The easiest way to check data is to look at the frequencies for each variable and graph the data using an appropriate technique for the scale of measurement. A "wild" value on a frequency distribution can tell you that data were entered incorrectly. Outliers on a box plot or histogram can tell you that you need to check extreme scores for incorrect entry, a problem with the study, or an unusual case. Data entry errors are easily corrected. Handling outliers or unusual cases may require eliminating data or applying transformations. If you violate test assumptions you might transform your data or choose a nonparametric statistic to test your hypothesis.

When students begin to conduct their own research, they often are uncertain about how to handle data problems. This is the perfect time to consult -- either your statistics text, your teaching assistant, or your course instructor.