An exposition is made of a method to elicit students' comprehension of the meaning of words commonly used in science, which may be influential in determining what they learn. The method's aim is to elucidate features in students' understanding of childrens' science (that found before formal teaching of science), student science (that found after formal teaching of science) and scientists' science (the consensus found among the community of scientists). The Interviewing-About-Instances Technique, which invites the subject to classify examples depicted as line drawings on cards, as instances or non-instances of a word, e.g. work, electric current, force, living, and to explain the reasons for each classification, is outlined. Guidelines for the planning of such a method and the designing of materials are given. The interview technique is based on a non-evaluative researcher orientation and seeks an elaborative reflective approach which seeks flexibility in use across age groups, in allowing reference back to earlier discussion, and in the analysis of unexpected comments. Techniques for the transcription, condensation and analysis of data, and for checking the representativeness of data are described. Also presented are outcomes of using the technique, illustrated by examples. (Author/AEF)