Abstract
A sample of fourth and fifth grade students (N = 317) from two suburban schools with differing socioeconomic characteristics was asked to write essays on what the phrase “an old person” meant to them. Content analysis of the essays revealed categories and response frequencies generally consistent with prior research. However, emphasis in the present study is placed on the implications of using non‐directive methods and on a critique of the variety of biasing and non‐comparable methods commonly used in the field. It is argued that more qualitative research making a minimum of prior assumptions is needed to ensure that attitudes toward the elderly are measured in terms of children's own cognitive constructions rather than those of researchers.