วิธีการเปรียบเทียบความแตกต่างค่าเฉลี่ยรายคู่หลังการวิเคราะห์ความแปรปรวน Post-hoc Multiple Comparison Tests

Abstract

In testing the difference of three or more population means by using the analysis of variance when the
null hypothesis is rejected at a predetermine alpha level, called the test result is significant at the a level, the
researcher have to test which population means are different from the rest. This method is called the posthoc
multiple comparison tests, which can help the researcher to make conclusion about the greatest
population mean. The researcher must select the appropriate method which can control the overall type I
alpha rate equal to the predetermined alpha level and also reserve the power of the test. This article presents
various methods of the post-hoc multiple comparison tests for the researcher to use appropriately in reporting
the research results.