A Bayesian Multilevel Regression Analysis of Gendered Livelihood Decision-Making Processes in in the Context of Climate Change

Authors: Faisal Islam*, University of Tennessee, Madhuri Sharma, Associate Professor, University of Tennessee, Knoxville
Topics: Human-Environment Geography, Behavioral Geography, Gender
Keywords: Climate Change, Livelihood, Bangladesh, Bayesian Multilevel model, women empowerment
Session Type: Virtual Paper
Day: 4/9/2021
Start / End Time: 11:10 AM / 12:25 PM
Room: Virtual 56
Presentation File: No File Uploaded


Climate change is increasingly challenging millions of people’s ability to sustain livelihood in the global south. While there exists extensive research on climate change hazards, impacts, and adaptation in the most vulnerable regions of the world, little attention has been given to gender roles in decision-making processes. This paper will explore how the power of livelihood decisions changes from men to women by different natural shocks (loss of crops due to flooding, cyclone, erosion) and life events (mobility decision, remittance, education) based on a composite livelihood decision scores in the vulnerable coastal regions of Bangladesh using the Bayesian Multilevel Regression Model. The livelihood-decision score for both men and women are calculated based on the women empowerment to agriculture index from the open-source dataset of Bangladesh Integrated Household Survey 2015. We first hypothesize that there is a significant gender gap in livelihood decision scores within the coastal regions. Secondly, we hypothesize that the gender gap in livelihood decision scores will increase with increasing negative life events (natural shocks including unemployment, death of a family member, paying a marriage dowry) and decrease with positive life events (such as receiving remittances, education, successful entrepreneurship, and access to credits). The study will be critical for creating new pathways for gender-based policies and intervention efforts, especially in the global south.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login