Kling Summary Index Aggregation in Multi-Outcome RCTs
RCTClinical Trial
Evaluating experimental interventions across multiple related outcome variables (such as transaction count, total amount, log amount, and binary indicators) increases the risk of false-positive statistical findings due to multiple hypothesis testing.
Picture this
A summary index standardizes and averages multiple related measurements into a single overall score, functioning like a composite GPA that summarizes a student's performance across individual class subjects into one overall mark.
What the evidence says
The overall labeling treatment effect on the aggregated Kling remittance index across the full sample was +0.013 standard deviations (p > 0.10), while for the low baseline remittance subsample it reached +0.027 standard deviations (p < 0.01).
- Who was studied
- N = 4,451 Filipino migrant workers in the UAE across 137,927 weekly observations.
- How
- Econometric evaluation utilizing Kling et al. (2007) equal-weighted standardized summary index combining binary remittance sending, transaction count, winsorized remittance amount, and log remittance amount.
What to do
Construct a Kling standard deviation normalized composite index when evaluating interventions with multiple correlated outcome variables to control false-positive rates.
From the source
"The last column of the table examines a 'labeling index', a Kling et al (2007) index of the four other outcomes in the table."
A Field Experiment among Filipino Migrant Workers in the UAE.pdf