CLIMR_power_report.md

Construal Level International Multilab Replication (CLIMR) Project: Power Analysis Report

CLIMR Team 2023-11-24

THESE RESULTS ARE SIMULATED. DO NOT INTERPRET.

Power Figure

knitr::include_graphics("./figures/climr_power_figure.png")

Power Analyses for Each Replication

Liberman & Trope (1998, Study 1)

What sample sizes are needed for 80% and 95% power to detect the replication effect?

power_80_n_temporal

## [1] 162.3499

power_95_n_temporal

## [1] 267.5077

What proportion of previous experiments had at least 80% power to detect effects the size of the replication effect?

smaller_prop_temporal

## [1] 0.14

What is the median power for the replication effect that the previous experiments samples provided?

median_power_temporal

## [1] 0.4408179

Fujita et al. (2006, Study 1)

What sample sizes are needed for 80% and 95% power to detect the replication effect?

power_80_n_spatial

## [1] 206.4215

power_95_n_spatial

## [1] 340.4778

What proportion of previous experiments had at least 80% power to detect effects the size of the replication effect?

smaller_prop_spatial

## [1] 0.08

What is the median power for the replication effect that the previous experiments samples provided?

median_power_spatial

## [1] 0.3609655

Social Distance (Paradigmatic Replication)

What sample sizes are needed for 80% and 95% power to detect the replication effect?

power_80_n_social

## [1] 189.1534

power_95_n_social

## [1] 311.887

What proportion of previous experiments had at least 80% power to detect effects the size of the replication effect?

smaller_prop_social

## [1] 0.1

What is the median power for the replication effect that the previous experiments samples provided?

median_power_social

## [1] 0.388382

Likelihood Distance (Paradigmatic Replication)

What sample sizes are needed for 80% and 95% power to detect the replication effect?

power_80_n_likelihood

## [1] 200.6022

power_95_n_likelihood

## [1] 330.8429

What proportion of effects for which previous experiments had 80% power is the replication effect smaller than?

smaller_prop_likelihood

## [1] 0.08

What is the median power for the replication effect that the previous experiments samples provided?

median_power_likelihood

## [1] 0.3697343