C. Statistical Study
To look at system outcomes to the IPV and paths, the study exploited the utilization of a cutoff endurance used on a continuing system qualification list (PMT rating) to understand therapy and you can review communities. Hence, results are legitimate toward society within the threshold and you may manage maybe not depict a full shipments out-of socioeconomic condition inside the investigation communities. Since a beneficial discontinuity is employed to your assignment of one’s therapy, i make certain the latest problems that usually need certainly to keep to possess a routine regression discontinuity build way of getting valid in our research, as well as zero control at tolerance otherwise expertise in the latest algorithm because of the possible beneficiaries to help you determine its qualifications standing. The new specialized sample of equilibrium between research and you will procedures organizations try and additionally reported into newest attempt in tables 2 and you can step three (col. 10). Regarding total centering on exercise held, the fresh PMT scores varied of ?3.294 to help you 4.601 z-scores, throughout investigations sample, this new PMT score start from ?0.571 in order to 0.324 z-ratings, towards the program qualifications cutoff within ?0.106 z-ratings ( standard deviation = 0.thirty two ). six Houses Popular datings dating apps had been methodically sampled inside the cutoff getting review objectives, while the resulting homes from the therapy shot is contained in this 0.46 important deviations of cutoff, as well as investigations domiciles was within this 0.43 simple deviations of cutoff. Because of the short threshold of distribution regarding sampled property to the latest PMT rating, an easy Performed was used to evaluate this new influences of program: eight
Control parameters found in the study integrated private- and home-level indicators and people-level repaired effects. 8 Within private height, we regulated for ages (in many years) and you may signs your certified studies off each other woman and you can partner. nine After that, for the data of complete take to, we manage to be inside the good polygamous connection (reference classification is actually nonpolygamous or, to have easy breakdown, an effective monogamous partnership). Household-level handle parameters included household size together with PMT get. 10 We report the brand new manner of all section indication details inside the detailed equilibrium and attrition tables; although not, while we become people-height fixed consequences, this type of indicators are not utilized in regression data (districts: Eastern Mamprusi, Karaga, and Yendi on North Area and Bongo and you may Garu Tempane regarding the Top Eastern Area). To assess differences by the relatives framework, i work together the new indicator having polygamous connection into the therapy signal, time indicator, and you will effect imagine (starting a multiple telecommunications 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.
