Bilingual Cooperation Experiment Replication in Uganda

A replication study finds participants playing a public goods game in Luganda contributed 28.9% more than those playing in Lugisu, matching the original effect within 0.1 percentage points. The research, led by Paul Clist, explores mechanisms by eliciting injunctive norms and empirical expectations for every possible action. The dataset was last updated in March 2026.

Use Cases

• Analyze the effect of randomized language assignment (Luganda vs. Lugisu) on contribution amounts in a public goods game.
• Test mechanisms by comparing elicited injunctive norms across the two language conditions.
• Compare empirical expectations for every possible action between language groups to rule out conditional cooperation.
• Replicate the original finding of a 28.9% difference in cooperation using the provided experimental data.

Strengths

• Replicates a key behavioral economics finding within 0.1 percentage points of the original effect.
• Experimental design excludes confounds like differences in grammar or comprehension.
• Elicits both injunctive norms and empirical expectations for mechanism analysis.

Limitations

• Sample size and row count are unknown, limiting statistical power assessment.
• Geographic scope is limited to Uganda, specifically comparing a national and a tribal language.
• The dataset's completeness and available variables beyond the described experiment are unknown.

Provenance

Source

Harvard Dataverse, author Paul Clist.

Collection Method

Experimental replication of a public goods game with bilingual subjects randomly assigned to a language condition.

Time Range

null

Freshness

Last updated March 2026.

Geography

Uganda.