The registered primary data are subsequently used for further analysis (see Section 4). Data analysis of this balanced
panel is based on a matched difference-in-difference approach.
We combine a ‘‘with and without”assessment of certification
comparing FT-, Utz- and non-certified cooperatives, and a
‘‘before and after”analysis of certification comparing baseline
with ex-post survey data. The diff-in-diff estimator compares
change in outcomes from the treatment group before and
after the intervention to the changes in outcomes in the control group. By analyzing differences, the estimator controls
for characteristics that do not change over time within the
treatment and control groups. The registered net change in
the control group is an estimate of the true counterfactual i.
e., what would have happened to the intervention group if
certification was not implemented. As a robustness check,
we also computed the minimax approach (Dehejia &
Wahba, 1999) that provided roughly similar outcomes.