Background Institutes for mental health care consider scientific research an important activity. A good way to stimulate research is by simplifying data collection. Creating a minimal data set for research purposes would be one way to achieve this, however, this would only be possible if the researchers use a limited variety of data types. This article will address the question whether or not this is the case. Methods Researchers working in Dutch mental health institutes were approached and asked to complete an internet questionnaire on the individual variables they collected for, and measurement instruments used in, their studies. Results In the 92 studies described by the researchers, 124 different variables were collected, and 223 different instruments were used. A total of 66% of the variables and 73% of the instruments were only used in one study. Conclusion There is little commonality among research data, hence flexibility will be a crucial factor in facilitating data collection for research in mental health institutes. Nevertheless, reducing the variety of variables and instruments used is important to increase the comparability of results.