AUTHOR=Abafe Ejovi Akpojevwe , Oduniyi Oluwaseun Samuel , Tekana Sibongile Sylvia 

TITLE=Beyond Least Squares Assumptions: Quantile Regression Estimates of the Heterogeneity in Commercialization–The Case of Sunflower-Producing Households in North West Province, South Africa

JOURNAL=Frontiers in Sustainable Food Systems

VOLUME=Volume 5 - 2021

YEAR=2022

URL=https://www.frontiersin.org/journals/sustainable-food-systems/articles/10.3389/fsufs.2021.758399

DOI=10.3389/fsufs.2021.758399

ISSN=2571-581X

ABSTRACT=Commercialization of smallholder farms in contemporary economic assessment is assigned a core role in rural development, wealth creation, and poverty reduction. This article examines the drivers of commercialization and the heterogeneity existing at the household level. Here the benefit of using a quantile regression technique is compared with the ordinary least square estimate to examine the relationships between smallholder commercialization and plausible predictor variables. Using quantile regression and ordinary least squares approach to analyze data from 177 sunflower-producing households in Ngaka Modiri Molema District, North West Province, South Africa; we find that the role of household size, market outlet, farm size, education, access to information, and gender of household head varies across the conditional distribution of smallholder commercialization. However, only market outlet and farming system significantly affected the upper and lower quantiles with a varying magnitude while the remaining predictors formed a particular cluster pattern exerting significant impacts at the lower quantiles. The result implies that quantile regression analysis is a good technique for investigating the drivers and mechanisms of heterogeneity in smallholder farmers' commercialization. The findings also demonstrate that any policy initiatives that fail to reverence the heterogeneity in smallholder commercialization and harnessed the regular blanket approach, if effective at all, may likely fail to achieve the overall welfare and behavioral responses across the targeted households.