These robustness monitors are offered regarding the standard review statement (Leap a thousand Comparison Group 2016)

C. Statistical Data

To look at program consequences on IPV and you may pathways, the research cheated the aid of a good cutoff threshold put on an ongoing system qualification directory (PMT score) to recognize medication and you may comparison teams. Hence, answers are valid to your population within the endurance and you will carry out maybe not portray the full delivery away from socioeconomic reputation during the study teams. Since an effective discontinuity is utilized into assignment of your therapy, we be certain that the brand new issues that typically must hold for a regular regression discontinuity framework way of getting valid within our study, as well as zero manipulation on endurance otherwise experience with the brand new algorithm by prospective beneficiaries in order to dictate the qualification condition. The fresh new official try away from equilibrium anywhere between evaluation and you can therapy organizations is actually also reported on current try inside the tables 2 and step 3 (col. 10). In the full concentrating on get it done presented, the new PMT score varied regarding ?step three.294 to 4.601 z-results, throughout the testing attempt, the new PMT results start around ?0.571 so you can 0.324 z-score, towards system eligibility cutoff from the ?0.106 z-scores ( fundamental deviation = 0.thirty-two ). 6 Homes had been methodically sampled in the cutoff to own assessment intentions, together with resulting domiciles regarding medication decide to try is within this 0.46 basic deviations of your own cutoff, and all sorts of review houses is actually contained in this 0.43 standard deviations of cutoff. Considering the brief tolerance of shipping regarding tested domiciles as much as the latest PMT score, an easy Performed was applied to evaluate new affects of program: 7

Manage details included in the study included personal- and you can house-top evidence and you will society-height fixed consequences. 8 During the personal height, i controlled to possess many years (in many years) and you can symptoms when it comes to official training off each other woman and you can spouse. 9 Subsequent, inside the data of your full take to, i manage for being in the a good polygamous union (resource category try nonpolygamous or, getting simple description, a monogamous connection). Household-height manage details included family size in addition to PMT get. ten We statement the latest a style of all district sign details within the descriptive harmony and you will attrition tables; yet not, as we is society-level fixed consequences, these indications are not included in regression research (districts: East Mamprusi, Karaga, and you can Yendi on the Northern Area and Bongo and you will Garu Tempane on the Higher East Area). To assess differences by friends build, i interact this new indicator getting polygamous commitment with the medication signal, go out indicator, and you can impact imagine (starting a multiple correspondence title): 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.