Registry · Priors

construct.helping_behavior predicts construct.counterproductive_work_behaviours

normal · uninformative · 1 studies · N = 50

Distribution

Uninformative prior. This prior is uninformative — too thin to dominate small-N posteriors. Treat as a placeholder until more evidence lands.

Intervals

Confidence interval (95%) — uncertainty about the mean ρ

[-0.51, -0.10]

Credibility interval (95%) — distribution of the true effect across settings (the Bayesian prior)

— needs ≥2 studies to estimate (k = 1)

I² (heterogeneity) — share of total variance from between-study differences

— needs ≥2 studies to estimate

k<2 — between-study heterogeneity not estimable; credibility interval / generalization not assessed

Evidence provenance

published ρ=-0.32 (k=1, replication: single); no primary-deployment evidence yet

• replication: single
• aging literature (freshness=0.37)

No primary-deployment evidence yet — this prior rests on published literature alone. As anonymized, aggregated effect sizes from real deployments are contributed, they appear here as a distinct, publication-bias-free source, fused with the literature into a posterior estimate.

Code

target += normal_lpdf(beta | -0.331647, 0.119523);

beta = pm.Normal("beta", mu=-0.331647, sigma=0.119523)

brms::prior(normal(-0.331647, 0.119523), class = "b")

# base R sample
rnorm(N, mean = -0.331647, sd = 0.119523)

np.random.normal(loc=-0.331647, scale=0.119523, size=N)

Parameters

Synthesis

Method

single_study

Informativeness

uninformative

Replication status

single

k_studies

1

n_total

50

Last updated

2026-05-29T15:21:31.824Z

Quality distribution

Contributing effect sizes

• effect.4f4ef62f58369ab0

Effect-size detail pages land with a later sub-ticket; for now, ids link to the filtered list. Browse all rows via /registry/effects.