Skill guide: /experiment-design — from a hypothesis to a defensible study

experiment-design turns a hypothesis into a full study design: participants, measures, protocol, constraints and sample plan — each choice grounded in how related studies did it.

How to call it

text

/experiment-design Design a two-arm randomized experiment testing whether a 4-week HRV biofeedback program lowers perceived stress (PSS-10) in university students compared with an active control. Include participants, measures, protocol flow, and a power analysis.

What happens

Asks what it cannot infer — control condition, delivery and monitoring, secondary measures, screening — with chips drawn from the literature. "No preference" and "Skip the rest" are always available.
Surveys how comparable studies designed theirs and cites the ones it borrows from.
Writes the design report and draws the figures.

The scoping survey: five questions with chips from the literature.

What you get

The design document: goal · framework · participants · measures · constraints · rationale and source per choice.
A protocol flow diagram (editable figure), an expected-results chart and a power curve, plus a summary of the reference studies' designs.

The protocol flow as an editable figure.

The power curve with the sample size that reaches .80 marked.

Tips

Give the effect size you expect (or the reference study) and the power curve is drawn for it.
On Deep effort the design is reviewed and revised once before it is delivered.