit.com/blog/2020/6/analyseit550to565recentimprovements
<p>We've been busy making lots of improvements to Analyseit recently, so we thought we'd highlight the significant new and improved features.</p>
<p>If you have <a href="https://analyseit.com/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a>, you can download and install the update now, see <a href="https://analyseit.com/docs/userguide/updating" target="_blank">updating the software</a>. If maintenance on your license has expired, you can renew it to get this update and forthcoming updates, see <a href="https://analyseit.com/store/maintenance/" target="_blank">renew maintenance</a>.</p>
<h2>Probit regression</h2>
<p>Fit Model now supports Probit regression. Probit regression is similar to logistic regression, as both use a link function to transform a linear model into a nonlinear relationship. A linear model uses the equation Y = α + β x, whereas both logit and probit equation use the form Y = <i>f</i>(α + β x). They only differ in the definition of the link function <i>f</i>(): the logit model uses the cumulative distribution function of the logistic distribution; the probit model uses the cumulative distribution function of the standard normal distribution. Both functions give a predicted probability, Y.</p>
<p>Health sciences, such as epidemiology, often use the logit model as the predictor coefficients are interpretable in terms of log oddsratios. The probit model coefficients cannot be interpreted as easily but may produce a better fitting model in other scenarios. For example, in method validation, probit regression is used to model the hit rate of a molecular test. You can then use the model to establish a detection limit or determine diagnostic cutoff points from an underlying continuous response. For a guided example, see our blog post <a href="https://analyseit.com/blog/2020/5/covid19calculatingthedetectionlimitforasarscov2rtpcrassayusinganalyseit" target="">Calculating the detection limit for a SARSCoV2 RTPCR test.</a><br>
</p>
<h2>Transform</h2>
<p>You can now apply a transformation to a variable during analysis. This feature is currently available on the Distribution and Fit Model (simple regressions) analysis, but it will be available in all analyses soon. To transform a variable, click the properties icon next to the variable selector dropdown, choose Transform, and then select the transformation function. <br>
</p>
<p><img src="/assets/article/5ed60a6f6eba9a1974e10fac/transformfn.png"></p>
<h2>Diagnostic performance</h2>
<p>A lot of customers buy Analyseit for ROC analysis, as it has always lead the way in diagnostic test analysis (<a href="https://pubmed.ncbi.nlm.nih.gov/12600955/" target="_blank">https://pubmed.ncbi.nlm.nih.gov/12600955/</a>). We recently extended the Binary (Sensitivity/Specificity) test to allow testing equivalence and noninferiority hypotheses tests. And, now you can calculate predictive values for different population prevalences – ideal for modelling the behavior of a test in different scenarios.</p>
<h2>Method Comparison</h2>
<p>Qualitative method comparison is now more prominent on the Method Comparison command menu, with clearer titles: Binary and SemiQuantitative. We added Average agreement measures, which are useful when there is no reference/comparative method (for example when comparing laboratories or observers). There is also an excellent new plot for visualizing agreement between qualitative methods: the Bangdiwala agreement plot. </p>
<p><img src="/assets/article/5ed60a6f6eba9a1974e10fac/bangdiwala.PNG"></p>
<h2>CLSI EP15A3 tutorial video</h2>
<p>Finally, we released a new <a href="https://analyseit.com/docs/video/verifyprecisionclsiep15" target="">video tutorial for the CLSI EP15A3 protocol</a>.</p>
<p>We're keen to hear what other tutorials you would like to see for CLSI EP protocols. <a href="http:// https://analyseit.com/company/contactus" target="">Contact us</a> to let us know what you would like to see covered, and we'll put together more tutorials over the next few months. <br>
</p>
<p><b>For more information, see the online documentation:</b><br>
<a href="https://analyseit.com/docs/userguide/methodcomparison/estimatingbinaryagreement" target="">Estimating agreement between two binary or semiquantitative methods </a><br>
<a href="https://analyseit.com/docs/userguide/methodcomparison/agreementplot" target="">Agreement plot </a><br>
<a href="https://analyseit.com/docs/userguide/fitmodel/logistic/fittingsimpleprobitregression" target="">Fitting a simple probit regression </a><br>
<a href="https://analyseit.com/docs/userguide/diagnosticperformance/testingsesp" target="">Comparing the accuracy of two binary diagnostic tests </a></p>
Releases
Thu, 04 Jun 2020 08:13:23 GMT

https://analyseit.com/blog/2020/5/covid19calculatingthedetectionlimitforasarscov2rtpcrassayusinganalyseit
COVID19: Calculating the detection limit for a SARSCoV2 RTPCR assay using Analyseit
https://analyseit.com/blog/2020/5/covid19calculatingthedetectionlimitforasarscov2rtpcrassayusinganalyseit
<p>Recent improvements to the <a href="https://analyseit.com/products/methodvalidation" target="">Analyseit Method Validation edition</a>, in version 5.50 and later, include the addition of probit regression. Probit regression is useful when establishing the detection limit (LoD) for an RTqPCR assay. </p>
<p>The <a href="https://clsi.org/standards/products/methodevaluation/documents/ep17/" target="_blank">CLSI EP17A2: Evaluation of Detection Capability for Clinical Laboratory Measurement Procedures</a> protocol provides guidance for estimating LoD and is recognized by the FDA. In this blog post, we will look at how to perform the relevant part of the CLSI EP17A2 protocol using Analyseit. </p>
<p>For details on experimental design, see section 5.5 in the CLSI EP17A2 guideline. In Analyseit, you should arrange the data in 2 columns: the first should be the concentration, and the second should be the result, positive or negative. You should have a minimum of 20 replicates at each concentration. We have put together a hypothetical example in the workbook <a href="/assets/article/5ebbd00c6eba9a230885c431/COVID19 Detection Limit.xlsx">COVID19 Detection Limit.xlsx</a> which you can use the follow the steps below:</p>
<p>To Analyseit:</p>
<ol>
<li>Select a cell in the dataset.</li>
<li>On the <b>Analyseit</b> ribbon tab, in the <b>Statistical Analyses</b> group, click <b>Fit Model</b>, and then click <b>Probit</b>.
<p>The analysis task pane opens.</p>
</li>
<li>In the <b>Y </b>dropdown list, select the Result variable, and then in the <b>Event</b> dropdown list, select the Positive state (1).
</li><li>In the <b>X</b> dropdown list, select the RNA variable, and then click the dropdown menu icon next to the variable dropdown, and in the <b>Transform</b> dropdown list, select <b>Log 10</b>.
</li>
<li>In the <b>Plot</b> dropdown list, select <b>Scaled  X against Y</b> to show the X axis as a logarithmic scale. </li>
<li>Clear the <b>Individual</b> checkbox, and select the <b>Aggregate</b> checkbox.</li>
<li>On the <b>Analyse</b><b>it</b> ribbon tab, in the <b>Fit Model</b> group, click <b>Predict</b>, and then click <b>X given Probability</b>.</li>
<li>On the analysis task pane, on the <b>Fit</b> panel, next to the <b>Predict X given Probability</b>, click <b>More</b>…, and then on the <b>Predict X given Probability</b> panel, click <b>Add</b> and select <b>LoD=95%</b>.
<p>NOTE: If using Analyseit preversion 5.65, on the <b>Fit</b> panel, in the <b>Predict X given Probability</b> edit box, type 0.95.</p>
</li>
<li>Click <b>Calculate</b>.</li>
<p>The analysis report shows the probit regression fit:</p>
<p><img src="/assets/article/5ebbd00c6eba9a230885c431/Probit regression limit of detection.PNG"></p>
<p>The Predict X for given Probability section shows the limit of detection (LoD) as the point where the probability of a positive result is 95%.</p>
<p>For more information, see the online documentation:<br>
<a href="https://analyseit.com/docs/userguide/measurementsystemsanalysis/detectioncapability/fittingdetectionlimitprobitregression" target="">Estimating the detection limit using a probit fit </a><br>
</p>
</ol>
Method validation
Statistics
Using Analyseit
Wed, 13 May 2020 13:10:44 GMT

https://analyseit.com/blog/2020/5/covid19calculatingppanpaagreementmeasuresusinganalyseit
COVID19: Calculating PPA/NPA agreement measures using Analyseit
https://analyseit.com/blog/2020/5/covid19calculatingppanpaagreementmeasuresusinganalyseit
<p>Following our last blog post, today, we will show how to calculate binary agreement using the <a href="https://analyseit.com/products/methodvalidation" target="">Analyseit Method Validation edition</a>. The <a href="https://clsi.org/standards/products/methodevaluation/documents/ep12/" target="_blank">CLSI EP12: User Protocol for Evaluation of Qualitative Test Performance</a> protocol is a useful companion resource for laboratories and diagnostic companies developing qualitative diagnostic tests.<br>
</p>
<p>In Analyseit, you should arrange the data in frequency or case form, as discussed in the blog post: <a href="https://analyseit.com/blog/2020/4/usinganalyseittoestablishthediagnosticaccuracysensitivityspecificityofacovid19test" target="">Using Analyseit to establish the diagnostic accuracy (sensitivity/specificity) of a COVID19 test</a>. You can find an example of both and follow the steps below, using the workbook <a href="/assets/article/5eb51377daedf737c040c7a1/COVID19 Agreement Example.xlsx">COVID19 Agreement Example.xlsx</a>.</p>
<p>To Analyseit:</p>
<ol>
<li>Select a cell in the dataset.</li>
<li>On the <b>Analyseit</b> ribbon tab, in the <b>Statistical Analyses</b> group, click <b>Method Comparison</b>, and then click <b>Binary (PPA/NPA)</b> or <b>PPA/NPA</b> depending on the version of Analyseit you are using.
<p>The analysis task pane opens.</p>
</li>
<li>In the <b>X (Reference / Comparative) </b>dropdown list, select the Comparative method variable.</li>
<li>In the <b>Y (Test /New)</b> dropdown list, select the Test method variable.</li>
<li>If the data are in frequency form, in the Frequency dropdown list, select the frequency variable # containing the counts.</li>
<li>Select the <b>Agreement by category</b> checkbox, and in the <b>Measure</b> dropdown, select <b>Reference</b> to calculate the proportion of the Test method results that agree with the Comparative method.
<p>NOTE: The <b>Average</b> method is useful when comparing two laboratories or observers where neither is considered a natural comparator. The reference method is asymmetric, and the result will depend on the assignment of the X and Y methods, whereas the average method is symmetric, and the result does not change when swapping the X and Y methods. </p>
<p>INFO: Older versions of Analyseit do not support the Average method, and the Agreement by category checkbox is called Agreement.</p>
</li>
<li>Click <b>Calculate</b>.</li>
</ol>
<p>The analysis report shows positive and negative agreement statistics.</p>
<p><img src="/assets/article/5eb51377daedf737c040c7a1/COVID19 Agreement Example.png"></p>
<p>The positive agreement is 93.2%, and the negative agreement is 86.0%. You can see from the contingency table that the Comparative method identified 59 positives and 43 negatives, whereas the Test method identified 61 positives, 41 negatives. There are 55 positive agreements, 37 negative agreements, and 4 disagreements where the Comparative method is positive and the Test method negative, 6 disagreements where the Comparative method is negative and the Test method positive.</p>
<p>The agreement plot is a visual representation of the agreement:</p>
<ul>
<li>Black rectangle outlines represent the maximum agreement possible (as represented by the marginal totals of the contingency table). If the marginal totals are the same, that is, both tests identify the same number of positives and negatives, the rectangles will be square and intersect with the diagonal identity line.</li>
<li>If one test tends to categorize more results as positive/negative than another, the rectangle will not be square and will not intersect the diagonal.</li>
<li>Blue filled rectangles represent the observed agreement (as represented by the diagonal cells of the contingency table). If you read down the vertical axis labeled Comparative method, this is an indication of the Test method agreement with the Comparative method. Likewise, if you read across the horizontal axis, this is the Comparative method agreement with the Test method.</li>
<li>The relationship between the area of the shaded to the unshaded rectangle is a measure of the average agreement.</li>
</ul>
<p>For more information, see our online documentation:<br><a href="https://analyseit.com/docs/userguide/methodcomparison/agreementmeasures" target="">Agreement measures for binary and semiquantitative data</a><br><a href="https://analyseit.com/docs/userguide/methodcomparison/agreementplot" target="">Agreement plot</a><br><a href="https://analyseit.com/docs/userguide/methodcomparison/estimatingbinaryagreement" target="">Estimating agreement between two binary or semiquantitative methods</a></p>
Method validation
Statistics
Using Analyseit
Fri, 08 May 2020 08:53:50 GMT

https://analyseit.com/blog/2020/4/diagnosticaccuracysensitivityspecificityversusagreementppanpastatistics
Diagnostic accuracy (sensitivity/specificity) versus agreement (PPA/NPA) statistics
https://analyseit.com/blog/2020/4/diagnosticaccuracysensitivityspecificityversusagreementppanpastatistics
<p>Due to COVID19, there is currently a lot of interest surrounding the sensitivity and specificity of a diagnostic test. These terms relate to the accuracy of a test in diagnosing an illness or condition. To calculate these statistics, the true state of the subject, whether the subject does have the illness or condition, must be known. </p>
<p>In recent FDA guidance for laboratories and manufacturers, <a href="https://www.fda.gov/regulatoryinformation/searchfdaguidancedocuments/policydiagnostictestscoronavirusdisease2019duringpublichealthemergency" target="_blank">“FDA Policy for Diagnostic Tests for Coronavirus Disease2019 during Public Health Emergency”</a>, the FDA state that users should use a clinical agreement study to establish performance characteristics (sensitivity/PPA, specificity/NPA). While the terms sensitivity/specificity are widely known and used, the terms PPA/NPA are not. </p>
<h2>Agreement statistics</h2>
<p><a href="https://clsi.org/standards/products/methodevaluation/documents/ep12/" target="_blank">CLSI EP12: User Protocol for Evaluation of Qualitative Test Performance</a> protocol describes the terms <i>positive percent agreement</i> (PPA) and <i>negative percent agreement</i> (NPA). When you have two binary diagnostic tests to compare, you can use an agreement study to calculate these statistics. <br>
</p>
<p><img src="/assets/article/54611655/PPANPA.PNG"></p>
<ul>
<li>Positive agreement is the proportion of comparative/reference method positive results in which the test method result is positive.</li>
<li>Negative agreement is the proportion of comparative/reference method negative results in which the test method result is negative.</li>
</ul><p>As you can see, these measures are asymmetric. That is, interchanging the test and comparative methods, and therefore the values of b and c, changes the statistics. They do, however, have a natural, simple, interpretation when one method is a reference/comparative method and the other a test method.</p>
<h2>What agreement statistics don’t tell us</h2>
<p>Although the formulae for positive and negative agreement are identical to those for sensitivity/specificity, it is essential to distinguish them as the interpretation is different. </p>
<p>We have seen product information for a COVID19 rapid test use the terms ‘relative’ sensitivity and ‘relative’ specificity when comparing against another test. The term ‘relative’ is a misnomer. It implies that you can use these 'relative' measures to calculate the sensitivity/specificity of the new test based on the sensitivity/specificity of the comparative test. That is simply not possible.</p>
<p>It is also not possible, from these statistics, to determine that one test is better than another. Recently a national UK newspaper ran an article about a PCR test developed by Public Health England and the fact it disagreed with a new commercial test in 35 out of 1144 samples (3%). Of course, to many journalists, this was evidence that the PHE test was inaccurate. There is no way to know which test is correct and which incorrect in any of those 35 disagreements. We simply do not know the true state of the subject in agreement studies. Only by further investigation of those disagreements would it be possible to identify the reason for the discrepancies.</p>
<p>To avoid confusion, we recommend you always use the terms positive agreement (PPA) and negative agreement (NPA) when describing the agreement of such tests.</p>
<p>In the next blog post, we show you how to use Analyseit to perform the agreement test with a worked example.</p>
For more information, see our online documentation:<br><a href="https://analyseit.com/docs/userguide/methodcomparison/agreementmeasures" target="">Agreement measures for binary and semiquantitative data</a><br><a href="https://analyseit.com/docs/userguide/methodcomparison/agreementplot" target="">Agreement plot</a><br><br>
Method validation
Statistics
Using Analyseit
Wed, 22 Apr 2020 14:06:43 GMT

https://analyseit.com/blog/2020/4/covid19establishingthediagnosticaccuracysensitivityspecificityofatestusinganalyseit
COVID19: Establishing the diagnostic accuracy (sensitivity/specificity) of a test using Analyseit
https://analyseit.com/blog/2020/4/covid19establishingthediagnosticaccuracysensitivityspecificityofatestusinganalyseit
In the last couple of blog posts, we’ve covered some of the statistics related to the diagnostic accuracy of COVID19 antibody tests. In this post, we will demonstrate how to establish these claims using the <a href="https://analyseit.com/products/methodvalidation" target="">Analyseit Method Validation edition</a>. A useful companion resource for laboratories and diagnostic companies developing qualitative diagnostic tests is <a href="https://clsi.org/standards/products/methodevaluation/documents/ep12/" target="_blank">CLSI EP12: User Protocol for Evaluation of Qualitative Test Performance.</a><br>
<p>It is important in diagnostic accuracy studies that the true clinical state of the patient is known. For example, in developing a SARSCoV2 antibody test, for the positive subgroup, you might enlist subjects who had a positive SARSCoV2 PCR test and clinically confirmed illness. Then, for the negative subgroup, you might use samples taken from subjects before the illness was in circulation. It is also essential to consider other factors, such as the severity of illness, as they can have a marked effect on the performance characteristics of the test. A test that shows high sensitivity/specificity in a hospital situation in very ill patients can be much less effective in population screening where the severity of the illness is less.<br>
</p>
<p>In cases where the true condition of the subject is not known, and only results from a comparative method and a new test method are available, an agreement measure is more suitable. We will cover that scenario in detail in a future blog post.<br>
</p>
<h2>Statistical analysis of diagnostic test study data</h2>
<p>In Analyseit, there are two ways to arrange your data for this analysis.<br>
</p>
<p>Frequency form data summarizes the frequency counts for each combination of true state, test result:<br>
<img src="/assets/article/54611654/testaccuracyfrequencyformdata.png"></p>
<p>Case form data lists the individual true state and test result for each subject:<br>
<img src="/assets/article/54611654/testaccuracylistdata.png"></p>
<p>You can find examples of both, and follow along with the steps below, using the workbook <a href="/assets/article/54611654/COVID19 Diagnostic Accuracy Example.xlsx">COVID19 Diagnostic Accuracy Example.xlsx</a></p>
<p>To Analyseit:</p>
<ol>
<li>Select a cell in the dataset.</li>
<li>On the <b>Analyseit</b> ribbon tab, in the <b>Statistical Analyses</b> group, click <b>Diagnostic</b>, and then click <b>Binary (Sensitivity / Specificity)</b>.<br>
The analysis task pane opens.</li>
<li>In the <b>True state</b> dropdown list, select the true state variable.</li>
<li>In the <b>Positive event</b> dropdown list, select the state that indicates the presence of the condition of interest.</li>
<li>In the <b>Y</b> dropdown list, select the test variable.</li>
<li>In the <b>Positive event</b> dropdown list, select the state that indicates a positive test.</li>
<li>If the data are in frequency form, in the <b>Frequency</b> dropdown list, select the frequency variable containing the counts.</li>
<li>Select the required statistics Sensitivity/Specificity, Likelihood ratios.<br>
<i>NOTE: Hover the mouse pointer over an option to see a contextsensitive help popup with more information and notes on when to use it.</i></li>
<li>Click <b>Calculate</b>.</li>
</ol>
<p>The analysis report shows the sensitivity/specificity and other statistics. </p>
<p><img src="/assets/article/54611654/covid19testreport.png"></p>
<p>To compute the predictive values given the prevalence of illness in a population:</p>
<ol>
<li>On the Analyseit task ribbon tab, in the <b>Diagnostic Accuracy</b> group, click <b>Predictive value</b><b>s</b>.<br>
The analysis task pane opens, and the Predictive value checkbox is selected.</li>
<li>In the <b>Prior probability</b> edit box, type the prevalence. If you have more than one scenario, click the <b>More</b> button, and then in the <b>Prior probabilities</b> grid, type the prevalences and optional scenario names.</li>
<li>Click <b>Recalculate</b>.</li>
</ol>
<p>If you have any questions about using the <a href="https://analyseit.com/products/methodvalidation" target="">Analyseit Method Validation edition</a> for diagnostic accuracy studies, please contact us.</p>
<p>For more information, see our online documentation:<br><a href="https://analyseit.com/docs/userguide/diagnosticperformance/diagnosticaccuracy" target="_blank">Measures of diagnostic accuracy</a><br>
<a href="https://analyseit.com/docs/userguide/diagnosticperformance/estimatingsensitivityspecificity" target="_blank">Estimating the sensitivity and specificity of a binary test</a></p>
Method validation
Statistics
Using Analyseit
Wed, 15 Apr 2020 18:39:29 GMT

https://analyseit.com/blog/2020/4/whythediagnostictestaccuracystatisticisuseless
Why the diagnostic test 'accuracy' statistic is useless
https://analyseit.com/blog/2020/4/whythediagnostictestaccuracystatisticisuseless
<p>In our last post, we mentioned that the <i>'accuracy'</i> statistic, also known as the probability of a correct result, was a useless measure for diagnostic test performance. Today we'll explain why. </p>
<p>Let's take a hypothetical test with a sensitivity of 86% and specificity of 98%.</p>
<p>As a first scenario we simulated test results on 200 subjects with, and 200 without, the condition. The <i>accuracy </i>statistic (TP+TN)/N is equal to (172+196)/400 = 92%. See below:</p>
<p><img src="/assets/article/54611653/accuracy1.png"></p>
<p>In a second scenario we again simulated test results on 400 subjects, but only 50 with, and 350 without, the condition. The <i>accuracy </i>statistic is (43+343)/400 = 96.5%. See below:</p>
<p><img src="/assets/article/54611653/accuracy2.png"></p>
<p>The <i>accuracy</i> statistic is effectively a weighted average of sensitivity and specificity, with weights equal to the sample prevalence P(D=1) and the complement of the prevalence (that is, P(D=0) = 1P(D=1)). </p>
<p>Accuracy = P(TP or TN) = (TP+TN)/N = Sensitivity * P(D=1) + Specificity * P(D=0)</p>
<p>Therefore as the prevalence in the sample changes so does the statistic. The prevalence of the condition in the sample may vary due to the availability of subjects or it may be fixed during the design of the study. It's easy to see how to manipulate the accuracy statistic to weigh in favor of the measure that performs best. <br></p><p>
</p><p>Finally, the 'accuracy' statistic is not a good way of
comparing diagnostic tests. As illustrated above, it is affected by the
prevalence in the samples of the tests. And, even given two tests, with sample prevalence
50%, suppose test A has a sensitivity of 100%, but specificity 0%, and test B
has a sensitivity of 0% and specificity 100%. Both tests give the same probability
of a correct result, yet the two tests are radically different.</p><p>The measures of the accuracy of a diagnostic test are sensitivity and specificity. The <i>accuracy </i>statistic, the probability of a correct test result, is <b>not</b> a measure of the intrinsic accuracy of a test.</p><p>For more information, see our online documentation:<br><a href="https://analyseit.com/docs/userguide/diagnosticperformance/diagnosticaccuracy" target="_blank">Measures of diagnostic accuracy</a><br></p>
Method validation
Statistics
Using Analyseit
Tue, 07 Apr 2020 12:27:28 GMT

https://analyseit.com/blog/2020/4/sensitivityspecificityandtheimportanceofpredictivevaluesforacovid19test
Sensitivity/Specificity and The Importance of Predictive Values for a COVID19 test
https://analyseit.com/blog/2020/4/sensitivityspecificityandtheimportanceofpredictivevaluesforacovid19test
<p>There’s currently a lot of press attention surrounding the fingerprick antibody IgG/IgM strip test to detect if a person has had COVID19. Here in the UK companies are buying them to test their staff, and some in the media are asking why the government hasn’t made millions of tests available to find out who has had the illness and could potentially get back to work.</p>
<p>We did a quick Google search, and there are many similarlooking test kits for sale. The performance claims on some were sketchy, with some using as few as 20 samples to determine their performance claim! However, we found a webpage for a COVID19 IgG/IgM Rapid antibody test that used a total of 525 cases, with 397 positives, 128 negatives, clinically confirmed. We have no insight as to the reliability of the claims made in the product information. The purpose of this blog post is not to promote or denigrate any test but to illustrate how to look further than headline figures.</p>
<p>We ran the data through the <a href="https://analyseit.com/products/methodvalidation">Analyseit Method Validation Edition</a> version 5.51. Here's the workbook containing the analysis: <a href="/assets/article/54611652/COVID19 IgMIgG Rapid Test.xlsx">COVID19 IgMIgG Rapid Test.xlsx</a></p>
<p><img src="/assets/article/54611652/covid19test.png"></p>
<h2>Sensitivity/Specificity and Predictive values</h2>
<p>We used Analyseit to determine the sensitivity/specificity and confirmed the performance claims on the website. The sensitivity is listed as 88.66% and the specificity 90.63%. Some websites also make an “accuracy” claim, usually a combination of (TP+TN)/Total, but that’s a useless statistic.</p>
<p>So, with reasonably impressive numbers around 90%, what’s the problem?</p>
<p>First, we need to look at the meaning of sensitivity and specificity:</p><p></p><ul><li>Sensitivity measures the ability of a test to detect the condition when it is present. It is the probability that the test result is positive when the condition is present.</li><li>Specificity measures the ability of a test to detect the absence of the condition when it is not present. It is the probability that the test result is negative when the condition is absent.</li></ul>
<p>These measures give the probability of a correct test result in subjects known to be with and without a condition, respectively. To understand how a test performs in the real world, we need to look at the predictive values. They tell us how useful the test would be when applied to a population. They indicate the probability of correctly identifying a subject's condition given the test result.</p>
<p>Here are the definitions of positive and negative predictive value:</p><p></p><ul><li>Positive predictive value is the probability that a subject has the condition given a positive test result</li><li>Negative predictive value is the probability that a subject does not have the condition given a negative test result.</li></ul>
<h2>The unknown!</h2>
<p>Here’s where things get a little trickier. To calculate predictive values, the important numbers that we need to make decisions, we need to know the prevalence of COVID19 in the population. And, at present, that’s an unknown.</p>
<p>We ran the numbers in Analyseit using four scenarios for the prevalence of the illness of 1%, 5%, 10%, 20%.<br></p>
<p><img src="/assets/article/54611652/covid19predictiveValues.png"></p>
<p>The positive predictive value, that is, the probability that someone with a positive test result from this test has had COVID19 illness is 8.7%, 33.2%, 51.2%, and 70.3%, respectively. Whilst the negative predictive values are 99.9%, 99.3%, 98.6%, 97.0%</p>
<h2>A practical example</h2>
<p>To understand what’s going on here, we’ll use an example to show how the positive predictive value works.</p>
<p>Let’s assume you have a workforce of 100 staff to test (the population), and that the true unknown prevalence of COVID19 illness in your workforce is only 5%. Therefore, 5 unknown staff have had the illness, and 95 have not. We apply the test to the population (all the staff) with the following results:</p>
<p></p><ul><li>5 who have had the illness are tested, and since the test has 88.7% sensitivity, 4 people test positive.</li><li>95 who have not had the illness are tested, and with the false positive rate (1specificity) of 9.4%, another 9 people test positive.</li></ul>
<p>In total, 9 + 4 = 13 people tested positive as having had the illness. But, in truth, only 4 of 13 = 30% (as we see in Scenario 2 above) of those with a positive test have actually had the illness!</p>
<h2>Conclusion</h2>
<p>As can be seen in the example above, things aren’t as simple as they appear when you read a headline quoting a test with “90% accuracy”.</p>
<p>When applied to a population with a low prevalence of illness, the false positives soon overwhelm the true positives and make the test less useful. If applied to populations with a higher prevalence of the illness, such as workers who have had symptoms and selfisolated, the usefulness of the positive test result increases (see, for example, Scenario 4 in the table above).</p>
<p>
</p><p>This paper presents just one example of a single
test, and others with higher sensitivity/specificity will perform better. It highlights the importance of predictive values in decision making, and in evaluating whether a particular test is really that helpful.</p><p>For more information, see our online documentation:<br><a href="https://analyseit.com/docs/userguide/diagnosticperformance/diagnosticaccuracy" target="_blank">Measures of diagnostic accuracy</a><br></p>
Method validation
Statistics
Using Analyseit
Wed, 01 Apr 2020 15:47:08 GMT

https://analyseit.com/blog/2019/8/latestanalyseitimprovementsroundup
Latest Analyseit improvements – roundup
https://analyseit.com/blog/2019/8/latestanalyseitimprovementsroundup
<p>We’ve been busy over the last few months adding
various new features and improvements to Analyseit. </p><p>There are always many
changes behind the scenes, which aren’t really newsworthy, but seek to make
Analyseit compatible with the latest changes to Microsoft Excel and Windows. Microsoft
have recently embraced a more aggressive continuousrelease deployment model
with Office and Excel, and these releases have caused quite a few headaches for
developers, notably with the <a href="https://analyseit.com/blog/2018/5/microsoftwindowsoffice2016multiplemonitordpiawarenessandanalyseitsmissinguserinterface" target="">multimonitor high DPI support</a>. </p><p>For these reasons we advise you have active maintenance so you get these
updates as they are released – if not we recommend you <a href="https://analyseit.com/store/maintenance" target="_blank">renew maintenance</a> on your license to get the latest updates and ensure compatibility
with the often fastpaced releases.</p><h3>So what have we added recently to
Analyseit?</h3><p>Starting with <b>version 5.20</b> we introduced a
new feature so you can set the scale (minimum, maximum and units) for a
variable that will then be used, where possible, to set the scaling for chart
axes. Normally Analyseit will choose excellent axis scaling, but sometimes you
know better and want a specific minimum, maximum, or units. Previously you would
have to set the scale manually on the charts output from the analysis, and any
changes would be lost on recalculate. Now you can control the scaling, see <a href="https://analyseit.com/docs/userguide/settingvariablescale" target="">Setting the minimum, maximum and units of a variable</a> </p><p>With <b>version 5.30</b> we focussed on improving
the Fit model analysis, with the addition of weighted regression and the
DurbinWatson statistic for testing autocorrelation in the residuals. For more
information see the updated documentation, <a href="https://analyseit.com/docs/userguide/fitmodel/linear/linearfit" target="">Linear fit</a> and <a href="https://analyseit.com/docs/userguide/fitmodel/linear/residualautocorrelation" target="">Residuals  independence</a>.</p><p>
</p><p>Finally, in <b>version 5.40</b> we added equivalence
tests. Unlike the usual equality of means hypothesis test which tests for a <i>difference</i>, an equivalence of means hypothesis
test tests if two population means are <i>equivalent</i>, that is, practically the
same. We’ve added the equivalence tests to Distribution, Compare Groups and
Compare Pair analyses. You can read more about these changes at <a href="https://analyseit.com/docs/userguide/distribution/continuous/parameterequivalencehypothesistest" target="">Distribution  Parameter equivalence hypothesis test</a>, <a href="https://analyseit.com/docs/userguide/comparegroups/equivalencemeanhypothesistest" target="">Compare groups – equivalence of means hypothesis test</a>,
and <a href="https://analyseit.com/docs/userguide/comparepairs/equivalencemeanhypothesistest" target="">Compare pairs – equivalence of means hypothesis test</a></p><p>If you have <a href="https://analyseit.com/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="https://analyseit.com/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your license has expired you can renew it to get this update and forthcoming updates, see <a href="https://analyseit.com/store/maintenance/" target="_blank">renew maintenance</a>.<br></p>
Excel
In development
Releases
Thu, 15 Aug 2019 12:54:40 GMT

https://analyseit.com/blog/2019/2/trainingandconsultancyforanalyseit
Training and consultancy for Analyseit
https://analyseit.com/blog/2019/2/trainingandconsultancyforanalyseit
<p>Our focus at Analyseit has always been on the development and improvement of our software. While we provide extensive help, tutorials, and technical support for Analyseit, one area we do not cover is training and consultancy. As many of you will know we are based in England in the United Kingdom, and providing training and consultancy is often done better locally, inperson. </p>
<p>Instead we partner with experts who can provide training and consultancy in various disciplines, in local language, and geographically near (or at least nearer) to our customers. You can always find a list of current consultant and training partners at <a href="https://analyseit.com/support/training">https://analyseit.com/support/training</a></p>
<h2>Dr. Keller & ACOMED statistik </h2>
<p>One of the experts we have had a long relationship with is Dr. Thomas Keller. Dr Keller is an independent statistician and has run <a href="https://www.acomedstatistiktools.de/" target="">ACOMED statistik</a> for 15 years. One his many areas of expertise is the planning and evaluation of experiments for method validation and he has been involved in international working groups (IFCC, CLSI) in the fields of clinical chemistry and laboratory medicine. Dr. Keller was actually a customer and started to provide training in Analyseit shortly after. His reputation is second to none in the industry and he has provided consultancy and training to many companies using Analyseit. See an example of a <a href="https://analyseit.com/support/20180906_statistical_seminar_analytical_performance_CLSI.pdf" target="">course on Method Validation according to CLSI guidelines</a> offered by Dr. Keller. He also provides <a href="https://www.acomedstatistiktools.de/workshops/" target="_blank">WebEx and telephone consultations</a> for anything from simple questions to full courses for individuals and small groups.</p>
<p>Dr. Keller is providing a training course in April on how to use Analyseit in biostatistical analysis. The course will be in Germany, in Leipzig, and details are below. If you would like to attend, would like to consult with Dr. Keller, or maybe even arrange a similar course for your colleagues, please contact Dr. Keller at <a href="https://analyseit.com/support/training#drThomasKeller">https://analyseit.com/support/training#drThomasKeller</a><br>
</p>
<h2>Training course: Biostatistics with Analyseit</h2>
<p>Course details taken from:<br>
<a href="http://www.biosaxonytools.de/limesurvey/index.php/764944/langde">http://www.biosaxonytools.de/limesurvey/index.php/764944/langde</a></p>
<p><b>3rd April 2019, 9am to 5pm</b><br>
<a href="https://goo.gl/maps/pLpP6hz7Sm32" target="_blank"><b>BIO CITY Leipzig, Deutscher Platz 5c, 04103 Leipzig, Germany</b></a><br>
</p>
<h3>Summary</h3>
<p>The evaluation of data from clinical laboratories and biomedical research requires a sound statistical basis. In addition to the generation of research results, statistical methods also play a central role in quality assurance and method validation. A suitable software for this is the ExcelAddOn AnalyseIt®, which provides basic statistical evaluations regarding estimation and testing as well as method validation. The software is based on the methodology described in the CLSI (Clinical Laboratory Standard Institute).</p>
<p>The seminar will introduce basic methods of biostatistics. Data from method validation experiments will mainly be used as examples. The aim of the seminar is to introduce scientific staff and laboratory assistants to basic statistical approaches and at the same time to apply and deepen this knowledge in exercises.</p>
<p> The event is aimed at scientists and laboratory assistants from biomedical research as well as clinical laboratories and comparable institutions who use Microsoft Excel and, if applicable, AnalyseIt® for their analyses or are planning to do so.</p>
<h3>Course overview</h3>
<p> The workshop will be provided in German language only. In the seminar exercises with the statistics software AnalyseIt® will form an important part. Participants should bring their own laptop with MS Excel installed. In addition, a temporary license for the AnalyseIt® software will be provided in advance.</p>
<p><b>Part 1: Parametric and nonparametric description of data, introduction to the AnalyseIt® software</b></p>
<p>Introduction to the AnalyseIt® software<br>
Basic operation<br>
Structure of the data for evaluations with AnalyseIt®<br>
Mean value and standard deviation<br>
Median and percentiles<br>
Suitable graphical representations<br>
Deviation from normal distribution<br>
<i> Application: Data from clinical studies, reference limits</i></p>
<p><b>Part 2: Statistical estimation</b></p>
<p>Standard error and confidence interval<br>
Statistical testing with confidence intervals<br>
Test for difference vs. test for equivalence<br>
<i> Application: Robustness investigations, method comparisons by means of difference plot</i></p>
<p><b>Part 3: Statistical hypothesis tests</b></p>
<p>How do hypothesis tests work?<br>
How do I select the appropriate test?<br>
How do I interpret the test results?<br>
<i> Application: Data from clinical studies</i></p>
<p><b>Part 4: Linear Regression and ANOVA</b></p>
<p>Linear and Polynomial Regression<br>
<i> Application: Linearity (Demonstration)</i><br>
ANOVA and variance components<br>
<i> Application: Precision investigations (demonstration)</i></p>
<h3>Pricing and booking</h3>
<p>For pricing and information on how to book please see <a href="http://www.biosaxonytools.de/limesurvey/index.php/764944/langde">http://www.biosaxonytools.de/limesurvey/index.php/764944/langde</a> and contact Dr. Keller directly at <a href="https://analyseit.com/support/training#drThomasKeller">https://analyseit.com/support/training#drThomasKeller</a></p>
Statistics
Business
Using Analyseit
Tue, 26 Feb 2019 13:49:21 GMT

https://analyseit.com/blog/2018/9/analyseit510releasedsaveandreapplyfilters
Analyseit 5.10 released: Save and reapply filters
https://analyseit.com/blog/2018/9/analyseit510releasedsaveandreapplyfilters
<p>It’s been a longrequested feature, and today we’re happy to announce that Analyseit version 5.10 now includes the ability to save the dataset filter with an analysis and reapply it on recalculation. </p>
<p>Analyseit always allowed you to use Excel autofilters to quickly limit analysis to just a subset of the data, but until now that filter wasn’t saved. Each time you recalculated the analysis it was based on the currently active filter rather than the filter ineffect when you created the analysis. </p>
<p>This <i>“active filter”</i> method had its uses in exploratory data analysis: you can easily create an analysis, adjust the filter criteria, click Recalculate to see the changes to the analysis, then repeat as necessary to explore the data. But it also had its limitations. For example, if you created two analyses to look at subjects where <i>Age > 20</i> and <i>Age <= 20</i>, simply clicking Recalculate on those analyses could get you in a mess if you didn’t reset the filter conditions on each analysis before recalculating. </p>
<p>Thankfully Analyseit will now take care of that for you.</p>
<h2>Saving the filter with the analysis</h2>
<p>In Analyseit version 5.10 and later, when you create an analysis (or edit an existing analysis), you will see a new command on the Analyseit ribbon called <i>Filter</i>. If you click the dropdown arrow alongside the command you will see two options:</p>
<p><img src="/assets/article/54611649/FilterOptions.png"><br>
</p>
<p>For new analyses, <i>Save Filter & Reapply</i> is selected by default so any filter criteria you apply during analysis will be saved and reapplied when you next recalculate or edit the analysis. </p>
<p>For existing analyses created before version 5.10, <i><i>Use active filter</i></i> will be the default and those analyses will continue to use the filter in effect when recalculating. That maintains the behavior users of earlier versions expect, so they’re not surprised by the new feature, but you can click Edit and change the Filter setting if you want to save the filter with the analysis instead.</p>
<h2>Identifying the filter used in an analysis</h2>
<p>The filter criteria used calculating an analysis shows in the header of an analysis report, see below: </p>
<p><img src="/assets/article/54611649/FilterHeader.png"><br>
</p>
<p>Space is limited in the worksheet tab name, but a short, truncated description of the filter criteria is included. Usually, you are better renaming the worksheet tab to better describe the filter in a clearer, succinct way. For example, Analyseit might label the tab <i>“Height (Age > 20 and Sex = ‘Mal..”</i>, truncated due to limitations on the length of a worksheet tab name, but you might rename it to <i>“Male Height over 20y/o”</i>. When you change a worksheet tab name Analyseit will respect the change and will leave it asis during recalculate.<br>
</p>
<h2>Changing or reusing a filter </h2>
<p>When a filter is saved with an analysis you will notice a new command on the Analyseit analysis ribbon called <i>Filter</i>, see below:</p>
<p><img src="/assets/article/54611649/ApplyFilterCommand.png"></p>
<p>When you click <i>Filter</i> it applies the filter criteria saved with the analysis to the dataset and shows the dataset. You now have a few options. You can see the dataset filtered by the criteria. Or you could change the filter criteria, switch back to the analysis, and click Recalculate to change the filter applied to the analysis. Or you could create a new analysis, based on the same filter criteria as the existing analysis. </p>
<p>The <i>Filter</i> command is disabled if the analysis is using the active filter, rather than a saved filter. </p>
<h2>Limitations of save filter</h2>
<p>Save filter works with almost all filter conditions: filter by value, by condition, by color/fill/icon, above/below average, top/bottom items, and so on. But there is one limitation: filtering to multiple dates, such as shown below:</p>
<p><img src="/assets/article/54611649/FilterByDates.png"></p>
<p>If you filter to more than 2 dates, addins such as Analyseit cannot retrieve the list of values. This is a limitation of Microsoft Excel that unfortunately hasn’t been remedied in the last few years so it must be low on Microsoft’s priorities. If this is seriously affecting you, please get in touch with Microsoft and ask them to resolve this limitation for addin developers.</p>
<p>If you have <a href="https://analyseit.com/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="https://analyseit.com/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your license has expired you can renew it to get this update and forthcoming updates, see <a href="https://analyseit.com/store/maintenance/" target="_blank">renew maintenance</a>.<br>
</p>
Using Analyseit
Releases
Mon, 10 Sep 2018 16:54:26 GMT

https://analyseit.com/blog/2018/5/microsoftwindowsoffice2016multiplemonitordpiawarenessandanalyseitsmissinguserinterface
Microsoft Windows, Office 2016, multiple monitor DPI awareness  and Analyseit's missing user interface
https://analyseit.com/blog/2018/5/microsoftwindowsoffice2016multiplemonitordpiawarenessandanalyseitsmissinguserinterface
<p><b><i>Update 19Sep2019: Unfortunately this continues to be an issue for some users and unfortunately there is currently no solution from Microsoft except to suggest the use of compatibility mode as detailed below. We have requested this be fixed, so please upvote it at <a href="https://officespdev.uservoice.com/forums/224641featurerequestsandfeedback/suggestions/38632276supportforhighdpitaskpanes">https://officespdev.uservoice.com/forums/224641featurerequestsandfeedback/suggestions/38632276supportforhighdpitaskpanes</a></i></b></p>
<p><b><i>Update 27Jun2018: Although we have a fix for this issue on an internal build, it appears that Microsoft Office version 1807 (which is currently only available on the Office Insider track) fixes this issue. The missing userinterface problem was caused by a bug in Microsoft Office 1805/1806 updates. We will release our fix shortly, but the 1807 version update will also become available to everyone over the next month or so. If you want to get it immediately see <a href="https://support.office.com/enus/article/howoffice365commercialcustomerscangetearlyaccesstonewoffice2016features4dd8ba4073c04468b778c7b744d03ead" target="_blank">How to get Office 365 insider builds</a>.</i></b></p>
<p>Microsoft has recently released updates to both <a href="https://docs.microsoft.com/enus/windows/whatsnew/whatsnewwindows10version1803" target="">Windows (the April 1803 update)</a> and <a href="https://support.office.com/enus/article/officeappsappearthewrongsizeorblurryonexternalmonitorsbc9f72794e424b15a94946ab8bcfe44f" target="">Microsoft Office 2016</a> to provide support for multiple monitor high DPI (dotsperinch) displays.<br>
</p>
<p>In the early days of Microsoft Windows, monitors were assumed to have 96 DPI and all applications worked on that assumption: with a user interface fixed on that assumption. In the last 15 years, monitors with higher DPI have started to appear with the benefit that onscreen text and graphics look much smoother because there are so many more dots per inch. That caused problems for many applications which were fixed to assume 96 DPI, causing their user interface to scale improperly on highDPI monitors. Applications like Analyseit supported high DPI monitors and adjusted their user interface appropriately... until now.</p>
<p>These days many of us have 2 or more monitors hooked up to a PC as it's a much quicker and more efficient way to work. Often these monitors are different sizes and often have different DPI, and the recent changes to Windows and Microsoft Office are designed to address varying DPIs. Instead of applications scaling their userinterface to one systemwide DPI setting, they now scale according to the monitor application is displayed on. If you have the latest updates to Windows and Office, and drag the Microsoft Excel main window between monitors with different DPIs you will see the application user interface rescale properly for the monitor.</p>
<p>This is a major new improvement and under the hood there are some major changes to Microsoft Windows to support it. Unfortunately, as with all major changes to Windows, it breaks some existing applications, especially addins like Analyseit that are at the mercy of what the host application (Microsoft Excel in this case) decides to do with respect to user interface scaling. In the case of Analyseit, when the new multiple monitor DPI awareness feature is enabled in Microsoft Excel 2016 the Analyseit user interface does not appear properly in the task panes.</p>
<h2>How to temporarily fix the issue in Microsoft Excel 2016</h2>
<p>We are now working on a fix for this issue, but in the meantime, if you are experiencing this problem follow these steps to temporarily disable multiple monitor DPI awareness in Excel:</p>
<ol>
<li>Start Microsoft Excel 2016.<br>
</li>
<li>Click <b>File</b> > <b>Options</b>.<br>
</li>
<li>In the <b>General</b> tab, at the very top of the options list, select <b>Optimize for compatibility</b>, then click <b>OK</b> (see screenshot below). If you do not see this option then you should not be affected by this issue  if you are, please <a href="https://analyseit.com/company/contactus" target="">contact us</a>.<br>
<br>
<img src="/assets/article/54611648/DisableDPIAwareness3.png"></li>
<li>Close and restart Microsoft Excel and the Analyseit user interface will now appear properly in the task panes.<br>
</li>
</ol>
<p>We will update this blog post when we have a fix available.</p>
Excel
Using Analyseit
Wed, 30 May 2018 12:04:13 GMT

https://analyseit.com/blog/2017/12/analyseitis20yearsold
Analyseit is 20years old!
https://analyseit.com/blog/2017/12/analyseitis20yearsold
<p>Today marks the 20<sup>th</sup> birthday of Analyseit. </p>
<p>It was December 1997 when we shipped the first disks containing Analyseit to paying customers. In some ways it seems just like yesterday, but in other respects software development and Analyseit has come so far in those 20years. </p>
<p>As many of you know Analyseit wasn’t our first foray into developing statistical software. My cofounder in Analyseit, Simon Huntington, had previously developed <a href="https://analyseit.com/blog/2014/10/asombrenoteprofessorrickjones" target="">Astute in 1992 from a spark of an idea from Dr Rick Jones</a>. Astute was the first statistical addin for Microsoft Excel, initially released in 1992 for Excel 4.0, and released a few months before Microsoft’s Data Analysis Toolpak which appeared in Excel 5.0.</p>
<p>We started developing Analyseit in 1996. Astute was no longer available and so we started work on developing a replacement. We were just a startup business, keen to get our first product to market and so worked 16 hour days, 6 days a week, for 18 months to build that first release of Analyseit. </p>
<p>Software development back then was pretty brutal. It had improved leaps and bounds since the 1980s, when we started developing software, but it was still timeconsuming and sheer hard (mental) work. The software development tools were relatively basic and computing power was a fraction of what it is today. We initially developed Analyseit in C++, the only commerciallyviable programming language back then, and although a very flexible programming language it was very easy to produce bugridden and hardtomaintain software. Each addition or change to the software code during development took 510 minutes to compile before the software could be run (or tested) to see the results. Not quite as slow as the mainframe era, but in an iterative process such as software development it was still tediously slow. Compilation often failed, due to typos in the source code, which then needed to be corrected and compilation restarted. Finally when we had an executable we could then start debugging and testing it. Windows 95 was the operating system of choice, but wasn’t the most resilient. It was easy to crash Windows resulting in a reboot, reload of the software development tools, and so on. Another 15 minutes gone.</p>
<p>Back then, the latest release of Excel was Microsoft Excel 95, which was basically Excel 5 with a new user interface and support for 32bit processors. More than half of our users at that time were still using Excel 5, Windows 3.1 and 16bit processors, so we had to support both processor architectures. We used Microsoft Visual C++ on Windows 95 and Borland C++ on Windows 3.1 to develop the 32 and 16bit versions. Constantly switching between compilers, testing on different versions of Excel and different operating systems took more time – 34 months in fact to convert Analyseit from a 32bit only to a 32bit and 16bit application! In comparison, it recently took just over a day to make the latest versions of Analyseit support both 32bit and 64bit processors!</p>
<p>Finally by October 1997 we had a saleable product. We had placed an advertisement in The Biochemist magazine a couple of months before and already had a backlog of orders. We had also launched the analyseit.com website, a very basic website by today’s standards, but (believe it or not!) relatively modern for the time. Most users back then had VGA resolution screens, which were very small, 640x512 pixels, supporting only up to 256 colors. The website design now looks ridiculously tiny (see below), but that was all the space available on the screen without scrolling. In other ways it was quite revolutionary – we offered a free downloadable trial of Analyseit, almost unheardof back then. In those days you called a company on the telephone, spoke with a salesman, and maybe, if they considered you a reasonable prospect, they would post a disk through the post for a trial. Analyseit was available for download directly from our website (albeit over a slow dialup internet connection for most people), with no strings attached, pretty much as it still is today.</p>
<p><img src="/assets/article/54611647/firstanalyseitweb.png"></p>
<p><i>A screenshot of the first analyseit.com website back in 1998</i></p>
<p>We finally sent out the first batch of orders for Analyseit on the 21<sup>st</sup> December 1997. Delivery was by post and on a single 3.5” disk. We used a few compression algorithms to get the software to fit onto a single disk, making postage cheaper and saving costs – important factors back then as a startup!</p><p>Feedback to the product was amazing. Customers raved about the product, on the telephone and by email, and we received so much great feedback that the product went from strength to strength to what it is today. From those humble beginnings, we’re now approaching more than 40,000 users of Analyseit and nearing version 5.0. Development is now much quicker due to modern software development tools and we’re able to add new features much more quickly. We’ve developed custom software solutions for the <a href="https://clsi.org" target="">CLSI</a> and other large companies based the reputation we've built over the last 20years. And most of our sales still come from word of mouth – from happy customers.</p>
<p>Here’s to the next 20 years!</p>
Business
Releases
Thu, 21 Dec 2017 12:07:29 GMT

https://analyseit.com/blog/2017/8/analyseit490releasedpredictionintervalsep14a3andep30a
Analyseit 4.90 released: Prediction intervals, EP14A3 and EP30A
https://analyseit.com/blog/2017/8/analyseit490releasedpredictionintervalsep14a3andep30a
<p>Prediction intervals on Deming regression are a major new feature in the Analyseit Method Validation Edition version 4.90, just released.</p>
<p>A prediction interval is an interval that has a given probability of including a future observation(s). They are very useful in method validation for testing the commutability of reference materials or processed samples with patient samples. Two CLSI protocols, <a href="https://clsi.org/standards/products/methodevaluation/documents/ep14/" target="">EP14A3: Evaluation of Commutability of Processed Samples</a> and <a href="https://clsi.org/standards/products/methodevaluation/documents/ep30/" target="">EP30A: Characterization and Qualification of Commutable Reference Materials for Laboratory Medicine</a> both use prediction intervals.</p>
<p>We will illustrate this new feature using an example from CLSI EP14A3:</p>
<p>1) Open the workbook <a href="/assets/article/54611646/EP14A3.xlsx">EP14A3.xlsx</a>.</p>
<p>2) On the <b>Analyseit</b> ribbon tab, in the <b>Statistical Analysis</b> group, click <b>Method Comparison</b> and then click <b>Ordinary Deming</b> regression.</p>
<p>The analysis task pane opens.</p>
<p>3) In the <b>X (Reference / Comparative)</b> dropdown list, select <i>Cholesterol: A</i>.</p>
<p>4) In the <b>Y (Test / New)</b> dropdown list, select <i>Cholesterol: B</i>.</p>
<p>5) On the <b>Analyseit</b> ribbon tab, in the <b>Method Comparison</b> group, click <b>Restrict to Group</b>.</p>
<p>6) In the <b>Group / Color / Symbol</b> dropdown list, select <i>Sample Type</i>.</p>
<p>7) In the <b>Restrict fit to group</b> dropdown list, select <i>Patient</i>.</p>
<p>8) In the <b>Prediction band</b> edit box, type <i>95%</i>.</p>
<p>NOTE: Select the <b>Familywise coverage</b> check box to control the probability of simultaneously for all additional samples rather than individually for each sample.</p>
<p>9) On the <b>Descriptives</b> task pane, select <b>Label points</b>, and choose <b>Additional groups only</b>.</p>
<p>10) Click <b>Calculate</b>.</p>
<p>The report shows the scatter plot with fitted regression line and 95% prediction interval (see image below). The regression line is only fitted to the points in the <b>Patient</b> group, as set in step 7 above, and additional points are colored depending on the type of sample, as set in step 6 above. </p>
<p>Any points outside the prediction band are not commutable with the patient samples, and in this case you can see sample ‘c’ is not commutable. The commutability table shows the additional samples and whether they are commutable or not with the patient samples. </p>
<p><img src="/assets/article/54611646/EP14A3.png"></p>
<p>The steps to perform an EP30 study are the same as described above. You should note that EP30 forms the prediction interval using the fit of the patient samples and the precision of the reference materials, whereas Analyseit uses the fit and precision of the patient samples. We chose to implement it like this since there are usually too few reference material samples to establish a reliable estimate of the precision.</p>
<p>We have extended the prediction intervals beyond the CLSI EP guidelines, so they support any number of replicates and are also available with Ordinary and Weighted Deming regression. This alleviates the need to log transform values as is recommended in EP14, which, although it corrects the constant CV, distorts the relationship between the two methods.</p>
<p>If you have <a href="https://analyseit.com/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="https://analyseit.com/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your license has expired you can renew it to get this update and forthcoming updates, see <a href="https://analyseit.com/store/maintenance/" target="_blank">renew maintenance</a>.<br>
</p>
Method validation
Using Analyseit
Releases
Statistics
Fri, 18 Aug 2017 09:23:39 GMT

https://analyseit.com/blog/2017/5/parameterestimation
Parameter estimation
https://analyseit.com/blog/2017/5/parameterestimation
<p>Often we collect a sample of data not to make statements about that particular sample but to generalize our statements to say something about the population. Estimation is the process of making inferences about an unknown population parameter from a random sample drawn from the population of interest. An estimator is a method for arriving at an estimate of the value of an unknown parameter. Often there are many competing estimators for the population parameter that differ based on the underlying statistical theory. </p>
<h2>Point estimate</h2>
<p>A point estimate is the best estimate, in some sense, of the population parameter. The most wellknown estimator is the sample mean which produces an estimate of the population mean.</p>
<p>It should be obvious that any point estimate is not absolutely accurate. It is an estimate based on only a single random sample. If repeated random samples were taken from the population the point estimate would be expected to vary from sample to sample. This leads to the definition of an interval estimator which provides a range of values defined by the limits [L, U].</p>
<h2>Confidence interval</h2>
<p>A confidence interval defines limits [L, U] constructed on the basis that a specified proportion of the confidence intervals include the true parameter in repeated sampling. How frequently the confidence interval contains the parameter is determined by the confidence level. 95% is commonly used and means that in repeated sampling 95% of the confidence intervals include the parameter. 99% is sometimes used when more confidence is needed and means that in repeated sampling 99% of the intervals include the parameter. It is unusual to use a confidence level of less than 90% as too many intervals would fail to include the parameter. </p>
<p>Again a confidence interval is formed using an interval estimator based on statistical theory and assumptions about the underlying population. The tbased confidence interval estimator for a population mean is an example taught in most introductory textbooks.</p>
<p>Many people misunderstand confidence intervals. A confidence interval is a frequentist concept, that is, the probability is defined in a series of repetitions of an experiment. <b>A confidence interval does not predict with a given probability that the parameter lies within the interval.</b> The problem arises because the word confidence is misinterpreted as implying probability. After drawing a sample and forming a confidence interval, it either does contain the population parameter or it doesn’t. This is the same as how the probability of obtaining a head or tails in the toss of fair coin is 50%, but after a toss has happened it is either a head or tail (the predicted event happened or it didn’t). Similarly, the 95% probability associated with a 95% confidence interval only applies to the method used to construct the interval, not to the individual realized intervals. </p>
<p>As an example, let’s say we draw a single random sample and construct a 95% confidence interval for the mean that results in an interval [15, 20]. We could say “We are 95% <i>confident</i> the population mean is between 15 and 20”. It would be incorrect to say “There is a 95% <i>probability</i> the population mean is between 15 and 20”. The switch here is a subtle change of the word confident to the word probability. The word confident means that the estimation method works 95% of the time, the other 5% of the time it is wrong. </p>
<h2>Illustration</h2>
<p>We can illustrate these concepts with a MonteCarlo simulation. </p>
<p>Let’s assume the population has a mean μ=0 and standard deviation σ=1, and we draw 100 random samples from the population each with a size n=40, then compute the 95% mean confidence interval for each sample.</p>
<p><img src="/assets/article/54611645/MeanConfidenceIntervalSimulation.png" alt="Mean confidence interval simulation: Analyseit blog"></p>
<p>The plot shows the limits of the confidence interval as an error bar, with the mean as a dot, for each of 100 samples. The horizontal red line indicates μ=0, the true population mean. Highlighted in red are repetitions where the mean confidence interval does not include μ=0. You can see that 5/100=5% are colored red and 95/100=95% are colored black, and therefore interval included the population mean 95% of the time. Proof that our estimation method works 95% of the time!</p>
<h2>Reporting results</h2>
<p>A point estimate of the parameter from a single sample is not a great way to present your findings. It gives no idea about uncertainty. A point estimate with a confidence interval provides more information, as the point estimate is the most likely value given the sample observed and the confidence interval expresses the uncertainty in this estimate. If the width of the confidence interval is large you should have less confidence in the estimate, and if the interval is narrow you can have more confidence. When the interval is wide you could repeat the experiment with a larger sample size to get a narrower interval. Repeating an experiment is always a good way to confirm that your results are real and is key to making good scientific judgments. </p>
<p><b><br>
</b></p>
<p><b>Related links:</b></p>
<p>A great interactive visualization of the confidence interval: <a href="http://rpsychologist.com/d3/CI/" target="_blank">http://rpsychologist.com/d3/CI/</a></p>
Using Analyseit
Statistics
Thu, 18 May 2017 15:40:39 GMT

https://analyseit.com/blog/2017/2/analyseit480indepththerevampeddocumentation
Analyseit 4.80 indepth: The revamped documentation
https://analyseit.com/blog/2017/2/analyseit480indepththerevampeddocumentation
<p>As we mentioned last week in the <a href="https://analyseit.com/blog/2017/2/analyseit480releasedanovaancovaandrevampeddocumentation" target="">post announcing Analyseit 4.80</a>, in this release we took the opportunity to revamp the documentation.</p>
<p>The revamp involved rewriting many topics to make the content clearer, adding new taskoriented topics, including refresher topics on common statistical concepts, and improving the indexing and links between topics so you can more easily navigate the help system. </p>
<p>The new taskoriented topics give you stepbystep instructions on completing common tasks. For example you will now find topics on how to <a href="https://analyseit.com/docs/userguide/processcontrol/creatingshewhartcontrolchartsubgroup" target="">create an XbarR Shewhart control chart</a>, <a href="https://analyseit.com/docs/userguide/fitmodel/linear/performinganova" target="">how to perform ANOVA</a>, <a href="https://analyseit.com/docs/userguide/fitmodel/linear/fittingadvancedmodels" target="">how to fit advanced models</a>, and even simple tasks like <a href="https://analyseit.com/docs/userguide/distribution/continuous/creatingunivariatedescriptives" target="">calculating descriptive statistics</a>. We have also fully documented the supported dataset layouts for each type of analysis so you can see how to arrange your data for Analyseit. The links in each topic help you more easily find related topics, for example links to topics on how to interpret the statistics, links to explain the pros and cons of the available statistical tests, links to topics for common tasks, and a link showing you how to arrange the dataset.</p>
<p>Of all the requests, the <b>most customerrequested improvement</b> is the new <a href="https://analyseit.com/docs/Analyseit_CompleteGuide.pdf" target="">printable PDF user guide</a>. Previously we offered EPUB and Kindle reader editions of the help, but not PDF. To be honest, producing a PDF user guide from the <a href="https://en.wikipedia.org/wiki/Darwin_Information_Typing_Architecture" target="">DITA</a> tools we use to write the help was a real technical challenge. The PDF produced just wasn’t good enough for us, and certainly not for our customers – the formatting and layout were poor, indexing was nonexistent, and there were so many other niggles. So we took the time to make the userguide both look good and be usable. Take a look and let us know what you think!</p>
<p>With the addition of the new PDF user guide, you can now find documentation in a number of formats suitable in all situations:</p>
<ul>
<li><b>Online:</b> <a href="https://analyseit.com/support/" target="">Browse and search the documentation</a> online.<br>
</li>
<li><b>In application:</b> Access the help directly from within Analyseit, even when you’re offline. To see the documentation, click the Analyseit icon on the Analyseit ribbon tab, then click Help. The Microsoft HTML Help will be shown, complete with full index and search capabilities. You can also see the contextsensitive help popup for any option or ribbon command in Analyseit simply by hovering the mouse over it.<br>
</li>
<li><b>For printing or reading on your computer:</b> To get a fully formatted book with table of contents and index that you can read or print, <a href="https://analyseit.com/docs/Analyseit_CompleteGuide.pdf" target="">download the PDF</a>.<br>
</li>
<li><b>For reading offline on an iPad or Kindle device:</b> For best reading experience on an <a href="https://analyseit.com/docs/Analyseit_CompleteGuide.epub" target="">iPad download the EPUB</a>, or on a <a href="https://analyseit.com/docs/Analyseit_CompleteGuide.mobi" target="">Kindle device download the MOBI</a>. The eBook versions are more flexible than the PDF as the content is reformatted and reflows depending on the screen size of your device and your preferred font size.<br>
</li>
</ul>
<p>Let us know your thoughts and how we can further improve the documentation.</p>
Releases
Using Analyseit
Tue, 28 Feb 2017 15:02:26 GMT

https://analyseit.com/blog/2017/2/analyseit480releasedanovaancovaandrevampeddocumentation
Analyseit 4.80 released: ANOVA, ANCOVA and revamped documentation
https://analyseit.com/blog/2017/2/analyseit480releasedanovaancovaandrevampeddocumentation
<p><img src="/assets/article/54611643/ANOVAANCOVAmenu.jpg" style="float:right; paddingleft: 16px;">Last week we released version 4.80 of Analyseit.</p>
<p>The new release includes multiway <a href="/docs/userguide/fitmodel/linear/performinganova" target="">ANOVA</a>, <a href="/docs/userguide/fitmodel/linear/performingancova" target="">ANCOVA</a>, and <a href="/docs/userguide/multivariate/itemreliability" target="">Cronbach’s Alpha</a> in the Standard edition, and since every licence includes the Standard edition, these features are available to all users. We also took the opportunity to revamp the <a href="/support/" target="">help for Analyseit</a> and develop a <a href="/docs/Analyseit_CompleteGuide.pdf" target="">PDF userguide</a>. We’ll go into more details on the improvements in the next few weeks.</p>
<p>If you have <a href="/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your license has expired you can renew it to get this update and forthcoming updates, see <a href="/store/maintenance/" target="_blank"> renew maintenance</a>.</p>
Using Analyseit
Releases
Tue, 07 Feb 2017 11:47:26 GMT

https://analyseit.com/blog/2016/1/analyseit460released3nestedfactorprecision
Analyseit 4.60 released: 3 nested factor precision
https://analyseit.com/blog/2016/1/analyseit460released3nestedfactorprecision
<p>Today we released version 4.60 of the Analyseit Method Validation edition.</p>
<p>The new release now includes 3 nestedfactor precision analysis, which extends Analyseit’s support for CLSI EP05A3 multilaboratory precision studies.</p><p><img src="/assets/article/54611642/3wayPrecision.png" style="cursor: default;"></p>
<p>If you have <a href="/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your license has expired you can renew it to get this update and forthcoming updates, see <a href="/store/maintenance/" target="_blank"> renew maintenance</a>.</p>
Method validation
Releases
Thu, 21 Jan 2016 12:20:27 GMT

https://analyseit.com/blog/2015/10/announcingtheanalyseitqualitycontrolimprovementedition
Announcing the Analyseit Quality Control & Improvement Edition
https://analyseit.com/blog/2015/10/announcingtheanalyseitqualitycontrolimprovementedition
<h3><b>Offical release</b></h3>
<p>The Analyseit Quality Control and Improvement Edition is now officially available. You can download a free trial and buy it online at <a href="https://analyseit.com/products/qualitycontrol/" target="">https://analyseit.com/products/qualitycontrol/</a><br>
</p>
<h3><b>Update 15Dec2015</b></h3>
<p>We have just released an update to the prerelease. The software is now available for anyone to test, and it now includes tutorials you can follow to quickly get started using Pareto analysis, Control charts, and Capability analysis. It does not require an active internet connection as we no longer need to monitor the software for error conditions.</p>
<p><strike>You can download the latest release and try the software from:</strike></p>
<p><a href="/download/Analyseit/4.51/Analyseit_4_51.EXE"><strike>/download/Analyseit/4.51/Analyseit_4_51.EXE</strike></a></p>
<p><strike>To activate the use the product key:</strike></p>
<pre><strike>PTESTD4NF4LXJSYTT</strike>
</pre>
<h3><b>Original article</b></h3>
<p>We are delighted to announce the addition of the <b>Analyseit Quality Control and Improvement Edition</b> to the range of Analyseit products.</p>
<p>The new edition includes the most impressive statistical process control (SPC) charts available in any Excel statistical software package, including Shewhart, LeveyJennings, CUSUM, and EWMA charts. Process capability statistics and plots help you ensure a process is able to meet specifications. And Pareto plots help you identify the qualityrelated problems that need the most attention and let you monitor efforts to reduce them.</p>
<p><i>Screenshot: XbarR chart of a quality characteristic by phase with stratification.</i></p>
<p><img src="/assets/article/54611641/controlchart2.png"></p>
<p><i>Screenshot: Process capability plots by phase.</i></p>
<p><img src="/assets/article/54611641/capability.png"></p>
<p><i>Screenshot: Pareto plot of failures with stratification and sorted by key plot.</i></p>
<p><img src="/assets/article/54611641/pareto.png"></p>
<h2>Download the prerelease now</h2>
<p>The new release is currently available for prerelease beta testing. Anyone can download and test the new release, though you will need an active internet connection to use it as we monitor usage and reliability of the prerelease.</p>
<p><strike>You can download the prerelease at:</strike></p>
<p><strike>/download/Analyseit/4.50/Analyseit_4_50.EXE</strike></p>
<p><strike>To activate the prerelease, use the product key:</strike></p>
<pre><strike>PTESTD4NF4LXJSYTT</strike>
</pre>
<br>
<p>We are now working to update the documentation to explain the new features of the software, but in the meantime these examples should help you get started</p>
<p><a href="/assets/article/54611641/Pareto.xlsx" class="excelfile">Pareto.xlsx: Pareto plots of failures stratified by date and before/after maintenance</a></p>
<p><a href="/assets/article/54611641/BurstingStrength.xlsx" class="excelfile">BurstingStrength.xlsx: Xbar R control charts and capability analysis of bursting strength of glass containers</a></p>
<p><a href="/assets/article/54611641/Cans.xlsx" class="excelfile">Cans.xlsx: Shewhart p and np control plots analysis of can seal attribute data</a></p>
<p>We will provide more information about the product, availability and pricing shortly, including pricing for existing users that want to upgrade to include the new quality control features.</p>
Releases
Business
In development
Tue, 20 Oct 2015 13:05:31 GMT

https://analyseit.com/blog/2015/9/analyseit420releasedsupportforexcel2016
Analyseit 4.20 released: Support for Excel 2016
https://analyseit.com/blog/2015/9/analyseit420releasedsupportforexcel2016
<p><img src="/assets/article/54611640/Office2016sm.png" style="float:right">
Microsoft officially released <a href="https://products.office.com/" target="_blank">Office 2016</a> a couple of days ago, and Analyseit version 4.20 now adds support for Excel 2016. </p>
<p>Over the next few weeks we will tweak the Analyseit user interface so it matches the new Office 2016 user interface styles. Line styles on the plots in Excel 2016 now also appear a lot thicker, due to antialiasing (smoothing), so we will decide whether to address that in a future update – let us know what you think.</p>
<p>If you have <a href="/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your licence has expired you can renew it to get this update and forthcoming updates, see <a href="/store/maintenance/" target="_blank">renew maintenance</a>.</p>
Releases
Excel
Thu, 24 Sep 2015 11:15:17 GMT

https://analyseit.com/blog/2015/8/thenumericalaccuracyofanalyseitagainsttheniststrd
The numerical accuracy of Analyseit against the NIST StRD
https://analyseit.com/blog/2015/8/thenumericalaccuracyofanalyseitagainsttheniststrd
<p>A critical feature of any analytical and statistical software is accuracy. You are making decisions based on the statistics obtained and you need to know you can rely on them.</p>
<p>We have documented our <a href="/blog/2014/1/oursoftwaredevelopmentandvalidationprocess">software development and validation process</a> previously, but another good benchmark to test statistical software against is the NIST StRD. The Statistical Engineering and Mathematical and Computational Sciences Divisions of NIST’s Information Technology Laboratory developed datasets with certified values for a variety of statistical methods against which statistical software packages can be benchmarked. The certified values are computed using ultrahigh precision floating point arithmetic and are accurate to 15 significant digits.</p>
<p>For more information about the NIST StRD see:</p>
<img src="https://www.itl.nist.gov/div898/strd/general/gifs/strd80.gif">
<p><a href="https://www.itl.nist.gov/div898/strd/" target="_blank">https://www.itl.nist.gov/div898/strd/</a></p>
<p>We tested version 4.00 of Analyseit against the NIST StRD on an Intel Xeon dual processor PC.<br>
</p>
<p>No statistical package achieves perfect accuracy for all the tests and no one package performs best for every test. Most statistical packages use IEEE754 double precision (64bit) floating point arithmetic and due to finite precision, roundoff, and truncation errors in numerical operations, are unable to obtain the exact certified value.</p>
<p>In the tests:</p>
<ul>
<li><b>Analyseit performed consistently amongst the best on all tests.</b></li>
<li><b>Analyseit performed better than some of the most popular wellknown statistical packages</b>.</li>
</ul>
<p>For more information on the performance of Analyseit against the NIST StRD, and to download the workbooks containing the analyses, see:</p>
<p><a href="/support/NISTStRD">/support/NISTStRD</a></p>
Statistics
Business
Tue, 11 Aug 2015 11:36:43 GMT

https://analyseit.com/blog/2015/6/analyseit40releasedsupportforclsiguidelinesandmeasurementsystemsanalysis
Analyseit 4.0 released: Support for CLSI guidelines, and Measurement Systems Analysis
https://analyseit.com/blog/2015/6/analyseit40releasedsupportforclsiguidelinesandmeasurementsystemsanalysis
<p>Today we released version 4.0 of the Analyseit Method Validation edition. This is a major new release with many new features and improvements.</p>
<h2>Support for 10 of the latest CLSI EP guidelines</h2>
<p>The latest release of the Analyseit Method Validation edition now supports 10 of the latest CLSI evaluation protocol (EP) guidelines. <a href="http://clsi.org" target="_blank">CLSI</a> guidelines are worldrenowned and are recognized by the College of American Pathologists (CAP), The Joint Commission, and the US Food and Drug Administration (FDA).</p>
<p>Analyseit has been a driving force in the adoption of statistics in method validation for over 15 years, has influenced many recommendations, and is the only software available with such extensive coverage for the latest CLSI guidelines.</p>
<p>CLSI guidelines supported in version 4.0 include:</p>
<table class="table small">
<tbody>
<tr>
<td class=""><a href="http://clsi.org/standards/products/methodevaluation/documents/EP05" target="_blank">EP05A3</a><br>
Evaluation of Precision of Quantitative Measurement Procedures</td>
</tr>
<tr>
<td class="redactorcurrenttd"><a href="http://clsi.org/standards/products/methodevaluation/documents/EP06" target="_blank">EP06A</a><br>
Evaluation of the Linearity of Quantitative Measurement Procedures</td>
</tr>
<tr>
<td class=""><a href="http://clsi.org/standards/products/methodevaluation/documents/EP09" target="_blank">EP09A3</a><br>
Measurement Procedure Comparison and Bias Estimation Using Patient Samples</td>
</tr>
<tr>
<td><a href="http://clsi.org/standards/products/methodevaluation/documents/EP10" target="_blank">EP10A3AMD</a><br>
Preliminary Evaluation of Quantitative Clinical Laboratory Measurement Procedures</td>
</tr>
<tr>
<td><a href="http://clsi.org/standards/products/methodevaluation/documents/EP12" target="_blank">EP12A2</a><br>
User Protocol for Evaluation of Qualitative Test Performance</td>
</tr>
<tr>
<td><a href="http://clsi.org/standards/products/methodevaluation/documents/EP15" target="_blank">EP15A3</a><br>
User Verification of Precision and Estimation of Bias</td>
</tr>
<tr>
<td class=""><a href="http://clsi.org/standards/products/methodevaluation/documents/EP17" target="_blank">EP17A2</a><br>
Evaluation of Detection Capability for Clinical Laboratory Measurement Procedures</td>
</tr>
<tr>
<td><a href="http://clsi.org/standards/products/methodevaluation/documents/EP21" target="_blank">EP21A</a><br>
Estimation of Total Analytical Error for Clinical Laboratory Methods</td>
</tr>
<tr>
<td class=""><a href="http://clsi.org/standards/products/methodevaluation/documents/EP24" target="_blank">EP24A2 (Replaces GP10A)</a><br>
Assessment of the Diagnostic Accuracy of Laboratory Tests Using Receiver Operating Characteristic Curves</td>
</tr>
<tr>
<td class=""><a href="http://clsi.org/standards/products/methodevaluation/documents/EP28" target="_blank">EP28A3C (Formerly C28A3C)</a><br>
Defining, Establishing, and Verifying Reference Intervals in the Clinical Laboratory</td>
</tr>
</tbody>
</table>
<h2>Measurement Systems Analysis (MSA)</h2>
<p>Measurement Systems Analysis (MSA) is a new feature in version 4.0. MSA unifies precision and linearity, which were available in earlier versions of Analyseit, but also includes trueness (bias) and detection capability so you can establish the limit of blank (LoB) and limit of detection (LoD). The unification of these analyses in MSA lets you dig deep to examine and understand the performance characteristics of a measurement procedure.</p>
<p><img src="/assets/article/54611638/Precision profile.PNG"></p>
<p><img src="/assets/article/54611638/Trueness.PNG"></p>
<p><img src="/assets/article/54611638/Detection limit.PNG"></p>
<h2>Partition Method Comparison and Reference Interval analyses</h2>
<p>A major improvement in version 4.0 is the ability to split a method comparison analysis into specific measuring intervals and apply a different analysis. For example, you might split the analysis into two measuring intervals: samples <1ug/L, and those above. For <1ug/L the relationship between measurements might be different from those >1 ug/L, so you can apply different fits and goals to each measuring interval to see how the methods compare.</p>
<p>Similarly, for reference intervals you can apply partitioning factors to generate reference intervals for different subgroups. For example, you might apply partitioning factors sex and age group so you can determine reference intervals for age groups 020, 2035, 3550, and 50+ for male / female subjects. For each subgroup you can change how the reference interval is calculated, for example using the robust reference interval calculation for a subgroup with very few subjects.</p>
<p><img src="/assets/article/54611638/Method comparision.PNG"></p>
<h2>Availability</h2>
<p>If you already have a licence for Analyseit, and you have active maintenance (access to software updates), you can download the latest version now. If maintenance on your licence has lapsed, now is a good time to renew maintenance to get this update and future updates before the price increase (see below).</p>
<p>To download the update, check whether you have maintenance, or to renew maintenance, see:</p>
<p><a href="/support/download" target="_blank">/support/download</a></p>
<p>If you are new to Analyseit you can download a free 30day trial at:</p>
<p><a href="/products/methodvalidation/try">/products/methodvalidation/try</a></p>
<h2>Pricing and licensing changes from 7th July 2015</h2>
<p>The cost of a 1user perpetual licence for the Analyseit Method Validation will increase from US$699 to US$749, and the cost of a 1 floatinguser perpetual licence will crease from US$1239 to US$1329 from the 7th July 2015. Since maintenance is based on a percentage of the licence price, the 1year maintenance renewal price for a 1user perpetual licence will increase slightly from US$69.90 to US$74.90.</p>
<p>We will also be trialling a new option to purchase a 1year annual subscription for the Analyseit Method Validation edition from the 7th July 2015. The cost will be US$ 449 for 1year and includes all software updates released within the year. At the end of the year you can renew your licence to continue using the software, or let the licence lapse if you no longer need it.</p>
<h2>Change summary</h2>
<p>Finally, below is a list of the major changes and improvements in version 4.0.</p>
<p><b>Reference interval</b></p>
<ul>
<li>Partition factors for separate reference interval analysis of subgroups.</li>
<li>Additional transformation functions: Manly exponential, 2stage Exponential/Modulus transformation.</li>
<li>Normal MVUE quantile estimator for reference limits.</li>
</ul>
<p><b>Diagnostic performance</b></p>
<ul>
<li>BiHistogram and dotpot of the distribution of positive/negative cases.</li>
</ul>
<p><b>Method comparison</b></p>
<ul>
<li>Adjust measuring interval to a linear range and extrapolate fits beyond the data to the measuring interval.</li>
<li>Partition into measuring intervals, with different relationships (e.g. constant / relative differences or variances) for each measuring interval.</li>
</ul>
<p><b>Linearity</b></p>
<ul>
<li>Improved report and plots.</li>
<li>Integrated into MSA to allow unified approach to evaluation of other components of error.</li>
</ul>
<p><b>Precision</b></p>
<ul>
<li>Improved report and plots.</li>
<li>Precision profile variance function fits with a choice of 7 fits, including Sadler 3parameter functions.</li>
<li>Limit of quantitation via prediction from CV% on precision profile plot.</li>
</ul>
<p><b>Trueness NEW</b></p>
<ul>
<li>Estimate bias with reference materials or PT/EQA material.</li>
</ul>
<p><b>Detection capability NEW</b></p>
<ul>
<li>Estimate the decision level critical value (LoB) using parametric and nonparametric estimator on blank material or via the precision profile variance function.<br>
</li>
<li>Estimate the detection limit (LoD) using nonblank material or via the precision profile variance function.</li>
</ul>
Method validation
Releases
Tue, 23 Jun 2015 14:08:02 GMT

https://analyseit.com/blog/2014/10/asombrenoteprofessorrickjones
A sombre note: Professor Rick Jones
https://analyseit.com/blog/2014/10/asombrenoteprofessorrickjones
<p>The recent of passing of Professor Rick Jones (see <a href="http://www.theguardian.com/education/2014/sep/09/rickjonesobituary" target="_blank">Rick Jones obituary</a>) caused me to reflect on the past.</p>
<p>I was very fortunate to earn a work placement with Dr Rick Jones at The University of Leeds in the summer of 1990. Rick was enthusiastic about the role of IT in medicine, and after securing funding for a fulltime position he employed me as a computer programmer. Early projects included software for automating the monitoring of various blood marker tests and software to diagnose Down’s syndrome. At the time many hospitals had inhouse solutions for diagnosing Down’s syndrome, and although the project took many years and the help of many other people to complete, it eventually gained widespread adoption.</p>
<p>Around 1992, Rick came up with the idea of a statistics package that integrated into Microsoft Excel. Armed with a ring bound folder containing the Excel SDK and a pile of medical statistics books, I set about the task of writing the software in C++. It wasn’t long before the first version of Astute was ready and commercially released.</p>
<p>Never short of ideas for new projects, Rick started leaving journal articles in my intray covering subjects such as Deming and PassingBablok regression, GalenGambino sensitivity/specificity and ROC analysis. Rick had widespread experience and knowledge of the statistics needed in the clinical laboratory, and was keen to make the subject less daunting and more accessible to clinicians. The plan was to extend Astute to include these new statistical techniques. It would be an entirely new type of statistics package, so new in fact that when we released the product we didn’t have a good name for it and it was simply called “Astute  Module 1”. </p>
<p><img src="/assets/article/54611637/AstuteCover.JPG"> <img src="/assets/article/54611637/AstuteMod1Contents.JPG"></p>
<p>Rick later coauthored the book <a href="http://www.amazon.co.uk/InvestigationStatisticsLaboratoryManagementTechnology/dp/0902429213" target="_blank">Clinical Investigation and Statistics in Laboratory Medicine</a> and it helped many clinicians better understand and appreciate the use of statistics in the laboratory.</p>
<p>Eventually, I moved on from Leeds University to cofound Analyseit, and develop the successor to Astute. The ‘Module 1’ product transformed into the Clinical Laboratory Module and later into the Method Validation Edition. From its conception in Astute, to its implementation to Analyseit, the product has transformed method validation in clinical laboratories and invitro diagnostic companies around the world. </p>
<p>Without Rick, Analyseit would not exist today. Rick was the original catalyst and visionary for the product. Although I am writing this with great sadness, reflecting on the past, Rick’s enthusiasm for life leaves me positive and inspired. His inspiration will live on through Analyseit.</p>
<p>Simon Huntington</p>
Method validation
Statistics
Wed, 29 Oct 2014 11:18:58 GMT

https://analyseit.com/blog/2014/8/analyseit380releasedprincipalcomponentanalysispca
Analyseit 3.80 released: Principal Component Analysis (PCA)
https://analyseit.com/blog/2014/8/analyseit380releasedprincipalcomponentanalysispca
<p>Today we released version 3.80 of the Analyseit Standard edition.</p>
<p>The new release includes Principal Component Analysis (PCA), an extension to the multivariate analysis already available in Analyseit. It also includes probably the most advanced implementation of biplots available in any commercial package. </p>
<p>New features include:</p>
<ul>
<li>Principal Components Analysis (PCA). </li>
<li>Scree plot. </li>
<li>Classical Gabriel and modern Gower & Hand biplots. </li>
<li>Covariance and Correlation monoplots. </li>
<li>Color maps for correlations and patterned matrices, based on sign and magnitude, to help identify patterns. </li>
<li><a href="/docs/tutorials/correlation/overview" target="_blank">Tutorial (with video)</a> on how to visualize the relationships between variables and similarities between observations. </li>
</ul>
<p><img src="/assets/article/54611636/BiPlot.png"></p>
<p>The tutorial walks you through a guided example looking at how to use correlation and principal component analysis to discover the underlying relationships in data about New York Neighbourhoods. It demonstrates the amazing new features and helps you understand how to use them. You can either follow the tutorial yourself, at your own pace, or <a href="/docs/video/correlationpca" target="_blank">sit back and watch the video now</a>.</p>
<p><a href="/docs/video/correlationpca" target="_blank"><img src="/assets/article/54611636/CorrelationPCA.jpg"></a></p>
<p>If you have <a href="/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your licence has expired you can renew it to get this update and forthcoming updates, see <a href="/store/maintenance/" target="_blank">renew maintenance</a>.</p>
Using Analyseit
In development
Releases
Statistics
Mon, 18 Aug 2014 12:46:05 GMT

https://analyseit.com/blog/2014/6/recentimprovementsinanalyseit376andourfirstvideotutorial
Recent improvements in Analyseit 3.76 and our first video tutorial!
https://analyseit.com/blog/2014/6/recentimprovementsinanalyseit376andourfirstvideotutorial
<p>If you <a href="https://www.facebook.com/analyse.it" target="_blank">follow us on Facebook</a> you will no doubt already know about the recent improvements in the Analyseit Method Validation edition and the release of our first video tutorial. If not, now is a good time to <a href="https://www.facebook.com/analyse.it" target="_blank">follow us</a> since we post short announcements and feature previews on Facebook, and use the blog only for news about major releases.</p>
<h4>What's new in release 3.76?<br>
</h4>
<p>The latest changes and improvements to the Analyseit Method Validation edition include:</p>
<ul>
<li>Extended BlandAltman Limits of Agreement (LoA) so you can calculate LoA using the ratio of measurements (for when difference between methods is related to magnitude of measurement), using linear regression (for when a transformation is not enough to eliminate the relationship between difference and magnitude), or using nonparametric percentiles (for when the differences are not normally distributed). </li>
<li>Added the <a href="/docs/userguide/methodcomparison/mountainplot" target="_blank">mountain plot</a> to method comparison so you can see the distribution of the differences. From the plot you can see the median of the differences, the central 95% interval, the range, and the percentage of observations outside the allowable error bands.</li>
<li>Added a new <a href="/docs/300/tutorials/blandaltman/overview" target="">indepth tutorial</a> to demonstrate how you can use Analyseit and the BlandAltman plot to determine the agreement between methods in various different scenarios. The tutorial is included in the software.</li>
</ul>
<p><img style="cursor: default;" src="/assets/article/54611635/BlandAltmanRegressionLoA.png"></p>
<p>If you have <a href="/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your licence has expired you can renew it to get this update and forthcoming updates, see <a href="/store/maintenance/" target="_blank">renew maintenance</a>.</p>
<h4>Watch our first video tutorial!<br>
</h4>
<p>Finally, we are delighted to release our first video tutorial. The tutorial is the video equivalent of the tutorial above. It walks and talks you through using Analyseit to determine the agreement between methods. Sit back and <a href="/docs/video/blandaltmanmethodagreement">watch the video now</a>. <br>
</p>
<p><a href="/docs/video/blandaltmanmethodagreement"><img style="cursor: default;" src="/assets/article/54611635/Intro1023.jpg"></a></p>
<p>We intend to produce more video tutorials in future, so let us know what you think: what you like, dislike, and how we can improve them in future.</p>
Method validation
Releases
Using Analyseit
Tue, 03 Jun 2014 13:15:29 GMT

https://analyseit.com/blog/2014/2/analyseit370releasedimprovedmultiplecomparisonsconfidenceintervaldatabarsmeanmeanplotandsavefitmodelvariablesbacktothedataset
Analyseit 3.70 released: improved multiple comparisons, confidence interval data bars, meanmean plot, and save fit model variables back to the dataset
https://analyseit.com/blog/2014/2/analyseit370releasedimprovedmultiplecomparisonsconfidenceintervaldatabarsmeanmeanplotandsavefitmodelvariablesbacktothedataset
<p>Today we released version 3.70 of Analyseit. <br>
</p>
<p>The new version includes many new features which some of you may have read about on our <a href="http://www.facebook.com/analyse.it" target="_blank">Facebook</a> page over the last few weeks:</p>
<p><img style="cursor: default;" src="/assets/article/54611634/MultipleComparisons.png"></p>
<p>New features include:<br>
</p>
<p>Compare groups</p>
<ul>
<li>Multiple comparisons: Hsu (with best), Scheffe (all contrasts), Steel (nonparametric against control), DwassSteelCritchlowFligner (nonparametric all pairs), Wilcoxon (individual comparisons).</li>
<li>pvalue on all multiple comparisons.</li>
<li>Confidence interval data bars on Excel 2010 and later.</li>
<li>MeanMean scatter plot for visualizing multiple comparisons. </li>
</ul>
<p>Fit Model</p>
<ul>
<li>Lack of fit test for all simple regression models: Line, Polynomials, Logarithmic, Exponential, Power.</li>
<li>Maximum attainable R2 in conjunction with lack of fit test.</li>
<li>Save variables back to the dataset: Fitted Y, Residuals, Standardized Residuals, Studentized Residuals, Leverage, Cook's Influence.</li>
<li>ttest on regression coefficients.</li>
<li>Standardized beta coefficients. </li>
</ul>
<p>Correlation</p>
<ul>
<li>Covariance matrix.</li>
<li>Correlation coefficient confidence interval data bars on Excel 2010 and later.</li>
</ul>
<p>If you have <a href="/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you will be notified an update is available when you next start Analyseit, or you can download and install the update now, see <a href="/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your licence has expired now is a good time to renew it to get this update and forthcoming updates, see <a href="/store/maintenance/" target="_blank">renew maintenance</a>.<br>
</p>
<p>We'll go into more detail on the new features over the next few weeks. In the meantime, <a href="http://www.facebook.com/analyse.it" target="_blank">like us on Facebook</a> to get the latest news, see announcements of forthcoming features, and get handy tips on how to use the myriad of features in Analyseit.</p>
Releases
Thu, 27 Feb 2014 13:34:49 GMT

https://analyseit.com/blog/2014/1/oursoftwaredevelopmentandvalidationprocess
Our software development and validation process
https://analyseit.com/blog/2014/1/oursoftwaredevelopmentandvalidationprocess
<p>Probably the greatest concern when using statistical software is reliability. Is the software producing accurate, numerically correct results that have been validated?</p>
<p>It’s a very important question. Many of you work in FDA and regulated environments where the penalties for mistakes are very high. And those of you outside such environments are still making important business and research decisions using Analyseit. It’s therefore imperative that the software you depend upon is developed to a professional standard, thoroughly tested and validated. </p>
<p>Today we are publishing details of the software development and validation process we use at Analyseit, see: </p>
<p><a target="_blank" href="/support/developmentandvalidationprocess">/support/developmentandvalidationprocess</a></p>
<p>For those of you working in regulated environments the document is essential to justify the use of offtheshelf, thirdparty software. </p>
<p>For everyone else it explains exactly how Analyseit is developed, tested and validated. You can use it as a checklist against which to judge the quality of alternatives. But be wary. Professional software development is complex and developing numerical software more so. We have over 25years software development experience (each developer, not aggregate), and over 20years developing statistical software. </p>
<p>For more about our track record, see:</p>
<p><a target="_blank" href="/company/aboutus">/company/aboutus</a></p>
In development
Business
Releases
Tue, 14 Jan 2014 14:59:07 GMT

https://analyseit.com/blog/2013/12/version360nowincludesrepeatmeasuresanovaandfriedmantests
Version 3.60 now includes repeatmeasures ANOVA and Friedman tests
https://analyseit.com/blog/2013/12/version360nowincludesrepeatmeasuresanovaandfriedmantests
<p>We have just released version 3.60 of the Analyseit Standard edition. It now includes repeatmeasures ANOVA and Friedman tests in the Compare Pairs analysis. </p>
<p>If you have active maintenance, Analyseit will notify you an update is available in the next few days, or you can download it immediately at:</p>
<p><a href="/support/download">/support/download</a></p>
<p>If you do not have active maintenance, now is a great time to extend maintenance to get this latest update and all updates for the next 1 or 3years, see:</p>
<p><a href="/store/FAQ#maintenance">/store/FAQ#maintenance</a></p>
Releases
Thu, 12 Dec 2013 15:42:01 GMT

https://analyseit.com/blog/2013/10/analyseitmethodvalidationeditionversion35isnowavailable
Analyseit Method Validation edition version 3.5 is now available
https://analyseit.com/blog/2013/10/analyseitmethodvalidationeditionversion35isnowavailable
<p>Today we released the Analyseit Method Validation edition version 3.5. The software is feature complete, validated, and includes documentation. It supports Excel 2007, Excel 2010 (32 and 64bit) and Excel 2013 (32 and 64bit). <br></p><p>We took this opportunity to rename the product from the Analyseit Method Evaluation edition to the Method Validation edition. The product is the same, but the new name better reflects the intended purpose of the product.</p><p><img style="cursor: default;" src="/assets/article/54611631/MethodValidation.PNG"></p>
<p>New features include:</p>
<p><b>Method Comparison</b></p>
<ul>
<li>All method comparison functions: scatters plots, difference plots, and fits, are now in one place so it’s easy to make decisions or choose fits after looking at the data.</li>
<li>Ordinary linear, Weighted linear, Ordinary Deming, Weighted Deming and Passing & Bablok regressions. </li>
<li>Constant difference and relative difference plots with mean difference, allowable difference, and limits of agreement.</li>
<li>Bootstrap confidence intervals for parameters and bias in Passing & Bablok regression.</li>
<li>Exact pvalues on CUSUM linearity test.</li>
<li>Predict mean difference at decision levels.</li>
<li>Equality and equivalence hypothesis tests at decision levels.</li>
<li>Kappa and Weighted Kappa qualitative method comparison.</li>
</ul>
<p><b>Precision</b></p>
<ul>
<li>GraybillWang / TingBurdickGraybillJeyaratnamLu modified large sample confidence interval estimators for variance components – produces confidence closer to the stated level than Satterthwaite confidence intervals, which can be too liberal.</li>
<li>Support for unbalanced 1 and 2factor nested random ANOVA models, for looking at precision between days, runs, laboratories, etc.</li>
</ul>
<p><b>Linearity</b></p>
<ul>
<li>HsiehLiu confidence interval estimator for degree of nonlinearity.</li>
<li>Equality and equivalence hypothesis tests for nonlinearity.</li>
</ul>
<p><b>Reference interval</b></p>
<ul>
<li>HarrellDavis, Bootstrap, and Robust Biweight quantiles.</li>
<li>BoxCox and other transformations.</li>
</ul>
<p><b>Diagnostic performance / ROC</b></p>
<ul>
<li>Compare up to 10 independent or correlated ROC curves. </li>
<li>Equality, equivalence and noninferiority hypothesis tests.</li>
<li>Estimate false positive fraction (FPF) at fixed sensitivity, sensitivity at fixed FPF, and sensitivity/FPF at fixed cutoff.</li>
<li>Find optimal decision threshold based on costs.</li>
</ul>
<p><b>Binary diagnostic tests</b></p>
<ul>
<li>Compare sensitivity and specificity of 2 independent or correlated tests.</li>
<li>Diagnostic oddsratio and Youden’s index.</li>
<li>Improved confidence interval estimators for likelihood ratio and predictive values.</li>
</ul>
<p>For more information about the new version, and to download a free 30day trial, see:</p>
<p><a href="/products/methodvalidation">/products/methodvalidation</a></p>
<p>Pricing for the Analyseit Method Validation edition starts at US$ 699 for a 1user perpetual licence. If you already have a licence you may qualify for a free upgrade, if you have active maintenance, otherwise you can extend maintenance to get the upgrade (and all updates for 1 or 3years) free of charge. To see if you qualify for a free upgrade, otherwise get a quote to extend maintenance, see:</p>
<p><a href="/support/download">/support/download</a></p>
In development
Method validation
Releases
Mon, 14 Oct 2013 13:44:55 GMT

https://analyseit.com/blog/2013/8/jointheanalyseitmethodevaluation35publicbetatestandearnafreeupgrade
Join the Analyseit Method Evaluation 3.5 public beta test and earn a free upgrade!
https://analyseit.com/blog/2013/8/jointheanalyseitmethodevaluation35publicbetatestandearnafreeupgrade
<h4>Update 14th October 2013</h4>
<p>The Analyseit Method Validation Edition version 3.50 has now been released, see <a href="/blog/2013/10/analyseitmethodvalidationeditionversion35isnowavailable">Analyseit Method Validation edition version 3.5 is now available</a>.</p>
<h4>Original article</h4>
<p>Today we released the first public beta test version of the Analyseit Method Evaluation edition, version 3.5. The software is feature complete and is validated – it is now only missing documentation.</p>
<p>We invite everyone to download the beta and try the new version of the software before it is finally released in September. You will need Excel 2007, 2010, or 2013 (32bit and 64bit versions are supported) and it can be installed and used alongside older versions of Analyseit so it won't interrupt your daytoday work.<br>
</p>
<p><img style="cursor: nwresize;" src="/assets/article/1/MethodComparisonScreenshot.png"></p>
<p>To download the beta version:</p>
<p><a target="_blank" href="/support/download/PTESTPKVAGWQKC7CS">/support/download/PTESTPKVAGWQKC7CS</a></p>
<p>To activate the software use the product key: </p>
<pre>PTESTPKVAGWQKC7CS</pre>
<h2>Will I qualify for a free upgrade?</h2>
<p>The software will be publically released at the end of September 2013. </p>
<p>If you purchased a licence in the last 12 months, the 12 months of maintenance included means you will qualify for a free upgrade to the new version. </p>
<p>If you are outside the 12 month free upgrade period you can purchase 12 months of maintenance, to get the upgrade (and all updates in the following year), for 20% of the cost of your licence. For example, if you have a 1user licence then the upgrade will cost 20% of the cost of a 1user licence. Similarly if you have a 3user licence the upgrade cost would be 20% of the cost of a 3user licence.</p>
<p>There is a way to reduce the cost…</p>
<h2>Earn a discount on your upgrade.</h2>
<p>The software is now validated but we would like your help to smoke out any differences in the statistics compared to Analyseit version 2.xx. To find differences, download the new release and compare the statistics against those shown by your current version of Analyseit. If you find a difference use the Send Feedback feature in the software to report the problem to us (see <a target="_blank" href="/docs/300/userguide/tasks/sendingfeedback">How to send feedback</a>).</p>
<p>If you find a difference in the statistics we will credit you $25 towards an upgrade (or new licence purchase), provided:</p>
<ul>
<li>The difference is not due to the use of a different algorithm or method. For example some statistics are now calculated using better formulas, so they will be numerically different.</li>
<li>The difference is not a small numerical difference due to a change in the order of calculation. Often a change in the order of calculation, or the splitting of computations between multiple processors, changes how rounding errors in floating point arithmetic accumulate, leading to a small difference.</li>
<li>Finally you must be using the latest release of the software. We will fix problems as they are reported and release an update that includes a fix, so please ensure you are using the latest version to avoid reporting problems that have already been fixed. Analyseit will automatically notify you of new releases, so ensure you install them as soon as possible.</li>
</ul>
<p>We will respond to every calculation problem reported. If it’s a genuine problem we will credit you US$ 25 that you can later use against the purchase of an upgrade or new licence. If it’s not a genuine problem we will tell you why (you will see the statistic does not change in later releases). There is no limit to the amount of credit you can earn, so please start testing now.</p>
<h2>Price increase from 1st September 2013.</h2>
<p>From the 1st September 2013 the price of the Analyseit Method Evaluation will increase slightly. A 1user perpetual licence will increase from US$ 649 to US$ 699, and a 1user concurrentuser perpetual licence will increase from US$ 1149 to US$ 1239. Pricing in other currencies will increase similarly.</p>
<p>The good news is you can buy a licence now, at the current prices, and the 1year maintenance included means you will get the new release (and all updates for the next year) free of charge. So if you’re considering buying a licence, or additional licences, buy now to beat the price increase.</p>
In development
Method validation
Releases
Wed, 14 Aug 2013 12:38:36 GMT

https://analyseit.com/blog/2013/6/analyseitmethodevaluationedition35alpha3nowavailable
Analyseit Method Evaluation Edition 3.5 alpha 3 now available
https://analyseit.com/blog/2013/6/analyseitmethodevaluationedition35alpha3nowavailable
<h4>Update 14th October 2013</h4>
<p>The Analyseit Method Validation Edition version 3.50 has now been released, see <a href="/blog/2013/10/analyseitmethodvalidationeditionversion35isnowavailable">Analyseit Method Validation edition version 3.5 is now available</a>.</p>
<h4>Original article</h4><p>Today we released the 3<sup>rd</sup> alpha release of the Analyseit Method Evaluation Edition 3.5. Alpha releases are versions of the software that are still in active development, but are released to small group of customers so we can identify and fix any problems before the public beta release.</p>
<p>This release now completes the package with method comparison, which includes Deming regression, PassingBablok regression, and BlandAltman difference plots. Linearity, precision analysis, diagnostic performance (ROC analysis and binary test performance) and reference intervals were already included in earlier alpha releases.</p>
<p>If you would like to take part in this and subsequent test phases reply to this post or <a href="/company/contactus" target="_blank">contact us</a>. The test releases will run alongside any existing version of Analyseit, so your daytoday work won't be interrupted or affected. And those who help during testing will receive a discount on the upgrade (a free upgrade for those who contribute the most) when the product is released later this year.</p>
<p><img style="cursor: nwresize;" src="/assets/article/54611629/methcomp.PNG"></p>
<p><br>
</p><p>New features in the
release include:</p>
<p><b>Method Comparison</b></p>
<ul>
<li>All method comparison functions: scatters plots, difference plots, and fits, are now in one place so it’s easy to make decisions or choose fits after looking at the data.</li>
<li>Ordinary linear, Weighted linear, Ordinary Deming, Weighted Deming and Passing & Bablok regressions. </li>
<li>Constant difference and relative difference plots with mean difference, allowable difference, and limits of agreement.</li>
<li>Kappa and Weighted Kappa qualitative method comparison.<br>
</li>
<li>Bootstrap confidence intervals for parameters and bias in Passing & Bablok regression.</li>
<li>Exact pvalues on CUSUM linearity test.</li>
<li>Predict mean difference at decision levels.</li>
<li>Equality and equivalence hypothesis tests at decision levels.</li>
</ul>
<p><b>Precision</b></p>
<ul>
<li>GraybillWang / TingBurdickGraybillJeyaratnamLu modified large sample confidence interval estimators for variance components – produces confidence closer to the stated level than Satterthwaire confidence intervals, which can be too liberal.</li>
<li>Support for unbalanced 1way and 2way nested random ANOVA models.</li>
</ul>
<p><b>Linearity</b></p>
<ul>
<li>HsiehLiu confidence interval estimator for degree of nonlinearity.</li>
<li>Equality and equivalence hypothesis tests for nonlinearity.</li>
</ul>
<p><b>Reference interval</b></p>
<ul>
<li>HarrelDavis, Bootstrap, and Robust Biweight quantiles.</li>
<li>BoxCox and other transformations.</li>
</ul>
<p><b>ROC</b></p>
<ul>
<li>Compare up to 10 independent or correlated ROC curves. </li>
<li>Equality, equivalence and noninferiority hypothesis tests.</li>
<li>Estimate false positive fraction (FPF) at fixed sensitivity, sensitivity at fixed FPF, and sensitivity/FPF at fixed cutoff.</li>
<li>Find optimal decision threshold based on costs.</li>
</ul>
<p><b>Binary diagnostic tests</b></p>
<ul>
<li>Compare sensitivity and specificity of 2 independent or correlated tests.</li>
<li>Diagnostic oddsratio and Youden’s index.</li>
<li>Improved confidence interval estimators for likelihood ratio and predictive values.</li>
</ul>
Releases
Method validation
In development
Wed, 19 Jun 2013 11:29:37 GMT

https://analyseit.com/blog/2013/5/analyseitmethodevaluationedition35alpha2nowavailable
Analyseit Method Evaluation Edition 3.5 alpha 2 now available
https://analyseit.com/blog/2013/5/analyseitmethodevaluationedition35alpha2nowavailable
<h4>Update 14th October 2013</h4>
<p>The Analyseit Method Validation Edition version 3.50 has now been released, see <a href="/blog/2013/10/analyseitmethodvalidationeditionversion35isnowavailable">Analyseit Method Validation edition version 3.5 is now available</a>.</p>
<h4>Original article</h4><p>Today we released the 2<sup>nd</sup> alpha of the Analyseit Method Evaluation Edition 3.5. <br></p><p>Alpha releases are prerelease versions of the software that are still in active development. We release them to a small group of customers so we can get feedback and quickly identify and fix any problems before the public beta release. If you want to take part in the test phase reply or comment on to this post or <a href="/company/contactus" target="">contact us</a>. You can use prerelease versions of Analyseit alongside your
existing version of Analyseit, so it won't disrupt your work. And, if you help during in the test phases you will get a discount on the upgrade (a free upgrade for those who contribute the most) when the product is released later this year.</p>
<p>This latest alpha release includes linearity and precision analysis, plus diagnostic test performance (ROC analysis and binary test performance) and reference intervals from the 1<sup>st</sup> alpha. <br></p>
<p><img style="cursor: default;" src="/assets/article/54611628/LinearityDifferencePlot.png"></p>
<p><br>Some of the new features included so far are:</p><p><b>Precision</b></p>
<ul>
<li>GraybillWang / TingBurdickGraybillJeyaratnamLu modified large sample confidence interval estimators for variance components – produces confidence closer to the stated level than Satterthwaire confidence intervals, which can be too liberal.</li>
<li>Support for unbalanced 1way and 2way nested random ANOVA models.</li>
</ul>
<p><b>Linearity</b></p>
<ul>
<li>HsiehLiu confidence interval estimator for degree of nonlinearity.</li>
<li>Equality and equivalence hypothesis tests for nonlinearity.</li>
</ul>
<p><b>Reference interval</b></p>
<ul>
<li>HarrelDavis, Bootstrap, and Robust Biweight quantiles.</li>
<li>BoxCox and other transformations.</li>
</ul>
<p><b>ROC</b></p>
<ul>
<li>Compare up to 10 independent or correlated ROC curves. </li>
<li>Equality, equivalence and noninferiority hypothesis tests.</li>
<li>Estimate false positive fraction (FPF) at fixed sensitivity, sensitivity at fixed FPF, and sensitivity/FPF at fixed cutoff.</li>
<li>Find optimal decision threshold based on costs.</li>
</ul>
<p><b>Binary diagnostic tests</b></p>
<ul>
<li>Compare sensitivity and specificity of 2 independent or correlated tests.</li>
<li>Diagnostic oddsratio and Youden’s index.</li>
<li>Improved confidence interval estimators for likelihood ratio and predictive values.</li>
</ul>
In development
Method validation
Releases
Fri, 10 May 2013 14:45:46 GMT

https://analyseit.com/blog/2013/4/announcingtheanalyseitmethodevaluationedition35
Announcing the Analyseit Method Evaluation Edition 3.5
https://analyseit.com/blog/2013/4/announcingtheanalyseitmethodevaluationedition35
<h4>Update 14th October 2013</h4>
<p>The Analyseit Method Validation Edition version 3.50 has now been released, see <a href="/blog/2013/10/analyseitmethodvalidationeditionversion35isnowavailable">Analyseit Method Validation edition version 3.5 is now available</a>.</p>
<h4>Original article</h4><p>We are now starting to release test previews of a major update to the Analyseit Method Evaluation edition. The new release will include many new features (we'll reveal more in the coming weeks) and will support 32 and 64bit versions of Excel 2007, 2010, and 2013.</p>
<p>During the initial test phases we release development versions of the application to a small group of customers to ensure it installs and runs as expected on a wide range of PCs and configurations. The official beta test phase stage then follows where more customers are invited to download and use the software, while we ironout the final few bugs before the official release. The official release is planned for summer 2013.</p>
<p><img style="cursor: default;" src="/assets/article/54611627/ReferenceIntervalPreview.png"></p>
<br>
<p><img style="cursor: default;" src="/assets/article/54611627/BinaryTestPerformancePreview.png"></p>
<p>If you want to take part in the test phase, reply to this post or <a href="/company/contactus">contact us</a> and let us know what aspects of Analyseit you use:</p>
<p>Analytical Linearity, Precision, Accuracy,<br>
Diagnostic performance (ROC, binary test performance),<br>
Reference ranges,<br>
Agreement (BlandAltman). </p>
<p>We will then invite you into the testing phase at the appropriate time over the next 46 weeks. You will be able to use the test version alongside your current Analyseit so it will not disrupt your daytoday work. </p>
<p>As a reward for your time, those who partipicate in the test phases will get a discount on the upgrade (or free upgrade for those who contribute the most) when the product is released later this year.</p>
In development
Releases
Method validation
Fri, 19 Apr 2013 15:39:36 GMT

https://analyseit.com/blog/2013/2/quantilespercentileswhysomanywaystocalculatethem
Quantiles, Percentiles: Why so many ways to calculate them?
https://analyseit.com/blog/2013/2/quantilespercentileswhysomanywaystocalculatethem
<p>What is a sample quantile or percentile? Take the 0.25 quantile (also known as the 25<sup>th</sup> percentile, or 1<sup>st</sup> quartile)  it defines the value (let’s call it <i>x</i>) for a random variable, such that the probability that a random observation of the variable is less than x is 0.25 (25% chance).</p>
<p>A simple question, with a simple definition? The problem is calculating quantiles. The formulas are simple enough, but a take a quick look on Wikipedia and you’ll see there are at least 9 alternative methods <a target="_blank" href="http://en.wikipedia.org/wiki/Quantile">http://en.wikipedia.org/wiki/Quantile</a>. Consequently, statistical packages use different formulas to calculate quantiles. And we're sometimes asked why the quantiles calculated by Analyseit sometimes don’t agree with Excel, SAS, or R.</p>
<h2>How are quantiles and percentiles calculated in Excel, SAS and R?</h2>
<p>Excel uses formula R7 (in the Wikipedia article) to calculate the QUARTILE and PERCENTILE functions. Excel 2010 introduced two new functions that use slightly different formulas, with different denominators: PERCENTILE.INC and PERCENTILE.EXC. </p>
<p>SAS, R and some other packages let you choose which formula is used to calculate the quantiles. While this provides some flexibility, as it lets you reproduce statistics calculated using another package, the options can be confusing. Most nonstatisticians don’t know when to use one method over another. When would you use the "Linear interpolation of the empirical distribution function" versus the "Linear interpolation of the modes for the order statistics for the uniform distribution on [0,1]" method? </p>
<h2>Why so many ways to calculate quantiles?</h2>
<p>Many of the formulas to calculate quantiles were developed when today's computing power wasn’t available. Believe it or not some are nearly 100 years old! Now they’re merely historical curiosities, but some remain in packages like SPSS that have their roots in the 1970s.</p>
<p>During the development of Analyseit we always ask: what’s the latest or best (in some sense) method we can use to calculate this statistic? </p>
<p>Filling Analyseit with all the formulas invented would be easy. But that doesn’t help the average user – users that don’t have the knowledge, or can’t invest the time needed to research the most suitable method. Instead we look at published research to find the best method. If there is no single best method we implement a small set of alternatives that provide the best solution in specific situations – situations we can clearly define and explain. </p>
<p>Every statistical test and estimator included in Analyseit has to pass this test to make the cut. We applied these principles when it came to quantiles.</p>
<p>Hyndman and Fan published a paper on calculating quantiles in 1996. It evaluated the methods used by popular statistics packages to calculate quantiles, with the intention to find a consensus on which all statistics packages could standardise. Of the 9 formulas used, 4 formulas satisfied five of the six properties desirable for a sample quantile and their derivations were deemed justified. Of those 4 formulas, Hyndman and Fan felt the "Linear interpolation of the approximate medians for order statistics" (method <a target="_blank" href="http://en.wikipedia.org/wiki/Quantile">R8 on the Wikipedia page</a>) formula was best due to the approximately medianunbiased estimates of the quantiles, regardless of the distribution. Of the remaining formulas, 2 were also distributionfree but were not unbiased, and the other was approximately unbiased only for the normal distribution. They concluded that formula R8 should be adopted as the standard across software packages. </p>
<p>That was 1996. Unfortunately little progress has since been made toward standardisation. Many statistical packages have a long history (even Analyseit is over 15years old now!) and most tend to stick to the same method to maintain backwards computability with older versions. Even R, a relative newcomer, doesn't use the recommended formula. It uses formula R7 by default, for compatibility with S (<a target="_blank" href="http://stat.ethz.ch/Rmanual/Rpatched/library/stats/html/quantile.html">http://stat.ethz.ch/Rmanual/Rpatched/library/stats/html/quantile.html</a>). Minitab, SPSS and SAS use R6.</p>
<h2>Consistency between the statistics and plots.</h2>
<p>To complicate the situation further quantiles and percentiles are also used in statistical plots. The Tukey boxplot, for example, uses the 1<sup>st</sup> and 3<sup>rd</sup> quartiles (0.25 and 0.75 quantiles) for the extent of the <i>box</i> element of the plot. </p>
<p>Frigge, Hoaglin and Iglewicz published a paper in 1989 that looked at how quantiles were calculated in 3 of the major statistical packages. They found that because each package used a different formula, each identified different observations as outliers. Confusing! They recommended statistical packages use the "Ideal or Machine Forths" formula for consistency, which is equivalent to using method R8 to calculate quartiles.</p>
<h2>What technique does Analyseit use?</h2>
<p>By now you can probably guess that we chose to use <a target="_blank" href="http://en.wikipedia.org/wiki/Quantile">R8 formula</a> in Analyseit.</p>
<p><br>
<img style="cursor: default;" alt="Analyseit uses the R8 formula (see Wikipedia) to calculate quantiles and percentile" src="/assets/article/54611626/Quantileformula.png"><br>
<br>
</p>
<p>Formula R8 is recommended as the standard in both papers cited above, for descriptive statistics and plots. And using a single formula avoids the confusing situation you’ll sometimes see with other statistics packages, where the 1<sup>st</sup> and 3<sup>rd</sup> quartiles used for the boxplot differ from sample .25 and .75 quantiles.</p>
<p>Of course using formula R8 can lead to some differences between the quantiles calculated by Analyseit and other packages, though often you'll only see it with small sample sizes. If possible, as you can in R, we recommend you change the quantile calculation to use the R8 formula. If not, you can be sure the statistics calculated by Analyseit are absolutely correct. And you can cite this article as to why!</p>
<p><b>Further reading:</b></p>
<p><a href="http://www.jstor.org/stable/2684934" target="_blank">Hyndman, R.J.; Fan, Y. (November 1996). "Sample Quantiles in Statistical Packages". The American Statistician 50 (4): 361–365. </a><br>
</p>
<p><a href="http://www.jstor.org/stable/2685173" target="_blank">Frigge, Michael; Hoaglin, David C.; Iglewicz, Boris (February 1989). "Some Implementations of the Boxplot". The American Statistician 43 (1): 50–54.</a><br>
</p>
Using Analyseit
In development
Statistics
Mon, 18 Feb 2013 14:39:54 GMT

https://analyseit.com/blog/2013/1/analyseit310nowincludesastatisticalreferenceguide
Analyseit 3.10 now includes a statistical reference guide
https://analyseit.com/blog/2013/1/analyseit310nowincludesastatisticalreferenceguide
<p>Yesterday we improved the help in the <a href="/products/standard/" target="">Analyseit Standard Edition</a> and added a statistical reference guide. The guide tells you about the statistical procedures in Analyseit, with help on using and understanding the plots and statistics. It’s a work in progress, and we intend to improve it further with your comments and feedback, but it’s important to understand the role of the guide.</p>
<p>Firstly, the guide is not intended to be a statistics textbook. While it covers key concepts in statistical analysis, it is no substitute for learning statistics from a good teacher or textbook. </p>
<p>Secondly, the guide does not include the mathematical formulas behind the statistics. While an understanding of the mathematics is useful, it is better to understand the practical application of statistics: when and where they can be used, and how to interpret the results. Software makes it unnecessary to know the exact formulas, and often the exact mathematics used in software differ from those in textbooks since optimised routines are used to ensure good performance and numerical precision. </p>
<p>Finally, the guide does not explain experimental study design  an area where many mistakes are often made. We recommend you invest in a good textbook on study design to save wasting time on research that is ultimately flawed and cannot be published.</p><p><br><img style="cursor: default;" src="/assets/article/54611625/StatisticalReferenceGuide.png"><br><br></p><p>The Statistical Reference guide is included in version 3.10 of the Analyseit Standard Edition. If you have active maintenance you will be notified the update is available when you next start Analyseit. When you have installed the update you can access the help from the Help icon on the Analyseit tab. You can provide feedback on any topic in the help using the Send Feedback button on the toolbar within help.</p>
<p>If you prefer to read the guide on your ereader we have published it in <a href="/docs/300/UserGuide.epub" target="">EPUB format</a> (suitable for Apple iBooks and other ereaders) and <a href="/docs/300/UserGuide.mobi" target="">MOBI format</a> (suitable for Kindle readers).</p>
Releases
Using Analyseit
Wed, 16 Jan 2013 15:28:32 GMT

https://analyseit.com/blog/2012/10/analyseitstandardedition30improvementsmakeitacosteffectivealternativetotraditionalstatisticalsoftware
Analyseit Standard Edition 3.0 Improvements make it a CostEffective Alternative to Traditional Statistical Software
https://analyseit.com/blog/2012/10/analyseitstandardedition30improvementsmakeitacosteffectivealternativetotraditionalstatisticalsoftware
<p>Leeds, England (PRWEB) October 03, 2012  Analyseit Software, Ltd. today announced a major new release of their popular <a href="/products/standard/">statistical analysis software</a>, Analyseit®. With support for Excel 2007, 2010 and the forthcoming Excel 2013, Analyseit transforms Microsoft Excel into a costeffective powerful statistical analysis and data visualization package. Statistics and plots are included for exploring and describing data, estimating parameters, testing hypotheses, uncovering relationships and fitting models. </p>
<p>"We’ve made major improvements to <a href="/products/standard/regression">model fitting and regression analysis</a> in Analyseit 3.0", said Simon Huntington, Director of Statistical Products at Analyseit. "The improvements make Analyseit a serious rival to larger statistical packages costing up to 5 times the price. Unlike the alternatives though, Analyseit lets you perform all your statistical analysis without having to leave, or export your data, from Microsoft Excel."</p>
<p>Amongst the hundreds of improvements, model fitting and regression analysis have been improved to support simple models such as linear, logarithmic, exponential and power regression and advanced models such as <a href="/products/standard/regression">multiple linear regression</a> and <a href="/products/standard/regression">logistic regression</a>. Relationships between variables can be visualized using the <a href="/products/standard/regression">scatterplot matrix</a> and partial residual leverage plots show the effect of each term when building the model. Residual plots, distribution plots, lag plots, and sequence plots are also included for checking model assumptions. And an influence plot helps to quickly identify outliers and influential points, based on Studentized residuals and Cook’s D.</p>
<p>Pricing for Analyseit 3.0 starts at just US$ 249 for a 1user perpetual licence, or US$ 99 for a 1user annual subscription. Customers with maintenance can download version 3.0 at <a href="/support/download/">/support/download/</a>, or can get the upgrade by extending maintenance from just US$ 50.<br>
</p>
<p>For more information and to download a 30day trial, visit:</p>
<p><a href="/products/standard">/products/standard/</a></p>
<p>To check if your maintenance includes a free upgrade to version 3.0, and if not to extend maintenance, visit:</p>
<p><a href="/support/download/">/support/download/</a></p>
Releases
Business
Press releases
Wed, 03 Oct 2012 08:15:53 GMT

https://analyseit.com/blog/2012/7/analyseitstandardedition30releasecandidatenowavailable
Analyseit Standard Edition 3.0 release candidate now available
https://analyseit.com/blog/2012/7/analyseitstandardedition30releasecandidatenowavailable
<h4>Update 3rd October 2012</h4><p>The Analyseit Standard Edition version 3.00 is now available, see <a href="/products/standard/" target="">Analyseit Standard Edition</a> to purchase a licence or download a trial.</p><h4>Original article</h4><p>Today we pushed the release candidate of the Analyseit Standard Edition v3.0 for Microsoft Excel 2007 & 2010, our statistical analysis software for Microsoft Excel. </p><p>The release candidate is feature complete and is intended to be the final, almost public release of the software. </p><p>The software is now validated against our library of thousands of tests to ensure the statistics and plots are accurate and correct. You can now use the software in your day to day work and, like previous versions, it can be used alongside your current version of Analyseit until you become more comfortable with it. </p><p>The release candidate also includes a userguide and five tutorials to guide you through using the software by example. We recommend you complete the tutorials first as they help you quickly understand how to use Analyseit 3.0 and demonstrate many of the types of statistical analysis included. The send feedback feature is available from the help toolbar so you can easily send your suggestions on how we can improve the help.</p><p> <img alt="Tutorial help for Fit Model analysis" style="" src="https://secure.analyseit.com/assets/FitTutorial.png"></p><p>Finally the performance of the software is much improved and installation is simpler as we’ve provided a combined installer for all supported versions of Excel: Excel 2007 32bit, Excel 2010 32bit and Excel 2010 64bit. </p><p>The release candidate is available to everyone. If you already have a prerelease of Analyseit v3.0 installed you should receive an update notification in the next few days. Alternatively you can download it now at:</p><p><a href="/support/test/rc/">/support/test/rc/</a></p><p>In the next blog post we will announce pricing for Analyseit 3.0, and upgrade pricing and eligibility for current users of Analyseit. Those of you who helped during the prerelease testing of Analyseit will also be rewarded with a discount on the upgrade.</p>
In development
Releases
Thu, 12 Jul 2012 09:35:06 GMT

https://analyseit.com/blog/2012/5/analyseitstandardedition30finalbetaandtechnicalspecification
Analyseit Standard Edition 3.0 final beta and technical specification
https://analyseit.com/blog/2012/5/analyseitstandardedition30finalbetaandtechnicalspecification
<h4>Update 3rd October 2012</h4><p>The Analyseit Standard Edition version 3.00 is now available, see <a href="/products/standard/" target="">Analyseit Standard Edition</a> to purchase a licence or download a trial.</p><h4>Original article</h4><p>We’re pleased to release the final beta of the Analyseit Standard Edition, v3.0. The beta is now publicly open to anyone as we iron out any final issues and conduct final testing before release. </p>
<p>To download the beta, please visit: </p>
<p><a href="/support/test/beta/join.aspx">/support/test/beta/join.aspx</a></p>
<p><img alt="Influence plot for spotting influential and outlying points in fitting model." src="https://secure.analyseit.com/assets/InfluencePlot680x404.png"></p>
<p><em>Screenshot: Analyseit fit model analysis includes an influence plot to identify points with a substantial effect on the fitted model.</em></p>
<p>Many of you have asked what statistical tests and plots are included in the new release. The full specification is shown below. In coming weeks we'll announce pricing, upgrade pricing, and award free licences to the beta testers who contributed most time and effort. </p>
<table class="table">
<tbody>
<tr class="divider">
<td colspan="3">
<p>Descriptive statistics</p>
</td>
</tr>
<tr>
<td><br>
</td>
<td colspan="2">
<p>Mean, Median, Variance, SD, Skewness, Kurtosis</p>
<p>Quantiles / Percentiles</p>
<p>Frequency distribution table</p>
<p>Contingency table</p>
<p>Correlation coefficients – r, rs, tau</p>
<br>
</td>
</tr>
<tr class="divider">
<td colspan="3">
<p>Plots</p>
</td>
</tr>
<tr>
<td><br>
</td>
<td colspan="2">
<p>Histogram</p>
<p>Frequency plot</p>
<p>Dot plot</p>
<p>Box plot</p>
<p>Mean error bar plot</p>
<p>CDF plot</p>
<p>Normal QQ plot</p>
<p>Scatter plot</p>
<p>Scatter plot matrix</p>
<p>Leverage plot</p>
<p>Residual plots</p>
<p>Outlier and influence plot</p>
<p>Difference plot</p>
<p>Mosaic plot</p>
<p>Grouped frequency plot</p>
<p>Stacked frequency plot</p>
<p>Pie frequency plot</p>
<br>
</td>
</tr>
<tr class="divider">
<td colspan="3">
<p>Hypothesis tests</p>
</td>
</tr>
<tr class="subhead">
<td><br>
</td>
<td colspan="2">
<p>Location</p>
</td>
</tr>
<tr>
<td><br>
</td>
<td><br>
</td>
<td>
<p>Z</p>
<p>Student t</p>
<p>Welch t</p>
<p>Wilcoxon</p>
<p>WilcoxonMannWhitney</p>
<p>Sign</p>
<p>ANOVA</p>
<p>Welch ANOVA</p>
<p>KruskalWallis</p>
<br>
</td>
</tr>
<tr class="subhead">
<td><br>
</td>
<td colspan="2">
<p>Dispersion</p>
</td>
</tr>
<tr>
<td><br>
</td>
<td><br>
</td>
<td>
<p>X<sup>2</sup></p>
<p>Fisher F</p>
<p>Bartlett</p>
<p>Levene </p>
<p>BrownForsythe</p>
<br>
</td>
</tr>
<tr class="subhead">
<td><br>
</td>
<td colspan="2">
<p>Normality</p>
</td>
</tr>
<tr>
<td><br>
</td>
<td><br>
</td>
<td>
<p>ShapiroWilk</p>
<p>AndersonDarling </p>
<p>KolmogorovSmirnov</p>
<br>
</td>
</tr>
<tr class="subhead">
<td><br>
</td>
<td colspan="2">
<p>Association</p>
</td>
</tr>
<tr>
<td><br>
</td>
<td><br>
</td>
<td>
<p>Kendall test</p>
<p>Spearman test</p>
<p>Pearson test</p>
<br>
</td>
</tr>
<tr class="subhead">
<td><br>
</td>
<td colspan="2">
<p>Proportions</p>
</td>
</tr>
<tr>
<td><br>
</td>
<td><br>
</td>
<td>
<p>Binomial exact</p>
<p>McNemar exact</p>
<p>Fisher exact</p>
<p>Score Z</p>
<p>Pearson X<sup>2</sup></p>
<p>Likelihood ratio G<sup>2</sup></p>
<br>
</td>
</tr>
<tr class="divider">
<td colspan="3">
<p>Estimators</p>
</td>
</tr>
<tr class="subhead">
<td><br>
</td>
<td colspan="2">
<p>Location</p>
</td>
</tr>
<tr>
<td><br>
</td>
<td><br>
</td>
<td>
<p>Mean</p>
<p>Median</p>
<p>HodgesLehmann pseudomedian</p>
<p>tbased confidence interval for mean</p>
<p>Zbased confidence interval for mean</p>
<p>ThompsonSavur confidence interval for median</p>
<p>Tukey confidence interval for HodgesLehmann pseudomedian</p>
<p>Mean difference</p>
<p>tbased confidence interval for mean difference</p>
<p>Zbased confidence interval for mean difference</p>
<p>WelchSatterthwaite tbased confidence interval for mean difference</p>
<p>HodgesLehmann location shift</p>
<p>Tukey confidence interval for HodgesLehmann location shift</p>
<p>TukeyKramer confidence interval for mean difference</p>
<p>Dunnett confidence interval for mean difference</p>
<br>
</td>
</tr>
<tr class="subhead">
<td><br>
</td>
<td colspan="2">
<p>Dispersion</p>
</td>
</tr>
<tr>
<td><br>
</td>
<td><br>
</td>
<td>
<p>Variance</p>
<p>X<sup>2</sup>based confidence interval for variance</p>
<p>Variance ratio</p>
<p>Fbased confidence interval for variance ratio</p>
<br>
</td>
</tr>
<tr class="subhead">
<td><br>
</td>
<td colspan="2">
<p>Proportions</p>
</td>
</tr>
<tr>
<td><br>
</td>
<td><br>
</td>
<td>
<p>Proportion</p>
<p>Odds</p>
<p>ClopperPearson exact confidence interval for proportion / odds</p>
<p>Wilson score confidence interval for proportion / odds</p>
<p>Proportion difference (risk difference)</p>
<p>Proportion ratio (risk ratio)</p>
<p>Odds ratio</p>
<p>MiettinenNurminen score confidence interval for proportion difference / proportion ratio / oddsratio</p>
<p>Newcombe score confidence interval for proportion difference</p>
<p>Conditional exact confidence interval for odds ratio</p>
<br>
</td>
</tr>
<tr class="subhead">
<td><br>
</td>
<td colspan="2">
<p>Correlation</p>
</td>
</tr>
<tr>
<td><br>
</td>
<td><br>
</td>
<td>
<p>Pearson r</p>
<p>Fisher Z confidence interval for Pearson r</p>
<p>Spearman rs</p>
<p>Kendall tau</p>
<p>SamaraRandles confidence interval for Kendall tau</p>
<br>
</td>
</tr>
<tr class="divider">
<td colspan="3">
<p class="redactorcurrenttd">Regression and model fitting</p>
</td>
</tr>
<tr>
<td><br>
</td>
<td colspan="2">
<p class="redactorcurrenttd">Linear regression</p>
<p>Polynomial 2nd to 6th order</p>
<p class="redactorcurrenttd">Logarithmic regression<br>
</p>
<p class="redactorcurrenttd">Exponential regression</p>
<p class="redactorcurrenttd">Power regression</p>
<p class="redactorcurrenttd">Multiple linear regression</p>
<p>Binary logistic regression</p>
<p class="redactorcurrenttd">Advanced multiple linear and logistic regression models with simple, crossed, polynomial and factorial terms, with categorical explanatory variables coded as dummy variables</p>
<br>
</td>
</tr>
</tbody>
</table>
Releases
In development
Thu, 24 May 2012 13:50:53 GMT

https://analyseit.com/blog/2012/3/jointheanalyseitstandardedition30beta
Join the Analyseit Standard Edition 3.0 beta!
https://analyseit.com/blog/2012/3/jointheanalyseitstandardedition30beta
<h4>Update 3rd October 2012</h4><p>The Analyseit Standard Edition version 3.00 is now available, see <a href="/products/standard/" target="">Analyseit Standard Edition</a> to purchase a licence or download a trial.</p><h4>Original article</h4><p>Today we released the first public beta of
Analyseit Standard Edition, v3.0.</p>
<p>Maybe we're biased, but it is an amazing product! It's been a heck of a lot of work, but the range
of statistical tests and plots in Analyseit v3 rival what the expensive, established statistical packages provide. In fact, in many cases we’ve surpassed what
they offer. And the statistical plots go beyond what's available in any
other Excel addin.</p>
<p><img style="width: 680px; height: 404px; cursor: default;" src="https://secure.analyseit.com/assets/MosaicPlot680.png" align="" border="0"></p>
<p>The betatest programme is open to everyone. Even if you don't currently use Analyseit you are welcome to join. If you do use Analyseit the beta will run alongside any existing version, so your daytoday work won't be interrupted or affected.</p>
<p>To join, drop an email to <a href="mailto:beta@analyseit.com">beta@analyseit.com</a> and <a href="/googleplus">follow us on Google+ to join the betatesting circle</a>. We will then send you a link with instructions on how to download, install and start using the software. As a reward for your time, all Google+ betatest followers will get a
discount on the upgrade when Analyseit v3.0 is launched.</p>
<p>In the coming weeks we will reveal more about Analyseit v3.0, including upgrade and new licence pricing. If you are currently considering whether to buy a licence for Analyseit v2.0, don’t hesitate – you’ll get the upgrade to version 3.0 free of charge and avoid the price increase.</p>
In development
Releases
Fri, 02 Mar 2012 14:57:22 GMT

https://analyseit.com/blog/2011/11/announcinganalyseitstandardedition30
Announcing Analyseit Standard Edition 3.0
https://analyseit.com/blog/2011/11/announcinganalyseitstandardedition30
<h4>Update 3rd October 2012</h4><p>The Analyseit Standard Edition version 3.00 is now available, see <a href="/products/standard/" target="">Analyseit Standard Edition</a> to purchase a licence or download a trial.<br></p><h4>Original article</h4><p>If you follow us on <a href="http://twitter.com/Analyse_it" target="_blank">Twitter (@Analyse_it)</a> you will have seen that we released a new version of the Analyseit Standard Edition, v3.0, to testing this week.</p>
<p>During the initial test phases we release development versions of the application to a small group of customers to ensure it installs and runs as expected on a wide range of PCs and configurations. The official beta test phase stage then follows where more customers are invited to download and use the software, while we ironout the final few bugs before the official release. The official release is planned for early 2012.</p>
<p><img style="bordercolor: rgb(102, 102, 102); width: 680px; height: 425px;" alt="Analyseit v3 scatter matrix with histograms and density ellipses" src="https://secure.analyseit.com/assets/AnalyseIt_v3_ScatterMatrix.png" border="1"></p>
<p><em>Screenshot: Analyseit v3 scatter matrix with histograms and density ellipses</em></p>
<p>In coming months we’ll reveal more about the many new features, statistics & plots in Analyseit Standard Edition 3 (see image above). Pricing and upgrade costs will also be announced, though many, including anyone buying a licence today, will receive the upgrade free. </p>
<p>To join, get involved, and start testing early releases of Analyseit 3 email <a href="mailto:support@analyseit.com">support@analyseit.com</a>. The only commitment is you need to use the application for a few hours over the coming weeks and will need either Excel 2007 or Excel 2010 (on any version of Windows). You can use the test version alongside your current Analyseit so it will not disrupt your daytoday work. </p>
<p>If you don’t already, you can <a href="/blog/subscribe.aspx" target="_blank">subscribe to the blog</a>, or follow us on <a href="http://twitter.com/Analyse_it" target="_blank"><font color="#004f84">Twitter (@Analyse_it)</font></a>, to get the latest information. We plan to use Twitter more in future to post small updates and notifications, and use the blog for longer articles. </p>
Releases
In development
Wed, 16 Nov 2011 14:47:30 GMT

https://analyseit.com/blog/2010/8/analyseit222addssupportforexcel2010
Analyseit 2.22 adds support for Excel 2010
https://analyseit.com/blog/2010/8/analyseit222addssupportforexcel2010
<h4>Update 4th April 2012</h4><p>The Analyseit Standard Edition version 3.00 now supports 32bit and 64bit versions of Excel 2010. For more information see <a href="/products/standard/" target="">Analyseit Standard Edition</a> .<br></p><h4><p> </p>Original article</h4><p>It's been a few months since we released Analyseit 2.22, which added compatibility with, what was then, the Excel 2010 release candidate. Now <a href="http://www.microsoft.com/showcase/en/us/details/d025b8ed721a47bf8b3ab104ffd31527" target="_blank">Excel 2010 is available</a> it seems many are upgrading from Excel 2003 and are contacting us to ask whether Analyseit is compatible. It is!</p>
<p>Interestingly, starting with Office 2010, Microsoft is providing 32 and 64bit versions of Microsoft Excel. Until now Excel has been a 32bit application only (going back to Excel 5 which was a 16bit application). <a href="http://blogs.msdn.com/b/excel/archive/2009/08/28/excel2010nowwithmorebits.aspx" target="_blank">Excel 64bit</a> natively supports the 64bit microprocessors becoming more common in desktop PCs and allows you to work with truly enormous quantities of data. </p>
<p>Unfortunately most Excel addins, such as Analyseit, are 32bit applications and are not compatible with the 64bit versions. It will take sometime until addins like Analyseit natively support 64bit – it has taken Microsoft many years to take the first step. For this reason, Microsoft recommends you only choose to install Excel 64bit if you really need it. And 99.99% of users don’t. Excel 32bit supports over 1million rows by over 65,000 columns and up to 2GB workbooks – that’s a lot of information! </p>
<p>If you have to choose whether to install Excel 32 or 64bit, and want to use Analyseit, ensure you choose to install the 32bit release. And before you think you'll have your cake and eat it – you can't install both 32bit and 64bit versions on the same PC. It's one or the other. See Microsoft’s advice: <a href="http://technet.microsoft.com/enus/library/ee681792.aspx" target="_blank">32bit vs 64bit editions of Office 2010</a>. </p>
<h4>How to download and install 2.22 </h4>
<p>You can download Analyseit version 2.22 now at: </p>
<p><a href="/support/download.aspx" target="_blank">/support/download.aspx</a> </p>
<p>When you’ve downloaded the update, simply install over your existing version of Analyseit. There’s no need to uninstall the old version and you won’t need your product key to reactivate Analyseit  unless you’re upgrading from Analyseit version 1.xx in which case you’ll need to <a href="/support/retrieve_product_key.aspx" target="_blank">request a product key</a> to activate Analyseit 2.20. </p>
<p>Network administrators can install Analyseit version 2.22 by repeating the server installation steps as described at <a href="/userguide/install_net.aspx" target="_blank">/userguide/install_net.aspx</a> </p>
<h4>What's changed in this release? </h4>
<p>Since many of you work in regulated environments, and need to revalidate changes made in each Analyseit update, we have described the changes in the change log at: </p>
<p><a href="/support/changes.aspx" target="_blank">/support/changes.aspx</a> </p>
<h4>Subscribe for news of the latest releases </h4>
<p>As we mentioned before, all major and minor updates to Analyseit on this blog. If you’re not already a subscriber, subscribe to the blog by email or via the RSS feed to be notified of new versions as soon as they’re released. </p>
Releases
Thu, 12 Aug 2010 19:41:27 GMT

https://analyseit.com/blog/2009/10/microsoftlauncheswindows7
Microsoft launches Windows 7
https://analyseit.com/blog/2009/10/microsoftlauncheswindows7
<P><IMG style="WIDTH: 120px; HEIGHT: 120px" alt="Windows 7" align=right src="/assets/Windows7.png">Yesterday Microsoft launched <A href="http://www.microsoft.com/windows/windows7/default.aspx" target=_blank>Windows 7</A>. It’s the next version of Windows, following on from Windows Vista, Windows XP, and Windows 2000.</P>
<P>As a software vendor we had early access to Windows 7 and have been using it daily for approximately 68 months. Our impressions are Windows 7 is very reliable, stable, much faster than Vista, and is <STRONG><EM>an upgrade we wouldn’t hesitate to recommend</EM></STRONG>.</P>
<P>Under the hood, dramatic improvements have been made. Startup time is reduced so Windows 7 is ready to use from coldstart or hibernation far more quickly than Vista and even XP. In normal use it also feels much faster and more responsive.</P>
<P>Usability is improved significantly, with many slick user interface improvements especially for managing applications on the task bar. Many users will be relieved to know that the learning curve is shallow though – the userinterface isn’t such a departure from Windows Vista that you’ll be less productive for the days after upgrading. </P>
<P>Probably the biggest concern when upgrading is backwards compatibility: will it run the software and work with the hardware you already have. Unlike Vista, the answer seems to be yes. </P>
<P>As you might expect, <A href="http://www.microsoft.com/windows/compatibility/windows7/enus/Details.aspx?type=Software&p=Analyseit%20Standard%20Edition&v=Analyseit%20Software%2c%20Ltd&uid=2&pf=0&pi=0&s=analyseit&os=32bit" target=_blank>Analyseit is fully compatible</A>. Office 2007 runs perfectly, as does Office 2003 so if you’re not ready for the Office 2007 Ribbon userinterface yet you won’t be forced to upgrade Office at the same time. </P>
<P>Generally it seems most applications released in the last few years will run without any problems. If you do have a particularly old Windows XP application that isn’t compatible, Windows 2007 Professional/Ultimate Editions offer <A href="http://windows.microsoft.com/enus/windows7/products/features/windowsxpmode" target=_blank>Windows XP mode</A>, a very slick feature whereby you can run an application in a virtual, yet fully fledged, Windows XP environment. The application runs in a true Windows XP environment, yet appears like any regular application as part of the Windows 7 desktop, on the start menu, and on the taskbar.</P>
<P>Finally, Windows 7 offers better outofthebox support for the myriad of PC hardware available, so in most cases you won’t be forced to search the internet for suitable hardware drivers.</P>
<P>To check the compatibility of any of your existing hardware or software, before upgrading, take a look at the <A href="http://www.microsoft.com/windows/compatibility/windows7/enus/default.aspx" target=_blank>Windows 7 Compatibility Center </A>.</P>
Releases
Fri, 23 Oct 2009 16:01:19 GMT

https://analyseit.com/blog/2009/5/analyseit220released
Analyseit 2.20 released
https://analyseit.com/blog/2009/5/analyseit220released
<P>Today we released the latest update to Analyseit, version 2.20, which includes some minor fixes and major improvements. The major improvements affect users of Excel 2007, though we recommend all users download this update for the minor fixes included.</P>
<H4>Microsoft Office/Excel 2007 Service Pack 2</H4>
<P>Late last week Microsoft started to release Microsoft Office 2007 Service Pack 2, and made it publicly available on Tuesday this week. Service Pack 2 includes many improvements, including some worthy performance improvements for users of Microsoft Outlook. </P>
<P>Unfortunately, the changes made to Excel 2007 caused Analyseit 2.14 (and earlier) to sometimes crash when you clicked Refresh or Collate on the Report toolbar. Analyseit 2.20 fixes the problem, working around the bug introduced in Excel 2007 Service Pack 2.</P>
<P>If you want to download Microsoft Office 2007 Service Pack 2, you can <A href="http://www.microsoft.com/downloads/details.aspx?FamilyID=b444bf1879ea46c68a819db49b4ab6e5&displaylang=en" target=_blank>download it from Microsoft</A> or install it automatically using <A href="http://windowsupdate.microsoft.com/" target=_blank>Windows Update</A>.</P>
<H4>Support for large Excel 2007 worksheets</H4>
<P>We've also taken the opportunity in this release to add support for large Excel 2007 worksheets. Before Excel 2007 a worksheet was limited to 65,536 rows and 256 columns. Excel 2007 expands this so you can now have a worksheet as large as 1,048,576 rows and 16,384 columns. </P>
<P>Analyseit now supports the same, so in Excel 2007 you can analyse very large datasets. Unfortunately, Excel charts are still limited to 32,000 data points though, so some Analyseit charts such as the dotplot, and scatterplot, won’t display for very large datasets.</P>
<H4>How to download and install 2.20</H4>
<P>You can download Analyseit version 2.20 now at: </P>
<P><A href="/support/download.aspx">/support/download.aspx</A> </P>
<P>When you’ve downloaded the update, simply install over your existing version of Analyseit. There’s no need to uninstall the old version and you won’t need your product key to reactivate Analyseit  unless you’re upgrading from Analyseit version 1.xx in which case you’ll need to <A href="/support/retrieve_product_key.aspx" target=_blank>request a product key</A> to activate Analyseit 2.20.</P>
<P>Network administrators can install Analyseit version 2.20 by repeating the server installation steps as described at <A href="/userguide/install_net.aspx">/userguide/install_net.aspx</A></P>
<H4>What's changed in this release?</H4>
<P>Since many of you work in regulated environments, and need to revalidate changes made in each Analyseit update, we have described the changes in the change log at: </P>
<P><A href="/support/changes.aspx">/support/changes.aspx</A></P>
<H4>Subscribe for news of the latest releases </H4>
<P>As we mentioned before, all major and minor updates to Analyseit on this blog. If you’re not already a subscriber, <A href="/blog/2008/6/theanalyseitblogislive.aspx" target=_blank>subscribe to the blog</A> by email or via the RSS feed to be notified of new versions as soon as they’re released. </P>
Releases
Fri, 08 May 2009 14:51:59 GMT

https://analyseit.com/blog/2008/12/pricechangevatreducedto15
Price change: VAT reduced to 15%
https://analyseit.com/blog/2008/12/pricechangevatreducedto15
<P>The British government recently announced a 2.5% reduction in VAT (sales tax) on goods purchased from the United Kingdom (see <A href="/blog/2008/7/youvespeltanalysewrong.aspx">where we are based</A>). UK VAT was previously 17.5%, but from the 1st December 2008 until the end of 2009 it has been reduced to 15%.</P>
<P>Like many businesses, last Monday, we implemented the change. </P>
<P>Customers in the United Kingdom who aren’t VAT exempt, and those of you in Europe who are not VAT registered and must pay VAT, will now pay only 15% VAT.</P>
<P>For customers outside Europe, including those of you in the USA, Canada, and Australia, UK VAT still does not apply and you can continue to purchase licences without paying UK sales tax.</P>
Business
Wed, 03 Dec 2008 13:35:04 GMT

https://analyseit.com/blog/2008/11/spotthedifferencenumbersstoredastext
Spot the difference: Numbers stored as text
https://analyseit.com/blog/2008/11/spotthedifferencenumbersstoredastext
<P>In clearly titling this blog post, we’ve probably already revealed the answer, but... Can you spot the difference between the two rows of values in the Excel spreadsheet shown below? </P>
<P><IMG style="WIDTH: 321px; HEIGHT: 42px" src="/assets/CompareNumbers.png" longDesc="Spot the difference: Numbers are stored as text"><BR><BR>Sorry, it’s a trick question, because (visually) there is no difference. The difference is how the values are stored by Microsoft Excel. The value 57 in the cell on second row is actually stored as a text string, not a number.</P>
<H4>When does Excel store numbers as text?</H4>
<P>When you type a value into a cell, Excel looks at what you’ve typed and decides whether it’s a valid number. If it is, the value is stored as a number, and if not it’s stored as text (a string of characters).</P>
<P>Considering this, how is it possible for Excel to store a value that looks like a number, as text? There are a few ways. Most common is when you copypaste data from another application, and the application providing the data <EM>fools</EM> Excel into believing the values should be stored as text. Similarly, if you import data from a database field that contained numbers stored as text, the numbers will be imported as text. Finally, you can force Excel to store a number as text by prefixing it with an apostrophe (‘).</P>
<H4>The sideeffects of numbers stored as text</H4>
<P>So what difference does it make? Venturing into computer science briefly, computers represent and store numbers and text values very differently – numbers are stored in a compact binary representation, and text strings are stored as a string of individual characters. The problem is that mathematical operators and functions can only be applied to values stored as numbers. Text strings, even those that look like numbers, cannot be operated on mathematically.</P>
<P>If you use Excel’s SUM function on the two rows, you can see the problem: </P>
<P> <IMG style="WIDTH: 417px; HEIGHT: 62px" src="/assets/CompareNumbersSUM.png" longDesc="Excel's SUM function doesn't work with numbers stored as text"></P>
<P>The SUM of the second row doesn’t match the first, because SUM works only on numeric values. The cell containing the text string 57 on the second row is ignored by SUM.</P>
<P>If you weren’t aware of this issue before, you might be surprised. Regardless of how Excel stores a value, if it looks like a number you might expect SUM and other worksheet functions to treat it as such!? Because Excel doesn't, this can lead to very subtle and difficult to spot errors.</P>
<H4>Microsoft’s attempt to mitigate the problem</H4>
<P>Microsoft recognises this problem and have tried to mitigate it since Excel 2002. Excel 2002, 2003 and 2007 now show a green triangular indicator in the topleft of any cell that contains a number stored as text. If you click the cell to activate it, a small popup menu appears so you can convert the cell content to a number. See below: </P>
<P><IMG alt="Excel's convert to number popup" src="/assets/ConvertToNumber.PNG"></P>
<P>For more information, see <A href="http://office.microsoft.com/enus/excel/HA010346511033.aspx" target=_blank>Convert numbers stored as text to numbers</A>. </P>
<H4>Analyseit’s solution to the problem</H4>
<P>We identified the problem shortly after releasing Analyseit in 1997. In all versions of Analyseit since, when Analyseit reads your data from the Excel worksheet it treats any <EM>numbers stored as text</EM> as numbers, so they are included in the analysis. </P>
<P>We feel this is a better approach than Excel’s current solution, but the different approaches can lead to confusion. Recently a customer was validating Analyseit and was surprised to find the Excel function CORREL gave a different answer to that shown by Analyseit’s Pearson correlation. You can probably guess why. Some numbers on his worksheet were actually stored as text, and just like Excel’s SUM function, the CORREL function ignored the text values. That meant the correlation coefficient was wrong, and didn't match Analyseit. Once we converted the numbers stored as text to actual numbers, Excel's CORREL function calculated the correlation coefficient properly. It then matched Analyseit.</P>
Using Analyseit
Excel
Thu, 27 Nov 2008 07:53:31 GMT

https://analyseit.com/blog/2008/11/analyseitspeedsmethodvalidationatnationallaboratory
Analyseit speeds Method Validation at National Laboratory
https://analyseit.com/blog/2008/11/analyseitspeedsmethodvalidationatnationallaboratory
<P>Today we’re delighted to publish the second case study into the use of Analyseit.</P>
<P>The case study features a national clinical laboratory in the USA that offers more than 2,000 tests and combinations to major commercial and government laboratories. They use Analyseit to determine analytical performance of automated immunoassays for some of the industry’s leading invitro diagnostic device makers  including Abbott Diagnostics, Bayer Diagnostics, Beckman Coulter and Roche Diagnostics.</P>
<P>Unfortunately we cannot name the enduser, or the organisation she works for, in the case study. Although she was delighted to feature in the case study, at final approval her organisation's committee preferred the names be withheld. Thankfully they have allowed us to use the case study, albeit anonymously.</P>
<P>You can <A href="/products/method_evaluation/case_study/NatClinLab.aspx" target=_blank>read the case study</A> online now or download the <A href="/products/method_evaluation/case_study/NatClinLab.pdf" target=_blank>Adobe PDF version</A>.</P>
<P>We would love to feature more customer stories in case studies. If you can get approval to participate – which we realise is very difficult in many industries – and have 20 minutes to spare for a telephone interview, please contact us at <A href="mailto:support@analyseit.com">support@analyseit.com</A>. </P>
Using Analyseit
Method validation
Case studies
Wed, 19 Nov 2008 11:27:47 GMT

https://analyseit.com/blog/2008/11/normalquantileprobabilityplots
Normal quantile & probability plots
https://analyseit.com/blog/2008/11/normalquantileprobabilityplots
<P>In a previous post, <A href="/blog/2008/8/testingtheassumptionofnormality.aspx">Testing the assumption of Normality</A>, we explained the tests provided in Analyseit to determine if a sample has normal distribution. In that post, we mentioned that although hypothesis tests are useful you should not solely rely on them. You should always look at the histogram and, maybe more importantly, the <EM>normal plot</EM>.</P>
<P>The beauty of the normal plot is that it is designed specifically for judging normality. The plot is very easy to interpret and lets you see where the sample deviates from normality. </P>
<H4>Interpreting the normal plot</H4>
<P>As an example, let’s look at the distribution of systolic blood pressure, for a random group of healthy patients. Analyseit creates the histogram (left) and normal plot (right) below: </P>
<P> <IMG alt="Normal Quantile plot" src="/assets/NormalQuantilePlot.png"></P>
<P>Looking at the histogram, you can see the sample is approximately normally distributed. The bar heights for 120122 and 122124 make the distribution look slightly skewed, so it’s not perfectly clear.</P>
<P>The normal plot is clearer. It shows the <EM>observations</EM> on the X axis plotted against the <EM>expected normal score</EM> (Zscore) on the Y axis. It’s not necessary to understand what an <EM>expected normal score </EM>is, nor how it’s calculated, to interpret the plot. All you need to do is check is that the points roughly follow the redline. The redline shows the ideal normal distribution with mean and standarddeviation of the sample. If the points roughly follow the line – as they do in this case – the sample has normal distribution.</P>
<P>And that’s the real beauty of the normal plot compared to the histogram – it's very easy to interpret. Visually, the human eye can better judge the points against a straightline. And, unlike the histogram, there’s less ambiguity. You don’t have to try judge histogram barheights against the normal overlay curve.</P>
<H4>Variations on the normal plot</H4>
<P>Analyseit creates what is technically called a Normal Quantile plot. Quantile is just another word for a normal or Zscore and refers to what’s shown on the Y axis (in the case of Analyseit).</P>
<P>There are actually four variations of the normal plot, or eight since depending on preference the X and Y axes are often swapped:</P>
<UL>
<LI><STRONG>Normal quantile plot.</STRONG> Observations plotted against expected normal score (Zscore, known as quantiles)
<LI><STRONG>Normal quantilequantile plot</STRONG> (also known as normal QQ plot). Normal score (Zscore, known as quantiles) of the observations plotted against expected normal score (Zscore, known as quantiles)
<LI><STRONG>Normal probability plot.</STRONG> Observations plotted against expected CDF (cumulative area under the normal curve, known as probability)
<LI><STRONG>Normal probabilityprobability plot</STRONG> (also known as normal PP plot). CDF (cumulative area under the normal curve, known as probability) of the observations plotted against expected CDF (cumulative area under the normal curve, known as probability) </LI></UL>
<P>By their nature, normal plots based on probability fail to emphasise nonnormality in extreme observations – in the tails of the distribution – as well as quantile based normal plots. Generally, probability/PP plots are better to spot nonnormality around the mean, and normal quantile/QQ plots to spot nonnormality in the tails.</P>
<P>Thankfully, whichever of variation of the normal plot you’re faced with, interpretation is the same. If the sample is normal you should see the points roughly follow a straightline. </P>
<P>In future posts we’ll show cases of skewed and peaked distributions, and explain how you can identify these problems from the histogram and normal plot. </P>
Using Analyseit
Plots
Statistics
Thu, 06 Nov 2008 15:26:28 GMT

https://analyseit.com/blog/2008/10/usingindirecttorefertocellsonanalyseitanalysisreports
Using INDIRECT to refer to cells on Analyseit analysis reports
https://analyseit.com/blog/2008/10/usingindirecttorefertocellsonanalyseitanalysisreports
<P>A customer contacted us last week to ask how to refer to cells on an Analyseit report worksheet, from a formula on another worksheet. The customer often used Analyseit's refresh feature, to repeat the statistical analysis and update the statistics, and direct references to cells on the report were being lost on refresh.</P>
<P>As an example, suppose you have used Analyseit linear regression to calculate the linear relationship between installation cost and the number of employees required, distance to the site, and the cost of machine being installed. Analyseit would calculate the effect of each variable on the final cost, technically known as regression coefficients, which you can then use to predict installation costs for jobs in future. </P>
<P>You might setup a worksheet to predict and quote installation costs for future jobs. You could use an Excel formula to reference the coefficients directly from the Analyseit report, for example:</P>
<P class=code>= Employees * CostAnalysis!C17 + Distance * CostAnalysis!C18 + MachineCost * CostAnalysis!C19</P>
<P>By directly referencing the coefficients calculated by Analyseit in your formula, you can be sure you’re using the exact values with no chance of error.</P>
<P>If you’ve used this technique before, you already know the problem. When you refresh the Analyseit report, to repeat the analysis and recalculate the statistics, references to cells on the report worksheet are broken. In the above example, the references to cells on the <EM>CostAnalysis</EM> worksheet become #REF!, for example:</P>
<P class=code>= Employees * #REF! + Distance * #REF! + MachineCost * #REF!</P>
<P>#REF! simply means the reference is broken and refers to a cell that no longer exists.</P>
<P>The reason the cells no longer exist is because when you click Refresh, Analyseit repeats the analysis, creates a new report worksheet to present the statistics and charts, then finally deletes the old worksheet and replaces it with the new. The cell references are broken when Analyseit deletes the old report worksheet.</P>
<P>Although not perfect there is a simple workaround for the problem using the INDIRECT function. Rather than refer directly to a cell, you wrap the cellreference in INDIRECT(“..”), to indirectly refer to it. For example, the installation cost prediction formula would become:</P>
<P class=code>= Employees * INDIRECT("CostAnalysis!C17") + Distance * INDIRECT("CostAnalysis!C18") + MachineCost * INDIRECT("CostAnalysis!C19")</P>
<P>If you now refresh the analysis, for example because you have collected more data, or found an error in the original data, the formula will still work. Cell references won’t be broken, and will refer to the latest calculated coefficients. </P>
Using Analyseit
Excel
Thu, 30 Oct 2008 15:19:34 GMT

https://analyseit.com/blog/2008/10/analyseitcutsprojecttimeinhalfatswisslab
Analyseit Cuts Project Time in Half at Swiss Lab
https://analyseit.com/blog/2008/10/analyseitcutsprojecttimeinhalfatswisslab
<P align=left><A href="http://www.eoc.ch/" target=_blank></A><A href="http://www.eoc.ch/" target=_blank><IMG style="WIDTH: 140px; HEIGHT: 78px" hspace=4 src="/assets/EOCLogo.PNG" align=right border=0></A>Today we’re delighted to publish the first case study into the use of Analyseit.</P>
<P align=left>Marco Balerna Ph.D., a Clinical Chemist at the <A href="http://www.eoc.ch/" target=_blank>EOC (Ente Ospedaliero Cantonale)</A> in Switzerland, used Analyseit when replacing the clinical chemistry and immunological analysers in EOC’s laboratories.</P>
<P align=left>Since the EOC provides clinical chemistry services to five large hospitals and three small clinics in the region, it was essential the transition to the new analysers went smoothly. Marco used Analyseit to ensure the analyser’s performance met the manufacturer’s claims, to ensure the reporting of patient results was not affected, and to comply with the regulations of the EOC’s accreditation.</P>
<P align=left>Overall the project involved comparing performance for 110115 parameters, comprising over 25,600 measurements with control materials and patient samples.</P>
<P align=left>Marco was so impressed with Analyseit and the time he saved, that he was very enthusiastic when we asked if we could feature his story in a case study. We would like to publically thank Marco for his cooperation in the case study. Grazie Marco! Salute!</P>
<P align=left>You can <A href="/products/method_evaluation/case_study/EOC.aspx">read the case study</A> online now or download the <A href="/products/method_evaluation/case_study/EOC.pdf" target=_blank>Adobe PDF version</A>.</P>
<P align=left>If you would like to feature in a future case study, on how you’re using Analyseit, please contact us at <A href="mailto:support@analyseit.com">support@analyseit.com</A> </P>
Using Analyseit
Method validation
Case studies
Tue, 21 Oct 2008 09:30:22 GMT

https://analyseit.com/blog/2008/10/analyseit212released
Analyseit 2.12 released
https://analyseit.com/blog/2008/10/analyseit212released
<P>Last Friday we released the latest update to Analyseit, version 2.12  a minor update, providing fixes to minor issues recently reported by customers. The update is available free.</P>
<P>If you're using <A href="/blog/2008/6/analyseit211released.aspx">version 2.11</A>, and not experiencing any of the issues fixed (see the <A href="/support/changes.aspx">change history</A>), then you can skip the the update if you wish. But if you're using an earlier version of Analyseit, version 2.10 or earlier, we recommend you get the update.</P>
<P>If you’re unsure which version of Analyseit you’re using, see our FAQ: <A href="/faqs/findinstalledversionofAnalyseit.aspx"><FONT color=#004f84>How to find which version of Analyseit you’re using</FONT></A>. </P>
<H4>How to download and install 2.12 </H4>
<P>Analyseit automatically checks for updates every 15 days on startup, and will tell you if an update is available to download. Firewalls can get in the way though, so if you haven’t got a notification yet, or want to download 2.12 right away, you can download at: </P>
<P><A href="/support/download.aspx"><FONT color=#004f84>/support/download.aspx</FONT></A> </P>
<P>When you’ve downloaded the update simply install over your existing version of Analyseit. There’s no need to uninstall the old version and you won’t need your product key to reactivate Analyseit  unless you’re upgrading from Analyseit version 1.xx in which case you’ll need to <A href="/support/retrieve_product_key.aspx"><FONT color=#004f84>request a product key</FONT></A> to activate Analyseit 2.12. </P>
<H4>What's changed in this release?</H4>
<P>We know many of you work in regulated environments and need to revalidate changes made in each Analyseit update. To see exactly what’s changed since the version you validated, so you can revalidate just the affected statistical tests, please see the change log at: </P>
<P><A href="/support/changes.aspx"><FONT color=#004f84>/support/changes.aspx</FONT></A> </P>
<H4>Subscribe for news of the latest releases</H4>
<P>As we mentioned last time, we’ll announce all major and minor updates on this blog. If you’re not already a subscriber, <A href="/blog/2008/6/theanalyseitblogislive.aspx"><FONT color=#004f84>subscribe to the blog</FONT></A> by email or via the RSS feed to be notified of new versions as soon as they’re released. </P>
Releases
Tue, 14 Oct 2008 10:15:19 GMT

https://analyseit.com/blog/2008/9/printanexcelchartfullpage
Print an Excel chart fullpage
https://analyseit.com/blog/2008/9/printanexcelchartfullpage
<P>Although the charts in Analyseit are large so they’re easy to read when printed, sometimes you need to print a chart to fill the full page. You can do so easily, without resizing the chart, in just a few steps:</P>
<OL>
<LI>Click anywhere on the chart you want to print. It doesn’t matter what part of the chart you select: the chart, a series, the legend – anywhere will do.
<LI><STRONG>Analyseit users:</STRONG> Click <EM>Print</EM> on the Analyseit toolbar. Analyseit will show a preview first, and from there you can print the chart.<BR><STRONG>Excel only users:</STRONG> Choose <EM>File</EM> > <EM>Print</EM> from the Excel menu bar to print, or choose <EM>File</EM> > <EM>Preview </EM>to preview it first. In Excel 2007, click the <EM>Office button</EM> then choose <EM>Print</EM>, or <EM>Print</EM> > <EM>Preview</EM>.
<LI>Excel will print the chart full page.</LI></OL>
<P>Chart size is only limited by the page size your printer supports. </P>
Plots
Using Analyseit
Excel
Thu, 25 Sep 2008 13:06:28 GMT

https://analyseit.com/blog/2008/9/euronowaccepted
EURO now accepted
https://analyseit.com/blog/2008/9/euronowaccepted
<P><IMG style="WIDTH: 153px; HEIGHT: 148px" alt=Euro hspace=2 src="/assets/eurosm.png" align=right vspace=2 border=0>We’re pleased to announce that from today we can accept payment in EUROs. You can now see prices for Analyseit in EUROs, as well as British Pounds sterling, and US dollars (for customers in the USA & Canada).</P>
<P>The EURO is now the primary currency in Europe and many of our customers have asked us to accept EUROs. We can accept payment in EUROs by VISA, MasterCard or AMEX credit card, or by cheque or wiretransfer.</P>
<P>When paying by credit card, your card will be charged the exact amount in EUROs. There are no currency conversions so no charges are incurred. The price you see, in EUROs, is the exact price we will charge to your credit card.</P>
<P>For established organisations we are happy to accept purchase orders in EUROs and invoice in EUROs. The invoice can be paid by cheque or wiretransfer, directly in EUROs, avoiding currency conversion costs. </P>
Business
Fri, 19 Sep 2008 09:11:59 GMT

https://analyseit.com/blog/2008/9/identifyingwhatwhenandwhoanalysedit
Identifying what, when and who analysed it
https://analyseit.com/blog/2008/9/identifyingwhatwhenandwhoanalysedit
<P>Identifying what was analysed, when, and by who, is the first step in understanding any Analyseit report. The top rows of each Analyseit report provide you with this information. The statistical test used, dataset and variables analysed, user who analysed, and the date and time last analysed, are included (see below). When you print the report the header is repeated at the top of printed page.<BR><BR><IMG alt="" src="/assets/ReportHeader.png"></P>
<P>The date the report was last updated is included so you can see when reports are out of sync with changes made to the dataset. It’s also useful if you archive analysis reports and need to know when the analysis was performed. For brevity Analyseit shows only the date, but the cell also contains the time of the last update to the report. To see the time, click the cell containing the date to activate it, and then look at the Excel formula bar to see the time (see screenshot above).</P>
<P>To aid traceability, Analyseit includes the name of the user who last updated the report. Analyseit gets the name from the Microsoft Office user name. The user name shared among all Microsoft Office applications, including Word, Excel, and PowerPoint. Office applications use the name to identify changes in documents, and store it in the document properties to identify who created, last edited, or modified an Office document. Analyseit includes the name in the report header so you can quickly see who last analysed the data, should you need to contact them.</P>
<P>To see, or change the Excel user name:</P>
<OL>
<LI>In <EM>Excel 2007</EM>, click the <EM>Office button </EM>at the topleft, then click the <EM>Excel Options </EM>button. <BR>In <EM>Excel 2003</EM> and earlier, choose <EM>Tools</EM> > <EM>Options</EM> from the Excel menu bar, then click the <EM>General</EM> tab.
<LI>Enter the <EM>User name</EM>. You can enter your name, department, group or company name, depending on how you want to be identified in the reports.
<LI>Click <EM>OK</EM>. Analyseit reports will now use the new user name. </LI></OL>
<P>If a few users share your PC, and each have a separate Windows user accounts, each user can set their own Excel/Office user name. Simply login to your account and then repeat the steps above. </P>
<P>When we first developed Analyseit, we did consider using the name of the Windows user currently loggedin. The problem is Windows account names are often first names, nicknames, or are abbreviated, and so aren’t ideal to clearly identify the person responsible for the analysis. It's also be difficult to change the name associated with a Windows user account, especially if you work at a large organisation where you need IT support to do it. For these reasons the Office/Excel user name is ideal.</P>
<P>A few consultants using Analyseit in their work with clients have asked if they can include their company name in the reports, or brand the reports with their company logo. You can enter your company name as the Excel user name, as we did for the screenshot above. Including a company logo is more complex, but is something we will consider if there are enough requests for the feature.</P>
<P>Post a comment and let us know what you would like to see in the report header. We’ll do our best to include any requests in future versions of Analyseit. </P>
Using Analyseit
Wed, 03 Sep 2008 16:57:52 GMT

https://analyseit.com/blog/2008/8/unfiledreportscollateandanalyseit3
Unfiled reports, Collate, and Analyseit 3
https://analyseit.com/blog/2008/8/unfiledreportscollateandanalyseit3
<P>In May this year, we surveyed users of the Analyseit Method Evaluation edition to gain insight into how we can improve Analyseit in future. Thank you to all those who responded.</P>
<P>In the responses, one issue became clear: the <EM>unfiled reports</EM> feature causes confusion.</P>
<P>When you run an analysis, Analyseit creates a new worksheet containing the statistics and charts for that analysis (what we call a report). Analyseit places the report in a temporary workbook called <A href="/userguide/Content/Using_Analyseit/Analysing_datasets/About_Unfiled_reports.htm"><EM>Unfiled reports</EM></A>. From there you can then decide what you want to do with the analysis: keep it, print it, email it, or discard it. If you want to keep it you click the <A href="/userguide/Content/Using_Analyseit/Analysing_datasets/Collating_reports.htm"><EM>Collate </EM>button</A> (see below), and Analyseit moves the report into the same workbook as your dataset.</P>
<P><IMG style="WIDTH: 640px; HEIGHT: 453px" src="/assets/UnfiledReports_CollateButton.png" longDesc="Analyseit Unfiled reports and Collate button"></P>
<P>You might wonder where the idea of <EM>unfiled reports</EM> originated. It was actually a carryover from <A href="/company/aboutus.aspx">Astute</A>, the predecessor to Analyseit. We implemented the same feature in Analyseit without really questioning it. We thought the feature would be useful to help you manage reports, plus most Astute users upgrading would expect it. </P>
<P>The survey revealed different. It seems the feature probably causes more confusion than it’s worth. It seems most uesrs tend to keep the analysis reports, and prefer Analyseit to place reports in the same workbook as the dataset. The survey revealed: </P>
<UL>
<LI>Some users didn’t realise the report worksheets could be moved from <EM>unfiled reports</EM> to their own workbook, and instead used copypaste to copy the statistics and charts to their own worksheet. This caused loss of formatting, and meant the reports couldn’t be edited or updated later.
<LI>Some users thought they needed to save the <EM>unfiled reports</EM> workbook containing the Analyseit analyses. That meant they had two workbooks: one containing the data, and another containing the Analyseit reports. Managing two workbooks caused more problems, especially if either were deleted. </LI></UL>
<P>We contacted the survey respondents reporting such concerns directly, to explain how they could better manage the Analyseit reports. Since we're now working on <A href="/blog/2008/6/analyseit30nowindevelopment.aspx">Analyseit 3.0</A>, we've been reconsidering the <EM>unfiled reports</EM> feature.</P>
<P>What do you think?</P>
<P>Do you like the <EM>unfiled reports </EM>feature? Would you prefer Analyseit place analysis reports in the same workbook as your dataset? From there you can choose whether to keep them, or delete them. Or do you think there’s a case for Analyseit 3.0 to support both options?</P>
<P>Post a comment below to let us know your thoughts, so we can improve this feature in <A href="/blog/2008/6/analyseit30nowindevelopment.aspx">Analyseit 3.0</A>. </P>
In development
Using Analyseit
Wed, 27 Aug 2008 16:29:02 GMT

https://analyseit.com/blog/2008/8/testingtheassumptionofnormality
Testing the assumption of normality
https://analyseit.com/blog/2008/8/testingtheassumptionofnormality
<P>The most used distribution in statistical analysis is the normal distribution. Sometimes called the Gaussian distribution, after <A href="http://en.wikipedia.org/wiki/Carl_Friedrich_Gauss" target=_blank>Carl Friedrich Gauss</A>, the normal distribution is the basis of much parametric statistical analysis.</P>
<P>Parametric statistical tests often assume the sample under test is from a population with normal distribution. By making this assumption about the data, parametric tests are more powerful than their equivalent nonparametric counterparts and can detect differences with smaller sample sizes, or detect smaller differences with the same sample size. </P>
<H4>When to check sample distribution </H4>
<P>It’s vital you ensure the assumptions of a parametric test are met before use.</P>
<P>If you’re unsure of the underlying distribution of the sample, you should check it.</P>
<P>Only when you know the sample under test comes from a population with normal distribution – meaning the sample will also have normal distribution – should you consider skipping the normality check.</P>
<P>Many variables in nature naturally follow the normal distribution, for example, biological variables such as blood pressure, serum cholesterol, height and weight. You could choose to skip the normality check these in cases, though it’s always wise to check the sample distribution. </P>
<H4>How to check the sample distribution </H4>
<P>You can use a statistical test and or statistical plots to check the sample distribution is normal. Analyseit includes three statistical tests for testing normality:</P>
<UL>
<LI><A href="/userguide/Content/Standard/Summary_statistics.htm#Normality"><STRONG>KolmogorovSmirnov test</STRONG></A> <BR>An EDFtype test based on the largest vertical distance between the normal cumulative distribution function (CDF) and the sample cumulative frequency distribution (commonly called the ECDF – empirical cumulative distribution function). <BR><BR>It has poor power to detect nonnormality compared to the tests below. <A href="http://www.amazon.com/gp/redirect.html?ie=UTF8&location=http%3A%2F%2Fwww.amazon.com%2FGoodnessfittechniquesStatisticsTextbooksMonographs%2Fdp%2F0824774876&tag=analyseitsoftwar&linkCode=ur2&camp=1789&creative=9325" target=_blank>D’Agostino and Stephens</A> say the KolmogorovSmirnov test is now really only of historical interest.
<LI><A href="/userguide/Content/Standard/Summary_statistics.htm#Normality"><STRONG>AndersonDarling test</STRONG></A> <BR>An EDFtype test similar to the KolmogorovSmirnov test, except it uses the sum of the weighted squared vertical distances between the normal cumulative distribution function and the sample cumulative frequency distribution. More weight is applied at the tails, so the test is better able to detect nonnormality in the tails of the distribution.
<LI><A href="/userguide/Content/Standard/Summary_statistics.htm#Normality"><STRONG>ShapiroWilk test</STRONG> <BR></A>A regressiontype test that uses the correlation of sample order statistics (the sample values arranged in ascending order) with those of a normal distribution. <BR><BR>It’s the most powerful normality test available and is able to detect small departures from normality. <BR><BR>The only limitation is it’s not suitable for very large sample sizes. Analyseit uses the latest algorithm and supports use on samples up to 5,000 observations, but some software limits use to 2,000, or as few as 50, observations. </LI></UL>
<P>While normality tests are useful, they aren’t infallible. </P>
<P>You shouldn’t rely on a normality test to exclusively to judge normality. You should look at the <A href="/blog/2008/11/normalquantileprobabilityplots.aspx">Normal plot</A>, or <A href="/userguide/Content/Standard/Summary_statistics.htm#Histogram">Frequency histogram</A> with normal overlay, to doublecheck the distribution is roughly Normal. The plots will also tell you why a sample fails the normality test, for example due to skew, bimodality, or heavy tails. </P>
<P>Small and large samples can also cause problems for the normality tests. </P>
<P>With small sample sizes of 10 or fewer observations it’s unlikely the normality test will detect nonnormality. If you know the population distribution is normal you should still use a parametric test, as it’s more powerful, but if you’re unsure a nonparametric alternative is usually more conservative. </P>
<P>Conversely, for large samples, for example 1000 observations or more, the normality test might conclude a small deviation from normality is significant. You should look at the normal QQ plot to see if the deviation from normality really is significant. </P>
<P>Many parametric tests, such as the ttest and ANOVA, use the mean of the sample so some nonnormality can be tolerated (due to the Central Limit Theorem). How large a sample you need depends on how skewed the sample distribution is – the more skewed the data, the larger the sample size should be – so it’s not possible to give hard and fast rules. You should first check the degree of nonnormality and, only after (careful!) consideration, decide if you can safely use the test. </P>
<H4>Using Analyseit to check normality </H4>
<P><A href="/">Analyseit</A> provides the normality tests, Normal QQ plot and Frequency histogram mentioned above. All are included on the single sample summary statistics (that’s a tongue twister!) report. </P>
<P>To display detailed summary statistics, plots, and the normality test for a sample: </P>
<OL>
<LI>Choose <EM>Describe</EM> > <EM>Summary</EM> from the Analyseit toolbar.<BR><BR><IMG alt="Describe, Summary, menu option" src="/assets/Describe_Summary.png">
<LI>Select the <EM>Variable</EM> to test.
<LI>Select the normality test to use from the <EM>Normality Test</EM> dropdown selector. When you choose a normality test, Analyseit assumes you are checking normality and will show Normal QQ plot.
<LI>Tick <EM>Overlay Normal distribution</EM>, below <EM>Histogram</EM>, to show the ideal normal distribution superimposed over the histogram bars. You can then judge the histogram barheights against the ideal Normal distribution.
<LI>Click <EM>OK</EM>. </LI></OL>
<P>In the forthcoming Analyseit 3.0 we’ve also made the normality tests available separately, directly from the Describe (to be renamed Distribution) menu.</P>
<H4>How to interpret the normality test </H4>
<P>Since the normality tests included in Analyseit are all hypothesis tests, they test a null against alternative hypothesis. For each test, the null hypothesis states the sample has a normal distribution, against alternative hypothesis that it is nonnormal. </P>
<P>The pvalue tells you the probability of incorrectly rejecting the null hypothesis. </P>
<P>When it’s significant (usually when lessthan 0.10 or less than 0.05) you should reject the null hypothesis and conclude the sample is not normally distributed. </P>
<P>When it is not significant (greaterthan 0.10 or 0.05), there isn’t enough evidence to reject the null hypothesis and you can only assume the sample is normally distributed. However, as noted above, you should always doublecheck the distribution is normal using the Normal QQ plot and Frequency histogram.</P>
<P><STRONG>On a technical note:</STRONG> Since we developed Analyseit over 10 years ago, a few users have asked about the pvalues calculated by Analyseit. When calculating the pvalue, Analyseit assumes the mean and standard deviation of the population are unknown and instead estimates them from the sample. Some software packages don’t make this assumption, and go on to calculate incorrect pvalues. </P>
Using Analyseit
Thu, 07 Aug 2008 15:02:50 GMT

https://analyseit.com/blog/2008/7/theanalyseitdatasetconcept
The Analyseit "Dataset" concept
https://analyseit.com/blog/2008/7/theanalyseitdatasetconcept
<P>For new and occasional Analyseit users, datasets can sometimes seem confusing. Today we’ll explain why we devised the 'dataset' concept, a concept now copied by some other Excel addins. </P>
<H4>Why can’t I select the cells containing the data to analysed?</H4>
<P>We introduced the dataset concept so Analyseit could automatically pickup the data and variables from your Excel worksheet. As we found with <A href="/company/aboutus.aspx">Astute</A>, the Analysis Toolpak, and other Excel addins, forcing you to select cells containing the data to be analysed can be problematic:</P>
<UL>
<LI><STRONG>Selecting cells can be tedious.</STRONG> <BR>When you need to run a few statistical tests you have to select the same cells, again and again. When there are lots of cells to select you have to wait while Excel autoscrolls the offscreen cells into view so you can select them.
<LI><STRONG>Selecting cells can be errorprone.</STRONG> <BR>It’s easy to select the wrong cells, too many cells, or too few cells. You have to carefully select cells, otherwise the statistics calculated won't be valid.
<LI><STRONG>Selecting a subset of the data often means reorganising and sorting the data.</STRONG> <BR>Most Excel addins require the data be in a rectangular contiguous range of cells. You might have to filter then sort it, or maybe even rearrange it manually, to bring data together into a selectable rectangular block of cells.
<LI><STRONG>Adding or removing cases and observations isn't easy.</STRONG><BR>You have to run the statistical test again and reselect the cells again, now to include new data you’ve added, or exclude data you’ve removed. </LI></UL>
<P>Statistics software is supposed to simplify what’s already a complex and errorprone subject area. Forcing you to select ranges of cells just seemed, to us, to introduce more potential for errors. </P>
<H4>What are the benefits of the Analyseit way? </H4>
<P>In Analyseit we tried to solve all these problems. We wanted the software to do more of the work, eliminating the need to select or reorganise your data. You should be able to:</P>
<UL>
<LI><STRONG>Select variables by name, not by selecting cells.</STRONG> <BR>Statistics and charts can then refer to variables by name, instead of cryptic meaningless cell references. Which is clearer? “1way ANOVA of Yield by Fertiliser”, or “1way ANOVA of Sheet1!$A$2:$A$350, Sheet1!$C$2:C$350”?
<LI><STRONG>Add, remove, or correct observations or cases in the dataset, and then easily rerun the analysis.</STRONG> <BR>The software should automatically recognise when data is added, removed, or changed. You should be able to update any statistics and charts to reflect the new data, without running the test and selecting data again.
<LI><STRONG>Slice and dice the data, to filter and limit analysis to a subset of data.</STRONG> <BR>Excel’s builtin AutoFilter provides a quick and easy way to filter data, providing a simple way to analyse a subset of the dataset without you having to rearrange data on the worksheet.</LI></UL>
<H4>How does Analyseit recognise the data for analysis?</H4>
<P>These requirements mean Analyseit has to know exactly how your data is arranged on an Excel worksheet  which cells contain data for analysis, and which cells contains the variable names.</P>
<P>This is when the real problem with spreadsheets (or benefit, if you like) becomes apparent. Spreadsheets enforce no structure to the data you place in the cells. You can use cells on a worksheet how you like, arranging data anywhere on the sheet, horizontally or vertically, with or without labels for each row/column. </P>
<P>To keep things simple, we chose to support a few layouts commonly used to organise data in Excel. We call these “datasets” and all are immediately recognisable. Over time we've found most users tend to naturally organise their data in the way Analyseit expects it – experienced Excel users more so. For more information datasets, and how to arrange your data for Analyseit, see the <A href="/userguide/default.htm" target=_blank>online help</A>.</P>
Using Analyseit
Wed, 30 Jul 2008 12:42:04 GMT

https://analyseit.com/blog/2008/7/handbookofparametricnonparametricstatisticalprocedures
Handbook of Parametric & Nonparametric Statistical procedures
https://analyseit.com/blog/2008/7/handbookofparametricnonparametricstatisticalprocedures
<P><A href="http://www.amazon.com/gp/product/1584888148?ie=UTF8&tag=analyseitsoftwar&linkCode=as2&camp=1789&creative=9325&creativeASIN=1584888148"><IMG style="PADDINGLEFT: 15px; PADDINGBOTTOM: 8px; WIDTH: 161px; HEIGHT: 240px" alt="Book cover" src="/assets/HandbookofParametricAndNonParametricStatisticalProceduresBookCover.jpg" align=right border=0></A>A few readers have emailed to ask for more information about the book by David J. Sheskin we alluded to in the comment reply re: the <A href="/blog/2008/7/choosingthecorrectstatisticaltest.aspx#comments">Statistical test advisor</A>, last week.</P>
<P>The book is the <STRONG>Handbook of Parametric & Nonparametric Statistical procedures</STRONG>, by David J. Sheskin, ISBN: 1584888148.</P>
<P>We have the third edition of the book which runs to over 1,200 pages  a phenomenal piece of work for a single (obviously very dedicated) author. While it’s not a book you would sit down and read covertocover, it is a very readable reference guide, covering all the parametric and nonparametric statistical procedures included in Analyseit. </P>
<P>For beginners the book starts at the very beginning, introducing summary statistics such as the mean, median, then moving on to explain concepts such as measurement scales, central tendency, variability, normal distribution, hypothesis testing, parametric and nonparametric statistics. The text is concise, but is clear, easy to read, and easy to understand  ideal for anyone needing a refresher course on statistics.</P>
<P>Over 130 statistical tests for univariate and bivariate data are then covered, including ANOVA, ChiSquare, t test, MannWhitney, Wilcoxon Signed Ranks, and many more. The historical background, purpose, use, assumptions and interpretation are explained, with examples to demonstrate realworld use of the test in business, education, life and environment science applications. Again, the text is very easy to read, and the extensive use of examples really helps to demonstrate how the tests are used in practical situations.</P>
<P>For experts the book includes formulas (without resorting to proofs) and explains how to calculate each test and statistic. References to further reading are included but aren’t needed as this book covers each procedure so thoroughly.</P>
<P>Of all the books we've read, this is handsdown, the definitive guide for univariate and bivariate statistical analysis. We haven't found another book so comprehensive and complete, yet easy to read and understand. The author really does succeed in his aim: to provide a comprehensive reference for readers at all levels.</P>
<P>The fourth edition of this book was published last year, with over 500 new pages, now covering topics such as multivariate analysis, clinical trials and survival analysis.</P>
<P>You can read more about the book and buy it online at <A href="http://www.amazon.co.uk/gp/product/1584888148?ie=UTF8&tag=analyseitsoftwar&linkCode=as2&camp=1789&creative=9325&creativeASIN=1584888148">Amazon (UK)</A> and <A href="http://www.amazon.com/gp/product/1584888148?ie=UTF8&tag=analyseitsoftwar&linkCode=as2&camp=1789&creative=9325&creativeASIN=1584888148">Amazon (US)</A>.</P>
<P> </P>
<TABLE class=blogtable>
<TBODY>
<TR>
<TD><STRONG>Contents (from the 3rd edition):</STRONG> </TD>
<TD></TD></TR>
<TR>
<TD width="70%">Introduction</TD>
<TD align=right width="10%">1</TD></TR>
<TR>
<TD width="70%">Outline of Inferential Statistical Tests and Measures of Correlation/Association</TD>
<TD align=right width="10%">107</TD></TR>
<TR>
<TD width="70%">Guidelines and Decision Tables for Selecting the Appropriate Statistical Procedure</TD>
<TD align=right width="10%">113</TD></TR>
<TR>
<TD width="70%">Inferential Statistical Tests Employed with a Single Sample</TD>
<TD align=right width="10%">119</TD></TR>
<TR>
<TD width="70%">The SingleSample z Test</TD>
<TD align=right width="10%">121</TD></TR>
<TR>
<TD width="70%">The SingleSample t Test</TD>
<TD align=right width="10%">135</TD></TR>
<TR>
<TD width="70%">The SingleSample ChiSquare Test for a Population Variance</TD>
<TD align=right width="10%">161</TD></TR>
<TR>
<TD width="70%">The SingleSample Test for Evaluating Population Skewness</TD>
<TD align=right width="10%">173</TD></TR>
<TR>
<TD width="70%">The SingleSample Test for Evaluating Population Kurtosis</TD>
<TD align=right width="10%">181</TD></TR>
<TR>
<TD width="70%">The Wilcoxon SignedRanks Test</TD>
<TD align=right width="10%">189</TD></TR>
<TR>
<TD width="70%">The KolmogorovSmirnov GoodnessofFit Test for a Single Sample</TD>
<TD align=right width="10%">203</TD></TR>
<TR>
<TD width="70%">The ChiSquare GoodnessofFit Test</TD>
<TD align=right width="10%">219</TD></TR>
<TR>
<TD width="70%">The Binomial Sign Test for a Single Sample</TD>
<TD align=right width="10%">245</TD></TR>
<TR>
<TD width="70%">The SingleSample Runs Test (and Other Tests of Randomness)</TD>
<TD align=right width="10%">337</TD></TR>
<TR>
<TD width="70%">Inferential Statistical Tests Employed with Two Independent Samples (and Related Measures of Association/Correlation)</TD>
<TD align=right width="10%">373</TD></TR>
<TR>
<TD width="70%">The t Test for Two Independent Samples</TD>
<TD align=right width="10%">375</TD></TR>
<TR>
<TD width="70%">The MannWhitney U Test</TD>
<TD align=right width="10%">423</TD></TR>
<TR>
<TD width="70%">The KolmogorovSmirnov Test for Two Independent Samples</TD>
<TD align=right width="10%">453</TD></TR>
<TR>
<TD width="70%">The SiegelTukey Test for Equal Variability</TD>
<TD align=right width="10%">465</TD></TR>
<TR>
<TD width="70%">The Moses Test for Equal Variability</TD>
<TD align=right width="10%">479</TD></TR>
<TR>
<TD width="70%">The ChiSquare Test for r x c Tables (Test 16a: The ChiSquare Test for Homogeneity; Test 16b: The ChiSquare Test of Independence (employed with a single sample))</TD>
<TD align=right width="10%">493</TD></TR>
<TR>
<TD width="70%">Inferential Statistical Tests Employed with Two Dependent Samples (and Related Measures of Association/Correlation)</TD>
<TD align=right width="10%">573</TD></TR>
<TR>
<TD width="70%">The t Test for Two Dependent Samples</TD>
<TD align=right width="10%">575</TD></TR>
<TR>
<TD width="70%">The Wilcoxon MatchedPairs SignedRanks Test</TD>
<TD align=right width="10%">609</TD></TR>
<TR>
<TD width="70%">The Binomial Sign Test for Two Dependent Samples</TD>
<TD align=right width="10%">621</TD></TR>
<TR>
<TD width="70%">The McNemar Test</TD>
<TD align=right width="10%">633</TD></TR>
<TR>
<TD width="70%">Inferential Statistical Tests Employed with Two or More Independent Samples (and Related Measures of Association/Correlation)</TD>
<TD align=right width="10%">665</TD></TR>
<TR>
<TD width="70%">The SingleFactor BetweenSubjects Analysis of Variance</TD>
<TD align=right width="10%">667</TD></TR>
<TR>
<TD width="70%">The KruskalWallis OneWay Analysis of Variance by Ranks</TD>
<TD align=right width="10%">757</TD></TR>
<TR>
<TD width="70%">The Van der Waerden NormalScores Test for k Independent Samples</TD>
<TD align=right width="10%">781</TD></TR>
<TR>
<TD width="70%">Inferential Statistical Tests Employed with Two or More Dependent Samples (and Related Measures of Association/Correlation)</TD>
<TD align=right width="10%">795</TD></TR>
<TR>
<TD width="70%">The SingleFactor WithinSubjects Analysis of Variance</TD>
<TD align=right width="10%">797</TD></TR>
<TR>
<TD width="70%">The Friedman TwoWay Analysis of Variance by Ranks</TD>
<TD align=right width="10%">845</TD></TR>
<TR>
<TD width="70%">The Cochran Q Test</TD>
<TD align=right width="10%">867</TD></TR>
<TR>
<TD width="70%">Inferential Statistical Test Employed with Factorial Design (and Related Measures of Association/Correlation)</TD>
<TD align=right width="10%">885</TD></TR>
<TR>
<TD width="70%">The BetweenSubjects Factorial Analysis of Variance</TD>
<TD align=right width="10%">887</TD></TR>
<TR>
<TD width="70%">Measures of Association/Correlation</TD>
<TD align=right width="10%">943</TD></TR>
<TR>
<TD width="70%">The Pearson ProductMoment Correlation Coefficient</TD>
<TD align=right width="10%">945</TD></TR>
<TR>
<TD width="70%">Spearman's RankOrder Correlation Coefficient</TD>
<TD align=right width="10%">1061</TD></TR>
<TR>
<TD width="70%">Kendall's Tau</TD>
<TD align=right width="10%">1079</TD></TR>
<TR>
<TD width="70%">Kendall's Coefficient of Concordance</TD>
<TD align=right width="10%">1093</TD></TR>
<TR>
<TD width="70%">Goodman and Kruskal's Gamma</TD>
<TD align=right width="10%">1109</TD></TR>
<TR>
<TD width="70%">Appendix: Tables</TD>
<TD align=right width="10%">1123</TD></TR>
<TR>
<TD width="70%">Index</TD>
<TD align=right width="10%">1175</TD></TR></TBODY></TABLE>
Using Analyseit
Publications
Tue, 15 Jul 2008 14:17:20 GMT

https://analyseit.com/blog/2008/7/choosingthecorrectstatisticaltest
Choosing the correct statistical test
https://analyseit.com/blog/2008/7/choosingthecorrectstatisticaltest
<P>Many Analyseit users readily admit their statistics knowledge is a little rusty, usually because it’s 10 years or more since their last statistics course. Should I use the ttest, MannWhitney, or Wilcoxon test? The names of the tests aren’t exactly helpful, nor do they give you any clue of the assumptions that must be met to use the test.</P>
<P>That’s why we devised the <A href="/support/advisor/testadvisor.aspx" target=_blank><STRONG>Statistical Test Advisor</STRONG></A>. </P>
<P>It’s a simple interactive wizard that asks what you want to do, what data you’ve observed, checks which pretest assumptions can be met, then tells you the best statistical test to use. Using the advisor you can be confident you’re using the correct statistical test  or even use it to check if your statistics knowledge really is as rusty as you think! </P>
<P>Try it for yourself now:</P>
<P><A href="/support/advisor/testadvisor.aspx" target=_blank>/support/advisor/testadvisor.aspx</A><A href="/support/advisor/testadvisor.aspx"></A></P>
<P>Bear in mind this is only a simple prototype at the moment. Eventually the advisor will integrate into Analyseit, leading you to help and tutorials showing you how to use the recommended test and interpret the statistics. You’ll also notice the advisor recommends tests that will be new to Analyseit 3. </P>
<P>So now it’s over to you. What do you think to the test advisor? Will you find this useful? Are the questions logical? Is the terminology clear or do we need to explain the terms used? Finally, have we missed anything you think should be included? </P>
<P>Let us know what you think, good or bad. Post your comments below. </P>
In development
Tue, 08 Jul 2008 09:55:50 GMT

https://analyseit.com/blog/2008/7/youvespeltanalysewrong
You’ve spelt “analyse” wrong!
https://analyseit.com/blog/2008/7/youvespeltanalysewrong
<P><IMG alt="Map of UK" src="/assets/Europe.png" align=right>Depending on where you’re located, the way we spell “Analyseit” may intrigue you. We chose the name in 1997 as it sounded active, a direction to analyse it! – similar to many product names of the time. </P>
<P>The name has served us well and hints as to what our business and product offers.</P>
<P>At the time we didn’t think such a simple name would cause so many headaches. Before you wonder, Analyseit doesn’t mean anything offensive in other languages, but it can be spelt different ways:</P>
<P>If you’re from the US, the obvious spelling is “Analyzeit”. If English isn’t your first language, “Analiseit” or “Analizeit” seems to be the natural spelling. Then there’s the hyphen – with or without? Combined there are 16 variations.</P>
<P>The variations have sometimes caused customers to send email to the wrong company, or made finding our website difficult. We’ve also had the occasional email suggesting we can’t spell! One of the notso polite emails went like this:</P>
<BLOCKQUOTE dir=ltr style="MARGINRIGHT: 0px">
<P><EM>You’ve spelt “analyse” wrong! “Licence” is wrong too! How can I trust your company and products if you can’t even use spell check!</EM></P></BLOCKQUOTE>
<P>Ouch! The spelling is a clue to our English heritage (see <A href="http://maps.google.com/maps/ms?ie=UTF8&hl=en&msa=0&msid=109722261619505924289.000450f72fea0e754c87a&ll=33.72434,11.25&spn=130.603569,286.171875&z=2" target=_blank>our location</A>). We’re based in England in the United Kingdom in Europe, and that meant we chose to use the English spelling – analyse, licence – rather than the arguably more logical AmericanEnglish spelling – analyze, license.</P>
<P>Thankfully most search engines now suggest our website regardless of which variation you use. More importantly, we’ve acquired many of the alternativespelled domain names such as www.analyzeit.com, www.analyzeit.com, with and without the hyphen. Owning many of the variations on the domain name means we now get misaddressed email (and browsers) – and all the spam for these 810 domains!</P>
Business
Tue, 01 Jul 2008 16:08:34 GMT

https://analyseit.com/blog/2008/6/howtogethelpusinganalyseit
How to get help using Analyseit
https://analyseit.com/blog/2008/6/howtogethelpusinganalyseit
<P>Most of you know where to find the help and examples provided with Analyseit, but if not, today we’d like to explain what’s available. If you're stuck we're always happy to help, and usually respond within a few hours, but it's always faster for you to check if the help answers your question first.</P>
<H4>Getting started tutorial</H4>
<P>If you’re new to Analyseit, or want a quick refresher, the best place to start is the Getting Started tutorial. It’s completely automated, no typing is required, so all you have to do is sit back and watch. In just 10 minutes it will demonstrate how to setup a dataset, how to filter the dataset, how to run a statistical test, and how to edit, refresh, and print the reports. </P>
<P>To watch the tutorial:</P>
<OL>
<LI>Browse to <A href="/userguide/gettingstarted.aspx">/userguide/gettingstarted.aspx</A>
<LI>The tutorial will start to play. It requires <A href="http://www.adobe.com/shockwave/download/download.cgi?P1_Prod_Version=shockwaveFlash">Adobe Flash</A>.
<LI>In full it's 10 minutes, but you can jump to any section  hover over the “Jump to section” link at the topleft of the tutorial, then choose the feature you want to see. </LI></OL>
<H4>Help when you’re using Analyseit</H4>
<P>The application help provided with Analyseit is a complete reference covering all aspects of Analyseit: how to install Analyseit, start it, layout datasets, manage reports, and how to use the statistical tests. You can either browse the contents to learn about Analyseit, or use the index or search to quickly find the right topic. Index and search were only recently added to the help, in Analyseit 2.10, so if you’re not using the latest version <A href="/support/download.aspx">download it now</A>.</P>
<P>To open the help: </P>
<OL>
<LI>Start Analyseit
<LI>Click the Help button on the Analyseit toolbar (see below):<BR><BR><IMG alt="" src="/assets/HelpHighlighted640.png">
<LI>Analyseit opens the help. Click the Contents, Index, or Search tab (see below) depending on how you want to navigate the help:<BR><BR><IMG alt="" src="/assets/HelpContentsIndexSearch640.png"></LI></OL>
<H4>Help on the statistical test you’re using </H4>
<P>Most questions arise when using a statistical test  what are the assumptions for this test, how do I use the test, and how do I interpret the statistics? Help topics for every statistical test in Analyseit are included, explaining what the test does, what the assumptions are, common tasks using the test, and references to further reading on how the test implemented and the statistics calculated. </P>
<P>To get help for the test you’re using: </P>
<OL>
<LI>Start Analyseit
<LI>Choose the test you want to use from the Analyseit toolbar, or click Edit on the Analyseit toolbar if you have an Analyseit report worksheet active.
<LI>Click the Help button on the test dialog box shown (see below):<BR><BR><IMG alt="" src="/assets/TestHelpButton.png">
<LI>Analyseit opens the help topic for the test.</LI></OL>
<H4>Learning by example</H4>
<P>Sometimes it’s easier to learn from an example  actually seeing how it's done is easier than reading about it. If you’re stuck and want to see an example dataset and report that you can then use as a template, take a look at the examples. We’ve included sample datasets and reports for almost every statistical test in Analyseit.</P>
<P>To access the examples:</P>
<OL>
<LI>Start Analyseit
<LI>On the Analyseit toolbar click the Further Resources button then choose Examples (see below):<BR><BR> <IMG alt="" src="/assets/FurtherResourcesExamplesButton640.png">
<LI>Analyseit opens the examples folder, containing the example Excel workbooks. </LI></OL>
<P>In future we plan to blog about how to use specific features and statistical tests in Analyseit. Your comments on these topics will help us ensure we concentrate our time and efforts on problem areas. Post a comment now and tell us what you think of the help and, more importantly, where we can improve it. </P>
Using Analyseit
Thu, 26 Jun 2008 15:04:55 GMT

https://analyseit.com/blog/2008/6/analyseit30nowindevelopment
Analyseit 3.0 now in development
https://analyseit.com/blog/2008/6/analyseit30nowindevelopment
<P>One of the primary reasons we launched the blog is to let you know what we’re currently working on, and give you to opportunity to feedback and influence the development.</P>
<P>At the moment we're spending most of our time developing version 3.0 of the Analyseit Standard edition. The improvements are based on what you've asked us to include, and through insights gleaned from recent customer surveys. Improvements and new features will include:</P>
<UL>
<LI>Test advisor to help choose the most appropriate statistical test – handy, if your statistics knowledge is a bit rusty.
<LI>New and improved statistical tests – including logistic regression, improved ANOVA with posthoc tests, new regression fits.
<LI>Simpler dataset setup – without the need to set measurement scales before running a test.
<LI>Transform variables, and easily choose groups to analyse, when running a test.
<LI>Clearer hypotheses and automatic interpretation of pvalues.
<LI>Improved help with stepbystep instructions for each test. </LI></UL>
<P>Version 3.0 will be a major upgrade and will be released later this year. As per our upgrade policy, you'll get the upgrade free if you purchased your licence within 2years of the release date – so if you're thinking of buying a licence now you will get the upgrade free. If you don't qualify for the free upgrade you'll be able to buy the upgrade for a small cost, necessary to cover development costs. We'll announce final pricing and upgrade pricing nearer release.</P>
<P>In the next few weeks we’ll post more about the new features, including screenshots, so you get a chance to comment on individual features. We'll then incorporate any requests and suggestions into the final version before release.</P>
<P>Start now by posting a comment to let us know what you’d like to see included in Analyseit 3.0.</P>
Thu, 19 Jun 2008 13:35:26 GMT

https://analyseit.com/blog/2008/6/analyseit211released
Analyseit 2.11 released
https://analyseit.com/blog/2008/6/analyseit211released
<P>Last Friday we released the latest update to Analyseit, version 2.11. It’s a minor update providing minor fixes to issues recently reported by customers.</P>
<P>The update is available free to everyone, including users still using Analyseit version 1. </P>
<P>In fact if you’re still using Analyseit version 1 you should get the update right away. You’ll be impressed with the improvements to both the application and help, and if you’re using Excel 2007 you’ll love the slick Analyseit RibbonBar . </P>
<P>If you’re unsure which version of Analyseit you’re using, see our FAQ: <A href="/faqs/findinstalledversionofAnalyseit.aspx">How to find which version of Analyseit you’re using</A>. </P>
<H4>How do I download and install the update? </H4>
<P>Analyseit automatically checks for updates every 15 days on startup, and will tell you if an update is available to download. Firewalls can get in the way though, so if you haven’t got a notification yet, or want to download 2.11 right away, you can download at: </P>
<P><A href="/support/download.aspx">/support/download.aspx</A> </P>
<P>When you’ve downloaded the update simply install over your existing version of Analyseit. There’s no need to uninstall the old version and you won’t need your product key to reactivate Analyseit  unless you’re upgrading from Analyseit version in which case you’ll need to <A href="/support/retrieve_product_key.aspx">request a product key</A> to activate Analyseit 2.11. </P>
<H4>Where can I find details of what’s changed? </H4>
<P>We know many of you work in regulated environments and need to revalidate changes made in each Analyseit update. To see exactly what’s changed since the version you validated, so you can just revalidate the affected statistical tests, please see the change log at: </P>
<P><A href="/support/changes.aspx">/support/changes.aspx</A> </P>
<H4>Subscribe to get the latest updates </H4>
<P>In future we’ll announce all major and minor updates on the blog. If you’re not already a subscriber, <A href="/blog/2008/6/theanalyseitblogislive.aspx">subscribe to the blog</A> then you’ll be notified of new versions as soon as they’re released. </P>
Releases
Mon, 16 Jun 2008 12:20:22 GMT

https://analyseit.com/blog/2008/6/theanalyseitblogislive
The Analyseit blog is live!
https://analyseit.com/blog/2008/6/theanalyseitblogislive
<P>Thanks for stoppingby to read the inaugural post to the Analyseit blog. </P>
<H4>Why launch a blog?</H4>
<P><IMG style="WIDTH: 128px; HEIGHT: 128px" height=30 hspace=4 src="/assets/rss128x128.png" width=28 align=right>We’ve launched the <A title="What's a blog?" href="http://en.wikipedia.org/wiki/Blog">blog</A> to keep you uptodate on what’s new at Analyseit, what we’re working on, and to notify you when updates and new products are released. We’ll also blog about using Analyseit, statistical analysis, and interpretation. </P>
<P>The best part about the blog is it’s a conversation  a conversation between you, our visitors and customers, and us. After reading a post you can add your thoughts, suggestions, and comments. We’ll reply when necessary or other visitors can post their opinions to ignite the debate. Click the links at the end of each post to read comments or add your own. </P>
<H4>How can I get the latest posts?</H4>
<P>To stay uptodate we recommend you subscribe so you get the latest posts as they're published, without having to check the website. You can subscribe in two ways</P>
<UL>
<LI><A href="/blog/feed.rss">Subscribe to the RSS feed</A><BR>New blog posts are delivered to your RSS reader as soon as they are published. Internet Explorer, Mozilla Firefox, Microsoft Outlook and Mozilla Thunderbird all let you subscribe to blogs via RSS feeds.
<LI><A href="/blog/subscribe.aspx">Subscribe by email</A><BR>New blog posts are sent by email directly to your inbox, just like a newsletter. You can unsubscribe at anytime.</LI></UL>
<P>That’s about it for the first blog post. Please remember to subscribe, and please post a comment now to introduce yourself or let us know how you want the blog and Analyseit to develop in future. </P>
Business
Wed, 11 Jun 2008 16:23:19 GMT