C. Statistical Study
To examine program consequences into IPV and you will paths, the study taken advantage of the employment of a cutoff threshold placed on a continuing program qualification directory (PMT score) https://datingranking.net/cuckold-dating/ to identify medication and evaluation organizations. Therefore, results are good for the population within endurance and you can create perhaps not show a full shipments regarding socioeconomic reputation in the research groups. While the a discontinuity is used to your assignment of procedures, i verify brand new problems that usually need keep for a regular regression discontinuity framework approach to getting valid in our investigation, including no manipulation from the threshold or knowledge of the algorithm by the prospective beneficiaries to help you influence the eligibility status. The official sample out of balance between investigations and you can cures organizations is and additionally said towards most recent take to in the dining tables 2 and step 3 (col. 10). About total targeting do it held, the PMT score ranged from ?step three.294 to help you 4.601 z-ratings, through the investigations decide to try, the new PMT score range between ?0.571 in order to 0.324 z-scores, into the system eligibility cutoff at the ?0.106 z-ratings ( standard deviation = 0.thirty two ). 6 Home was indeed systematically tested within cutoff to have research purposes, together with ensuing properties on procedures decide to try try in this 0.46 important deviations of your cutoff, and all of analysis homes are within 0.43 basic deviations of your cutoff. Because of the quick tolerance of your distribution out of tested properties doing the PMT rating, a simple Performed was utilized to test this new affects of one’s program: 7
Control variables found in the research integrated personal- and home-height evidence and you will community-peak repaired outcomes. 8 From the individual height, we controlled to possess decades (in years) and evidence for the formal education off both lady and spouse. nine Further, for the studies of full try, we control to be inside a polygamous union (resource category are nonpolygamous or, to have simple malfunction, a beneficial monogamous relationship). Household-top manage details provided house dimensions therefore the PMT get. ten I declaration brand new means of the area sign variables from inside the detailed balance and attrition tables; not, once we tend to be society-peak repaired outcomes, this type of evidence aren’t used in regression studies (districts: East Mamprusi, Karaga, and you may Yendi throughout the Northern Part and you may Bongo and you can Garu Tempane on Upper Eastern Area). To evaluate variations by relatives construction, we come together this new signal for polygamous union for the medication signal, big date indicator, and you will impression guess (undertaking a multiple correspondence title): eleven
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.