The results from our analyses begin to disentangle the complex relationships between several psychosocial risk factors (i.e., gender, acculturation, and sensation seeking tendencies) and two behaviors associated with sensation seeking – a health-compromising behavior (experimenting with cigarettes), and a health-enhancing behavior (being physically active, as defined by meeting physical activity recommendations for adolescents [26
]). In this sample of Mexican origin youth, aged 12–17, just less than one in four youth have experimented with cigarettes and just over one in four were physically active for at least 60 minutes on at least five days a week.
In order to compare the rates of ever smoking cigarettes (i.e. experimenting) and being physically active reported by participants in our sample to national data we focus on those in high school only, among whom rates were 40.3% and 28.9% respectively (data not shown). In 2009, the overall prevalence of ever smoking cigarettes among U.S. Hispanic high school students nationwide was 51.0% and 33.1% of Hispanic high school students reported engaging in at least 60 minutes of physical activity on at least five days a week [27
]. We note that the prevalence rate of ever smoking cigarettes in our sample is somewhat lower than among Hispanics nationwide, while the prevalence of those who engage in at least 60 minutes of physical activity on at least five days a week is comparable to Hispanics nationwide. That said, it is important to note that national data are for Hispanics overall, while our data focus on a subgroup of Hispanics, which might account for differences in reported rates of ever smoking cigarettes.
Overall, a higher proportion of boys have experimented with cigarettes and are physically active for 60 minutes or more on five days or more a week compared to girls. Thus our findings underscore the continued need to develop interventions designed to deter Mexican origin youth, especially boys, from trying cigarettes and from adopting sedentary lifestyles. In addition, boys report higher levels of sensation seeking tendencies than girls for all three subscales. Among girls, as linguistic acculturation increased, sensation seeking tendencies also increased. Among boys, linguistic acculturation was related to social disinhibition only; the most acculturated boys reported being the least inhibited.
Contrary to our first hypothesis, participants who experiment with cigarettes or who meet physical activity recommendations are not more acculturated than their peers who have not experimented or who are less active. This was the case overall and when examined separately by gender (data not shown). However, consistent with our second hypothesis, the gender disparity in experimentation with cigarettes and engagement in physical activity declined as acculturation increased. Among the most acculturated, the odds of experimenting with cigarettes and of being physically active were the same for both genders.
With regards to cigarette use among Hispanic adolescents, researchers have reported negative associations between increased acculturation [13
], bi-cultural stress [32
], as well as experiences of discrimination and cigarette use [33
], but that enculturation – that is, identification with one’s culture of origin, is theorized to have protective influences on smoking behavior among Hispanic youth [34
]. The overall low rate of experimenting among girls underscores the possibility that the majority experience more enculturation than acculturation. Moreover, consistent with our findings, several researchers have noted that the relationship between acculturation and smoking related behavior is stronger among girls and women compared to boys and men [20
]. These results suggest that although additional research is needed to assess potential protective benefits associated with identification with Mexican culture and to understand how the acculturation process varies by gender to impact behavior, gender-specific anti-smoking and pro-exercise messaging might be effective among less acculturated Mexican origin youth.
With regards to physical activity, a closer inspection of the data revealed that those in the medium category of linguistic acculturation had a lower prevalence of participants meeting physical activity recommendations when compared to the lowest and highest categories of linguistic acculturation. Overall many Hispanics in the U.S. are of low SES [36
]; Hispanic adolescents of low SES report lower physical activity levels in school [37
] and reduced access to safe recreational facilities [38
]. In the current study, parental educational attainment is significantly and positively correlated with linguistic acculturation (r
0.01; data not shown). Thus it is possible that the higher proportion of participants meeting physical activity recommendations among the most acculturated reflects both increased participation in school-based physical activity as well as increased disposable income that is used to gain access to gyms or similar fee-for-access community-based opportunities for physical activity.
We further explored the U-shaped relationship between physical activity and acculturation by examining minutes of physical activity in the community separately from minutes of physical activity at school. A higher but equal proportion of participants in the lowest and highest categories of linguistic acculturation reported meeting physical activity recommendations in the community compared to those in the medium group (25% vs. 17%; p
0.05). Among those in the lowest acculturation group, a higher proportion of boys met physical activity requirements compared to girls (35% vs. 15%), which may reflect participation in pick-up community-based sports on the part of the boys. In school, although the proportions meeting physical activity recommendations did not differ significantly by level of acculturation (p
0.17), the proportion of those meeting physical activity recommendations was highest in the highest acculturation group (21%), followed by the lowest acculturation group (18%), and lowest among those in the medium acculturation group (15%). Perhaps the lower rate of physical activity among those in the medium acculturation group reflects that, in terms of acculturation, they are in transition. In other words, these adolescents may not feel sufficiently connected to the school to engage in school-based sports, come from families with insufficient disposable income to gain access to gyms or similar fee-for-access community-based opportunities, and no longer participate in the more traditional community-based pick-up sports due to either lack of interest on their part or lack of opportunity created by their family circumstances.
Consistent with our third hypothesis, we found that higher levels of social disinhibition were associated with experimenting with cigarettes among the least acculturated only; and that higher levels of TAS were associated with meeting physical activity recommendations only among the least acculturated, as well. Perhaps this reflects the point during the acculturation process that attitudes, beliefs and norms that influence behavior begin to change. However, inconsistent with our third hypothesis, positive attitudes towards drugs and alcohol, the last of the 3 sensation seeking subscales we considered, were related to increased cigarette experimentation across all levels of acculturation, but not related to meeting physical activity recommendations, for any level of acculturation. Theory from social psychology posits that attitudes are most strongly associated with behavior when the attitude and behavior are assessed at a similar level of specificity [39
]. Thus, consistent with our results, we would expect that the subscale on the sensation seeking measure we used in this study that assesses attitudes towards drugs and alcohol are associated with tobacco use and not physical activity. Regardless, the overall pattern of results call for further examination of the mechanisms underlying the relationships between acculturation and sensation seeking.
The current study has both strengths and limitations. The participants were from a population-based cohort and included roughly equal numbers of girls and boys. In addition, all covariates were assessed using validated measures, and the data were collected in the participants’ homes using PDAs to ensure their privacy. The high retention rate over 30
months – 87% of the youth provided data at both contacts – is strength of this study. A final strength of the study is the participants, who represent a large ethnically homogenous and predominantly low-income sample of Mexican origin youth, an understudied population. The households in the population-based cohort from which our participants are drawn are representative of the Mexican origin population in Houston, Texas [22
Conversely, a limitation of this study stems from the fact the participants were all of Mexican origin, and therefore the results may not generalize to youth from other ethnic backgrounds, including Hispanics from different countries of origin. Also, the instrument that we used to assess sensation seeking in this study was designed for use in elementary and middle school children while the participants in the current study were attending elementary, middle or high schools. However, we report that the instrument has good internal validity and note that three of the dimensions of sensation seeking assessed by the instrument we used are consistent with items in the Brief Sensation Seeking Scale-4 [40
], which has been reported as a valid measure among Hispanic adolescents aged 12–17 in the identification of adolescents at risk for smoking [41
]. In addition, smoking status in this study was self-reported and unverified by biological samples, and hence were possibly under-reported. However, studies indicate that the validity of self-reported data increases if participants believe they may be asked to provide a biological sample [42
], which was the case in our study. Physical activity data were self-reported; therefore, derived estimates may be subject to reporting bias. This limitation notwithstanding, unlike many previous studies of children, these data were obtained from the participants, using items adapted from the 2005 YRBSS, and not from a parent/guardian proxy [43
]. Although the psychometric properties of this instrument have been previously evaluated in similarly aged children [44
] the reliability and validity of the adapted questions used in this study have not been examined. Future studies utilizing more direct measurements of physical activity might yield different findings. Furthermore, the language-based acculturation measure used assesses language use only rather than a broader range of factors that are influenced by the acculturation process (e.g. values, behaviors), and implicitly assumes that greater English usage is associated with increased orientation to US society and decreased orientation to Latino culture. Finally, the data are cross-sectional; therefore we can simply examine associations between the constructs of interest and cannot establish directionality or causality.