Comparing responses in repeated cross-sectional and panel studies: Results across eight weeks during the first COVID-19 lockdown in Denmark

It is often important to study people's thoughts, feelings, and behaviors over time. To this end, researchers have relied on repeated cross-sectional (RCS) studies, in which different people from the same population participate on different measurement occasions. Also, researchers have relied on panel studies, in which the same group of people participate on different measurement occasions. However, few studies have directly tested whether participants' responses in RCS studies were similar to those found in panel studies. To address this gap, we compared the responses to 33 items, 28 of which were further grouped into four aggregates (Affections, Worries, States, Health concerns), over 8 weeks during the first COVID-19 lockdown in Denmark in a RCS study (overall N = 5,993, per measurement occasion 616

≤

n

≤

964)

with the responses in a largely equivalent panel study (993

≤

n

≤

2,546 across measurement occasions).

The study participants were randomly drawn from the same quota-representative participant pool and

responded to the same items on the same measurement occasions. Results indicated a few differences

between the study samples on the

fi

rst measurement occasion (i.e., selection effects between studies).

Further, we found statistical support for different trajectories in 21 aggregates/items. However, visual

inspection of the trajectories suggested subtle differences between the studies at large. The results thus raise

awareness that the trajectories of people

’

s thoughts, feelings, and behaviors can differ between survey

methods, especially when only a few measurement occasions are considered. Nevertheless, such differences

might not be substantial over time.