“If I studied past projects, then I could identify areas of improved performance, but I’d like an overview of the process before beginning.”
Ok, I’ve got a real fancy pants set of words for you today, “Developmental Research.” Yes, it is a R&D method, but we already are doing this less formally at some time. In fact, that’s the case with most of the methods in this series.
The purpose for developmental research methods are to investigate patterns and sequences of growth and/or change as a function of time. Of course, this becomes an essential method of most project performance improvements initiatives.
Some examples of Developmental Research would include:
Some common characteristics of developmental research include focuses on the study of variables and their development over a period of months or years. It asks, ''What are the patterns of growth, their rates, their directions, their sequences, and the interrelated factors affecting these characteristics?"
The sampling problem in this method is complicated by the limited number of subjects it can follow over the years; any selective factor affecting attrition biases the study. If the threat of attrition is avoided by sampling from a stable population, this introduces unknown biases associated with such populations. Furthermore, once underway, the method does not lend itself to improve¬ments in techniques without losing the continuity of the procedures. Finally, this method requires the continuity of staff and financial support over an extended period and typically is confined to a specific business unit or program that can maintain such an effort.
Cross-sectional studies involving developmental research usually include more subjects, but describe fewer growth factors than single unit studies. While the latter is the only direct method of studying project team development, the cross-sectional approach is less expensive and faster since the actual passage of time is eliminated by sampling different project managers across organizations.
Sampling in the cross-sectional method is complicated because the same project managers are not involved with the same projects and may not be comparable. To generalize intrinsic developmental patterns from these sequential samples of projects runs the risk of confusing differences due to development with other differences between the groups that are artifacts of the sampling process.
Trend studies are vulnerable to unpredictable factors that modify or invalidate trends based on the past. In general, long-range prediction is an educated guess while short-range prediction is more reliable and valid.
The steps for conducting developmental research are: