sajor2000/methodsmith / plugins/methodsmith/skills/method-power
method-power
VerifiedCalculates sample size and power for clinical, survival, cluster, and prediction-model studies, with a script, sweep, and SAP text. Use when the user asks for power or sample size.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 11.9 KB |
| references/evidence-access.md | markdown | 2.2 KB |
| references/formula-cheatsheet.md | markdown | 8.2 KB |
| references/sensitivity-sweep-template.R | other | 2.3 KB |
| references/sensitivity-sweep-template.py | script | 2.3 KB |
| scripts/pmsampsize_runner.R | other | 6.3 KB |
references/sensitivity-sweep-template.R
View on GitHub# Sensitivity sweep template -- two-sample t example
# Edit the params block; rest is fixed.
# Never install from a template: a silent install.packages() can break a locked renv.
for (pkg in c("pwr", "ggplot2")) {
if (!requireNamespace(pkg, quietly = TRUE)) {
stop(sprintf("R package '%s' is not installed; run install.packages('%s') with the user's approval, then re-run", pkg, pkg))
}
}
library(pwr)
library(ggplot2)
params <- list(
design = "two-sample-t",
effect_point = 0.40, # example placeholder, not a literature value: replace with the method-meta-analysis anchor or a PMID-cited study
effect_grid = seq(0.20, 0.60, by = 0.05),
alpha = 0.05,
power_target = 0.80,
sided = "two",
dropout = 0.15 # assumption: replace with the observed loss-to-follow-up rate
)
point <- pwr.t.test(d = params$effect_point, sig.level = params$alpha,
power = params$power_target, type = "two.sample",
alternative = ifelse(params$sided == "two", "two.sided", "greater"))
n_per_arm_point <- ceiling(point$n)
n_total_point <- 2 * n_per_arm_point
n_with_dropout <- ceiling(n_total_point / (1 - params$dropout))
cat(sprintf("Point estimate: n per arm = %d, total = %d, with %.0f%% dropout = %d\n",
n_per_arm_point, n_total_point, 100 * params$dropout, n_with_dropout))
sweep <- do.call(rbind, lapply(params$effect_grid, function(d) {
res <- pwr.t.test(d = d, sig.level = params$alpha, power = params$power_target,
type = "two.sample",
alternative = ifelse(params$sided == "two", "two.sided", "greater"))
data.frame(effect_size = d, n_per_arm = ceiling(res$n), n_total = 2 * ceiling(res$n))
}))
out_stem <- file.path("analysis/power", paste0(params$design, "-", format(Sys.Date(), "%Y-%m-%d")))
dir.create(dirname(out_stem), recursive = TRUE, showWarnings = FALSE)
write.csv(sweep, paste0(out_stem, "-sweep.csv"), row.names = FALSE)
p <- ggplot(sweep, aes(effect_size, n_total)) +
geom_line() + geom_point() +
labs(x = "Effect size (Cohen's d)", y = "Total N",
title = "Required N as a function of effect size",
subtitle = sprintf("alpha=%.2f, power=%.2f, %s-sided",
params$alpha, params$power_target, params$sided)) +
theme_minimal()
ggsave(paste0(out_stem, ".png"), p, width = 7, height = 5)
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |