„pairfam set new standards“

Karsten Hank is one of the five Principle Investigators for the German Family Panel pairfam and has accompanied the project since 2014. An interview about the start and the special features of pairfam – and about the panel’s future as part of the FReDA research project.

Why was pairfam launched?

For a long time, there was no data infrastructure in Germany that would have made it possible to study family processes longitudinally, and thus be able to analyse the various levels of partnership, parenthood, intergenerational relationships, etc. together and comprehensively.

Moreover, earlier studies usually focused only on very specific aspects, such as the "Mannheim Divorce Study."

In the early 2000s, it became clear that more in-depth research on family processes would quickly reach its limits with the existing datasets. The need for good scientific data was obvious. In the DFG priority programme preceding the long-term pairfam project, 21 projects – at that time virtually the entire German family research community – finally joined forces and launched pairfam with a mini-panel. This set new standards, also internationally.

What is special about pairfam?

Of course, the longitudinal perspective of pairfam with its annual surveys immediately catches the eye. Especially in the phase of early adult life, there are numerous events that happen almost simultaneously: Many people complete their education, they get married, have children – and all this in a comparatively very short period of time. If the intervals between interviews were longer, it would be difficult to work out which event occurred first and what followed. It's great to be able to continue this close interval under FReDA.

The other thing is pairfam's multi-actor design. I cannot think of any other study with such a sophisticated multi-actor design that includes the children and parents of the anchor respondents. In the case of the partner survey, it is important to note that partners not living in the household were also included. This design element of pairfam was also adopted by FReDA.

Overall, we tried to cover as many areas of family life as possible: partnership, parenting, and intergenerational relationships. For me, it is particularly exciting to discover interactions between sub-areas of the family "system" that I would not have suspected at first.

Another special feature of pairfam is, of course, the cohort design. This makes it possible to analyse "critical" phases in the life course of the respondents very well. For example, the transition to adulthood in the youngest cohort, the nuclear family formation age in the middle cohort, and the processes of late parenthood in the oldest cohort. By focusing on selected cohorts, one obtains sufficiently large numbers of cases to be able to investigate life-stage-specific questions with the necessary statistical power. In the past, we could only dream of such a large sample as we have now realized in the first survey wave of FReDA (three times as large as the original pairfam sample)!

How was the scientific response to pairfam?

Very quickly, it became clear that a broad national and international scientific community uses the data. What we are particularly pleased about is that pairfam is now so firmly established in the research landscape: The quality of the data is apparently so convincing that not only German researchers can use it to publish in international journals, but that there are also quite a number of very good international colleagues who say, "Hey, we don't have such data in our country at all – so let's work with the German data!"

After DFG funding ends: What happens next?

The value of panel studies increases with each additional wave. In pairfam, the respondents are now entering a new phase of life: The oldest are now entering the "empty nest phase"; we observe, for example, how the children move out and what effect this has on the parents' partnership or relationship with the children.

This is why it is so enormously important that the pairfam sample was integrated into FReDA and continues to run. Currently, the 15th wave is running for the pairfam sample – now as sub-waves W2a and W2b of FReDA. My great hope is that the large, established pairfam user community will continue to work with the FReDA data. Because that is the great thing about FReDA: We have a very fresh, very large representative sample – and we already have a panel with 14 waves. Plus, the 15th wave, which is in the field right now.

But the continuation of the sample will (hopefully) not only make pairfam users happy. By covering the German Generations and Gender Survey (GGS) in addition to the pairfam survey, FReDA will open up new user groups interested in international comparative questions. So, something completely new is emerging here. That is the cool thing about FReDA: We go one step further.

The community of researchers continues to have a high-quality data source at its disposal. This serves scientific purposes – and, of course, provides policy advice: Only with FReDA can solid statements be made about the situation of families in Germany.

Particularly in the last two years, we have seen what empirical social research can achieve. All the major surveys have managed to respond extremely quickly to the challenge of the COVID-19 pandemic. I have never seen it happen before that there was such an immediate need for social science evidence. I was very pleased that the social sciences were able to make such an important contribution to understanding what was happening in this crisis situation and to provide guidance to policymakers on how to better deal with the challenges facing people.