Getting started with testflow

library(testflow)
cardio <- make_cardio_data()
x <- test_two_groups(sbp_3m ~ sex, data = cardio)
x
#> Statistical test workflow
#> 
#> Outcome: sbp_3m
#> Group: sex
#> Design: two independent groups
#> 
#> Assumptions
#> * Independence of observations: assumed: Assumed from study design.
#> * Normality: sbp_3m (female): acceptable: Approximate normality looks reasonable. (method=Shapiro-Wilk; statistic=0.99; p=0.913)
#> * Normality: sbp_3m (male): acceptable: Approximate normality looks reasonable. (method=Shapiro-Wilk; statistic=0.98; p=0.233)
#> * Variance homogeneity: acceptable: Variance homogeneity looks reasonable. (method=Levene test; statistic=1.57)
#> * Extreme outliers: warning: 4 potential outlier(s) flagged by IQR. (IQR rule, n = 4)
#> * Variance ratio check: acceptable: Variance ratio looks reasonable. (statistic=1.27)
#> 
#> Recommended test
#> Student independent t-test
#> 
#> Result
#> H0: the population mean or location of sbp_3m is equal across levels of sex.
#> statistic = -1.91, df = 178.00, p = 0.058, 95% CI [-11.22, 0.18]
#> 
#> Effect size
#> Cohen's d: -0.29, small
#> 
#> Report
#> The two independent groups workflow for sbp_3m did not show a statistically significant result using Student independent t-test, statistic = -1.91, df = 178.00, p = 0.058. The 95% confidence interval was [-11.22, 0.18]. The effect size was small (Cohen's d = -0.29). H0: the population mean or location of sbp_3m is equal across levels of sex.
report(x)
#> [1] "The two independent groups workflow for sbp_3m did not show a statistically significant result using Student independent t-test, statistic = -1.91, df = 178.00, p = 0.058. The 95% confidence interval was [-11.22, 0.18]. The effect size was small (Cohen's d = -0.29). H0: the population mean or location of sbp_3m is equal across levels of sex."
plot(x)