Causal inference of the effect of a treatment on an outcome is usually done on the group or subgroup level. Although the typically reported average treatment effect may be positive, suggesting that the treatment is effective, at the level of individual participants, the treatment effect may be zero or even negative-the treatment may even harm some individuals. For making decisions on whether a specific person should take the treatment, information on the probability of benefiting or being harmed by the treatment for a single person is necessary. Estimating the probability of possible benefit or possible harm for a person involves counterfactual reasoning and thus strong assumptions about unobservable events. Precise statements about the causal effect of a treatment for an individual are only possible to a limited extent. This tutorial introduces the method of causal attribution to psychology that allows for estimating bounds in which the probability of benefit or harm lies. These bounds can be calculated using data at the group level, which can come from experimental or observational studies. The bounds can be narrowed by simultaneously using data from both randomized trials and observational studies and by using information from pretreatment covariates. R functions are provided for calculating these bounds from binary data and are illustrated with examples from basic laboratory research and clinical intervention research. (PsycInfo Database Record (c) 2026 APA, all rights reserved).
Kaiser et al. (Thu,) studied this question.