Join the network
Research Brief
Three key insights for research on social mobility in developing countries
The volume, Social Mobility in Developing Countries: Concepts, Methods, and Determinants, brings together leading scholars from a range of social science disciplines working on a variety of issues related to social mobility. Three motivations guide this joint effort: identifying important knowledge gaps; bringing together innovations and improvements in research practice; and offering policy advice to accelerate social mobility in developing countries. In this brief, we identify three key lessons for improving research practice.
The emerging field of research on social mobility in developing countries can benefit from more coherent research practice.
There are several lessons for research practice emanating from a recent UNU-WIDER volume on the topic.
Research practice in the current literature often appears ad hoc, with considerable variation in the research questions asked, in the concepts and measures used to answer them, and in the articulation and inference of what empirical findings mean.
A plea for greater precision in scholarly work on social mobility
First, there are many possible combinations of concepts and measures that are available to compare intergenerational mobility within countries, across countries, and over time. Because such comparisons are often shrouded in ambiguity, the volume proposes a simple checklist to improve social mobility research practice (for details see the chapter by Gary Fields).
Researchers should start with being more explicit about key preliminaries, such as the outcome of interest, context, and the level of analysis, and then follow these steps: (1) articulate the question they hope to answer, (2) define the mobility concept(s) and the mobility measure(s) they will use, and (3) present the empirical findings.
Account for context
Second, echoing the observation made by Florencia Torche in a review of social mobility research on Latin America, the measures and methods developed, debated, and used to study intergenerational mobility are often borrowed from analyses of high-income countries without enough scrutiny of how well they perform in other contexts. This highlights the need for a deeper understanding of the properties, and of the strengths and limitations, of different social mobility concepts and measures in the analysis of developing country settings.
Among the most widely-used measures of intergenerational mobility in developing countries, the intergenerational correlation (IGC) was found to be less vulnerable to biases than the intergenerational regression coefficient (IGRC), but also that biases are less pronounced in some countries than in others (for example, in Bangladesh compared to India). The chapter by Emran and Shilpi highlights the more robust properties of the intergenerational rank correlation (IRC) and the value added of the intercept term for cross-country comparisons.
The volume also notes an oft-neglected weakness in correlation measures of intergenerational mobility. A lower association between parental characteristics and offspring outcomes (known as less origin-independence) is usually interpreted as greater mobility, but both greater upward and greater downward mobility for children can account for a weakening of the association. This means that a modest prevalence of moderate or large descents into poverty may register as increased social mobility.
In another illustration of how context matters, the occupational classifications standardized for use in fully industrialized countries are less applicable elsewhere. A chapter by Anthony Heath and Yizhang Zhao proposes that anthropological insights about the institutions of the country under study can be used to significantly improve efforts to align occupational rankings and classifications with realities on the ground. For example, farmers remain a dominant occupational group in low-income settings and there is often a need to disaggregate it to a more granular level of classification.
The value of interdisciplinary conversations and collaboration
Third, research and policy dividends can arise from interdisciplinary conversations. While claims of such gains are so regularly encountered that they risk becoming a trope, many of the chapters in the volume provide important real examples of value added from interdisciplinary research. A chapter on the ethnographic approach by Divya Vaid underscores the subjective nature of how the outcomes that matter are understood, with highly-localized variation. In a chapter by Florencia Torche, the utility of theoretical explanations from sociology for why educational inequality persists is discussed, including the distinction between so-called primary and secondary effects. While the former captures the association between an individual’s socioeconomic background and educational attainment, the latter captures class-based choices net of educational attainment.
Educational attainment is an important example of how applying this distinction improves our understanding. While developing country evidence remains sparse, research points to educational aspirations, access to information and guidance, self-esteem, and self-efficacy as major drivers of secondary effects which are critical to securing higher levels of schooling among disadvantaged children. For example, Luana Marotta finds that such secondary effects account for about half of the inequality in secondary-school completion in Brazil. Further, and as Torche makes clear, the relevance of secondary effects is often greater in developing countries, including and possibly in relation to gender.
Interdisciplinary conversations bring about exposure to new ideas and lines of inquiry. A quantitative analysis of intergenerational mobility in the chapter by Yaojun Li introduces readers to sociological measures of mobility, but also to unusually rich, new evidence on intergenerational mobility in China — including granular findings on mobility variation across different generations of women and men. Adding to this, Himanshu and Peter Lanjouw use the multiple-decade Palanpur village panel dataset to show how high-quality, granular, longitudinal data can answer and inspire new questions and theoretical ideas among economists using a macro-lens and among scholars working on social mobility from other disciplinary backgrounds. Other and similarly valuable insights are provided in the chapters by Nancy Luke, Emily Rains and Anirudh Krishna, Patricia Funjika and Rachel Gisselquist, and others.
For researchers involved in studies of social mobility, the book offers a checklist with three suggestions that can improve research practice coherence.
First, clearly define the research question, specify which concept(s) and measure(s) will be used to answer that question, and articulate the empirical findings. Refer to the chapter by Gary Fields for an in-depth review of the concept(s) and measure(s) used in social mobility studies.
Second, be aware of the context of your study, especially as it relates to your choice of concept(s) and measure(s) and their various strengths and weaknesses in different settings.
Third, look for inspiration and relevant insights from outside your own discipline.
If taken onboard by researchers, these insights and recommendations can reduce the risk of fragile or erroneous claims and translate into higher-quality, more reliable policy advice. Heeding the recommendations found in the chapter by Gary Fields will also increase uptake and understanding among policy makers and other interested readers, by making research findings easier to navigate and interpret.
