We are receiving a lot of questions about relevant analyses in the Analyse-it Method Validation edition to help in evaluating new diagnostic tests in the fight against COVID-19. Below are some quick links that will help, but contact us if you have questions - we are working as normal.
Also see our latest blog post: Sensitivity/Specificity and The Importance of Predictive Values for a COVID-19 test
Missing values occur when no data is recorded for an observation; you intended to make an observation but did not. Missing data are a common occurrence and can have a significant effect on the statistical analysis.
Missing values arise due to many reasons. For example, a subject dropped out of the study; a machine fault occurred that prevented a measurement been taken; a subject did not answer a question in a survey; or a researcher made a mistake recording an observation.
A missing value is indicated by an empty cell or a . (full stop) or * (asterisk) in the cell. These values always indicate a missing value, regardless of measurement scale of the variable. Similarly, cell values that are not valid for the measurement scale of the variable are treated as missing. For example, a cell containing text or a #N/A error value is treated as missing for a quantitative measurement scale variable, since neither are valid numeric values.
Some applications, especially older statistical software, use special values such as 99999 to indicate missing values. If you import such data, you should use the Excel Find & Replace command to replace the values with a recognized missing indicator.
In the presence of missing values you often still want to make statistical inferences. Some analyses support techniques such as deletion, imputation, and interpolation to allow the analysis to cope with missing values. However, it is still important before performing an analysis that you understand why the data is missing. When the reason for the missing data is not completely random, the study may be biased, and the statistics may be misleading.