Search

Begin New Search Topic Survey Author Rosa Abraham

1. What did they say? Respondent identity, question framing and the measurement of employment

Author:
Rosa Abraham, Nishat Anjum, Rahul Lahoti, and Hema Swaminathan
Publication Date:
09-2023
Content Type:
Working Paper
Institution:
Centre for Sustainable Employment, Azim Premji University
Abstract:
In developing countries, a precise approach to measuring women’s employment remains elusive. Emerging evidence underscores the pivotal role of survey methodology, encompassing respondent selection and question framing, in shaping the assessment of women’s employment. Drawing from two labor market experiments in rural India, this study offers insights on the influence of survey design on the measurement of women’s employment. The first experiment contrasts self-reported women’s and men’s employment figures with proxy-reported data from spouses. Women’s self-reported workforce participation surpasses proxy-reported estimates by six percentage points, while men’s estimates exhibit negligible differences. There are significant differences in the type of employment activities reported by self and proxy for both women and men. These divergences emanate from asymmetric measurement errors, stemming from gender-based norm disparities between spouses, and divergent interpretations of employment. Additionally, information asymmetry between spouses concerning women’s marginal activities and disparities in spousal characteristics contribute to these self-proxy differences. The second experiment investigates if framing of questions and recall period has an impact on reporting of labor market outcomes. We find that employing multiple questions to capture weekly employment status yields a 10-percentage-point increase in reported women’s workforce participation, but men’s participation rate decreases by six percentage points. Furthermore, when a distinct employment query is directed at each day of the preceding week as opposed to a single query for the entire week, reported women’s workforce participation increases by seven percentage points, and men’s by four percentage points.
Topic:
Development, Employment, Survey, Methods, and Gender
Political Geography:
South Asia and India

2. Telephone surveys for data collection – some reflections

Author:
Rosa Abraham and Mridhula Mohan
Publication Date:
09-2023
Content Type:
Working Paper
Institution:
Centre for Sustainable Employment, Azim Premji University
Abstract:
Phone-based surveys have emerged as important survey tool with their use increasing particularly during the pandemic-induced lockdowns. However, the use of phone-based surveys comes with concerns around sampling and reporting – the former refers to issues in terms of reaching the intended respondents and adequate representation of different groups while the latter relates to the quality of data collected. Based on a phone-survey of a pre-existing cross-section of households in Karnataka and Rajasthan conducted during the lockdowns of 2020, this paper looks at how sampling and reporting issues play out especially with regard to speaking to women. Access to women, via phone surveys, is limited since most women do not have their own phone, and even if they are reached, they may not have the freedom to respond to all questions asked. We recommend a detailed protocol with multiple callbacks and rescheduling of calls that reduce sampling errors and allow for greater reach of women respondents and minimise discretionary decisions from the enumerators . We also recommend a set of questions to enumerators to help reflect on the quality of data collected. While phone surveys cannot replace in-person surveys, in the event they are used, a well designed protocol along with the use of meta and para data can go a long way in enhancing the quality and reliabilty of the data.
Topic:
Women, Employment, Survey, Methods, and Data Collection
Political Geography:
South Asia and India

3. Improving Survey Quality using Para Data: Lessons from the India Working Survey

Author:
Deepti Goel, Rosa Abraham, and Rahul Lahoti
Publication Date:
08-2021
Content Type:
Working Paper
Institution:
Centre for Sustainable Employment, Azim Premji University
Abstract:
The efficacy of survey-based policy recommendations is primarily dictated by the quality of data collected in the first place. Is the survey truly representative of the population it claims to characterise? Are respondents voicing their true opinions or are they playing to the gallery? Did enumerator bias creep into the data? These are questions that most users of surveys have, but are typically brushed aside in the race to get the analyses out. While there are no foolproof measures to ensure that survey data are authentic, certain steps can be taken to improve their dependability. One such is the use of what is called ‘para data’ (data about the process of data collection), to streamline enumerator practices, and thereby improve the reliability of the data being collected. This report details our experience of using para data to improve the quality of the India Working Survey (IWS).
Topic:
Economics, Employment, Survey, Data, and Workforce
Political Geography:
India