The rationale for striving to achieve double-blinding is to avoid biased research outcomes caused by the placebo effect. In the medical arena, where RCTs are well established, double-blinding can be ensured by running RCTs in laboratories, where the environment can be fully controlled, but the case is different for studies in the area of the social sciences and international development in particular. For example, RCTs evaluating the impact of education, social services or microfinance programmes are usually not even single-blinded but essentially ‘zero-blinded’ (Scriven, 2008, p. 12). In other words, individuals usually discover whether they belong to treatment or control groups, which undermines the notion of double-blindedness. 

Another challenge is the prevalence of pseudo-random methods which often occurs during the process of assigning individuals to treatment and control groups. It pays to investigate how exactly individuals were assigned to their respective groups; was the underlying process truly random? For example, the evaluation of the Girl’s Education Challenge in Mozambique, funded by the UK’s Department for International Development (DFID), claimed to be a RCT but upon further investigation and discussions with the evaluators, it became apparent that some non-random elements had crept into the allocation of individuals to treatment and control groups through challenges encountered during fieldwork.