What question did this study set out to answer?

Assess the reliability of multilevel (hierarchical) linear and logit models when estimating country-level effects in multi-country survey datasets.

May 8, 2015European Sociological ReviewOpen Access

Multilevel Modelling of Country Effects: A Cautionary Tale

View Full Paper

Ask AI

Bookmark

Authors

MBMark L. BryanUniversity of Sheffield SJStephen P. JenkinsStatistical Service

Discussion

Member takes

Implication

Simulation study reveals unreliable country-effect parameter estimates in multi-country datasets with few nations, highlighting limitations of hierarchical models despite large individual samples.

Key Points

Assess the reliability of multilevel (hierarchical) linear and logit models when estimating country-level effects in multi-country survey datasets.
Conducted Monte Carlo simulation analyses using multilevel linear and logit models.
Evaluated model performance under conditions featuring large individual sample sizes within countries combined with small total numbers of countries.
Individual-level effects were reliably estimated when sample sizes within each country were large.
Country-level parameter estimates were unreliable when the number of sampled countries was small.

Cite This Study

Bryan et al. (2015) studied this question.

synapsesocial.com/papers/6a103ad72badbc352aff953a https://doi.org/10.1093/esr/jcv059

View Full Paper

Ask AI

Bookmark

Implication

Simulation study reveals unreliable country-effect parameter estimates in multi-country datasets with few nations, highlighting limitations of hierarchical models despite large individual samples.

Multilevel Modelling of Country Effects: A Cautionary Tale

Discussion

Member takes

Implication

Key Points

Cite This Study

Multilevel Modelling of Country Effects: A Cautionary Tale

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider