C. Mathematical Research
To examine program consequences towards IPV and paths, the analysis rooked the usage of good cutoff tolerance placed on a continuing system qualifications list (PMT score) to spot cures and you may analysis groups. For this reason, answers are legitimate on the population inside the threshold and you may would not represent an entire shipments out of socioeconomic condition from inside the investigation communities. While the a discontinuity is used into assignment of one’s medication, i be certain that brand new issues that usually have to hold to have a normal regression discontinuity construction method to end up being appropriate in our study, and additionally zero control at the tolerance otherwise expertise in the algorithm because of the potential beneficiaries so you can determine their eligibility updates. The fresh specialized decide to try regarding balance ranging from investigations and you may cures teams are plus advertised into newest shot into the dining tables 2 and you may 3 (col. 10). About total emphasizing exercise presented, brand new PMT ratings varied from ?step 3.294 to help you cuatro.601 z-score, through the evaluation decide to try, the newest PMT ratings are normally taken for ?0.571 so you’re able to 0.324 z-scores, for the program qualifications cutoff at the ?0.106 z-scores ( simple deviation = 0.32 ). 6 House was indeed systematically sampled in the dating site Hookup singles only cutoff having comparison intentions, and the ensuing domiciles on cures test is actually within this 0.46 simple deviations of one’s cutoff, and all sorts of comparison households is inside 0.43 important deviations of cutoff. Considering the quick tolerance of your own distribution of sampled properties doing this new PMT rating, a straightforward Did was used to check on brand new affects of your own program: seven
Handle parameters utilized in the study incorporated private- and household-peak indications and area-height fixed consequences. 8 During the personal top, we controlled to possess age (in years) and indications when it comes to certified education out of both girl and you will companion. nine Further, for the study of one’s complete take to, we control to be for the a good polygamous connection (reference category was nonpolygamous otherwise, for ease of dysfunction, a great monogamous union). Household-level handle variables provided house size therefore the PMT score. 10 We declaration the new means of most of the section indicator details in the detailed equilibrium and you can attrition tables; however, while we were area-top repaired effects, these types of indicators commonly used in regression research (districts: Eastern Mamprusi, Karaga, and you may Yendi on the North Region and Bongo and you can Garu Tempane from the Upper Eastern Part). To assess distinctions because of the family framework, i work together the new sign to own polygamous union for the medication indicator, date sign, and you will effect imagine (carrying out a multiple telecommunications label): 11
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.