Panel study drop-out

Panel studies enable us to monitor people over a long period and measure any changes in them. Data from panel studies is extremely valuable, but there is one major drawback: drop-outs. We are using statistical analyses to investigate why people stop participating in panel studies.

Valuable data

If panel studies constitute a representative reflection of the population, you can use them to measure different changes:

  • at individual level (for example: what kind of unemployed people quickly find a new job?)
  • at group level, for the population as a whole (for example: is unemployment increasing or decreasing?

Why is drop-out such an issue?

The people who stop participating in a panel study are usually specific groups. In medicine, these are often people with a poorer state of health, and in the social sciences, those who often move house, are young or have less interest in social problems. If increasing numbers of this type of people drop out, we are no longer able to provide good answers as to why change happens using data from these panel studies.

We are researching why people drop out of panel studies and how we could prevent this. Objective: obtain more and better data for measuring changes.

Understanding why people drop out

Although we know the type of people who drop out, we do not know exactly why this happens and how we could prevent it. What is clear is that the motives for participating in panel studies or stopping doing so vary from person to person. In order to chart these motives, we are conducting statistical analyses to distinguish unobserved groups ('latent classes') from people who differ in their motives.

Further research

In further research, we are trying to design a different approach strategy for each of the classes. For example, some people like to have a lot of contact with the research and information about the results, whereas others will only participate in research if it is conducted on the internet (or if it is not).

Associate Professor