Definition:  Qualitative and quantitative assessments of a person or group by another person about an educational or psychological construct.

            Description.  If self-reports are subject to distortion, an obvious avenue to pursue is raters who have some experience in gathering information and who do not share the biases of test-takers.  Thus, the interview is the most commonly employed method outside of self-reports.

Interviews have been referred to as a conversation with a purpose.  Interviews can be categorized according to their degree of structure.  Structure here refers to an interviewer’s predetermination of such elements as the information to be obtained, order of questions, coding of questions and answer, and guidelines for probing responses.  Research suggests that the addition of structure to interviews often improves their reliability and validity (e.g., Conway, Jako, & Goodman, 1995).  In the realm of employment interviews, Wright, Lichtenfels, and Pursell (1989) maintain that such structured interviews work well because they:  (a) are closely based on a job analysis of the employment position, thus reducing error from information irrelevant to the specific job; (b) assess individuals’ work intentions, which are often linked to work behavior; and (c) use the same set of questions and standards for scoring answers, thereby increasing reliability.

Hoshmand (1994) summarized another set of guidelines for interviewers.  She suggested, for example, that interviewers need to manage interviewees’ anxiety so as to facilitate communication.  Open questions that require elaboration (e.g., “Tell me more about that experience”) produce better information than closed questions that produce one- or two-word questions (e.g., “Were you satisfied with that job?”).

Structured interviews begin with a set of items or questions that the interviewer poses to the participant.  For example, Hood and Johnson (1991) described the SAD PERSONS scale, developed by Patterson et al. (1983) to assess suicide risk.  With relevant training, researchers interested in suicide could assess risk using the following interview questions:

S ex (Males more likely to commit suicide)

A ge (Persons under 25 or over 45 more likely)

D epression

P revious attempts

E thanol abuse

R ational thinking loss

S ocial support loss

N o spouse

S ickness

One risk point is awarded for each of these 10 risk factors.  Particularly with factors such as depression and rational thinking loss, interviewers would probe beyond an initial question before making a yes/no judgment.