ataset: meta-regression_joined_data_heterogeneity_in_replication_projects.csv

Domain: Meta-Science
Source: Meta-regression study on heterogeneity in replication projects
Shape: 86 rows × 60 columns
Missing values: Partial — pm_belief (68 missing), endprice, transactions, trading_volume (39 missing each), aggregated and lab_id (fully empty)

Columns:

study — Name/ID of the original study
project.x — Replication project name (e.g. Psychology)
ro, rr — Correlation coefficients for original and replication studies
fiso, fisr — Fisher z-transformed effect sizes for original and replication
se_fiso, se_fisr — Standard errors of the Fisher z scores
po, pr — P-values for original and replication
po1, pr1 — Alternative p-value measures
pm_belief — Prediction market belief score
nr, no — Sample sizes for replication and original
authors.o — Authors of the original study
pub_year — Publication year of original study
journal — Journal of publication
discipline — Scientific discipline
length — Length of original paper
citations — Citation count
power.o, power.r — Statistical power of original and replication
power_planned.r — Planned statistical power of replication
experiment_country.o/.r — Country where experiment was conducted (original/replication)
experiment_language.o/.r — Language of experiment (original/replication)
online.o/.r — Whether experiment was online (0/1)
compensation.o/.r — Participant compensation type
subjects.o/.r — Subject population type
endprice, transactions, trading_volume — Prediction market variables
replicated — Whether the study was successfully replicated (0/1)
relative_es — Relative effect size (replication vs original)
n_authors.o/.r — Number of authors (original/replication)
author_citations_avg/max.o/.r — Author citation metrics
authors_male.o/.r — Proportion of male authors
seniority.o/.r — Author seniority level
same_country, same_language, same_online, same_subjects — Whether original and replication matched on these dimensions (0/1)
us_lab.o/.r — Whether the lab was US-based (0/1)
es_80power — Effect size needed for 80% power


"Coffee doesn't ask questions, coffee understands."