This article argues that scanned information exposure is a useful concept, offers guidance about its measurement, and provides evidence that these measures are valid indicators of the construct. The proposed measures have good convergent validity–scanners about one behavior also tend to scan about others (supporting H1)–and discriminant validity, distinguishing scanning from both general media use (supporting H2) and seeking exposure in terms of mean correlations and relative frequency (supporting H3, H4). The measures also appear to have reasonable nomological validity, being associated with the volume of media coverage for at least some behaviors (partially supporting H5) and five out of six of the relevant behaviors (partially supporting H6) and showing substantial test–retest association over 1 year (supporting H7). Other evidence establishing scanned exposure as a construct discrete from sought exposure includes findings that it occurs more prevalently. For all of the six behaviors, scanning occurred more often than seeking. This is logical, as scanning does not require the same level of motivation or purpose as does seeking and can occur in the normal course of media use.
Explanations for Partially Supported Hypotheses
Overall weekly news coverage volume was associated with scanned information exposure about only mammography and colonoscopy. Tests of H5 may have been constrained by a lack of weekly variation in news coverage during the observation period. Major spikes in coverage, catalyzed by news events, may be necessary to ensure sufficient weekly variation in coverage. There are two major news events that were associated with large spikes in coverage: Breast Cancer Awareness Month (October) and Colon Cancer Awareness Month (March). Mammography coverage was much higher in October relative to other months (t test p <0.001), while colonoscopy coverage was much higher in March (t test p <0.001). Since variation in the independent variable is a necessary condition for detecting associations with another variable, a lack of weekly variation in news coverage about the other behaviors may have constrained our ability to detect a significant association.
Scanning information was related to having engaged in five of the six behaviors but not mammography. The large majority of women had engaged in this behavior within the last 2 years (71%). This is a behavior that is widely accepted and has good compliance, compared with the others. It may be that the behavior is so institutionalized that scanned information exposure does not have much influence.
The scanned exposure measures were limited in quantity. Due to time and space constraints on the survey we were able to ask only about the three screening behaviors (two for males and two for females) and three prevention activities. In addition, we were limited in the number of sources about which we could ask. Family and friends were the only interpersonal source besides the doctor, and they were included in the same question. Future analyses might explore differences between scanning from family members versus friends, as well as others who may be less closely tied to the respondent, such as coworkers or acquaintances.
Recall is always an issue when dealing with self-report of any type of information exposure. Logically, it may be easier to remember a topic about which you actively engaged in a search than one about which you came across information incidentally. The active search was likely predicated by some occurrence or interest in the topic that led to the need or desire for additional information. Such an occurrence may enhance recall.