I have 79 total observations in my dataset, spread over four experimental conditions (18, 20, 20, and 21 observations, respectively). I ran the data/split file command so I could test normality distributions in each of the four groups, and then ran the analyze/descriptive statistics/explore command to get the plots. At the top of the output it shows the correct total and the correct amount of observations in each group. It also confirms there are no missing datapoints.
But instead of seeing 18, 20, or 21 datapoints in the Q-Q plots as expected, in some cases there are only ten or so. Any guidance about why this is happening and how to fix it would be greatly appreciated!
------------------------------
Jonathan Graffeo
------------------------------