Abstract: This case study is part of a broader investigation into the effects word choice and context have on survey outcomes in interview-based research – and the risk they pose to researchers who, if unaware, may gather unrelated data in respect to their research aims. That is what I call ‘semantic surprises’: poor question wording and superficial context analysis create a gap between the questioner's intention and the interviewee's answer. On the one hand, the inherent complexity of the multi-faceted concept of context calls on researchers to clearly map the sociological coordinates of the research field so that they can avoid asking inappropriate or irrelevant questions. Here contextual effects are shown to change the way questions are interpreted by interviewees, leading them to answer different questions to the ones the researcher intended. On the other hand, questions that are convoluted, ambiguous or overly technical are shown to cause problems surrounding interpretation for both the interviewer and interviewee. The impact of word choice and contextual effects on questions is largely the same: the question's basic structure remains the same but is interpreted differently by the interviewees and researchers. The subject is illustrated using examples taken from my own experience in empirical research, as well as those of respected social scientists.