동영상이 없는 텍스트 강의 자료입니다. 사후분석(post hoc)
FWER
대표적인 사후분석 방법유의수준을 보정하여 FWER을 0.05로 고정시킴
사후분석을 위한 준비분산분석을 실시: m = aov(weight ~ group, data = PlantGrowth)
피셔의 LSD
PostHocTest(m, method='lsd') Posthoc multiple comparisons of means : Fisher LSD 95% family-wise confidence level $group diff lwr.ci upr.ci pval trt1-ctrl -0.371 -0.94301261 0.2010126 0.1944 trt2-ctrl 0.494 -0.07801261 1.0660126 0.0877 . trt2-trt1 0.865 0.29298739 1.4370126 0.0045 ** --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
봉페로니 교정
PostHocTest(m, method='bonferroni') Posthoc multiple comparisons of means : Bonferroni 95% family-wise confidence level $group diff lwr.ci upr.ci pval trt1-ctrl -0.371 -1.0825786 0.3405786 0.5832 trt2-ctrl 0.494 -0.2175786 1.2055786 0.2630 trt2-trt1 0.865 0.1534214 1.5765786 0.0134 * --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
투키의 HSD
PostHocTest(m, method='hsd') Posthoc multiple comparisons of means : Tukey HSD 95% family-wise confidence level $group diff lwr.ci upr.ci pval trt1-ctrl -0.371 -1.0622161 0.3202161 0.3909 trt2-ctrl 0.494 -0.1972161 1.1852161 0.1980 trt2-trt1 0.865 0.1737839 1.5562161 0.0120 * --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
셰페의 방법
PostHocTest(m, method='scheffe') Posthoc multiple comparisons of means : Scheffe Test 95% family-wise confidence level $group diff lwr.ci upr.ci pval trt1-ctrl -0.371 -1.0930531 0.3510531 0.4241 trt2-ctrl 0.494 -0.2280531 1.2160531 0.2265 trt2-trt1 0.865 0.1429469 1.5870531 0.0163 * --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 여기서도 @목차 Tukey
Scheffe
Games-Howell
Bonferroni
참고 자료 |