PhD research
Rerun your analyses across every reasonable model choice.
Antinuous maps the papers around your question, reproduces published results and reruns your analyses across every model choice. Each result is code you can check line by line.
Use cases
Plan how many participants you need
You have to read how similar studies turned out before you can pick a sample size.
You ask
Find published studies of mindfulness training on test anxiety in undergraduates, pull their reported effect sizes, and run a power analysis for a two-group design.
What it does
- Antinuous searches papers on the topic and reads the full text of the relevant ones.
- It records each study's design, sample size and effect size in a sheet with a citation per value.
- It runs a power analysis in code across a range of plausible effect sizes.
- It charts statistical power against sample size so you can see the tradeoff.
You get
A cited sheet of prior effect sizes and a power chart for preregistration.
Chance of detecting the effect (%)Participants per group
Small effect (d = 0.2)
Medium effect (d = 0.5)
Large effect (d = 0.8)
Power of a two-sided, two-group t-test at a 5% significance level, by effect size (Cohen's d).
Who it's for
PhD studentsMap the literature around your question and find open gaps.
Master's studentsPlan how many participants your study needs.
PostdocsReproduce published results and list where they differ.
Research assistantsRun and rerun analyses in code you can check.
Works with
PubMed
arXiv
GitHub
Gmail
Google Calendar
Google Slides