The 1448 participants ranged from grades 4-6, with approximately one-half of the sample (48.6%) girls (). Nearly two-thirds (63.2%) of the tweens lived in a rural setting, with the majority (61.9%) reporting three or more TV sets per home. Approximately two-thirds (62.6%) of the tweens in this study reported having a television in their bedroom, with 18.6% of those also reporting cable and/or satellite reception. Nearly two-thirds (64.3%) of the respondents reported watching ≤ two hours of TV per day during the school week, with no differences in viewing frequency by either gender (), or by school location (χ2=0.01; P=0.96). Boys were more likely than girls (P<0.001) to have a TV with cable or satellite reception accessible in their bedroom. Additionally, 65.3% of boys reported a TV in their bedroom with or without cable or satellite, compared to 59.7% of girls with or without cable or satellite (χ2=4.95; P=0.03). More tweens reported their parents were routinely involved in regulating the timing of snacks (64.5%) versus the content (48.8%), in which parents/guardians told them which snacks they “had to eat/drink.” There were no respective differences in either “low” or “high” parental involvement of snack timing or content by gender ().
Characteristics of tweens by gender.
Snacking frequency did not differ by gender, with greater than two-thirds of respondents (69.2%) reporting they snacked “sometimes or always” with TV (). However, those that reported watching more TV (n=514; 35.7%) also reported snacking “sometimes or always” more often (n=407; 78.7% versus n=595; 59.4%), χ2=34.22; P<0.001. Overall, the sample reported a total of 873 (62.9%) food and 514 (37.1%) beverage snacks. Nearly two-thirds of tweens (n=934; 64.5%) did not report drinking beverages while watching TV. However, among those who indicated consuming beverages (n=514), girls reported drinking more juice (P=0.02), whereas boys reported drinking sugar-sweetened beverages (SSB) more frequently (P=0.006). Although girls reported drinking low or no calorie beverages more frequently than boys (41.8% versus 35% respectively), there was no statistical difference (P=0.11). Milk intake did not vary by gender, and comprised 14.8% of all beverages in total. However, the most nutrient dense choices, milk and juice, made up less than one-quarter (24.1%) of the total beverages consumed ().
Favorite snack beverage categories of tweens by girls (pink/gray) and boys (blue/black) while watching TV (n = 514). SSB=Sugar sweetened beverage. * Chi-square P value < 0.05. ** Chi-square P value < 0.01
By far, both boys (49%) and girls (46.6%) reported consuming salty snacks most often (n=418), including chips, pretzels, and popcorn (). The second most popular category included a range of fruits and vegetables (n=161), and girls (22.6%) reported fruit or vegetable consumption more often than boys (14.7%), P=0.003. Other than desserts/sweet snacks for boys (10.7%), all remaining food categories for both genders were less than 10% of the total. Quick-serve restaurant (ie, fast-food) responses were negligible (n=3). Boys reported eating significantly more meats (P=0.03) and desserts (P=0.04) than girls. Girls, on the other hand, reported snacking on ”other foods” (i.e., mainly dinner, or not otherwise specified), P=0.02, as compared to boys.
Favorite snack food categories of tweens by girls (pink/gray) and boys (blue/black) while watching TV (n = 873). * Chi-square P value < 0.05. ** Chi-square P value < 0.01
Results from this survey show snacking prevalence “sometimes” or “always” while watching TV to be high (69.2%), and that those watching more TV are more likely to report snacking more frequently. Tween girls tended to select nutrient-dense, lower-energy foods (ie, more fruits and vegetables) more often than boys for snacking while watching TV. However, both genders favored salty food snacks over all other categories. Similar to adolescent reports (5
), tweens reported selecting salty snacks over fruits and vegetables nearly three-to-one, and a staggering 32-to-one of salty snacks to dairy products, like yogurt or cheese. For adolescents in 2005-2006, the top-rated snacks of grain, vegetable, and oils groups were tortilla, corn chips, and/or potato chips respectively, with carbonated soft drinks leading the groups of discretionary calories and added sugars (5
). Of note, boys in this sample also reported drinking SSBs more often than girls, with girls consuming juice more often. These results are consistent with other research reporting adolescent boys drink more soft drinks than girls (23
), and that females consume more juice than males (35
). There were no differences between stated parental rules surrounding snacking with TV by gender.
Although the preference of nutrient-poor, energy-dense foods and beverages of children and adolescents is consistent with other research (4
), this study offers hope that some tweens might be open to interventions encouraging healthful eating, given that fruits and vegetables were a distant second in the food category. These results align with Caine-Bish and Scheule’s study (25
) that found girls tended to select fruits and vegetables at school, and boys were more likely to report choosing meat products. In this study, nine students selected meat or nut products (); boys reported 100% (n=8) of the meat products, with the remaining answer “nuts” (n=1) reported by a girl.
Differences in snacking behaviors could be attributed to a number of possibilities. Recently, experts concluded that food and beverage advertising alters children’s dietary preferences and choices (22
), which may affect boys and girls differently. Anschultz, Engels, and Van Strien recently (12
) found that boys 8-12 years old consumed more food when exposed to these cues on commercials, as compared to non-food commercials and female peers. Another possibility could be that pre-adolescent girls perceive worse health than their male peers, which could lead them towards positive lifestyle changes earlier (36
). Yet another possibility relates to the prevalence of dieting (i.e., restrictive eating patterns) and negative eating attitudes in pre-adolescent females. Previous research reports the prevalence of female tween dieting ranges anywhere in between 30-50% (37
), with age-related increasing trends. Although restrictive eating patterns, including vegetarianism, have been associated with increased fruit and vegetable consumption, they have also been correlated with an increased risk of disordered eating behaviors, including binging and unhealthful weight control activities (40
). Thus, care needs to be taken in promoting healthful dietary patterns to emphasize a healthy lifestyle instead of an ideal image or weight. Finally, research suggests that for both genders, eating in front of a TV may minimize physiological cues that signal satiety to the individual’s body, which could influence discretionary intake and increase risk for obesity (41
The present study should be considered in the context of limitations. First of all, although our survey measures contain face validity from focus groups, they have not been measured against a gold standard, given the lack of research on tween segments. Media questions were limited to TV and did not include cell phones, computers, or other media devices. Because TV viewing frequency estimates did not take into account these other non-traditional media platforms, the frequency of watching TV may be underestimated. Secondly, generalizability may be limited due to the regional population sampled. Although rural oversampling was representative in this area, this would not extend to all geographic locations. Future research should encompass a wider scope of children, including body mass indices, from different geographic regions. We also rely on the accuracy of self-reported measures and perceptions, which lend themselves to memory biases and may interject errors of measurement (43
). For example, we were unable to discern whether “juice” responses were 100% juice or otherwise, or the accuracy of tween perceptions regarding parental involvement. However, other researchers use similar measures to garner information from youth (9
), and although social desirability may bias responses, answers may be more truthful because total dietary recall or intake was not tallied (45
). Additionally, we did not assess socioeconomic status, which has been shown to correlate to food accessibility and dietary quality (2
). However, we surveyed students in public schools, with greater than 99% participation, suggesting that we captured the general population within these areas. Lastly, amounts or portion sizes of snacks were not collected, and specific nutritional information could not be calculated. Therefore, “nutrient-poor, energy-dense” assumptions come from general categories of foods and beverages. Future research should include quantitative measurements of snacks in the home to determine the absolute and relative contribution of intake during TV viewing, as compared to overall eating patterns in context of parental regulation. Also, future studies should examine the frequency and amount of food intake based on TV food advertising exposure (i.e., amount, type, and timing) in the tween population. Strengths of this study include a large rural sample size of the tween population, as well as snacking habits in conjunction with TV viewing.