Abstract
The Role Construct Repertory Test (reptest; Kelly, 1955) has a longstanding tradition of application and variation within the personal construct literatures. This study examined the impact of different kinds of personal construct examples in the instructional set for eliciting constructs in otherwise identical reptests. Within a sample of 30 university students, results indicated that the use of physical construct examples (e.g., "tall vs. short") resulted in the elicitation of significantly more physical constructs, and significantly lower levels of construct system organization, than did the use of psychological construct examples (e.g., "warm vs. cold"). Results are discussed in relation to the growing literature on the significant impact of sometimes subtle variations in repertory grid procedures.