Abstract

An earlier paper from the same parent survey documented overlapping value-chain roles and descriptive sex differences in processor and seller participation. This secondary analysis extends that work by assessing whether the contrasts remained under multiplicity and covariate adjustment and whether sex differentiation was evident across selected access and decision indicators. Data came from an April 2025 survey of 266 shea actors in Arilo Sub County, Uganda (150 women and 116 men). Non-exclusive role indicators were reconstructed from occupation fields. Within-sex percentages, Pearson chi-square and exact tests, post hoc Holm and Benjamini-Hochberg adjustments, binary models with robust covariance, a multinomial market information model and a married-subsample credit-use sensitivity model were applied. Estimates are unweighted sample-level associations. The previously reported processor role contrast (62.7% of women versus 41.4% of men) remained after Holm correction and covariate adjustment (odds ratio = 1.94, 95% CI 1.13-3.34, p = 0.017). The seller role contrast was not sustained under stricter inference, and no adjusted sex associations were detected for land ownership, income sufficiency and market-information access. The credit use pattern differed descriptively but was not confirmed in the adjusted sensitivity model. Sex differentiation was therefore concentrated in processing. The findings support role-responsive services that attend both to participation and to control over resources and benefits.

This work is licensed under a Creative Commons Attribution 4.0 License.