Generic Longitudinal Business Process Model

The intention of this document is to provide a generic model that can serve as the basis for informing discussions across organizations conducting longitudinal data collections, and other data collections repeated across time. The model is not intended to drive implementation directly. Rather, it is primarily intended to serve as a reference model against which implemented processes are mapped, for the purposes of determining where they may be similar to or different from other processes in other organizations. It may also prove useful to those designing new longitudinal studies, providing reminders of steps which may need to be planned. This is a reference model of the process of longitudinal and repeat cross-sectional data collection, describing the activities undertaken and mapping these to their typical inputs and outputs, which would then be described using DDI Lifecycle.  With early roots in the social sciences, this model is grounded in human science. Elements such as anonymizing data (step 5.8 in Figure 5) and managing disclosure risk (step 8.6) relate directly to research on people, whether a biomedical study or a study on political attitudes. The model was developed with longitudinal surveys being the archetypal study type so many of the examples in this paper relate to surveys. Nevertheless, the model described here is intended to be applicable to a wider range of study types. This model should be just as applicable to a longitudinal series of experiments as a survey (see Block et al. 2011). This model is not intended to be comprehensive. It is intended to be descriptive of a generalized view of longitudinal data collection. This model may be extended or specialized to describe specific processes within an organization. Appendix A provides one example of extending this model by incorporating elements from another process model.