CASE STUDY THREE: The Desire to Stop or Delay: Historical Trends and Regional Patterns
- chen
- March 8, 2022
Background
Fertility in Sub-saharan Africa is the highest in the world and remains relatively stable comparing to all other developing countries. These high birth rates are largely a function of both desired fertility and unintended fertility.
Research Questions
Why is the total fertility rate (TFR) in Sub-Saharan Africa persistently high? Are the nature and causes of fertility decline in this major region fundamentally different (i.e. “African exceptionalism”)?
Study Goals
To provide new insight into this question, we examine the country and regional differences in their desire to avoid pregnancy at particular stages of fertility transition (i.e., the fertility from high to low levels).
My role in the team:
- Multinational Database Management
- Longitudinal Data Analysis
- Texting Drafting
- Data Visualization
- Conference Presentation
Study design
2016 Survey of Chinese Families' Fertility Decision-Making Processes
Methodology
We pulled data from multiple national demographic surveys covering all major developing regions from 1975 to present. The dataset contains over 5 million women.
Sample
The following rules are applied in selecting surveys:
- Countries with minimum total population of 500,000 in 2000
- Countries with at least two surveys
- Surveys with TFR > 2.2
- Surveys with more than 2000 women
- Surveys with effective fertility preferences measurement
Analysis
The analysis is confined to currently married women. We construct two dichotomous indicators: “stop” (not wanting another child), and “avoid” (not wanting another child or wanting to delay next birth 24+ months). Region-specific trend lines are fit via regression, with adjustment for woman’s age and random effects for country. Decomposition via standardization assesses the contribution to trends in preferences of changes in parity composition.
Findings & Implications
Findings
(1) The fraction of married women wanting to stop or delay changes surprisingly little during fertility decline. (2) One explanation for finding (1) is change in parity composition that works against increase in the fraction wanting to stop or delay. (3) There are large regional differences in both levels and trends. (4) East & Southern Africa is distinctive in the magnitude of change.
Implications
Finding (1) challenges common assumptions about contemporary fertility declines. Findings (3) and (4) reinforces accumulating evidence that African declines are different in character, specifically more “demand-driven”. This makes African declines more consistent with classic demographic transition theory, i.e. “exceptional” because they conform more closely to long-standing theory than declines in other regions.