Ideally, the decision will be based more on the notion of saturation. Saturation occurs when continued data collection produces no new information or insights into the phenomenon you are studying. For example, as you continue to interview you begin to realize that you are hearing the same things you've heard earlier; no new information is forthcoming, or some tidbit is relatively minor in comparison to the effort spent collecting the information. Also, the ongoing analysis of your data has produced categories, themes, or findings robust enough to cover what emerges in later data collection