Interviewing institutionalized elders: threats to validity.

In the course of an ethnographic study of an intergenerational Geriatric Remotivation Program in a Southeastern U.S. nursing home, we encountered difficulties in interviewing institutionalized elders that threatened the validity of the data. The purpose of this paper is to explicate the difficulties and to recommend strategies for overcoming them. Four main clusters of elder characteristics threatened the validity of data: (1) physical characteristics; (2) cognitive characteristics; (3) affective characteristics; and (4) personal characteristics. Problematic data were categorized as "insufficient," "unclear," "nice" and "emotionally charged." Strategies for increasing validity when interviewing impaired institutionalized elders included increasing the sample size, returning to the setting frequently, lengthening observation periods, recognizing the value of stories, recognizing the value of socializing, using videotapes, collecting both interview and observation data, and having elders view and respond to videotaped recordings.