Package: testflow 1.0.2

Introduction | How test selection works | Descriptive statistics | Assumption checks | One numerical variable | Two independent groups | Paired measurements | More than two groups | Post-hoc comparisons | Factorial designs | Repeated measurements (numeric) | Categorical outcomes | Association between two categorical variables | One-sample proportion | Multinomial goodness of fit | Paired categorical measurements | Repeated categorical measurements (3+ time points) | Correlation | Regression | Linear regression | Logistic regression | Survival analysis | Log-rank test | Cox proportional hazards regression | Diagnostic test accuracy and agreement | Diagnostic accuracy | ROC / AUC | Inter-rater agreement (Cohen's kappa) | Intraclass correlation coefficient | Outlier screening | Effect-size reference | Magnitude labels | Sample-size planning | Continuous endpoints | Binary endpoints | Survival endpoints | Ordinal endpoints | Bioequivalence | Precision-based sample size | Cluster-randomized adjustment and dropout | Summary of supported settings | References

Purpose | Continuous endpoints | Parallel superiority | Paired superiority | Repeated measures (3+ time points) | Non-inferiority and equivalence | Binary endpoints | Paired (discordant-pairs) superiority | Survival endpoints | Required events | Converting events to a total sample size | Uniform accrual | Ordinal endpoints | Bioequivalence | Precision-based sample size | Rare-prevalence planning: Wilson and exact confidence intervals | Unequal allocation for two proportions | Cluster-randomized adjustment | Dropout adjustment and reporting | Summary of supported settings | Reference

Purpose | sample_size_continuous() | design = "parallel" | objective = "superiority" | objective = "noninferiority" | objective = "equivalence" | design = "paired" | design = "repeated" | sample_size_binary() | objective = "superiority", method = "pooled" | objective = "superiority", method = "anticipated" | design = "paired" / design = "repeated" (n_time = 2) | sample_size_survival() | objective = "superiority", method = "exponential" | objective = "superiority", method = "ph_only" | Converting events to a total sample size | Uniform accrual | sample_size_ordinal() | sample_size_bioequivalence() | method = "iterative_tost" (default) | method = "normal_approx" | sample_size_precision() | sample_size_cluster_adjust() | sample_size() dispatcher | sample_size_adjust_dropout() | Methods on sample_size objects | References

Purpose | What "validated" means here | Two defects found and fixed | Bioequivalence sample size: iterative_tost was not exact | Cohen's kappa: the hypothesis test used the wrong standard error | Everything else checked, with nothing found | What this does and does not guarantee

Getting started