Whether the scale of 84 samples of graduation thesis is appropriate depends on the number of remaining samples with dummy variable of 0, whether effective methods are used to make full use of the information of samples, whether the data of other control variables are sufficient, and so on. Finally, the remaining effective sample size may be too small and inappropriate.

The sample size of graduation thesis is about 300~500. If a questionnaire survey is involved in the graduation thesis, there must be a sample for investigation, and the sample size should not be too small. The sample size is too small to explain the problem, so the basic sample size should be controlled at around 300~500. Only in this way can it be the support of the data in the paper and can it be passed in the evaluation process.