Effects of a smoking prevention simulation on students’ smoking attitudes
Elizabeth S. Winge
Abstract: The purpose of this study was to test a smoking prevention program’s effectiveness in strengthening fifth grade students’ negative attitudes toward smoking cigarettes. The participants consisted of fifty-five students in three classes. The pre-and post-surveys measured behavioral beliefs, perceived subjective norm and predicted level of self-control concerning a smoking addiction. The results of the study showed that the intervention group shifted in their beliefs, perceived subjective norm and level of self-control to form a stronger intention not to have a cigarette if an older child were to offer them one within the following year.
The Youth Risk Behavior Surveillance System Report, released every two years by the Centers for Disease Control, seeks to identify important trends in the risky behaviors of adolescents. Of the six areas of health risk behavior the report studies, tobacco use has demonstrated the most consistent rise, resulting in a 30 percent increase over the course of the last decade (CDC, 2000). Although rates among this group have been declining since 1997, more than one in four U.S. high school students still smoke cigarettes (MMWR, 2002).
Millions of dollars are allocated to school systems across the United States for the purpose of prevention and intervention (Herrmann & McWhirter, 1997). Applying methods of behavior prediction is of particular importance due to the interventions’ purpose of being effective. Behavior, however, cannot always be realistically or ethically measured. Therefore, constructs involved in the decision-making process of the impending behavior have been identified. Of the most acclaimed models, the theory of planned behavior has successfully predicted a multitude of subsequent behaviors.
The theory of planned behavior is a result of an evolving concept created by researchers Ajzen and Fishbein (Ajzen, 1988). The premise of the model is to specify the predictors of behavior through assessing an individual’s particular attitudes and beliefs regarding the behavior (Fishbein & Ajzen, 1975). Three components are of essential importance: behavioral beliefs, normative beliefs, and control beliefs. In effect, these three components merge to form intention. Intention is the direct antecedent of behavior and equates to an individual’s readiness to perform the behavior. The theory of planned behavior has been widely utilized to predict behaviors with considerable success (Cheung & Chan, 1999).
The curriculum and simulation included in the NICoteen[R] Program fit nicely into the constructs of the theory of planned behavior. The curriculum seeks to evoke thoughtful discussion as it teaches consequences and corrects any prior misgivings. The basis of the smoking simulation is to attempt to give students a sense of what it feels like to lose control of their lives in hopes of providing a form of negative reinforcement. Lepper and Malone (1987) argue that control is an essential part of intrinsic motivation. Humans want to have direct experience with the outcomes of one’s own actions and choices, not merely act as a ‘Pawn’ of external forces. The first activity included in the curriculum uses this concept, as students discuss as a class or in small groups how addictions make a person lose control of their lives. Students are then asked to discuss the characteristics of an addiction in order to gain an understanding of the lack of control it creates.
Day two begins with the “Breathing Activity.” Students insert a drinking straw into their mouths and are asked to hold their noses and breathe solely through the straw. Students then stand and run in place. This provides individuals first hand experience of the physical discomforts smokers feel when they lose the ability to hold enough oxygen in their lungs.
Students then discuss how smoking hinders physical fitness and how the activities in which they enjoy participating would be affected. This discussion leads to the “Smoking Makes Me Sick” activity. Students have different levels of awareness of the health effects associated with smoking. To broaden their awareness, students are asked as a group to think of diseases that can be caused by smoking. Once diseases are listed and understood, the instructor distributes the “Portrait of a Teen Smoker” worksheet. Students are asked to sketch a face of a person and relate the previously listed diseases’ effects. Students specifically indicate some of the characteristics on their portraits. This gives them a very concrete association to illustrate the consequences of smoking.
Day two concludes with handing out the “Interview with a Smoker” worksheet. The purpose of the worksheet is to show how an addiction controls an individual’s life through social proof. Students must interview an adult smoker to find the impact smoking has had on his/her life. The worksheet is to be handed in at the program’s completion.
The program’s third day begins with a brief discussion of the media, its mediums and how they seek to persuade individuals. This information is applied to tobacco companies to help students understand how cigarette advertising is targeted to them. Cartoons, giveaways, and promotional products are analyzed in how they attempt to get teenagers to smoke.
The “Behind Media’s Smoke Screen” worksheet is then distributed. Students are asked to bring newspapers and magazines with smoking advertisements in them. They reference the advertisements while answering the questions on the worksheet. The worksheet brings attention to cigarette advertisements and their intent. Students are confronted with what the advertisements imply and how this is targeted to them. They then share their advertisements along with their findings with the class, specifically concentrating on what the advertisements portray in comparison to reality.
On day four, students are asked to find out if their favorite movie shows people smoking. This is discussed as a precursor to the “Up In Smoke” activity Actual costs of selected brands of cigarettes are written on the blackboard. Students then complete the worksheets, based on the prices provided, to determine the monetary costs of smoking given different brands and amounts of cigarettes smoked. This makes students more aware of immediate costs of smoking cigarettes, which has been demonstrated to be more powerful than long-term health costs (Lynch & Bonnie, 1994). Students then spend class time familiarizing themselves with the operation of the NICoteen[R] Packs.
Beginning day five of the NICoteen[R] Program, students are asked to hand in their signed Student Contracts. The Student Video is played again to refresh the operational components of the simulation. Students take the “Ready to Burn” quiz to answer any last questions, and students are given a journal to record activities during the smoking simulation. Wristbands are then attached to each student’s wrist.
Following the completion of the curriculum, students carry a small interactive device (the NICoteen[R] Pack) that resembles a pack of cigarettes. Students must carry the Pack throughout the weekend. The Pack requires the student to simulate the estimated amount of time an addicted smoker spends smoking each day by subjecting the student to the smoking regimen of an addicted smoker. Upon the Pack’s demand of “a cigarette,” the student must interact with it by touching the identification chip (ID) connected to his/her wristband to the back of the device (this identifies the correct person to the Pack). The Pack may demand “a cigarette” anywhere from ten to sixty times throughout each day, each lasting an average of three minutes. Using an identification chip (ID) and voice recognition technology, students must repeat an anti-smoking message into the Pack for the amount of time it would take to smoke a cigarette. The NICoteen[R] Pack records the student’s interaction during the smoking simulation. When the student returns the Pack to the instructor, the instructor hears a report of the student’s simulation, including the student’s percentage of participation for the simulation. The simulations main purpose is to demonstrate “the loss of control a smoking addiction has on an individual’s life” (NICoteen[R] Program Guide, 2000).
The purpose of the present investigation was to assess whether participating in the NICoteen[R] Program would affect fifth-graders’ attitudes towards cigarette smoking. The attitudes measured specifically centered on behavioral belief, normative beliefs, and control beliefs in accordance with Ajzen’s theory of planned behavior (1988).
It was assumed that students should have general knowledge regarding the risks associated with cigarette smoking, as well as awareness of the existence of pressure from outside individuals for them to smoke. It was expected that differences within the intervention group and between comparison group would depend on the item. It was hypothesized that those individuals that participated in the intervention would record a stronger intent to not try smoking upon completion of the intervention than would the comparison group.
Due to the smoking simulations focus on reaching the loss of control that cigarette smoking has on an individual’s life, the section of the survey that centers on control issues was expected to show the greatest change in responses within the intervention group (item numbers eighteen through twenty-three). It was hypothesized that the NIcoteen[R] Program would give the intervention students a more realistic view of the amount of control the addiction would have over their lives. Therefore, individuals who have participated in the intervention were expected to demonstrate an increase in the amount of control they believe a smoking addiction would have on their lives.
The participants were enrolled in a relatively small public school district located in Wright County, Minnesota. The three fifth grade teachers of the school had interest in the NICoteen[R] Program and volunteered their class time for students to participate in the study. The participants consisted of fifty-five students with a mean age of 134.45 months, or 11.2 years. Of the three classes, only seven did not complete the surveys due to either not participating in the entire program if the individual was in the intervention group or either student or parent refusal to complete the survey in the control group. As a result, the control group consisted of 19 individuals and the intervention group, 36.
A survey was developed by the researcher for the purpose of this study, based on the basic components of Ajzen’s theory of planned behavior (1980). The same instrument was used at both the pre- and post-intervention assessments. Gender, age, and parental smoking behavior were included in the questionnaire. A Cronbach alpha reliability analysis of the instrument yielded a score of .7195. Two items did not meet item-total criterion of .4, and thus were removed from the data analysis.
Items one through nine assessed behavioral beliefs and attitudes toward smoking. Items focused on students’ beliefs on whether they would smoke, along with health and image beliefs.
Normative beliefs were assessed in items ten through thirteen. The influence of parents, friends, advertisements and older, “cool” kids, were included in separate items.
The subjective norm was assessed in items fourteen through seventeen. The same group of influential others identified in items ten through thirteen was listed (parents, friends, cigarette companies, and older, cool kids). The subjective norm seeks to determine how important it is to perform behaviors congruent with what influential others believe. For example, item number fourteen states “I want to do what my parents think I should do.”
Items eighteen through twenty assessed control beliefs. They specifically addressed young people’s underestimation of the addictiveness and control of nicotine. Items twenty-one through twenty-three assessed how important it was for individuals to be in control of their lives as a whole. This included financial control as well as being in control of one’s health.
The three fifth grade classes were randomly placed by class into either the comparison or intervention groups. Teachers in the intervention group were trained before the study began to minimize biases and unify the message of the Program. Consent forms were given to the students to take home to parents/caregivers prior to any mention of the NICoteen[R] Program.
Beginning on the Monday after the consent forms were returned, the six-day NICoteen[R] curriculum was introduced into the classroom. Approximately one hour of class time was set aside each day to perform activities included in the program.
On the weekend of the smoking simulation, participating students brought the NICoteen[R] Packs home until the following Monday. When students returned the Packs to school, the teacher met individually with each student to listen to the read-out of how well the simulation was followed. The day following completion of the entire program, instructors issued the post-intervention survey using the same procedures used for the pre-intervention survey. The control group was given the opportunity to participate in the NIcoteen[R] Program at a later stage.
The intervention assessment was a quasi-experiment with a mixed design. Due to the mixed design, the pre- and post-intervention survey responses were entered as separate variables. Then separate Repeated Measures Analyses of Variance (ANOVAs) were used to test group effects and pre- and post-intervention effects, one for each survey item. The within-subjects factor was time of testing (pre and post) and intervention (present and absent) was the between-subjects factor for each analysis.
When interactions were found within items, paired samples t-tests and independent t-tests were used to determine specific differences. When differences were found between the pre- and post-survey responses of the intervention group, a paired samples t-test that filtered out the comparison group were used. When differences existed between the post-intervention survey responses of the comparison and intervention groups, independent t-tests were used. The Bonferroni test was applied in the specific comparisons tests, due to the large number of comparisons made. Therefore, the probability of error for the specific comparisons was .008.
As expected, there was no main effect for any of the twenty-five items tested. This finding confirmed the hypothesis that there were no significant differences between the groups that were due to factors other than the intervention.
Four of the twenty-three items showed an interaction of some sort, according to Pillai’s Trace, Wilk’s Lambda, Hotelling’s Trace, and Roy’s Largest Root. The interaction for item number one, “If an older kid offers me a cigarette within the next year I will say no,” was significant, F(1,53)=4.308, p=.043. As hypothesized, the specific comparison test showed that the difference from the pre- and post-survey responses in the intervention group was significant, t(35)=2.534, p=.008.
The interaction for item number twelve, “Cigarette companies would like it if I smoked cigarettes,” was also significant, F(1,53)=5.330, p=.025. The specific comparison test showed a significant difference between the pre- and post-survey responses in the intervention group, t(35)=3.162, p=.0015.
The interaction for item number thirteen, “The older, cool kids would like it if I smoked cigarettes,” was also significant, F(1,53)=5.975, p=.018. The specific comparison test showed that the difference between the pre- and post-survey responses approached significance in the intervention group, t(35)=1.641, p=.055.
The interaction for item number twenty-one, “It is important to me to be able to buy the things I like,” was also significant, F(1,53)=3.759, p=.029. The specific comparison test showed that the difference between the pre- and post-survey responses in the intervention group was significant, t(35)=2.788, p=.0045.
Subsequent specific comparison tests showed additional differences between the pre- and post- survey responses of the intervention group that approached significance for items 5, 8, 20, 22, and 23.
As expected, there was no main effect for either the intervention or time of testing factors for any of the twenty-three items. However, the means of the comparison group were consistently higher than those of the intervention group on the pre-intervention survey. This trend can be explained by the teachers’ assessments of their classes. The teachers of the intervention group stated that their classes included students with greater needs. These students have demonstrated fewer skills, greater behavioral problems, and showed more signs of engaging in risky behavior than the students in the comparison group demonstrate.
The within groups differences followed expectations to a large degree. Most of the students tended to be very knowledgeable about the consequences of smoking. The means in the pre-intervention survey suggested little room for improvement in understanding for both comparison and intervention groups.
Consistent with the hypothesis that parents and friends were the most influential others in the students’ decision-making, little difference was found between the pre- and post-intervention responses in these items. Of great importance, however, was the intervention group’s increased agreement in the post-intervention surveys that cigarette companies would like it if they smoked cigarettes. Approaching significance was the intervention group’s post-intervention agreement that older, cool kids would like it if they smoked cigarettes. This implies that the intervention was effective in demonstrating both how adolescents are targeted by cigarette companies and adolescents’ ability to recognize the onset of peer pressure.
Of recondite importance were the findings in items that measured control beliefs. Of the six items, four pre- to post-intervention response differences demonstrated some level of significance in the intervention group. The NICoteen[R] Program’s hypothesized basis for effectiveness centers on demonstrating the loss of control that results from an addiction. The expected effect is that upon experiencing the simulated loss of control, students would be less attracted to experimenting with smoking cigarettes.
On the post-intervention survey, the intervention group agreed more strongly than on the pretest that “if an older kid offered me a cigarette within the next year, I would say no.” This result suggests the program has helped alter the students’ intentions. However students’ intentions to try a cigarette at some point in their life (item number two) did not show significant change following the intervention. Perhaps this is due to the item being more abstract. It seems most effective to explicitly state the specific elements of the assessment items in terms of action, target, context, and time, as suggested by Fishbein and Ajzen (1980).
Due to the small size of the sample, findings can not be generalized to the population as a whole. In addition, the lack of significance between the experimental and control groups at posttest makes the findings more difficult to interpret. The lack of significance could be due to the initial difference between the groups, as previously discussed. Due to the fact that both groups were enrolled in the same school, the novelty of the NICoteen[R] Program may also have provided a carryover effect into the control. But it is also possible that the lack of significant difference was due to a maturation effect rather than solely the influence of the NICoteen[R] Program.
The results were fairly consistent with predictions and seemed to support students’ greater intent to not smoke a cigarette within the next year if offered a cigarette by an older child. It is recommended that future studies include more diversity in age, gender and geographic location in their population.
In addition, further assessments should be longitudinal in order to determine how accurately intentions predict behavior. A study that follows students from fifth through twelfth grade with multiple “refreshers” of the program would provide the dearest assessment of the Program’s effectiveness as a long-term deterrent of cigarette smoking.
Table 1. Significance of Post-Survey Responses of the Intervention
If an older kid offers me a cigarette
within the next year I will say no. t(35)=2.534, p=.008 ***
Being healthy is important to me. t(35)=1.821, p=.0385 *
It would be nice to look more mature. t(35)=-1.972, p=.0285 *
The older, cool kids would like it if
I smoked cigarettes. t(35)=1.641, p=.055 *
Cigarette companies would like it if
I smoked cigarettes. t(35)=3.162, p=.0015 ***
If I started smoking, it would control
my life. t(35)=2.019, p=.0255 *
It is important for me to be able to
buy the things I like. t(35)=2.788, p=.0045 ***
It is important to me to be able to
play the activities I like to play. t(35)=2.127, p=.0205 *
It is important for me to be in control
of my life. t(35)=1.642, p=.055 *
*** Significant interaction
** Marginally significant interaction
* Interaction approaching significance
The author gratefully acknowledges Dr. Helen Swanson for her assistance in conducting this research study.
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting behavior New Jersey: Prentice-Hall, Inc.
Ajzen, I. (1988). Attitude, personality and behavior. Milton Keynes: Open University Press.
Centers for Disease Control (CDC). (2000). Tobacco use among middle and high school students–National Youth Tobacco Survey, 1999. Morbidity and Mortality Weekly Report, 47, 49-53.
Centers for Disease Control (CDC). (2002). Trends in cigarette smoking among high school students–United States, 1999-2001. Morbidity and Mortality Weekly Report, 51(19), 409-412.
Cheung, S., & Chan, D. (1999). Reexamining the theory of planned behavior in understanding recycling. Environment & Behavior, 31, 587-613.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introduction to theory and research. Massachusetts: Addison-Wesley Publishing Company.
Herrman, S. D., & McWhirter, J. J. (1997). Refusal and resistance skills for children and adolescents: A selected review. Journal of Counseling & Development, 75, 177-188.
Lepper, T., & Malone, L. (1987). Intrinsic motivation and instructional effectiveness in computer-based education. In R. E. Smow and M. J. Farr (Eds.) Aptitude, learning and instruction: Conative and affective process analyses. New Jersey: Erlbaum.
Lynch, B. S., & Bonnie, R.J. (Ms.). (1994). Growing up tobacco free. Washington DC: National Academy Press.
NICoteen[R] program guide. (2000). BTIO Educational Products, Inc., Eau Claire, WI.
HEALTH EDUCATION RESPONSIBILITY AND COMPETENCY ADDRESSED
Responsibility IV–Evaluating Effectiveness of Health Education Programs
Competency C–Interpret results of program evaluation.
Sub-competency 3–Report effectiveness of educational programs in achieving proposed objectives.
Elizabeth S. Winge, MS is Research Analyst for Realityworks, Inc. Address all correspondence to Elizabeth S. Winge, MS, 2709 Mondovi Road, Eau Claire, WI 54701, PHONE: 715.858.7189, FAX: 715.830.2066 E-MAIL: Beth.Winge@realityworksinc.com
COPYRIGHT 2003 University of Alabama, Department of Health Sciences
COPYRIGHT 2003 Gale Group