Health Education Research Advance Access originally published online on May 20, 2004
Health Education Research 2004 19(5):551-560; doi:10.1093/her/cyg077
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Health Education Research Vol.19 no.5, © Oxford University Press 2004; All rights reserved
Smoking initiation among Gambian adolescents: social cognitive influences and the effect of cigarette sampling
1 Department of Health Education and Health Promotion, Universiteit Maastricht, P. Debijeplein 1, PO Box 616, 6200 MD Maastricht, The Netherlands, 2 School of Psychology, Open Universiteit Nederland, Postbus 2960, 6401 DL Heerlen, The Netherlands and 3 PO Box 81, Banjul, The Gambia
4 Correspondence to: S. P. J. Kremers; E-mail: s.kremers{at}gvo.unimaas.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionReferences |
|---|
In this study, determinants of tobacco use among Gambian youths were examined. A cross-sectional school-based survey was conducted among 282 students, aged 14–18. The potential determinants were derived from the core constructs of the Theory of Planned Behavior and the ASE model. Moreover, various measures of knowledge and media influence were added to the research model. The results show that free cigarette offers by representatives of tobacco companies were mostly responsible for the difference between smokers and non-smokers. Other variables associated with smoking behavior were greater intention to smoke in the future, lower self-efficacy expectations regarding emotional situations, smoking behavior of the respondent's best friend, the mother having a job and the absence of other family members living in the house. Our study supports current efforts to prohibit tobacco advertising and promotions in The Gambia, especially the distribution of free samples. Additionally, we recommend prevention activities in schools and communities to aim at making children less vulnerable to ‘sampling’, increasing their self-efficacy expectations regarding emotional situations and making them aware of the influence of their best friend.
In this study, determinants of tobacco use among Gambian youths were examined. A cross-sectional school-based survey was conducted among 282 students, aged 14–18. The potential determinants were derived from the core constructs of the Theory of Planned Behavior and the ASE model. Moreover, various measures of knowledge and media influence were added to the research model. The results show that free cigarette offers by representatives of tobacco companies were mostly responsible for the difference between smokers and non-smokers. Other variables associated with smoking behavior were greater intention to smoke in the future, lower self-efficacy expectations regarding emotional situations, smoking behavior of the respondent's best friend, the mother having a job and the absence of other family members living in the house. Our study supports current efforts to prohibit tobacco advertising and promotions in The Gambia, especially the distribution of free samples. Additionally, we recommend prevention activities in schools and communities to aim at making children less vulnerable to ‘sampling’, increasing their self-efficacy expectations regarding emotional situations and making them aware of the influence of their best friend.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Smoking prevalence in Africa is relatively low, with tobacco use being more common among men than among women. The estimated smoking prevalence in Africa for adults aged 15 years and older is 29% for men and 4% for women (Corrao et al., 2000
). However, a steady and rapid increase in the proportion of smokers amongst adolescents and women is taking place in Africa (WHO, 2000a). Current trends show that only time separates Africa from the rest of the world, with the African tobacco epidemic expected to peak in the middle of the 21st century (Chapman and Leng, 1990). According to WHO, ‘smoking diseases will appear in developing countries before communicable diseases and malnutrition have been controlled and the gap between rich and poor countries will thus be further expanded’ (WHO, 1983; Chapman and Leng, 1990).
Increased smoking rates in Africa are caused by a combination of westernization, urbanization and increased disposable income, while other causes include the marketing and promotion strategies of the tobacco industry (Yach, 1996). While the advent of stricter control measures and increased awareness of smoking-related health problems have led to a reduction in tobacco consumption in developed countries, the opposite is happening in developing countries. Tobacco companies are turning to developing countries to restore their profitability.
Africa is regarded as a virgin territory with great potential for tobacco marketers (Corrao et al., 2000). Awareness of tobacco's effect on health is generally poor in developing countries and difficult to raise because of the high levels of illiteracy. Moreover, most African countries barely have any legislative control measures in place (Chapman and Leng, 1990), making people more vulnerable to sales efforts (Yach, 1996). Tobacco use among young people continues to rise, as tobacco companies aggressively promote their products to a new generation of potential smokers (Pucci and Siegel, 1999). Tobacco companies are promoting cigarettes through every thinkable medium, including magazines and newspapers, television and radio, billboards, and the Internet. Two other popular examples of promotion are brand stretching and cigarette sampling. Brand stretching refers to selling or giving away non-tobacco products that carry a tobacco brand name. Cigarette brand names are being printed on T-shirts, hats, rucksacks and other merchandise popular with adolescents, turning the wearers into walking billboards. Cigarette sampling involves giving away free cigarettes in areas where young people gather, such as rock concerts, discos and shopping malls (WHO, 2000b).
The fact that many developing countries have no legislation on tobacco control and have not undertaken any specific public education or information programmes on this subject makes it easier for tobacco companies to penetrate the market. The most common form of tobacco regulation concerns smoking in public places, especially schools and health care facilities. About 20 African countries have such regulations, but enforcement mechanisms rarely exist (Corrao et al., 2000). Tobacco products are also actively promoted in The Gambia, with little or no restrictions (Centers for Disease Control and Prevention, 2001). There is no ban on selling cigarettes to children under 16 years (Corrao et al., 2000). Currently, the only legislation on tobacco in The Gambia is the ‘Prohibition of Smoking (Public Places) Act, 1998’. This act was introduced in 1998 by one of the Members of Parliament. The act prohibits smoking in any public place, workplace, hospital, public vehicles or government premises, although it lacks sanctions or other instruments to enforce the rules. Consequently, little has been done to enforce the act.
Generally speaking, two types of interventions can be designed to tackle the tobacco epidemic—those aimed at getting people to quit smoking and those aimed at the prevention of smoking initiation. Systematically designing prevention programmes requires the determinants of smoking initiation to be examined. The present study aimed to explore the determinants of smoking initiation in The Gambia. The potential determinants were derived from the core constructs of the Theory of Planned Behavior (Ajzen, 1988) and the ASE model (De Vries et al., 1988), which have been shown to be associated with smoking initiation [for reviews, see (Conrad et al., 1992; Tyas and Pederson, 1998)]. Since we suspected that lack of knowledge and media influence would play a role in taking up smoking in The Gambia, we added various measures of knowledge and media influence to the research model.
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Subjects
The Gambia is divided into five divisions and two municipalities, including a total of 104 secondary schools (Jallow, 2001
). We attempted to obtain representative data by selecting one school from every division/municipality, resulting in a total of seven schools. The sample included four government, one private and two Roman Catholic schools (Latrikunda, Farafenni, Bansang, Kwinella, Banjul, Basse and Lamin), thereby representing the distribution of schools in The Gambia. Every school that was approached agreed to take part and all students agreed to take part in the study. One class per school was selected at random, with the average of 40 students per class. This resulted in a sample of 282 Gambian students, aged 14–18.
Questionnaire
Data were collected at schools, using a self-administered questionnaire. The teacher left the classroom after introducing the researcher, who explained that students did not have to write their names on the questionnaires and assured strict confidentiality of the responses. Consequently, the questionnaires were distributed in the class by the researcher. Most of the questions were based on the ESFA questionnaire (European Smoking prevention Framework Approach) (De Vries et al., 2003). The content of the ESFA questionnaire is based on a review of the literature and on work regarding adolescent smoking behavior done over the last 15 years (De Vries and Kok, 1986; De Vries et al., 1988; De Vries and Backbier, 1994; Dijkstra et al., 1999; Kremers et al., 2001). The questionnaire included items regarding demographic and cognitive characteristics. Media and knowledge items were based on the Global Youths Tobacco Survey (GYTS), which has been developed by WHO and Centers for Disease Control and Prevention (Centers for Disease Control and Prevention, 2002). In order to increase the reliability, two pre-test methods have been used. First, health education experts in The Gambia reviewed the questionnaire. Second, the questionnaire was completed by small groups of Gambian adolescents, both in a rural and an urban area, using the plus-and-minus, or Content Response Code, method (Vroom, 1994). Respondents were asked to complete the questionnaire, while adding a plus or minus to questions they thought were good/bad or clear/unclear. When the questionnaires were completed, focus group discussions took place to clarify and discuss the meaning of the plusses and minuses. The pre-test resulted in alterations of specific words and phrases in the questionnaire.
Smoking status was measured as the dependent variable and was divided into four categories: weekly (3), monthly (2), less than monthly (1) and never smokers (0) [for more information, see (Kremers et al., 2001)]. Attitude was measured with 13 questions on a seven-point scale. For example: If I smoke (or were to smoke) it will make me feel relaxed (3) to it will make me very stressed (–3). The questions could be divided into two scales: pros of smoking (
= 0.63), measuring the advantages of smoking, and cons of smoking ( = 0.56), measuring the disadvantages of smoking. Social influences were measured by assessing the subjective norm, perceived behavior of others and social pressure. Subjective norm was measured with seven items on a seven-point scale. For example: My brother thinks that I definitely should smoke (+3) to my brother thinks that I definitely should not smoke (–3). Two scales were formed, one for family ( = 0.74) and one for peers ( = 0.66). Perceived behavior of the social environment was measured by seven items, asking whether an important other smoked or not. For example: Does your father smoke? (no = 0; yes = 1). Social pressure was measured by seven items on a five-point answering scale. For example: Have you ever felt pressure to smoke from your best friend? Answering options included very often (4), often (3), sometimes (2), a few times (1) and never (0). One reliable scale was formed ( = 0.77). Six items on a five-point scale assessed self-efficacy, by measuring the ability of adolescents to resist the temptation to smoke in various situations. Three scales were made: social self-efficacy ( = 0.62), including when being with friends who smoke and when being offered a cigarette; opportunity self-efficacy ( = 0.82), relating to when you are on your way home from school and when you are doing homework; and emotional self-efficacy ( = 0.72), relating to when you feel nervous and when you feel upset. Intention was assessed by one question, measuring adolescents' intention to smoke in the next year on a five-point scale, ranging from definitely not (–2) to definitely (+2). Knowledge was measured by six items on a five-point scale. For example: Smoking is harmful to your health, answering options ranging from definitely (+2) to definitely not (–2). Media influences were measured by eight questions on a three-point scale. For example: When you watch TV, how often do you see actors smoke? Answering options included never (1), sometimes (2) and a lot (3). No reliable scales could be formed with respect to the concepts of knowledge and media influences. Demographics included in this study were district, age, gender, disrupted family and job situation of father and mother. Job situation was measured dichotomously by asking ‘Does your father/mother have a job?’. Disrupted family was measured by asking whether or not both the father and mother lived in the same house with the adolescent.
Statistical analyses
Data analysis included basic descriptive statistics of the respondents. Frequencies were run for all the variables. Differences in demographics as well as in smoking-specific cognitions for smokers (weekly smoking) and non-smokers (non-weekly smoking) were analyzed univariately using an independent samples t-test for continuous variables and 
2-tests for dichotomous variables. In order to gain in-depth insights into belief structures and salient social influences, the univariate analyses were executed at the item level. Linear regression was performed to test multivariately which combination of concepts would explain the variance in smoking behavior. All variables were forced into the regression equation, entering all demographics, knowledge and media influences in block 1, while all scales of attitude, social influence and self-efficacy were entered in block 2, and intention to smoke was entered in block 3 (Tabachnick and Fidell, 1996).
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Demographics and smoking behavior
The sample consisted of 282 adolescents, of whom 63.8% were boys and 36.2% were girls, thereby representing the lower enrolment ratios of girls in Gambian schools (UNICEF, 2003
). The mean age of the adolescents was 16.0 years (SD = 1.33). The prevalence of smoking at least once a week was 11.3%, while that of monthly smoking was 3.9% and that of less than monthly smoking was 32.0%. Never smokers represented more than half of the participating adolescents (52.8%). Of the total sample, 41.8% lived in urban areas and 58.2% in rural areas. The percentage of adolescents who lived with both father and mother was 43.3%. The percentage of fathers that had a job was 67%, while 39% of the mothers had a job.
Differences between smokers and non-smokers
Table I shows that there was a significant difference between smokers and non-smokers in terms of the items ‘other male guardian living in the same house’, ‘sister living in the same house’, ‘other people living in the same house’ and the item ‘district’. A total of 87.5% of the smokers and 54.4% of the non-smokers lived in rural areas.
|
Smokers were significantly more convinced than non-smokers that smoking is pleasant and that smoking would help them to calm their nerves. Non-smokers were significantly more convinced than smokers that smoking is bad for their health, stupid of them, that the behavior is wrong and that cigarettes taste horrible. The largest differences were found for the recognition that smoking is pleasant, stupid and bad for your health. Although smokers had a lower score on ‘bad for my health’, their score was still high, suggesting that they were aware of the harmful effects of smoking.
The differences between smokers and non-smokers proved to be significant in all the social norm items. Overall, the negative scores on this concept indicated a social norm that is directed towards non-smoking. However, smokers did score significantly higher than non-smokers on each item, with the largest differences for best friend and teacher. As for perceived behavior, smokers scored significantly higher than non-smokers on the items ‘most friends’, ‘best friend’ and ‘teacher’, with the largest difference for ‘best friend’. Smokers felt significantly more frequent pressure to smoke than non-smokers from all the important persons in their environment with the exception of the father. The greatest differences were found for brother and best friend.
As for self-efficacy expectations, a significant difference was found for emotional self-efficacy. Smokers had significantly less confidence than non-smokers in their ability not to smoke when they felt nervous and when they felt upset. Additionally, although both groups had a negative intention to smoke in the next year, non-smokers were more definite in their intention not to smoke in the next year than smokers.
No significant differences were found for knowledge and media influences, except for one concept. Namely, a large difference between smokers and non-smokers was found for being offered a free cigarette by a representative of a tobacco company. The percentage of smokers that had been offered a cigarette by a cigarette representative was 59.4%, compared to 14.5% of the non-smokers.
Regression analyses
In Table II, the results of the regression analysis are shown, depicting only the variables with significant ß weights (P < 0.05). Model 3 represents the final model. Additionally, Model 1 represents the regression equation in which only the demographics, knowledge and media were included. In Model 2, attitude, social influence and self-efficacy scales were added to Model 1. The final model consisted of six concepts.
|
The highest ß weights were found for cigarette offers by the tobacco industry. Thus, whether or not the adolescents had been offered a cigarette by a tobacco industry representative was the best predictor of smoking behavior. Since this factor remained highly significant in the final model, the results show that the influence of cigarette sampling on smoking behavior bypasses the investigated cognitive constructs. Smoking behavior was also associated with emotional self-efficacy, perceived behavior of best friend and mother having a job. There was a negative association between smoking behavior and other family members living in the same house. The more people lived in the same house, the less respondents were inclined to smoke. Approximately 32.9% of the variance between smokers and non-smokers could be explained by demographic variables (Model 1). When attitude, social influences and self-efficacy variables were added to the model, more than 44.1% of the variance between smokers and non-smoker was explained. When intention to smoke in the next year was added to the model, it accounted for 46.9% of the variance in smoking behavior.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Information about tobacco use in Africa is sparse. Only a few African countries systematically collect data on smoking prevalence. The surveys of smoking behavior, when available, are seldom representative of the whole country, as they are usually scattered and cover only urban populations or selected regions. Unfortunately, the lack of reliable data undermines the perceived necessity for comprehensive control strategies (Corrao et al., 2000
). Such a lack of reliable data about smoking behavior is also found in The Gambia. The present study is the first to examine determinants of smoking initiation among Gambian adolescents. The results show that free cigarettes offered by tobacco companies were mostly responsible for the difference between smokers and non-smokers.
In many countries, including The Gambia, cigarette companies give away free cigarette samples in areas where young people gather, such as rock concerts, discos and shopping malls. This popular form of marketing is called ‘sampling’ and involves free samples of an addictive product being handed out to possible consumers in circumstances where they are more likely to try them out (Popper, 1986). Giving away cigarettes costs cigarette companies little and ensures for them a steady stream of new customers (WHO, 2000b). According to the US Department of Health and Human Services (US Department of Health and Human Services, 1994), ‘Distribution of free samples is one of the most powerful devices available to marketers’.
Although the tobacco industry asserts that sampling is not intended for non-users or minors, there is little evidence of distribution control (Davis and Jason, 1988). Its effectiveness in getting non-smokers to experiment with tobacco and getting smokers to try a different tobacco brand is reflected by the industry's growth. From 1975 to 1990, expenses for distributing samples increased 4-fold, from US$25 to 100 million (US Department of Health and Human Services, 1994). In a George H. Gallup International Institute survey, held nationwide in the US in 1992, about half of 1125 adolescent smokers reported that they had received promotional items from tobacco companies, as had one in four non-smoking adolescents (Gallup Organization, 1993). These figures are comparable to those found in the present study. Sampling encourages children to experiment with tobacco and puts them at great risk of becoming addicted to it (Garner, 1986).
Other external variables that were associated with smoking behavior in the present study were whether the mother had a job and the number of other family members who lived in the house. As for household size, other studies have also found that larger families are protective against smoking (Burchfiel et al., 1989; Boyle and Szatmari, 1993). The reason for this could be greater social control in larger families. This may also explain the association between smoking and whether or not the mother has a job.
Cognitive concepts associated with smoking behavior were the intention to smoke in the future, emotional self-efficacy and the smoking behavior of the respondent's best friend. An association between emotional self-efficacy and smoking behavior has also been found in other studies. Smoking is sometimes used as a means of dealing with stress (Mates and Gordon, 1992) and several studies have shown that stress is indeed associated with smoking behavior (Bonaguro and Bonaguro, 1987; Byrne et al., 1995). Peer smoking is also known to be related to adolescent smoking behavior and intentions (Botvin et al., 1992; Biglan et al., 1995).
The present study was subject to various limitations. (1) The representativeness of the sample needs to be addressed. Our school-based design resulted in a sample that consisted for more than 60% of boys, caused by the low enrolment rates for girls in Gambian schools. (2) Smoking behavior was measured using adolescents' self-reports. There have been extensive investigations of the validity and reliability of self-reports. Several studies have found little discrepancy between biochemical assessments and self-reports of adolescent smoking behavior (Stacy et al., 1990; Komro et al., 1993), and research shows that adolescents' self-reports can be accurate provided confidentiality is assured (Hansen et al., 1985; Dolcini et al., 1996), which was the case in the present study. (3) The present study used a cross-sectional design and the interpretation of cross-sectional data on cognitive concepts is known to be problematic. Cognitions may well have been formed after the execution of the behavior in question. For example, adolescents may have formed certain intentions and self-efficacy expectations after they had started smoking. Additionally, several studies have shown that adolescents select peers with similar smoking behavior (Bauman and Ennett, 1996). Furthermore, smoking adolescents might be more aware of, and have a better memory for representatives of tobacco companies than non-smoking adolescents. Moreover, students who are more likely to pick up smoking may also be more likely to attend social events at which free cigarettes are offered. These reciprocal influences might partly explain the associations between the cognitive constructs and smoking behavior found in the present study.
All in all, two findings of the present study add to the current state of scientific knowledge. (1) The findings show the major impact of cigarette sampling as the influence of this practice was found to outweigh cognitive constructs. Usually, it is social cognitive constructs that are found to have the highest associations with smoking behavior in studies with a comparable design and theoretical model (Conrad et al., 1992; Petraitis and Flay, 1995; Tyas and Pederson, 1998). (2) The finding of a limited importance of smoking-specific cognitions may indicate the unplanned nature of smoking initiation. Recent studies have shown that smoking initiation is not a planned action to such an extent as has often been assumed (Sutton, 1992; Engels et al., 1999; Kremers, 2002). According to Kremers (Kremers, 2002), adolescents' uptake of smoking is not exclusively the result of a rational decision to smoke or not. Many adolescents do not have concrete plans to start smoking, but they are assumed to experiment with tobacco in a more spontaneous manner. This notion of ‘unplanned’ smoking initiation might explain the great effectiveness of marketing techniques like cigarette sampling (Kremers, 2002). Therefore, the present study would benefit from qualitative follow-up studies into the exact situations in which students accept the sample cigarettes.
If children live in an environment in which tobacco use is condoned and promoted, and if tobacco is cheap and easily accessible, they will be much more likely to smoke (Tobacco-Free Kids, 1999). One important step that can be taken involves prohibiting tobacco companies from handing out their products to pedestrians in the street and attendants at public events. The WHO recommends that countries:
...prohibit all tobacco advertising and promotions, including free samples and other giveaways, sale of non-tobacco products that carry a tobacco brand name, point of sale advertising and tobacco company sponsorship of sporting and cultural events. (WHO, 1998)Our study supports current efforts to prohibit cigarette sampling in The Gambia. Further, we recommend prevention programmes to be developed and implemented in The Gambia, aimed at making children less vulnerable to cigarette sampling. The most effective of such programmes have been found to be integral prevention programmes that address both the school and the community. The present study has shown that emotional self-efficacy beliefs and social influences, especially from peers, should be particularly emphasized in future smoking prevention programmes in The Gambia. Prevention workers should also keep in mind that children who live in a house with few other residents are more at risk of becoming smokers. This contextual factor should be further examined.
|
Acknowledgments |
|---|
This research was made possible by The Gambia College School of Public Health and the WHO in The Gambia.
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Ajzen, I. (1988) Attitudes, Personality and Behavior. Milton Keynes, Open University Press.
Bauman, K.E. and Ennett, S.T. (1996) On the importance of peer influence for adolescent drug use: commonly neglected considerations. Addiction, 91, 185–198.[CrossRef][ISI][Medline]
Biglan, A., Duncan, T.E. and Ary, D.V. (1995) Peer and parental influences on adolescent tobacco use. Journal of Behavioral Medicine, 18, 315–330.[CrossRef][ISI][Medline]
Bonaguro, J.A. and Bonaguro, E.W. (1987) Self-concept, stress symptomatology and tobacco use. Journal of School Health, 57, 56–58.
Botvin, G.J., Baker, E. and Goldberg, C.J. (1992) Correlates and predictors of smoking among black adolescents. Addictive Behaviors, 17, 97–103.[CrossRef][ISI][Medline]
Boyle, M.H., Szatmari, P. and Offord, D.R. (1993) Substance Use among Adolescents and Young Adults; Prevalence, Sociodemographic Correlates, Associated Problems and Familial Aggregario. Ontario Ministry of Health, Toronto.
Burchfiel, C.M., Higgins, M.W. and Keller, J.B. (1989) Initiation of cigarette smoking in children and adolescents of Tecumseh, Michigan. American Journal of Epidemiology, 130, 410–415.
Byrne, D.G., Byrne, A.E. and Reinhart, M.I. (1995) Personality, stress and the decision to commence cigarette smoking in adolescence. Journal of Psychosomatic Research, 39, 53–62.[CrossRef][ISI][Medline]
Centers for Disease Control and Prevention (2001) National Tobacco Information Online Systems (NATIONS). Available at: http://apps.nccd.cdc.gov/nations/index.asp; retrieved 23 March 2004.
Centers for Disease Control and Prevention (2002) Global Youth Tobacco Survey (GYTS). Available at: http://www.cdc.gov/tobacco/global/GYTS.htm; retrieved: 23 January 2003.
Chapman, S. and Leng, W. W. (1990) Tobacco Control in the Third World: A Resource Atlas. International Organization of Consumers Unions, Penang.
Conrad, K.M., Flay, B.R. and Hill, D. (1992) Why children start smoking cigarettes: predictors of onset. British Journal of Addiction, 87, 1711–1724.[CrossRef][ISI][Medline]
Corrao, M.A., Guindon, G.E., Sharma, N. and Shokoohi, D.F. (2000) Tobacco Control Country Profiles. American Cancer Society, Atlanta, GA.
Davis, R.M. and Jason, L.A. (1988) The distribution of free cigarette samples to minors. American Journal of Preventive Medicine, 4, 21–26.[ISI][Medline]
De Vries, H. and Backbier, E. (1994) Self-efficacy as an important determinant of quitting among smoking pregnant women. Preventive Medicine, 23, 167–174.[CrossRef][ISI][Medline]
De Vries, H. and Kok, G. J. (1986) From determinants of smoking behavior to the implications for a prevention programme. Health Education Research, 1, 85–94.
De Vries, H., Dijkstra, M. and Kuhlman, P. (1988) Self-efficacy: third factor besides attitude and subjective norm as a predictor of behavior intention. Health Education Research, 3, 273–282.
De Vries, H., Mudde, A.N., Leijs, I., Charlton, A., Vartiainen, E., Buijs, G., Pais Clemente, M., Storm, H., González Navarro, A., Nebot, M. Prins, T., Kremers, S.P.J. (2003) The European Smoking Prevention Framework Approach (ESFA): an example of integral prevention. Health Education Research, 18, 611–626.
Dijkstra, M., Mesters, I., De Vries, H. and Parcel, G. S. (1999) Effectiveness of a social influence approach and boosters to smoking prevention. Health Education Research, 14, 791–802.
Dolcini, M.M., Alder, N.E. and Ginsberg, D. (1996) Factors influencing agreement between self-reports and biological measures of smoking among adolescents. Journal of Research on Adolescence, 6, 515–542.
Engels, R.C.M.E., Knibbe, R.A. and Drop, M.J. (1999) Predictability of smoking in adolescence: between optimism and pessimism. Addiction, 94, 115–124.[CrossRef][ISI][Medline]
Gallup Organization (1993) The Public's Attitudes toward Cigarette Advertising and Cigarette Tax Increase. Poll conducted for the Coalition for Smoking OR Health. Gallup, Princeton, NJ.
Garner, W. (1986) Tobacco sampling, public policy and law. Journal of Health Politics, Policy and Law, 11, 423–443.[ISI][Medline]
Hansen, W.B., Mallotte, K.C. and Fielding, J.E. (1985) The bogus pipeline revisited: the use of threat as a means of increasing self-reports of tobacco use. Journal of Applied Psychology, 70, 789–792.[CrossRef][ISI][Medline]
Jallow, C.O.A. (2001) Country presentation; the Gambia. A presentation at the GYTS Training Workshop. WHO, The Gambia.
Komro, K.A., Kelder, S.H., Perry, C.L. and Klepp, K.I. (1993) Effects of a saliva pipeline procedure on adolescent self-reported smoking behavior and youth smoking prevention outcomes. Preventive Medicine, 22, 857–865.[CrossRef][ISI][Medline]
Kremers, S.P.J., Mudde, A.N. and De Vries, H. (2001) Subtypes within precontemplation stage of adolescent smoking acquisition. Addictive Behaviors, 26, 237–251.[CrossRef][ISI][Medline]
Kremers, S.P.J. (2002) On your Marks; Revising, Testing and Integrating Stage Models of Smoking Initiation. Unigraphic Maastricht, Maastricht.
Mates, D. and Gordon, J, R. (1992) Sources of stress and coping responses of high school students. Adolescence, 27, 461–474.[ISI][Medline]
Petraitis, J. and Flay, B.R. (1995) Reviewing theories of adolescent substance use: organizing pieces in the puzzle. Psychological Bulletin, 117, 67–86.[CrossRef][ISI][Medline]
Popper, E.T. (1986) Sampling and Couponing Promotional Activity in the Domestic Cigarette Market: A Report of the Office on Smoking Health. US Department of Health and Human Services, Washington, DC.
Pucci, L.G. and Siegel, M. (1999) Features of sales promotion in cigarette magazine advertisements, 1980–1993: an analysis of youth exposure in the United States. Tobacco Control, 8, 29–36.
Stacy, A.W., Flay, B.R. and Sussman, S. (1990) Validity of alternative self-reports indices of smoking among adolescents. Psychological Assessment, 2, 442–446.[CrossRef]
Sutton, S.R. (1992) Is taking up smoking a reasoned action? Addiction, 87, 21–24.
Tabachnick, B. G. and Fidell, L. S. (1996) Using Multivariate Statistics, 3rd edn. Harper Collins, New York.
Tobacco-free Kids (1999) Overview: The Global Crisis. Prepared for the International Policy Conference On Children and Tobacco. Available at: http://tobaccofreekids.org/campaign/global/crisis.shtml; retrieved: 23 January 2003.
Tyas, S.L. and Pederson, L.L. (1998) Psychosocial factors related to adolescent smoking: a critical review of the literature. Tobacco Control, 7, 409–420.
UNICEF (2003) At a Glance: Gambia. Available at: http://www.unicef.org/infobycountry/gambia_statistics.html; retrieved 24 March 2004.
US Department of Health and Human Services (1994) Preventing Tobacco use among Young People: A Report of the Surgeon General. Public Health Services, Centers for Disease Control and Prevention, Office on Smoking and Health, Atlanta, GA.
Vroom, B. (1994) The Text Tested. Manual for Pretesting Written Material. Van Gorcum, Assen.
WHO (1983) Smoking Control Strategies in Developing Countries. Report of a WHO Expert Committee. Technical Report Series 695. WHO, Geneva.
WHO (1988) Guidelines for Controlling and Monitoring the Tobacco Epidemic. WHO, Geneva.
WHO (2000a) Tobacco Control Situation in the African Region: A Challenge for the Future. WHO, Kenya.
WHO (2000b) The WHO International Conference on Global Tobacco Control: Towards a WHO Framework Convention on Tobacco Control. WHO, Geneva.
Yach, D. (1996) Tobacco in Africa. World Health Forum, 17, 29–36.[Medline]
Received on May 20, 2003; accepted on November 10, 2003
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. J. Wellman, D. B. Sugarman, J. R. DiFranza, and J. P. Winickoff The Extent to Which Tobacco Marketing and Tobacco Use in Films Contribute to Children's Use of Tobacco: A Meta-analysis Arch Pediatr Adolesc Med, December 1, 2006; 160(12): 1285 - 1296. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. DiFranza, R. J. Wellman, J. D. Sargent, M. Weitzman, B. J. Hipple, J. P. Winickoff, and for the Tobacco Consortium, Center for Child Healt Tobacco promotion and the initiation of tobacco use: assessing the evidence for causality. Pediatrics, June 1, 2006; 117(6): e1237 - e1248. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||