Sample Size Seems Too Small

From Q
Jump to navigation Jump to search

Common causes of this problem are:

  • The category with the small sample size is a combination of different categories (e.g., has been created by merging), and, these categories have different sample sizes (as shown by Statistics - Cells > Base n). In this situation, Q computes any statistics on merged cells (including the NET) using only the observations with no missing data. To fix this problem, modify the Missing Data settings in Value Attributes. See also How to Interpret Column n with Missing Data.
  • The sample size truly is small.
  • The Weight has been computed erroneously, containing either missing values or 0s (to see this, select it in the Blue drop-down menu).
  • The settings in the Value Attributes are not correct.
  • There is a problem with the data.
  • Back-coding or merging variables, where the variables have different base sizes. Where this is the case, the best solution is usually to right-click on a category, select Values and un-check the Missing Data selection. And, if this changes the overall Base n to an inappropriate value: select the cells showing the value of the NET, right-click and create a Filter.

See also

Column n Is Too Small