Meanings assigned by undergraduates to frequency statements of condom use.

Studies of condom use must rely upon self-report data which may not be reliable and valid. This investigation examined how 192 undergraduates (17-48 years) interpret response categories (never, rarely, sometimes, most of the time, always) used in some surveys to assess frequency of condom use. Subjects completed a questionnaire that described a scenario of a couple who had engaged in sexual intercourse 20 times during the past 3 months. As part of the survey, there were 21 statements in which the stated number of times condoms were used varied from 0 to 20 (e.g., condoms were used 18 out of the 20 episodes of sexual intercourse). For each statement, subjects were instructed to circle the category they believed best matched the frequency with which condoms had been used: 31% indicated that using condoms 1 out of 20 times was an example of never using condoms; similarly, 23% indicated that using condoms 2 times out of the 20 encounters was an example of never using condoms, 40% indicated that condom use for 19 out of 20 encounters was always using condoms, whereas 23% applied the always label to condom use for 18 out of 20 encounters. These results generally support the validity of this type of condom use measurement, but suggest that caution is needed in interpretation. Implications of these findings are discussed with respect to health messages and research methodology.