In 1900, people ages 65 years and older comprised only 4% of the population of the United States. By 2000, their share of the population had tripled to 12%, and is predicted to reach 20% by 2050. Nearly 75% of today’s adults over 65 say they are in good to excellent health, 42% report they have attended some college and the median income is $33,848. Older Americans are living active lives in which many still hold careers and participate in community and religious groups; some seniors are even caring for older parents. According to research by the AAA Foundation for Traffic Safety, 86% of Americans ages 65 and older still drive, meaning that there are presently approximately 36 million drivers ages 65 and older. If current population projections and licensure rates hold, this number will grow to 48 million by 2020. Drivers ages 65 and older drove an estimated 219 billion miles in the one-year period from May 2008 through April 2009. As older people remain healthier for longer, driving will undoubtedly continue to play a major role in their ability to stay mobile, independent, and engaged in their community. Hence, there is a need to better understand and delineate the changes that occur in the attitudes and perceptions of older drivers as they continue to age. The majority of previous studies have treated drivers aged 65 and older as a single group. However, “older senior drivers” (e.g., ages 80 and older) might differ in important ways from “younger senior drivers.” The purpose of the research reported here was to examine the extent and nature of the variability in driving behaviours and safety-related attitudes among drivers ages 65-69, 70-74, and 75+, using data from the AAA Foundation’s annual Traffic Safety Culture Index survey. The data analysed for this study were from the AAA Foundation’s annual Traffic Safety Culture Index survey. The survey comprises questions about Americans’ attitudes about traffic safety issues, social norms, and self-reported driving-related behaviours. Each year, it is administered online to a sample of U.S. residents aged 16 years and older who were enrolled in GfK’s KnowledgePanel®, a nationally representative sample of members of U.S. households recruited by standard probability-based survey methods (address-based mail and random-digit dial telephone sampling). If a recruited household lacks Internet access, GfK provides an Internet connection and hardware at no cost to the household. The current study analysed data from the 2011, 2012,and 2013 Traffic Safety Culture Index surveys, administered June 6 — 28, 2011, September 7 — 24, 2012, and September 18 — October 3, 2013, respectively. Data were weighted to account for differences in individual respondents’ probabilities of a person being recruited into KnowledgePanel®, differences in respondents’ probabilities of being selected for these specific surveys, and non-response at both stages. Post-stratification weights aligned the demographics of the sample to that of the U.S. population with respect to age, gender, race and ethnicity, education, census region, urban versus rural residence, household size, and household income. Respondents were asked questions related to the extent that they believed specific driving behaviours of other drivers were a threat to their (the respondent’s) personal safety. Response categories were very serious threat, somewhat serious threat, minor threat, and not a threat; these were dichotomized as very or somewhat serious threat versus minor threat or not a threat. To assess social norms, respondents were asked whether they considered it to be completely acceptable, somewhat acceptable, somewhat unacceptable, or completely unacceptable for a driver to perform a variety of specific behaviours; these were dichotomized as acceptable versus unacceptable. In questions related to support for a number of specific traffic safety interventions, respondents were asked whether they would strongly support, somewhat support, somewhat oppose, or strongly oppose each one; these were dichotomized as support versus oppose. Respondents who reported having driven in the past 30 days were asked how often they had engaged in several specific behaviours; response options were regularly, fairly often, rarely, just once, and never. These data are displayed collapsed into 3 categories for brevity but were data were treated as ordinal and were not dichotomized or otherwise recoded. (These were collapsed into 3 categories for brevity in Table 5; however, analysis was based on all five categories.) Number of moving violations over the last 2 years was coded into the categories: 0, 1, and 2 or more. For crashes, the categories were: 0, 1, 2, and 3 or more. This was done to ensure adequate numbers of observations in all categories, as very few respondents reported more than 2 violations or more than 3 crashes. One observation, in which a respondent reported having zero moving violations but 9 crashes in the past 2 years (18 standard deviations from the mean number of crashes in the full sample), was suspected to have been a coding error and was recoded as missing. Respondents were coded as living in a metropolitan ornon-metro area according to the United States Office of Management and Budget classification of the ZIP code of the respondent’s home address. (Author/publisher) For more information on the LongROAD Study, see http://www.longroadstudy.org/
Samenvatting