Descriptive research and investigative research: accounting for 10% of the total. Except for a few cases, the sample size of investigative research should not be less than 100.

There should be at least 30 satisfactory samples for related research and comparative research.

Experimental research: research with strict condition control, such as psychological experiment, each group 15 people; The educational experiment with relaxed conditions is preferably a natural teaching class with no less than 30 students.

It cannot be absolutely understood as "the larger the population, the larger the sample size". You can't sample absolutely in proportion.