# Koji Yatani's Course Webpage

### Site Tools

hcistats:ttest_jp

# Differences

This shows you the differences between two versions of the page.

 hcistats:ttest_jp [2017/03/08 01:47]Koji Yatani [Effect size for an unpaired t test] hcistats:ttest_jp [2017/03/08 01:52] (current)Koji Yatani [A paired t test] Both sides previous revision Previous revision 2017/03/08 01:52 Koji Yatani [A paired t test] 2017/03/08 01:47 Koji Yatani [Effect size for an unpaired t test] 2017/03/08 01:45 Koji Yatani [Effect size for a paired t test] 2017/03/07 06:57 Koji Yatani [Effect size] 2017/03/07 06:50 Koji Yatani [対応があるかないか] 2017/03/07 06:30 Koji Yatani created 2017/03/08 01:52 Koji Yatani [A paired t test] 2017/03/08 01:47 Koji Yatani [Effect size for an unpaired t test] 2017/03/08 01:45 Koji Yatani [Effect size for a paired t test] 2017/03/07 06:57 Koji Yatani [Effect size] 2017/03/07 06:50 Koji Yatani [対応があるかないか] 2017/03/07 06:30 Koji Yatani created Line 65: Line 65: ---- ---- - =====A paired ​t test===== + =====対応のあるt検定===== - You should use a paired ​t test if you do a within-subject design. What a paired ​t test does is to take differences between data in the two groups, and see whether the distribution of the differences is too different from the t distribution. Because it uses the differences between the groups, ​**a paired ​t test does not assume the variances of the population of the two groups are equal**. But it still assumes the normality. The null hypothesis is there is no significant difference in the means between the two groups. If the p value is less than 0.05, you reject the null hypothesis, and say that you find a significant difference. + 対応のあるt検定は被験者内要因があるときに使います．対応のあるt検定がやっていることをおおざっぱに言うと，2つのグループの差を取って，その差の分布がt分布とどれくらい違っているかを計算しています．この「差を取る」という作業があるため，**対応のあるt検定では2つのグループの母集団の分散が同じであるという仮定は必要としません．**しかし，正規性は必要です．帰無仮説として2つのグループに差がない(つまり，差の平均が0である)こととして，検定を行います． ---- ----