Developing the Questionnaire and Collecting
Data
To assess the adequacy of the remaining items, we constructed
a questionnaire that directed respondents to think
about an emotionally charged service encounter they had
experienced and to respond to questionnaire items about that
specific encounter (we refer to this sample as the ‘‘initial sample’’).
We adopted the common practice of using convenience
samples (e.g., Menon and Dube´ 2004), which featured respondents
from two populations: 144 questionnaires from college
students and 167 from staff members of a business school at a
Belgian university. After removing unusable questionnaires,
we retained 112 questionnaires from the students and 135
from the staff members (n ¼ 247). The mean age of the
respondents was 34 years, and 66% were women. On average,
the reported incident occurred 1.5 years before our study took
place. Six sectors accounted for 75% of the reported critical
incidents: medical services (29%), retailing (16%), public services
(10%), home repair services (8%), hotels/restaurants
(7%), and banking/insurance services (5%). Finally, face-toface
interactions were the most frequent type of communication
represented (82%), compared with voice-to-voice (17%)
or electronic (1%) interactions