This type of robustness monitors are given in the baseline comparison statement (Dive 1000 Review Group 2016)

C. Analytical Research

To look at system effects towards the IPV and you can paths, the analysis taken advantage of using a cutoff threshold applied to an ongoing program qualifications list (PMT get) to identify cures and assessment groups. Ergo, email address details are valid on population within endurance and manage maybe not show a complete distribution out-of socioeconomic status from inside the investigation teams. Because the an effective discontinuity is employed to your project of one’s cures, i ensure new conditions that normally have to keep to own a typical regression discontinuity structure method to feel valid in our research, and additionally no control on tolerance otherwise experience in the new formula by the prospective beneficiaries to influence their qualification reputation. The specialized take to regarding equilibrium between review and procedures teams are and additionally reported on the current try for the tables dos and you will step three (col. 10). On the complete targeting do it held, the newest PMT results varied regarding ?step 3.294 so you’re able to cuatro.601 z-ratings, during analysis attempt, the newest PMT ratings cover anything from ?0.571 to help you 0.324 z-scores, on the program qualifications cutoff from the ?0.106 z-scores ( fundamental departure = 0.thirty-two ). six Property was basically systematically sampled around the cutoff getting investigations objectives, plus the resulting domiciles about cures test was within 0.46 simple deviations of one’s cutoff, as well as review properties try contained in this 0.43 basic deviations of one’s cutoff. Considering the brief tolerance of your own delivery out-of tested homes up to the latest PMT score, an easy Did was used to check the latest has an effect on of one’s program: eight

Control parameters utilized in the study included private- and domestic-height evidence and you will people-level fixed outcomes. 8 On individual peak, we regulated to possess years (in many years) and evidence for any formal education out of both lady and you may partner. 9 Then, in the study of one’s full take to, i handle if you are BHM dating within the good polygamous connection (site category is actually nonpolygamous or, to own easy dysfunction, a beneficial monogamous connection). Household-peak manage variables incorporated home size in addition to PMT get. ten I report the brand new a style of the section signal parameters within the descriptive harmony and you will attrition tables; not, as we is community-peak fixed outcomes, such signs are not used in regression studies (districts: East Mamprusi, Karaga, and Yendi about North Part and you may Bongo and you will Garu Tempane on the Top Eastern Area). To assess distinctions by friends build, i work together this new indication getting polygamous relationship into the medication indication, go out indicator, and you can impression guess (carrying out a multiple telecommunications 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.