Registry · Priors

construct.training predicts construct.task_performance

normal · informative · 3 studies · N = 17,363

Distribution

normal prior
Prior PDF · normalnormal distribution. normal prior. 95% CI ≈ [0.349, 0.701]; mean ≈ 0.525.0.1660.3460.5250.7050.884x0density

mean ≈ 0.525 · 95% CI ≈ [0.349, 0.701]

Intervals

Confidence interval (95%) — uncertainty about the mean ρ
[0.34, 0.61]
Credibility interval (95%) — distribution of the true effect across settings (the Bayesian prior)
[0.30, 0.63]
I² (heterogeneity) — share of total variance from between-study differences
54%

The true effect varies across settings — moderators likely matter.

SD_ρ>0 — true effect varies across settings; likely moderated (observed-score scale until artifact correction, PRN-058)

Evidence provenance

published ρ=0.48 (k=3, replication: meta-analytic); no primary-deployment evidence yet

Sourceρ (r)Scope
Published literature0.48k = 3 · N = 17,363

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

Drop this prior straight into your model. Snippets generated from the synthesized distribution + parameters.

target += normal_lpdf(beta | 0.525185, 0.0897184);
beta = pm.Normal("beta", mu=0.525185, sigma=0.0897184)
brms::prior(normal(0.525185, 0.0897184), class = "b")
# base R sample
rnorm(N, mean = 0.525185, sd = 0.0897184)
np.random.normal(loc=0.525185, scale=0.0897184, size=N)

Parameters

FamilyParameters
normalI2 = 0.5399, mu = 0.5252, sigma = 0.08972, k_studies = 3.000, tau_squared = 0.01173

Synthesis

Method
random_effects_meta
Informativeness
informative
Replication status
meta-analytic
k_studies
3
n_total
17,363
Last updated
2026-05-31T01:07:48.894Z

Quality distribution

GradeCount
A2
B1
C0
D0

Contributing effect sizes

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