Stats Tests (code + APA style) Flashcards
Independent / unrelated t-test
between subjects
interval / ratio data
normal distribution
data must have homogeneity of variance
PARAMETRIC
Code for independent t-test
t.test(VN~VN)
APA style: independent t-test
t(df) = t-test value, p-value, cohen’s d, confidence interval
Paired / Related t-test
within subjects design
interval / ratio data
normal distribution
PARAMETRIC
Code for paired samples t-test
t.test(VN, VN, paired = TRUE)
APA style: paired samples t-test
(mean diff= x) t(df) = t-test value, p-value, cohen’s d, confidence interval
Mann Whitney U
between subjects
ordinal data
or interval / ratio that’s skewed
NON-PARAMETRIC
Code for Mann Whitney U
wilcox.test(VN~VN)
if it cannot compute exact p-value with ties use code:
wilcox.test(VN~VN, exact=F)
APA style: Mann Whitney U
Mdn = x.xx, IQR = x.x, W = x.xx, N = x.xx, p= x.xx, A = x.xx
Wilcoxon Signed Rank test
within subjects design
ordinal data
or interval / ratio that’s skewed
NON-PARAMETRIC
Code for Wilcoxon Signed Rank
wilcox.test(VN, VN, paired =T)
APA style: Wilcoxon Signed Rank
(Mdn= x.xx, IQR= x.xx), V= x.xx, N=x.xx, p= x.xx, A= x.xx
Friedman’s test
Within subjects design
3+ conditions
ordinal
or interval / ratio that’s skewed
NON-PARAMETRIC
Code for Friedman’s test
friedman.test(VN~VN |ID)
APA style: Friedman test
X2(df) = Friedman chi-square value, p= x.xx, W= x.xx
Kruskal-Wallis test
between subjects design
3+ conditions
ordinal
or interval / ratio that’s skewed
NON-PARAMETRIC
Code for Kruskal-Wallis test
kruskal.test(DV~IV)
APA style: Kruskal-Wallis test
X2(df) = kruskal-wallis value, p= x.xx, n2=1 N² x.xx (mdn= x, IQR= x)
Code for Post hoc testing
pairwise.wilcox.test(VN, VN, paired=T, exact=F, p.adjust.method = “none” or “bonferroni”)
