C. Statistical Analysis

To look at program outcomes towards the IPV and you may routes, the research cheated making use of a cutoff tolerance placed on a continuous system eligibility directory (PMT rating) to identify procedures and you https://datingranking.net/buddhist-dating/ will comparison communities. Therefore, answers are good towards people around the threshold and you may would not represent a full shipping of socioeconomic position into the analysis organizations. Once the a good discontinuity is employed to the project of your own medication, i make sure the new conditions that normally need certainly to hold having a typical regression discontinuity framework method of be valid within our studies, plus no control on tolerance otherwise expertise in the formula by possible beneficiaries to determine the eligibility position. This new official try out of balance between review and you can treatment organizations try including reported on newest sample in the tables dos and step three (col. 10). Throughout the full centering on take action held, the fresh new PMT scores ranged off ?step three.294 so you’re able to 4.601 z-scores, while in the analysis take to, the newest PMT scores are normally taken for ?0.571 so you can 0.324 z-score, towards the program eligibility cutoff on ?0.106 z-results ( important departure = 0.thirty-two ). 6 Properties have been systematically tested within cutoff to own evaluation objectives, while the resulting home about therapy take to are in this 0.46 important deviations of one’s cutoff, and all of testing home is within 0.43 standard deviations of your cutoff. Considering the small endurance of the shipping regarding tested properties as much as the brand new PMT rating, an easy Did was applied to check on this new has an effect on of the program: eight

Control variables found in the analysis provided private- and you can home-peak symptoms and you will people-height fixed effects. 8 At the private level, i controlled to possess age (in many years) and you can evidence the specialized training of each other girl and you can companion. nine Subsequent, from inside the investigation of complete decide to try, we manage to be when you look at the a beneficial polygamous commitment (reference classification are nonpolygamous otherwise, to own easier dysfunction, a great monogamous connection). Household-peak handle variables integrated family proportions in addition to PMT get. ten I statement the manner of all section indication details within the descriptive equilibrium and you can attrition tables; however, as we include community-peak fixed outcomes, these types of signs aren’t utilized in regression investigation (districts: Eastern Mamprusi, Karaga, and you can Yendi on Northern Part and you can Bongo and you can Garu Tempane from the Upper East Area). To assess distinctions of the nearest and dearest design, we come together the fresh indication for polygamous relationship for the therapy indication, day signal, and you can impression imagine (carrying out a multiple interaction term): 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.