Health Education Research, Vol. 18, No. 4, 439-452,
August 2003
© 2003 Oxford University Press
Cigarette smoking in teenage girls: exploring the role of peer reputations, self-concept and coping
School of Psychology, Deakin University, Burwood Highway, Burwood, 3125 Australia
*To whom correspondence should be addressed at Department of Public Health, School of Health and Environment, La Trobe University, PO Box 199, Bendigo, 3552, Australia e-mail: p.snow{at}bendigo.latrobe.edu.au
 |
Abstract |
|---|
 Top Abstract IntroductionAimsMethodProcedureResultsDiscussionNotesReferences |
|---|
Reducing the prevalence of smoking among teenagers is an important goal for health and education professionals. In the present study, self-reported cigarette smoking status was examined among 241 adolescent females from four metropolitan, independent girls’ schools in suburban Melbourne, Australia. This study was particularly concerned with the role of peer reputations, coping and self-concept as influences on teenage girls’ decision making with respect to smoking. Coping strategies, levels of self-concept and reputation enhancement were assessed using the High School Student Activity Questionnaire. Three separate multivariate analyses of variance (MANOVAs) revealed that six of the 11 reputation variables, three of the four self-concept variables and two of the three coping variables contributed to the main effect of smoking status. Post hoc analyses allowed some profiling of current cigarette smokers as compared to never smokers. This study offers support to reputation enhancement theory with regard to cigarette smoking in adolescent girls, and also provides continued evidence for the importance of particular aspects of self-concept and coping skills with respect to cigarette smoking. Suggestions for further research are discussed and some possible implications of the findings for school-based health education programmes are raised.
|
Introduction |
|---|
|
TopAbstractIntroductionAimsMethodProcedureResultsDiscussionNotesReferences |
|---|
Reducing the prevalence of cigarette smoking is a high priority on Australia’s public health agenda (Hill et al., 1998
). Although historically more men than women have smoked (van Roosmalen and McDaniel, 1992), during the past 25 years there has been a convergence in the prevalence of female and male smoking, such that currently more teenage girls than boys smoke in Australia (Hill et al., 1999). There is evidence to indicate that if smoking has not been established as a daily routine by age 20, the potential for its initiation drops significantly (Bechtle et al., 1998). If education programs can prevent the uptake of cigarette use during adolescence, therefore, the likelihood of smoking during the adult years is lessened.
A number of different influences have been examined in an attempt to account for adolescent smoking behavior. In particular, these include self-concept, coping skills and peer group influence. Self-concept and self-esteem have been central to investigations of adolescent substance use in recent years; however, distinctions between these notions are not always clear. Hattie (Hattie, 1992) noted that self-concepts (conceptions of self) are cognitive appraisals which are integrated across the various dimensions that we attribute to ourselves. He noted that this integration occurs via processes such as self-verification and self-enhancement, and is sensitive to confirmation/disconfirmation received from both others and ourselves. Hattie argued that the individual’s self-concept, as a cognitive appraisal of oneself, may guide, mediate and regulate behavior, as self-concept:
...interprets and organizes self-reliant actions and experiences; it has motivational consequences, providing the incentives, standards, plans, rules, and scripts for behavior; and it adjusts in response to challenges from the social environment. [(Hattie, 1992), p. 99]
This definition emphasizes the dynamic nature of self-concept, and the salience of situational/social context for determining the influences which operate between self-concepts and behavior. Self-esteem, however, refers to the individual’s judgment of the ‘worthiness’ of certain aspects of self and takes account of the fact that different individuals place premiums on different attributes (Hattie, 1992). Hattie observed that ‘To have high self-esteem implies both that we consider aspects of our life as important, and that we have the confidence to fulfill our expectations’ [(Hattie, 1992), p. 54].
Available evidence suggests that there is an established link between low scores on a range of self-concept dimensions and an increased risk for smoking. Thornton et al. (Thornton et al., 1999) examined five different aspects of self-concept among smokers and non-smokers. They found that physical, social and family self-concept were all related to uptake of cigarette smoking among adolescents. Further, they found that these dimensions of self-concept were also related to smoking cessation. Similarly, McInman and Grove (McInman and Grove, 1991) found significant self-concept differences between regular and non-smokers.
Smoking has also been investigated in adolescents and adults with respect to its role as a mechanism for coping with stress. In recent years, Fydenberg and Lewis (Fydenberg and Lewis, 1991, 1994) have argued that the ways in which adolescents cope with stress should not necessarily be examined using only adult-centric models which deal with problem versus emotion-focused approaches to stress. These workers have identified three main types of stress experienced by adolescents: achievement (e.g. academic), social issues (e.g. fear of nuclear war) and relationships (e.g. redefining one’s relationship with parents), and have proposed that adolescents’ coping can be viewed as productive (strategies which are helpful in adjusting to or dealing with a situation1) or non-productive (those which may be ineffective, or actually exacerbate a situation2). Employing Frydenberg and Lewis’ (Frydenberg and Lewis, 1993) Adolescent Coping Scale, Odgers et al. (Odgers et al., 1996), in their analysis of the coping strategies used by substance (including tobacco) users, found that both productive coping strategies and non-productive coping strategies were able to differentiate between current, experimental and non-substance users, whereas relying on others could not. They found that non-substance users were more likely to use productive coping strategies to deal with stress than either experimental or current users. This nexus between stress, coping style and cigarette smoking has continued to be confirmed, particularly in female adolescents [e.g. (Byrne and Mazanov, 1999, 2001)], but does not fully account for variance in adolescent smoking behavior.
In addition to research examining the influence of self-concept and coping skills on cigarette usage in adolescents, the peer group and peer influences on cigarette smoking have also been examined by a number of workers [e.g. (Fisher and Baumann, 1988; van Roosmalen and McDaniel, 1992; Ennett and Bauman, 1994; Charlton et al., 1999; Glendinning and Ingliss, 1999; Koval and Pederson, 1999; Alexander et al., 2001)]. The study by van Roosmalen and McDaniel, for example, examined the effects of gender and peer group among a large sample of adolescents, and found that when none of their friends smoke, a smaller proportion of females than males smoke. These workers concluded that smoking intention and continuation might be as much a function of social context as of individual choice, and this may be truer for females who are part of peer and kinship networks that have considerable influence on their behavior. As Alexander et al. (Alexander et al., 2001) have recently observed, however, it is difficult to determine whether peer influence acts via a form of social contagion (transmission of norms from popular students to other students) or social influence (adoption of a behavior by an individual, in an effort to conform to a perceived norm).
Thus although the link between cigarette smoking and peer group is one that has been confirmed empirically by a number of workers, the mechanism of this group influence on smoking is still debated (Kandel, 1998; Odgers, 1998; Schofield et al., 2001). One theory that has attempted to bridge this gap with a range of other substances is Reputation Enhancement Theory (RET) (Emler, 1984). This theory holds that much of adolescent behavior is motivated by the need to manage how the self is presented to the peer community. The goal of this presentation is to induce the peer group to credit one with a particular chosen social identity and in turn to manage one’s own long-term reputation. Thus people are both students of reputation, in that they assiduously investigate the reputations of others, and promoters of their own reputations (Houghton et al., 1998). RET further suggests that individuals display behaviors that are consistent with their desired reputations. These behaviors are displayed in the company of friends, are witnessed by peers and recounted when there is an audience present. According to this theory, during adolescence a limited range of people provide a regular audience and feedback, the most influential being the peer group (Odgers et al., 1996). The attitudes and expectations of the peer group are important in the formation and enhancement of a reputation because peers make judgments about an individual’s behavior and in turn influence behavior. Emler (Emler, 1984) proposed that not only do adolescents have reputations, but they care about the kind of reputation they have. Thus, not only are adolescents aware of the consequences of their involvement in drug usage, but they actively pursue the attainment of their ideal reputations through the use/non-use of drugs. This may help to explain why some adolescents engage in socially approved behaviors such as working hard or gaining an education, but others engage in behaviors that are not socially approved, such as substance use and/or delinquent behavior.
RET has contributed to the understanding of adolescents’ involvement in a number of risk-related activities, including maintenance and cessation of substance use (Odgers et al., 1996), delinquency (Carroll et al., 1999), disruptive behavior in school (Houghton and Carroll, 1996), volatile substance use (Houghton et al., 1998a), and alcohol misuse (Houghton et al., 1998b). These studies indicate that decisions to use drugs and substance of choice are related to the type of reputation an individual wishes to affirm in the eyes of his/her peers (Houghton et al., 1998a,b), and that the direction of the effect of reputation enhancement variables can change depending on the drug of choice (Carroll et al., 1999). Odgers et al. (Odgers et al., 1996) have shown that reputation mediates peer influences upon delinquency and drug use among adolescents more closely than factors relating to coping or self-concept. However, this theory has not yet been examined specifically in relation to cigarette usage.
Applied to adolescent females specifically, RET suggests that members of this population may become involved in cigarette usage as a means of developing and maintaining a particular reputation, and to gain membership of a specific peer group. This theory also implies that individuals who admire the activities of a certain group will strive to attain a reputation congruent with that group. This hypothesis has implications for the development of prevention programmes aimed at different subgroups of adolescent females. Given that (1) RET has not yet been applied to the study of tobacco use by young people and (2) a number of studies have found that influences on cigarette smoking interact with gender (Heaven, 1989; Mingawa et al., 1993; Abernathy et al., 1995; Charlton et al., 1999; Crisp et al., 1999; Lucas and Lloyd, 1999; Moon, 2000), it would seem logical to begin the exploration of this theory’s implications for tobacco use by studying males and females separately.
|
Aims |
|---|
|
TopAbstractIntroductionAimsMethodProcedureResultsDiscussionNotesReferences |
|---|
The aim of this study was to investigate the role of RET in relation to self-concept and coping strategies employed by a sample of adolescent females attending private, single-sex schools, according to their status as current, experimental or non-cigarette smokers. Single-sex schools were selected in order to reduce the confounding influence of male–female peer group interactions. Although there are a small number of government single-sex schools in Melbourne, these are greatly out-numbered by private settings. In order to achieve a suitable (and demographically homogeneous) sample size, therefore, only private girls’ schools were approached.
|
Method |
|---|
|
TopAbstractIntroductionAimsMethodProcedureResultsDiscussionNotesReferences |
|---|
Settings
Letters were sent to 14 independent single-sex girls schools in the eastern metropolitan region of Melbourne, Australia. According to the 1996 Victorian census, of the four schools that agreed to participate, the median weekly income of families in those areas ranged from $AUS317.00 to 444.00. The median weekly income across the whole of metropolitan Melbourne was $AUS311.00. The total student enrolments (i.e. kindergarten to Year 12) at the participating schools ranged from 536 to 1013. According to the Australian Bureau of Statistics (Australian Bureau of Statistics, 2001
), in 2000, 86.4% of female students in the State of Victoria were retained at school between Years 10 and 12. (Students may not normally leave school before age 15 in Australia.) With the exception of the Australian Capital Territory, these are the highest retention rates for female students in Australia. In 2000, 95.4% of 15-year-old girls in Victoria were participating in some type of formal schooling. Further, 88.3% of students in private schools were retained across these years.
Participants
The initial sample comprised 249 female adolescents. Six participants were excluded from the analysis because of failure to indicate their smoking status. A further two participants were excluded after being identified as multivariate outliers (one smoker and one non-smoker). This left a total of 241 students in the final analysis (representing a response rate of 35%). The age range of respondents was 13–16 years (M = 15, SD = 0.69) and all were in Year 9 or 10 at the time of testing (Year 12 signifies the end of secondary schooling in Australia). Participants were classified as either current cigarette smokers, experimental cigarette smokers or non-cigarette smokers, according to their responses on a self-report cigarette usage item. Definitions of these terms are provided below.
Instrumentation
The High School Student Activity Questionnaire [HSSAQ (Odgers, 1996)] is a paper and pencil instrument with strong psychometric properties (Odgers et al., 1996). The HSSAQ has been used in previous research to examine the effects of reputation, self-concept and coping on substance users (Houghton et al., 1998; Odgers et al., 1996). The HSSAQ comprises five parts, covering: self-concept, adolescent coping strategies, reputation enhancement, self-reported substance use and demographic variables. The present authors replaced the item on self-reported substance use with an item pertaining to cigarette usage.
Self-concept measures
A 20-item modified version of Song and Hattie’s (Song and Hattie, 1994) About Myself Scale was used to provide measures of four self-concept variables: family, physical, social and peer self-concept. This instrument has been shown to have sound psychometric properties. Items were scored such that the lower the score, the higher the self-concept.
Coping strategies measures
This section sought information pertaining to the coping strategies used by individuals to deal with stress. Data were gathered using the general short form of the Adolescent Coping Scale (Frydenberg and Lewis, 1993), which is a reliable and valid self-report instrument, comprising 18 items pertaining to productive coping strategies, non-productive coping strategies and coping by relying on others. The scale was coded such that the more a respondent used a given coping strategy to deal with stress, the higher the score.
Reputation enhancement measures
Eleven reputation enhancement variables (sociability, admiration of law-abiding activities, admiration of law-breaking activities, deviant social identity, conforming social identity, activity self-description, ideal activity self-description, power self-description, ideal power self-description, communication of conforming events and communication of non-conforming events) were examined across 124 items (Carroll et al., 1999). The scoring of the sociability variable, and the two power and activity variables was such that the higher the score attained, the more pro-social a participant.
On the remaining six variables the scoring was such that the higher the score the more a participant displayed that particular characteristic.
Cigarette use inventory
This section sought information pertaining to self-reported cigarette usage. Participants were asked to indicate if they had ever smoked even part of a cigarette. Responses available were: (1) yes, I now smoke cigarettes; (2) yes I used to smoke at least once a week, but I quit; (3) yes, but I have only smoked a few times; (4) yes, but I have only smoked part of a cigarette; and (5) no, I have never even smoked part of a cigarette. Participants were classified as either non-smokers, experimental smokers (former and occasional users) or current smokers. Given the evidence that fewer than 1% of secondary students misrepresent their smoking status (Lader and Matheson, 1991), physiological validation of smoking status was not employed.
|
Procedure |
|---|
|
TopAbstractIntroductionAimsMethodProcedureResultsDiscussionNotesReferences |
|---|
Approval to conduct the study was gained from the Deakin University Human Research Ethics Committee, and letters were sent to school principals outlining the nature, purpose and relevance of the study, along with a copy of the instrument to be used. Principals were provided with a stamped self-addressed envelope and a slip on which they could indicate their willingness to participate. A follow-up phone call was made 3 weeks later to those schools that had not responded to ascertain interest in being involved in the study.
Consent for individuals to participate was obtained from the students and their parents. All girls were provided with a plain language statement for their parents, along with a parental consent form to take home. Plain language statements contained an instruction to return the forms to their daughters’ school before the designated date of testing. After receipt of parental consent forms, only those students (1) whose parents had consented and (2) who were willing to take part were surveyed. Students were surveyed during class time, under simulated test conditions. Upon completion of the questionnaire, participants placed their questionnaire into the envelope provided and sealed it. Participants then placed the envelope into a box at the front of the classroom. This was used to minimize under-reporting of cigarette use and to ensure confidentiality. The questionnaire took approximately 25 min to complete.
|
Results |
|---|
|
TopAbstractIntroductionAimsMethodProcedureResultsDiscussionNotesReferences |
|---|
The main focus of this study was the question of differences in cigarette smoking status in relation to four dependent variables concerning self-concept variables, three coping variables and 11 reputation enhancement variables. Three separate multivariate analyses of variance (MANOVAs) were used to examine differences in smoking status on the four variables relating to self-concept, the three variables relating to coping strategies and the 11 variables relating to reputation enhancement.
Prevalence of cigarette smoking
Half of the sample (49.6%) had smoked a cigarette at some stage (current and experimental); of these, 15.8 % could be classified as current smokers and 33.8% were experimental smokers. Just over half of the respondents (50.4%) had never smoked. These findings suggest a slightly lower prevalence rate of smoking in this sample than in the population of Victorian female secondary students as a whole. In 1996 (most recent data available), 56.7% of girls in Years 7–10 (roughly ages 12–16) reported ever-use of cigarettes and 21.1% had smoked in the last week (Department of Human Services, 1999).3
An analysis of variance (ANOVA) to investigate differences in cigarette status as a function of year level revealed no significant differences thus the sample was examined as one group [F(1,240) = 0.97, P > 0.05].
Self-concept
A MANOVA to determine whether respondents’ self-concept would differ according to their cigarette smoking status showed the combined dependent variables for self-concept were significantly affected by user group [F(8,468) = 11.339, P < 0.001]. The results reflected a moderate association between user group (current, experimental and non-cigarette smoker) and the combined dependent variables (n2 = 0.16).
Three of the self-concept variables were identified as significant at the 0.01 level in the analysis. These were family self-concept, physical self-concept and social self-concept. Peer self-concept was not significant. The summary statistics, F statistics, Scheffé post hoc comparisons and effect sizes for the main effect of cigarette usage are presented in Table I. Lines connecting values in Table I represent post hoc significant differences between groups.
|
Coping strategies
A MANOVA to examine respondents’ coping skills as a function of cigarette smoking status showed that the combined dependent variables for coping skills were significantly affected by user group [F(6,470) = 12.14, P < 0. 001]. The results reflected a weak to moderate association between user group (current, experimental and non-cigarette smoker) and the combined dependent variables (n2 = 0.13).
Two of the coping variables were identified as significant in the analysis—non-productive coping strategies and relying upon others as a coping strategy.
The summary statistics, F statistics, Scheffé and effect sizes for the main effect of cigarette usage are presented in Table II. Lines connecting values in Table II represent post hoc significant differences between groups. Productive coping strategies did not differ significantly between user groups.
|
Reputation enhancement variables
A MANOVA to determine whether respondents’ perceived reputations would differ according to their cigarette smoking status was carried out. Wilk’s
converted to an approximate F statistic was used to evaluate multivariate significance. The combined reputation variables were significantly affected by user group [F(22,454) = 6.4, P < 0. 001]. The results reflected a moderate association between user group (current, experimental and non-cigarette smoker) and the combined dependant variables (n2 = 0.24). Six of the reputation variables were identified as significant in the analysis: sociability, deviant social identity, admiration of law-breaking activities, communication of negative events, activity self-description and ideal activity attributes. The summary statistics, F statistics, Scheffé post hoc comparisons and effect sizes for the main effect of cigarette usage are presented in Table III. Lines connecting values in Table III represent post hoc significant differences between groups. The variables of ideal power attributes, conforming social identity, admiration of conforming activities and communication of positive events were not significant in discriminating between user group.
|
Profile of current cigarette smoker
The findings from this study permit some profiling of Year 9 and 10 female cigarette smokers (in comparison to experimental smokers and non-cigarette smokers). First, these individuals tend to be less confident in themselves and believe they are less liked by their families. They are also less confident in terms of physical self-concept than never smokers and less confident in terms of social self-concept than either experimental or never smokers. They do not, however, differ from non-smokers in terms of peer self-concept. Current smokers tend to use more non-productive coping strategies than do experimental or non-smokers. They also tend to rely on others as a coping strategy more than experimental or non-smokers. There are no differences, however, in the extent to which they utilize productive coping strategies to deal with their problems. Current smokers also place a higher value on friendship and group membership than do never smokers, although they do not differ from experimental smokers in this respect. Current smokers have higher rates of approval and regard for delinquent-type activities, and perceive themselves and want their friends to perceive them as being more non-conforming than either never smokers or experimental smokers. Current smokers were found to perceive themselves as ‘mean and nasty’, to cause trouble, break rules and to be unreliable, and to desire to be this way more than either never smokers or experimental smokers. Finally, current smokers communicate to more people about the negative events they engaged in than adolescents who had never smoked.
|
Discussion |
|---|
|
TopAbstractIntroductionAimsMethodProcedureResultsDiscussionNotesReferences |
|---|
The aim of the present study was to examine the extent to which variables relating to perceived reputations, self-concept and coping strategies could distinguish the cigarette smoking status of a group of adolescent females. The findings offer support to previous research which has demonstrated that an association exists between cigarette smoking and self-concept [e.g. (Thornton et al., 1999
)], and cigarette smoking and inadequate coping strategies [e.g. (Shiffman and Wills, 1985)]. However, the present findings also suggest that differences between current, experimental and non-smoking teenage girls are partly accounted for by factors derived from RET.
Three of the four self-concept variables (family, social and physical) contributed to the significant main effect of user group, whereas the fourth self-concept variable (peer) did not. This suggests that the cognitive appraisals that female adolescent smokers have about themselves in relation to their family, their body and their social selves are more negative than those of non-smokers. This finding is consistent with previous research by Thornton et al. (Thornton et al., 1999) who examined five different aspects of self-concept among smokers and non-smokers. They found that physical, social and family self-concept were all related to uptake of cigarette smoking among adolescents, whereas peer self-concept was not.
In light of the strong influence that peer group has on adolescent behavior, including substance use, the finding in the current study that peer self-concept is not related to cigarette use requires some consideration. This may reflect the instrumentation used, given that other workers who have employed the HSSAQ (Odgers et al., 1996; Houghton et al., 1998a,b; Thornton et al., 1999) have reported similar findings. Given that smoking may confer a desired reputational profile for a teenage girl, however, it must be considered that peer self-concept can be enhanced via the adoption of this behavior. Alternatively, the role of peers may be less significant than other influences on girls’ decisions relating to smoking, as was suggested by findings reported by Charlton and Blair (Charlton and Blair, 1989a). These workers found that while having a best friend who smokes was a significant predictor of girls’ smoking, this was less important than girls’ positive beliefs about smoking (e.g. notions that smoking confers confidence). Further research examining the relative role of peer influence in relation to drug use by a range of teenage subgroups may shed light on the ways in which peers and peer self-concept do and do not influence decisions to become involved in cigarette usage.
Post hoc analyses revealed that both social and family self-concept differentiated between current smokers and experimental smokers. This suggests that experimental smokers have a more positive self-concept in these areas than current smokers. This sheds light on experimental smokers as a group and is congruent with findings reported by Thornton et al. (Thornton et al., 1999) that not only did self-concept decrease upon initiation of cigarette smoking, but it also increased upon quitting smoking. An alternative possibility that could be examined in future research is that self-concept in the social and family domains differentiates smokers who begin smoking at a young age but do not continue on (experimental) and those that do continue smoking (current). Longitudinal research examining whether self-concept operates as a causal mechanism, or an outcome, could shed light on this relationship.
Coping skills
Two of the three coping variables were shown to contribute to the main effect of user group. These variables were non-productive coping strategies and reliance on others as a coping strategy. Post hoc analyses revealed that current smokers utilized more non-productive coping strategies than either non-smokers or experimental smokers. This points to the possibility that smoking may, in itself, be a non-productive coping strategy used by this group. This is in line with smoking consistently being reported as a mechanism people use to help cope with stress (Shiffman and Wills, 1985).
The results also suggest that current smokers rely more on others to help cope with stress than do experimental smokers, who in turn show greater reliance on others than do those who had never smoked. This points to the possibility that for some people, a mechanism underpinning abstention from smoking is the ability to independently deal with stress. This again highlights the importance of the peer group to young smokers and suggests that the level of involvement with a peer group when dealing with stress can influence smoking status.
Our results suggest that differences in the amount that productive coping strategies are utilized to deal with stress did not differentiate between user group. Although groups did not differ in terms of the productive coping strategies used, they did differ with respect to the use of non-productive coping strategies. It may be that once smokers adopt smoking as a method of coping, their use of non-productive strategies increases. It should also be recognized, however, that the ‘productiveness’ of a coping mechanism may vary according to one’s perspective. If, for example, smoking is adopted as a means of dealing with social anxiety and is seen by the individual as effective in achieving this end, it is arguably a ‘productive’ mechanism—albeit one that health and welfare educators would not endorse or promote. ‘Productive’ coping, in this instance, then, might ideally be seen to encompass approaches which are associated with enhanced psychosocial well-being, without compromised physical health status. Frydenberg [(Frydenberg, 1999), p.18] has addressed this issue by acknowledging that ‘...coping need not be a "successfully completed" act but an attempt to deal with the problem. The concern is with the attempt, rather than the effectiveness of the outcome’. This suggests that students should be encouraged to identify a range of coping mechanisms and to evaluate these along a range of ‘cost–benefit’ dimensions. Results of the present study suggest that prevention programmes need to teach girls different ways of defining ‘productive’ and ‘non-productive’ coping skills, in order to foster the ability to generate and employ health-promoting approaches when dealing with stressful life events.
Cigarette smoking status was differentiated by the reputation variables of sociability, admiration of law-breaking activities, deviant social identity, ideal activity description, activity self-description and communication of negative events. Smokers as a group scored more highly than non-smokers on each of these variables, except sociability. Compared to never smokers, the smokers in this study were more sociable, admired law-breaking activities, perceived themselves as ‘deviant’ and ideally want to be this way, perceive themselves as rule breakers, nasty, mean, lazy and as being a drop-out; further, smokers ideally want to be that way and communicate to more people about the illegal activities in which they had engaged.
This profile of a current smoker lends support to RET (Emler, 1984) in a number of ways. First, it attests to the importance of an audience in enhancing reputations among adolescents, confirming previous findings on adolescent drug use (Emler, 1984; Houghton and Carroll, 1996; Odgers et al., 1996) and supporting the proposition that adolescents are active promoters of their own reputations (Emler, 1984). Second, it suggests that current smokers want to portray a reputation that is non-conforming and that this non-conforming reputation is something that they both admire and recognize as part of the identity they wish to promote to others. Third, it supports the proposition that current smokers place a higher value on being part of a particular social group. This is consistent with previous research examining the effect of peers on smoking intention and action. Alexander et al. (Alexander et al., 2001) have recently reported that having one or two best friends who are current smokers significantly increases the likelihood of an adolescent smoking and Schofield et al. [(Schofield et al., 2001), p. 2] have recently observed that ‘...forces of social influence, either overt or covert, within the [peer] group may produce behavioral homogeneity’. These findings are consistent with trends reported by Minagawa et al. (Minagawa et al., 1993) that levels of social acceptance may be higher among smokers than non-smokers, and also raise the possibility that this effect is particularly so for females who are embedded in peer and kinship networks that have considerable influence on their behavior. Fourth, it supports the proposition that the peer group that is admired subsequently influences the self-perception of an individual. Finally, this profile supports the proposition that current smokers will go to considerable lengths to attain a given reputation within the peer group. This is all congruent with adolescent smokers and non-smokers desiring a reputation that is similar to that of the peer group they admire, and with group action not simply being an expression of identity, but an active attempt to ground group identity.
According to RET, individuals strive to encourage the audience of peers to attribute certain characteristics to them. Individuals choose to inform an audience of their actions and, through this, hope to attain their ‘ideal’ image (Emler, 1984). Thus the fact that current smokers tell more people about the deviant activities that they engage in would suggest that they are trying to inform an audience of their actions in order to achieve and maintain their ideal image.
In the present study, smokers emerged as more deviant than non-smokers in what they admire, their reputations, activities, self-perception and communication. This was accompanied, however, with there being no differences in the extent to which they disliked conforming activities, reputations, perceptions and communications. This suggests that it is specifically the non-conforming reputation that cigarette smokers strive to attain which distinguishes them from experimental users and non-users, rather than a dislike for conforming activities. This may be due to the fact that teenage cigarette usage, which is typically regarded as undesirable by adults, may be interpreted as a conforming activity by some adolescent peer groups (Houghton et al., 1998a). Thus individuals who engage in cigarette usage, although portraying and seeking a non-conforming reputation in the eyes of adults, may not necessarily want to be seen as less conforming amongst adolescent peers.
The present findings sit well alongside the two qualitative reports on adolescent perceptions of smoking in relation to peer groups associations. Michell (Michell, 1997) found that teenagers themselves thought smoking was most likely among three different subgroups—‘top’ girls, for whom relationships with boys were more important than academic success, low status pupils, particularly girls, who smoke to ‘buy’ popularity and trouble-making boys. The present findings need further examination to determine specifically what type(s) of non-conforming reputation(s) are admired by different groups and in what circumstances. It could be that in this age group, neither user group wants to be seen as behaving in a ‘conforming’ manner (i.e. fulfilling the expectations of a variety of adult authority figures), whereas current users specifically want to be seen to be more ‘deviant’ with respect to conventional expectations. Either way, it should be noted that it appears to be an actively non-conforming/deviant reputation that distinguishes current smokers from never smokers, not a lesser desire to be conforming. As Michell’s data reminds us, the reasons for desired non-conformity may vary between peer groups, and this has implications for the way in which public health approaches to smoking prevention and cessation are conceived and delivered.
More recently, Denscombe (Denscombe, 2001) has reported that young people (aged 15–16) reject the notion of ‘peer group pressure’ as a mechanism which accounts for teenagers’ smoking uptake. The fact that this notion is rejected because it casts teenagers as passive ‘victims’ of an outside force, rather than as autonomous agents of their own identity, is consistent with the central tenet of RET, i.e. that individuals select and promote a particular desired reputation for themselves.
An important qualification on these findings concerns the power and ideal power attributes variable, which did not differentiate user groups. There are certain attributes that are measured by this variable to which most young people aspire, such as being popular, powerful or good-looking, irrespective of their status as smokers or non-smokers. Although these attributes are held in common, they may be expressed in different ways amongst different groups of adolescents (Carroll et al., 1999). Further qualitative research could examine such a possibility by examining the ways specific peer groups act out these ideals in day-to-day interactions.
Limitations and implications for further research
A number of limitations need to be considered with respect to these results. There is some evidence that self-report methods alone result in under-reporting of cigarette smoking status despite requests for honesty (Aguinis et al., 1993). Given the social prestige associated with smoking among some peer groups, it is also possible that over-reporting occurs in some instances. Second, because the sample consisted entirely of school students, usage rates obtained are likely to be underestimates of the smoking prevalence of this age group, as smokers are more likely than non-smokers to leave school at an early age (Hill et al., 1999), or be absent from school (e.g. due to truancy) on a given day (Charlton and Blair, 1989b; Charleton et al., 1997). Third, generalization cannot be made beyond the population from which these data were gathered (i.e. single-sex, female, private schools) and further research on a range of subpopulations (e.g. by manipulating school sector, gender, socio-economic status) would be useful to extend these findings. Fourth, in the current study, the sample size did not allow differentiation between different types of current smokers (e.g. occasional versus daily). It is possible that some of the variables under consideration would vary as a function of the extent to which a young person has become physically dependent upon smoking. Further, the present study was cross-sectional in nature, thus it is not possible to determine what, if any, causal relationships exist between coping, self-concept and reputation, on the one hand, and cigarette use, on the other. Longitudinal studies are thus required to address questions arising from these findings.
Finally, surveys were completed by only a little over a third of those who were approached, thus a systematic bias may exist in the characteristics of those who did/did not choose to participate. Given the ethical and logistical constraints within which studies such as these are conducted, it is disappointing, but perhaps not surprising, that response rates are not higher. In order to enlist a student’s participation in this study, for example, a number of conditions needed to be satisfied—(1) the school principal needed to respond favorably to the invitation to participate; (2) relevant staff had to ensure that students in the target age group received information about the study; (3) the student had to read this information and make a decision as to her interest/willingness to participate; (4) students had to ensure that their parent or guardian was provided with the plain language statement and consent form; (5) the parent/guardian needed to read the documents and consider whether or not they consented to their daughter participating; (6) signed documentation needed to be returned to the relevant staff member by a given due date; and (7) the student needed to be present at school on the day of testing. Clearly, many factors can interfere with one or more steps in this process and conspire to produce a spuriously low participation rate.
If replicated by future workers, the results of this study have implications for education and prevention programmes regarding uptake of cigarette smoking by teenage girls. The finding that family, social and physical self-concept are related to teenage girls’ smoking suggests that these different aspects of self-concept should be targeted from an early age, through interventions aimed both at strengthening family well-being, and the capacity of schools to foster and value positive peer relationships.
This study has highlighted the continuing importance of teaching students to evaluate a variety of coping skills and enhancing self-concept levels in young people in a bid to curb cigarette uptake. The findings also have implications for peer-led approaches to drug education, reinforcing that different peer leaders may be credible for different subgroups of students. Further longitudinal studies are required to assess the replicability of these findings, in order to determine the role that might be played by RET in both understanding cigarette uptake and developing programmes which address this. Clearly, however, no single construct or theory can be expected to comprehensively address a behavior as psychologically and sociologically complex as teenagers’ smoking.
The finding that some adolescents use cigarette smoking as a way of achieving a particular reputational status amongst their peers and others provides for a better understanding of youth culture and the reasons young people engage in substance use. It also highlights the continuing importance of the peer network and provides a deeper understanding of how this is managed by groups of current, experimental and never smokers. Drug education that recognizes the active role played by young people in defining their reputational profile should be better able to successfully target specific subgroups of adolescents. A significant challenge in this respect is the fact that smoking may be an important part of the behavioral repertoire of a young person who is striving to be seen as ‘non-conformist’ with respect to societal expectations. This raises the question then of what behavior such young people might substitute for cigarette smoking in order to achieve their desired reputation?
|
Acknowledgements |
|---|
The authors wish to thank three anonymous reviewers for the helpful and constructive comments on the manuscript.
|
Notes |
|---|
|
TopAbstractIntroductionAimsMethodProcedureResultsDiscussionNotesReferences |
|---|
1. Items on Frydenberg and Lewis’ (Frydenberg and Lewis, 1993
) Adolescent Coping Scale which reflect productive coping include: (1) Seek social support (e.g. ‘Talk to other people to help me sort it out’) and (2) Focus on systematically solving the problem (e.g. ‘Work on solving the problem to the best of my ability’).
2. Items on Frydenberg and Lewis’ (Frydenberg and Lewis, 1993) Adolescent Coping Scale which reflect non-productive coping include: (1) Self-blame (e.g. ‘Accept that I am responsible for the problem’) and (2) Keep to self and withdrawal from others (e.g. ‘Keep my feelings to myself’).
3. The sample for this survey was drawn from government, Catholic and private schools, Victoria-wide.
|
References |
|---|
|
TopAbstractIntroductionAimsMethodProcedureResultsDiscussionNotesReferences |
|---|
Abernathy, T.J., Massad, L. and Romano-Dwyer, L. (1995) The relationship between smoking and self-esteem. Adolescence, 30, 899–907.[Web of Science][Medline]
Aguinis, H., Pierce, C. and Quigley, B. (1993) Conditions under which a bogus pipeline procedure enhances validity of self-reported smoking: a meta-analytic review. Journal of Applied Social Psychology, 23, 352–373.[CrossRef]
Australian Bureau of Statistics (2001) Schools. Australian Bureau of Statistics, Canberra.
Alexander, C., Piazza, M., Mekos, D. and Valente, T. (2001) Peers, schools, and adolescent cigarette smoking. Journal of Adolescent Health, 29, 22–30.[CrossRef][Web of Science][Medline]
Bechtle, J., Tinley, V.L. and Dickey, D. (1998) Cigarette smoking and inhaled drug use among Virginia high school students. The Nurse Practitioner, 23, 8.
Byrne, D.G. and Mazanov, J. (1999) Sources of adolescent stress, smoking, and the use of other drugs. Stress Medicine, 15, 215–227.[CrossRef]
Byrne, D.G. and Mazanov, J. (2001) Self esteem, stress and cigarette smoking in adolescents. Stress and Health: Journal of the International Society for the Investigation of Stress, 17, 105–110.
Carroll, A., Houghton, S., Hattie, J. and Durkin, K. (1999) Adolescent reputation enhancement: differentiating delinquent, non-delinquent and at-risk youths. Journal of Child Psychology and Psychiatry and Allied Disciplines, 40, 593–606.[Medline]
Charlton, A. and Blair, V. (1989a) Predicting the onset of smoking in boys and girls. Social Science and Medicine, 29, 813–818.[CrossRef][Medline]
Charlton, A. and Blair, V. (1989b) Absence from school related to children’s and parental smoking habits. British Medical Journal, 298, 90–92.
Charlton, A., Mingawa, K. and While, D. (1999) Saying ‘no’ to cigarettes: a reappraisal of adolescent refusal skills. Journal of Adolescence, 22, 695–707.[CrossRef][Web of Science][Medline]
Crisp, A., Sedgwick, P., Halek, C., Joughin, N. and Humphrey, H. (1999) Why may teenage girls persist in smoking? Journal of Adolescence, 22, 657–672.[CrossRef][Web of Science][Medline]
Denscombe, M. (2001) Peer group pressure, young people and smoking: new developments and policy implications. Drugs: Education, Prevention and Policy, 8, 7–32.[CrossRef][Web of Science]
Department of Human Services (1999) School Students and Drug Use. 1996 Survey of Alcohol, Tobacco and other Drug use among Victorian Secondary School Students. Drug Treatment Services Unit, Aged, Community and Mental Health Division, Melbourne.
Emler, N. (1984) Delinquency and reputation. Progress in Experimental Personality Research, 13, 174–230.
Ennett, S.T. and Baumann, K.E. (1994) The contribution to influence and selection to adolescent peer group homogeneity: the case of adolescent cigarette smoking. Journal of Personality and Social Psychology, 67, 653–663.[CrossRef][Web of Science][Medline]
Fisher, L.A and Baumann, K.E. (1988) Influence and selection in the friend–adolescent relationship: findings from studies of adolescent smoking and drinking. Journal of Applied Social Psychology, 18, 289–314.[CrossRef][Web of Science]
Frydenberg, E. (1999) Understanding coping: towards a comprehensive theoretical framework. In Fydenberg, E. (ed.), Learning to Cope. Developing as a Person in Complex Societies. Oxford University Press, Oxford, pp. 9–30.
Frydenberg, E. and Lewis, R. (1991) Adolescent coping styles and strategies: is there functional and dysfunctional coping? Australian Journal of Guidance and Counselling, 1, 35–42.
Frydenberg, E. and Lewis, R. (1993) Adolescent Coping Scale. Australian Council for Educational Research, Melbourne.
Frydenberg, E. and Lewis, R. (1994) Coping with different concerns: consistency and variation in coping strategies used by adolescents. Australian Psychologist, 29, 45–48.
Glendinning, A. and Inglis, D. (1999) Smoking behavior in youth: the problem of low self esteem? Journal of Adolescence, 22, 673–682.[CrossRef][Web of Science][Medline]
Hattie, J. (1992) Self-concept. Lawrence Erlbaum, Hillsdale, NJ.
Heaven, P.C.L. (1989) Adolescent smoking, tough-mindedness and attitudes to authority. Australian Psychologist, 24, 27–35.
Hill, D.J., White, V.M. and Scollo, M.M. (1998) Smoking behaviors of Australian adults in 1995: trends and concerns. Medical Journal of Australia, 168, 209–213.[Web of Science][Medline]
Hill, D.J., White, V.M. and Letcher, T. (1999) Tobacco use among Australian secondary students in 1996. Australian and American Journal of Public Health, 23, 252–259.
Houghton, S. and Carroll, A. (1996) Enhancing reputations: high school adolescent males’ effective use of teacher behavior intervention strategies. Scientia Pedagogica Experimentalis, 33, 227–244.
Houghton, S., Carroll, A., Odgers, P. and Allsop, S. (1998a) Young children, adolescents and alcohol Part II: reputation enhancement and self-concept. Journal of Child and Adolescent Substance Abuse, 7, 31–56.
Houghton, S., Odgers, P. and Carroll, A. (1998b) Reputations, self-concepts and coping strategies of volatile substance users. Journal of Drug Education, 28, 199–210.[Web of Science][Medline]
Kandel, D.D. (1998) Persistent themes and new perspectives on adolescent substance use: a lifespan perspective. In Jessor, R. (ed.), New Perspectives on Adolescent Risk Behavior. Cambridge University Press, Melbourne, pp. 43–89.
Koval, J. and Pederson, L. (1999) Stress coping and other psychosocial risk factors: a model for smoking in grade 6 students. Addictive Behaviors, 24, 207–218.[CrossRef][Web of Science][Medline]
Lader, D. and Matheson, J. (1991) Smoking Among Secondary School Children in 1990. An enquiry carried out by the Social Survey Division of OPCS on behalf of the Department of Health, the Welsh Office and the Scottish Office Home and Health Department. HMSO, London.
Lucas, K. and Lloyd, B. (1999) Starting smoking: girl’s explanations of the influence of peers. Journal of Adolescence, 22, 647–655.[CrossRef][Web of Science][Medline]
McInman, A.D. and Grove, J.R. (1991) Multidimensional self-concept, cigarette smoking and intentions to smoke in adolescents. Australian Psychologist, 26, 192–196.
Michell, L. (1997) Loud, sad or bad: young people’s perceptions of peer groups and smoking. Health Education Research, 12, 1–14.
Minagawa, K., While, D. and Charlton, A. (1993) Smokers and self-perception in secondary school students. Tobacco Control, 2, 215–221.
Moon, A.M. (2000) Traditional smoking programs don’t help women. Clinical Psychiatry News, 28, 1–2.
Odgers, P. (1998) Adolescent substance use. In Hamilton, M., Kellehear, A. and Rumbold, G. (eds), Drug Use in Australia. A Harm Minimisation Approach. Oxford University Press, Melbourne, pp. 111–127.
Odgers, P., Houghton, S. and Douglas, G. (1996) Reputation enhancement theory and adolescent substance use. Journal of Child Psychology and Psychiatry and Allied Disciplines, 37, 1015–1022.[Medline]
Schofield, P.E., Pattison, P.E., Hill, D.J. and Borland, R. (2001) The influence of group identification on the adoption of peer group smoking norms. Psychology and Health, 16, 1–16.
Shiffman, S. and Wills, T.A. (eds) (1985) Coping and Substance Use. Academic Press, New York.
Song, I.S. and Hattie, J.A. (1994) Home environment, self-concept and academic achievement: a causal modelling approach. Journal of Educational Psychology, 76, 1269–1281.[CrossRef]
Thornton, W., Douglas, G. and Houghton, S. (1999) Transition through stages of smoking: the effect of gender and self-concept on adolescent smoking behavior. Journal of Adolescent Health, 25, 284–289.[CrossRef][Web of Science][Medline]
van Roosmalen, E.H. and McDaniel, S.A. (1992) Adolescent smoking intentions: gender differences in peer context. Adolescence, 27, 87–103.[Web of Science][Medline]
Received on August 31, 2001; accepted on June 24, 2002
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. R. Stevens and W. Caan Gender, economics and culture: diversity and the international evolution of smoking prevalence Perspectives in Public Health, May 1, 2008; 128(3): 113 - 116. [Abstract] [PDF]
|
|
|
|
 |
|
 R. C. Plotnikoff, K. Bercovitz, R. E. Rhodes, C. A. Loucaides, and N. Karunamuni Testing a conceptual model related to weight perceptions, physical activity and smoking in adolescents Health Educ. Res., April 1, 2007; 22(2): 192 - 202. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||