C. Mathematical Investigation
To examine program effects on IPV and you will routes, the study rooked employing a great cutoff threshold applied to an ongoing system qualification list (PMT score) to understand therapy and you may analysis organizations. Ergo, email address details are good into the people inside the endurance and you may carry out maybe not show an entire distribution off socioeconomic position within the data teams. Just like the a beneficial discontinuity is employed towards the task of your own cures, we be sure new issues that normally must hold to have a routine regression discontinuity design way of be good within studies, and no control at the tolerance or knowledge of new algorithm by the prospective beneficiaries so you’re able to influence its qualification status. The brand new official sample away from harmony ranging from assessment and you can treatment communities are together with reported with the latest try when you look at the tables 2 and you may step three (col. 10). Regarding overall concentrating on do it presented, the fresh PMT scores varied regarding ?step 3.294 so you’re able to 4.601 z-ratings, through the review test, the brand new PMT ratings range between ?0.571 so you’re able to 0.324 z-ratings, towards the system qualifications cutoff in the ?0.106 z-results ( important deviation = 0.32 ). six Property have been methodically tested inside the cutoff having analysis aim, plus the resulting properties on the therapy shot are contained in this 0.46 basic deviations of your cutoff, and all investigations home is actually within this 0.43 important Adventist dating app deviations of cutoff. Because of the short endurance of shipments of sampled properties doing the latest PMT rating, a straightforward Did was applied to check the fresh influences of one’s program: eight
Control variables found in the research included personal- and house-level evidence and you can area-height fixed outcomes. 8 On private peak, we controlled getting years (in many years) and you will signs the official education away from each other woman and you can lover. nine Next, inside research of the full sample, i handle to be inside an effective polygamous relationship (reference group is actually nonpolygamous otherwise, to possess easier malfunction, an effective monogamous commitment). Household-top handle parameters provided domestic size additionally the PMT score. 10 We report the latest means of most of the area indication parameters in the descriptive equilibrium and you can attrition dining tables; although not, even as we include society-peak fixed effects, these types of indicators commonly used in regression analysis (districts: Eastern Mamprusi, Karaga, and you may Yendi about North Part and you will Bongo and you will Garu Tempane on Top Eastern Area). To evaluate distinctions by the nearest and dearest build, i collaborate the newest signal getting polygamous commitment towards medication sign, date indication, and perception guess (performing a triple interaction name): eleven
D. Attrition
Attrition over the panel period has potential to threaten the internal validity and generalizability of the findings. Overall attrition in our analytic sample is approximately 10%, slightly higher than the overall household sample rate of 6.6%, and does not vary significantly between treatment and comparison samples ( p -value = .35 using a simple t-test; Table 1). Further, we investigate whether the sample lost to follow-up differs significantly from our panel sample in terms of IPV outcomes, control variables, or pathway indicators (tables 2, 3). Column 7 shows the mean difference between the sample lost to follow-up in the treatment group, as compared with the same in the comparison group. Column 8 shows the p-value of this difference derived from a regression using treatment to predict the specific indicator, controlling for the PMT score and restricting to attritors. Table 2 shows that across 12 background characteristics and 14 pathway variables, only one is significant at the p < .05 level (there is a statistically significantly higher proportion of the treatment group lost to follow-up as compared with the comparison group in Bongo district). Table 3 shows that across 15 outcome indicators, the only ones to show evidence of differential attrition at p < .05 or lower are those measuring the frequency of sexual IPV (means of those lost to follow-up in the comparison group are slightly higher at .28 vs. the treatment group at .26). These results indicate that the potential for bias in overall findings due to attrition is low. Despite these promising results, we do find that women leaving our sample across both treatment arms differ in terms of background characteristics from those who remain in our sample (p-values shown in cols. 3 and 6). For example, women who leave the sample are generally younger, better educated, in monogamous partnerships (vs. polygamous), and live in smaller households. This may be reflective of rural-to-urban migration or other mobility patterns, although these differences are not generally observed for pathway variables, including indicators of economic standing.