In qualitative research, data saturation refers to the point at which sufficient data have been collected to represent all possible emergent themes within the phenomena under investigation. Despite numerous guidelines, researchers unnecessarily continue collecting additional data or, alternatively, cease data collection prematurely. The aim of this study was to present the complexities associated with determining the point of saturation, using multiple questions, groups and researchers, as well as actual responses. The research was conducted within a leadership educational setting. Responses were solicited from two cohorts of young adult students enrolled in an intensive 4-week leadership programme. Participants in the study originated from 14 different Southern African countries. A qualitative research method was followed. Three open-ended questions, varying in complexity, were posed to participants in two similar cohorts. In Cohort 1 the scope of the themes in the text was independently determined by three researchers through applying summative content analyses. These themes were then tested in Cohort 2, assessing the prevalence of the themes in the control group. Given the themes identified through this process, the researchers independently determined the saturation point by counting the number of cases which needed to be read in order to account for all the observed themes. Saturation in Cohort 1 was reached much earlier than in Cohort 2. The variance in Cohort 1 was found to be dependent on the complexity of the question, while in Cohort 2 the variance was found to be dependent on the researcher identifying additional themes that did not emerge in Cohort 1. It is concluded that saturation was dependent on both the complexity of the question, the group, and the researcher, and saturation was reached at a margin much higher than those proposed in general guidelines. A policy of ‘more is better’, in terms of cases, seems an appropriate approach.

Username
Password