A Microcredit Alternative in South Asia by Shahrukh Rafi Khan Natasha Ansari

Author:Shahrukh Rafi Khan, Natasha Ansari [Shahrukh Rafi Khan, Natasha Ansari]
Language: eng
Format: epub
Tags: Social Science, Ethnic Studies, General, Regional Studies
ISBN: 9781351174572
Google: lnlTDwAAQBAJ
Publisher: Routledge
Published: 2018-03-28T03:36:47+00:00

Research design and sampling

Initially we requested Akhuwat to provide us with a sampling frame that would only include individuals whose loan contract with Akhuwat had ceased, i.e. those who received one or more loans but had none outstanding at the time of the interview and did not plan to get a loan in the future. Our reasons for this selection were two-fold. First, current recipients may not have had the time to fully process their interaction with Akhuwat. Second, they might be more favorably biased in their responses since they would be borrowers at the time of interview. A bias might also have been introduced by their desire for and expectation of additional loans. We thought we could get a full picture of how borrowers benefited from the loans they received and of their reflections regarding their interactions with Akhuwat once the loan contract was completed.

We confronted several problems during the pre-tests in tracing ex-borrowers and that resulted in our changing our research design. First, since there is staff turnover, the new unit managers often did not know the earlier borrowers. Second, in early 2015, the Government of Pakistan decided to block cell phone sims registered without biometric data. Many people decided not to register their biometric data, and so many old cell contacts in Akhuwat’s records were unreachable. Third, some individuals migrated or passed away. Finally, we found that since ex-borrowers were no longer beholden to the organization, they were less responsive to current unit manager requests for interviews, especially since they do not know the new unit managers.4

We therefore decided to opt for a sampling frame which included both old and new borrowers. In retrospect, we realized that we could have oversampled ex-borrowers. However, diminishing marginal returns in securing interviews set in very fast once current borrowers had been interviewed, and we sought interviews with ex-borrowers. Spending more time per site was commensurate with higher cost per interview because of higher fixed costs such as automobile rent and hotel stay. It was also expensive in terms of the organization staff members’ time because they had to trace ex-borrowers ahead of our arrival. Since our arrival virtually suspended the normal business of the branch, we had reason not to overstay our welcome despite support from the head office.

While we have explored to see, in relevant cases, if there are systematic differences in responses of borrowers and ex-borrowers, there may be response biases. The responses of ex-borrowers could vary based on whether they completed a loan cycle and did not need another one, planned to get another loan but did not have one yet or were denied an additional loan after having applied.

The sampling frame contained 47,914 respondents from 253 branches. This represents the population that we generalized to. We first randomly selected 5 percent of the branches (13 branches out of the 253 – Appendix 6.1 lists the sample distribution by branch). Given our budget, we randomly selected 1.5 percent borrowers and ex-borrowers from each branch for a total sample size of 458 respondents: 43 percent females and 57 percent males.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.