Analyze research questions to recommend experimental, quasi-experimental, or survey designs.
npx skills add https://github.com/HosungYou/Diverga --skill c1
Please help me install this Agent Skill. Skill: c1 Source: https://github.com/HosungYou/Diverga/tree/main/skills/c1 Command: npx skills add https://github.com/HosungYou/Diverga --skill c1
This Skill helps researchers select the most appropriate and rigorous quantitative research designs, moving beyond common or obvious choices to optimize for validity, feasibility, and research goals.
Use the c1 skill to propose quantitative research designs for a study investigating the impact of remote work on employee productivity, assuming causal inference is highly needed and random assignment is feasible.
High-intent search queries and answers about installing and using this skill.
A power analysis for experimental design calculates the minimum sample size required to detect a statistically significant effect. It uses your expected effect size, significance level, and desired power to ensure your study yields valid and reliable results.
Yes, you can use survey design when random assignment is not feasible, though it limits causal inference claims. The Skill recommends quasi-experimental alternatives like regression discontinuity or rigorous sampling strategies to control threats to internal and external validity.
Common threats to validity in experimental and quasi-experimental designs include selection bias, history, and maturation. Identifying these threats allows researchers to implement specific controls and choose optimal designs that protect the integrity of causal inferences.
Calculating sample size for a quantitative study requires a power analysis based on your expected effect size and variance. The Skill assists with these calculations and recommends optimal sampling strategies tailored to your research design and constraints.
You should diverge from a simple pretest-posttest design when facing complex research questions or high threats to internal validity. The Skill integrates advanced principles to recommend factorial or quasi-experimental designs that better isolate treatment effects.
zivtechOperationalize research questions into PICO frameworks and causal DAGs.
ai-analyst-labEstimate treatment effects from observational data with assumption checks and sensitivity analysis.
ClawBioSearch PubMed and preprint servers, summarize literature, and build citation graphs.
NeuroAIHubPerform bootstrap-based SEM mediation analysis of creative self-efficacy in R with lavaan.
