Generate formatted results and effect sizes for manuscripts

After fitting models, use summaryh() in place of summary() to get APA (American Psychological Association) formatted output that also includes effect size estimates for each effect (r effect size). Currently supports lm, glm, aov, anova, lmer, lme, t-test, chisq.test, and cor.test. Unfortunately, this function won't write your entire results section for you (yet).

summaryh(model, decimal = 2, showTable = FALSE, tbl_es = FALSE, es = 'r', ...)

Arguments

model	a fitted model
decimal	round output to decimal places
showTable	show results in table format (returns list)
tbl_es	show other effect sizes computed using es function
es	effect size measure to report (for now, it's always r but options will be introduced later)
...	further arguments passed to or from other methods

Value

A datatable or a list of datatables (if showTable = TRUE or tbl_es = TRUE)

Note

Cohen's d: 0.20 (small), 0.50 (medium), .80 (large) (Cohen, 1992)
correlation r: .10 (small), .30 (medium), .50 (large)
R-squared: R2: .02 (small), .13 (medium), .26 (large)

Author

Hause Lin

Examples

summaryh(lm(mpg ~ qsec, mtcars))
#>           term                                                   results
#> 1: (Intercept) b = −5.11, SE = 10.03, t(30) = −0.51, p = .614, r = −0.09
#> 2:        qsec     b = 1.41, SE = 0.56, t(30) = 2.53, p = .017, r = 0.42
summaryh(aov(mpg ~ gear, mtcars))
#> For one-way between subjects designs, partial eta squared is equvilant to eta squared.
#> Returning eta squared.
#>    term                             results
#> 1: gear F(1, 30) = 9.00, p = .005, r = 0.48
summaryh(cor.test(mtcars$mpg, mtcars$gear), tbl_es = TRUE)
#>                          term    d    r   R2    f oddsratio logoddsratio  auc
#> 1: mtcars$mpg and mtcars$gear 1.09 0.48 0.23 0.55      7.28         1.98 0.78
#>    fishersz
#> 1:     0.52