The Dynamics of Microcredit Borrowings in Cambodia

Making and justifying evidence claims: evidence synthesis of impact evaluations and systematic reviews in international development

“…Similarly, eight studies checked potential biases or performed balance checks to prove the effective randomisation and internal validity of the evaluation (Aker et al, 2017;Angeles et al, 2019;Armand et al, 2019;Bandiera et al, 2017;Barnett et al, 2018;Diagne and Cabral, 2017;Källander et al, 2021;Pellegrini, 2018). Several studies also conducted sub-group analysis to assess heterogeneous effects (Gibbs et al, 2020;Khwaja et al, 2020;Najy et al, 2018;Roth et al, 2017;Saboya et al, 2018;Zegarra et al, 2017).…”

“…Approaches to address the issues of the counterfactual were one of the main justifications when making decisions on the study design. Implementation constraints such as the time and resources available, local conditions, ethical concerns, sample size, costs, and data availability, were often cited as factors when considering whether CEDIL syntheses working paper 10: Making and justifying evidence claims: Evidence synthesis of impact evaluations and systematic reviews in international development cedilprogramme.org 30 it was feasible and appropriate to conduct RCTs or other types of quasiexperimental methods (Angeles et al, 2019;Barnett et al, 2018;Béné et al, 2020;Berhane et al, 2017;Pellegrini, 2018;Roth et al, 2017;Saboya et al, 2018;Zegarra et al, 2017). Béné and colleagues (2020) 'A major factor in the choice of a quasi-experimental design rather than a randomised controlled trial is the stated policy of the Ministry that all eligible households will be enrolled in the programme once a district enters the programme.…”

“…Other studies discussed the representativeness of the study population, which they claimed made the study findings generalisable to different settings (Asunka et al, 2019;Cocciolo et al, 2020;Edjekumhene et al, 2019;Hirvonen et al, 2017;Roth et al, 2017;Saboya et al, 2018). Another justification relating to generalisability was that the intervention programmes had been evaluated and CEDIL syntheses working paper 10: Making and justifying evidence claims: Evidence synthesis of impact evaluations and systematic reviews in international development cedilprogramme.org 40 proved to be transferable to multiple settings, which made the authors more confident that the programme 'can be successfully implemented in contexts other than Bangladesh and by organisations other than BRAC' (Bandiera et al, 2017, p. 27).…”

“…All impact evaluations used a variety of tables to communicate the findings of statistical analyses. For seven studies, these tables were included in the appendices, as opposed to being placed alongside the main text (Amirapu et al, 2020;Bandiera et al, 2017;Brudevold-Newman et al, 2017;Källander et al, 2021;McKenzie, 2017;Najy et al, 2018;Roth et al, 2017). One study, Parker et al (2019), acknowledged that 'some readers prefer a formal table of results', and explicitly provided a visual alternative to communicate claims.…”