4. Method
4.1. Participants
We conducted a survey among 3531 participants (42% female)
consisting of both customers and non-customers, all residing in The
Netherlands. The age distribution of our sample was 55: 33%. To be classified
as “customer” in this study, a participant must have used the ser-
vices (for private or business use) of KLM in the two years before
the start of the study. With this definition, 1912 participants (54% of
N) were classified as “customer”. In order to ensure that we had
enough customers in the sample and a variety in social media usage
and online engagement in the company's social media activities, we
determined the sample size and composition in advance. The par-
ticipants were selected in four different ways, which resulted in
four subsamples. In all these subsamples, there were KLM cus-
tomers based on the customer criterion described above. To be able
to test our research hypotheses H3 and H4, the total sample was
categorized as customers or non-customers (54% versus 46% of N).
After data collection, we did not exclude any of the participants.
The goal of the first and largest subsample was to provide us
with a general and representative picture of the (adult) Dutch
population. This sample was provided by the Dutch market
research firm Motivaction, which administers an online panel of
20,000 members, representative for the adult Dutch population.
The members of this panel regularly participate in market research
on a voluntary basis, often on behalf of companies. The participants
from this subsample were rewarded for taking part in this survey,
as a regular agreement of their membership of the panel. A total of
6650 members from the panel were invited by e-mail and, with a
hyperlink, directed to the online questionnaire. This led to a
response of 31% (n 1⁄4 2077; 59% of total N), of which 538 partici-
pants (26% of this subsample) were classified as “customer” based
on the criterion described above.