This paper complements earlier work about incentive effects in wave 5 of the Panel Survey of the Swiss Election Study (Selects) about the possibilities to decrease incentives for high estimated response propensity respondents. In the present paper, we study possibilities to decrease incentives in wave 6 for the complementary group, the low-propensity respondents. For high-propensity respondents, continuing an (expensive) conditional CHF 10 (cash) produced only slightly higher response rates and a similar sample composition compared with an (inexpensive) lottery (5 x 300 CHF) in wave 5. For low-propensity respondents, however, switching to the lottery produced an 8% points lower response …

Little is known about the trade-off between response rates and sample selection on the one hand and costs of different incentives on the other hand in mature online panel surveys. In wave 5 of the Panel Survey of the Swiss Election Study (Selects), a conditional CHF 20 (cash) is used for the politically least interested, while the remaining sample is randomized in two incentive groups: a conditional CHF 10 (cash) and a lottery (5×300 CHF). In the two experimental groups, there are only small differences regarding sample composition, and response rates are only slightly higher in the more expensive cash …

This paper compares the usability of data stemming from probability sampling with data stemming from nonprobability sampling. It develops six research scenarios that differ in their research goals and assumptions about the data generating process. It is shown that inferences from data stemming from nonprobability sampling implies demanding assumptions on the homogeneity of the units being studied. Researchers who are not willing to pose these assumptions are generally better off using data from probability sampling, regardless of the amount of nonresponse. However, even in cases when data from probability sampling is clearly advertised, data stemming from nonprobability sampling may contribute …