Similarly to other domains of the social sciences, behavioural science has grappled with a crisis concerning the effect sizes of research findings. Different solutions have been provided to answer this challenge. This paper will discuss analytical strategies developed in the context of computational social science, namely causal tree and forest, that will benefit behavioural scientists in harnessing heterogeneity of treatment effects in RCTs. As a mixture of theoretical and data-driven approaches, these techniques are well suited to exploit the rich information provided by large studies conducted using RCTs. We discuss the characteristics of these methods and their methodological rationale and provide simulations to illustrate their use. We simulate two scenarios of RCTs-generated data and explore the heterogeneity of treatment effects using causal tree and causal forest methods. Furthermore, we outlined a potential theoretical use of these techniques to enrich behavioural science ecological validity by introducing the notion of behavioural niche.

India’s main food and nutrition security programme, the Public Distribution System (PDS), provides subsidised rice and sugar to deprived households. Using longitudinal data from Young Lives for Indian children (n = 2,944) aged 5 to 16 years, we assessed whether PDS subsidies skewed diets towards sugar and rice consumption, increasing risk of stunting (low height-for-age). Linear regression models were used to quantify additional rice and sugar consumption associated with accessing the PDS, and the association with stunting linked to consumption. Controlling for sociodemographics, accessing the PDS was positively, significantly associated with consumption of rice (30g/day) and sugar (7.05g/day). There was no evidence that this increase corresponded to nutritional improvements. Each 100g increase in daily rice intake was associated with a lower height-for-age z-score (HAZ) and no decline in stunting. Results were robust to alternative model specifications. There was no evidence that receipt of PDS rice and sugar was associated with improvements in child nutrition.

In this paper, we assess whether or not organic agriculture has a positive impact on local economies. We first identify organic agriculture hotspots (clusters of counties with positively correlated high numbers of organic operations) using spatial statistics. Then, we estimate a treatment effects model that classifies a county's membership in an organic hotspot as an endogenous treatment variable. By modeling what a hotspot county's economic indicators would have been had the county not been part of a hotspot, this model captures the effect of being in a hotspot on a county's economic indicators. We perform the same analysis for general agricultural farm hotspots to confirm that the benefits associated with organic production hotspots are, in fact, due to the organic component. Our results show that organic hotspot membership leads to a lower county-level poverty rate and a higher median household income. A similar result is not found when investigating the impact of general agriculture hotspots. On the other hand, our result is robust to alternative hotspot definitions based on type of organic operations to alternative methods of estimating average treatment effects on the treated. These results provide strong motivation for considering hotspots of organic handling operations, which refers to middlemen such as processors, wholesalers and brokers, and hotspots of organic production to be local economic development tools, and may be of interest to policymakers whose objective is to promote rural development. Our results may incentivize policymakers to specifically focus on organic development, rather than the more general development of agriculture, as a means to promote economic growth in rural areas, and may further point them in the direction of not only encouraging the presence of organic operations, but of fostering the development of clusters or hotspots of these operations.

Using data on joint annuities, we conduct an analysis of the inter-couple lifetime dependence. We estimate a mixed proportional hazards model with treatment effects, which disentangles the broken-heart syndrome from the spurious risk dependence induced by unobserved heterogeneities. We use a flexible semi-parametric distribution for the unobserved heterogeneities to allow for either positive or negative spurious risk dependence. We find that the effect of losing one's spouse is asymmetric for the two sexes. Moreover, although the broken-heart syndrome explains a large portion of the dependency, there is evidence of positive spurious risk dependence. These findings have important implications for the pricing of joint insurance products. Failure to take into account either of these two effects leads to pricing error that can be either positive or negative, depending on the characteristics of the couple.

Background: We aimed to develop a standardization method to pool data recorded on different activities of daily living (ADL) scales in order to reduce variability of functional outcome data from Alzheimer's disease (AD) clinical trials and to better evaluate the effect of donepezil treatment on function in patients with AD.

Methods: Based on pre-specified criteria, six studies were selected from among all donepezil clinical trials in AD. Individual items from nine ADL scales used in these trials were mapped to a standardized functional scale comprising 12 domains (six basic, six instrumental); scores were transformed to a 0–100 scale. External validation of this scale yielded a concordance rate of 90.8%. For each domain, mean change from baseline to 24 weeks in the placebo and donepezil groups was compared for the total population and for subgroups stratified by baseline disease severity. Study settings included outpatient, assisted living, and skilled nursing facilities. Participants comprised 2183 patients (donepezil, 1261; placebo; 922) with baseline Mini-mental State Examination (MMSE) scores 5–26.

Results: Significant treatment differences favoring donepezil were observed for five items (two instrumental and three basic). Patients with moderate AD at baseline (MMSE 10–17) demonstrated the greatest treatment effect.

Conclusion: Functional data were successfully pooled using standardizing methodology. A beneficial effect of donepezil treatment on function was demonstrated using this standardized functional scale. Similar analyses from studies with other anti-dementia drugs may help to determine the generalizability of these findings and potentially encourage use of functional assessment as a clinical tool.

Objective: To assess the effect of the use of cochlear implants (CI) on the health status of postlingually deaf adults.

Methods: Participants comprised 45 postlingually deaf adult multichannel CI users and 46 deaf candidates on the waiting list for a CI. The latter group acted as control subjects to corroborate the validity of retrospective completion of the questionnaires by the CI recipients. Three HRQOL instruments were used: a) a specially developed CI questionnaire (NCIQ); b) a generic HRQOL questionnaire (SF-36); and c) a health-state classification system (HUI-2) suited to estimate single preference scores.

Results: Retrospectively estimated pre-implant scores in the CI user group corresponded very well with the scores in the control group. Postimplant scores in the CI users were substantially higher in all six domains (p < .001) of the NCIQ than the scores in the control group. Effects due to a CI were also observed with the SF-36 in five of the seven domains (p < .01). Statistically significant differences between the two groups (p = .001) were observed in two of the six domains of the HUI-2.

Conclusions: All three questionnaires detected improvements in HRQOL due to CI use. To make a detailed assessment of the effect of a CI on functional outcomes and well-being, a special purpose HRQOL instrument is far more adequate than a general HRQOL instrument. This study also showed that a CI affects several other health domains besides auditory performance. The effect of CI use on general functioning and well-being proved to be considerable.

Despite some evidence that genotype-environment interaction (G×E) effects may be involved in the variation observed in behavioral and biological traits, few attempts have been made to detect and quantify this component of genetic variation in humans. We propose that one way to achieve this goal is to challenge several genotypes in a similar manner, submitting both members of several MZ twin pairs to an ethically acceptable experimental treatment capable of inducing an adaptative response. In this situation, the G×E effect can be assessed with a two-way analysis of variance for repeated measures on one factor, the treatment effect. In this design, twins are considered nested within the pair, whereas the treatment effect is considered a fixed variable. The intrapair resemblance in the response to the treatment is quantified with an intraclass correlation coefficient computed with between-sibhips and within-sibhips means of squares. To illustrate this approach, changes induced by long-term endurance training were studied in 10 MZ twin pairs. Significant intrapair resemblance in the response of maximal oxygen uptake was observed, with about 7 to 8 times more variance between pairs than within pairs. This design with MZ twins may be helpful in the study of human variation for multifactorial phenotypes.