Parental factors and adolescents' smoking behavior: an extension of The theory of planned behavior
Introduction
Adolescence is a period in which many youngsters begin to experiment with smoking. In most of the Western countries, there is an increase in the prevalence of smoking in the adolescence period [2], [3]. In many of these countries, the prevalence of regular smoking for 13 year olds varies from 3.5% to 12.5% and increases to 17% to 24.5% for 15 year olds [4]. In the Netherlands, the highest increase of ever smoking among adolescents occurs in 12- to 14-year-old children [5].
It is relevant to focus attention on the first phases of smoking onset because experimenting with smoking by adolescents is not without risk. Because of the physiological dependence on nicotine, once adolescents start to smoke, it is very hard for them to quit and as a consequence they are more likely to develop a regular smoking pattern [6], [7], [8], [9], [10], [11]. Regular smoking has been found to be the annual cause of 540,000 deaths in the European Union, 461,000 deaths in the USA, and 457,000 deaths in the former USSR [12]. Because of this addictive and harmful nature of smoking, it is important to examine the precursors of the first phase of smoking onset in adolescents. The aim of the current study is to investigate whether parental factors add to The theory of planned behavior in predicting adolescents' smoking onset.
The theory of planned behavior is designed to predict and explain human behavior in specific contexts [1]. In terms of smoking, The theory of planned behavior postulates that smoking-related cognitions (i.e., attitude, self-efficacy, and social norm) predict intention to start smoking, and intention in its turn predict actual smoking onset. In addition, self-efficacy also directly predicts actual smoking onset. Various studies have found support for the predictive validity of The theory of planned behavior with respect to smoking ([13], [14], [15], [16], [17], [18], see also reviews, [19], [20]). For example, a study of De Vries et al. [21] among early adolescents showed that intention to smoke was the most powerful predictor in explaining adolescents' future smoking behavior.1 Their findings also demonstrated that the impact of smoking-related cognitions (attitude, self-efficacy, and social influences) was largely exerted through intention. The cognitions also made small unique contributions in the prediction of adolescents' smoking behavior. A positive attitude toward smoking or adolescents' perception of the social influence to smoke predicted an increased risk for adolescents to smoke. Among the smoking-related cognitions, self-efficacy was the best predictor of adolescents' smoking behavior with a high self-efficacy being negatively related to adolescents' smoking behavior.
The theory of planned behavior only includes the role of proximal cognitive factors. Petraitis et al. [19] argued that it is important to include distal factors besides the proximal factors in The theory of planned behavior to adequately predict experimental substance use (e.g., tobacco, alcohol, and marijuana use). They suggested that distal factors affect experimental substance use through proximal, cognitive factors. Since several studies have showed the relevance of parental factors, Petraitis et al. [19] proposed to add them to a potential set of distal factors. The theory of planned behavior, however, does not take into account the influence of parents when predicting adolescents' smoking behavior, although studies have showed that parents do have an influence on the smoking behavior of adolescents [13], [18], [19], [22], [23], [24], [25], [26], [27], [28], [29]. To our knowledge, only a study of Flay et al. [18] investigated the mediating effects on adolescents' smoking behavior by extending The theory of planned behavior by including parental factors as distal factors, although the parental factors were limited to parental smoking behavior and they did not examine parenting.
Some researchers have argued that parenting can be divided into two dimensions: the warmth (affect) and control dimension [30], [31]. The affect dimension of parenting encounters the quality of the parent–child relationship. The control dimension of parenting refers to manipulative, suppressing control (i.e., psychological control) or to more supervision and monitoring (i.e., strict control and parental knowledge). The present study is the first study to investigate the indirect influence of parenting factors (i.e., the quality of the parent–child relationship, psychological control, strict control, and parental knowledge) on adolescents' smoking onset through smoking-related cognitions. Most of the studies on parental factors looked mainly at the direct effects of these factors on adolescents' smoking onset. Therefore, the indirect effects as well as the direct effects of the quality of the parent–child relationship, psychological control, strict control, parental knowledge, and parental smoking behavior on adolescents' smoking behavior will be examined. Studies have investigated the direct effects of these five parental factors before and the influence of each parental factor on adolescents' smoking onset will be briefly discussed below.
First, conflicts between the parent and child might result in a negative parent–child relationship, which in turn will increase adolescents' risk to smoke [25], [26], [29]. In contrast, if parents have a positive, supportive, and stimulating relationship with their child, it decreases adolescents' likelihood to smoke [27]. Second, the influence of psychological control on adolescents' smoking behavior has been barely investigated. A study by Engels et al. [32] is one of the few studies investigating the relation of manipulative, suppressing control and adolescents' smoking onset. Findings of this study showed that higher levels of psychological control were related to higher levels of smoking onset, but only for boys. Third, several studies have examined the relation between strict control and adolescents' smoking behavior, and mixed results were reported. Studies by Chassin et al. [22], [23] indicated that higher levels of strict control were related to lower involvement in smoking. In contrast, a study of Engels et al. [32] among Dutch and Swedish adolescents demonstrated that strict control was not related to adolescents' smoking onset. Fourth, longitudinal studies have found that children of parents who have knowledge about their whereabouts and activities are less likely to start smoking [24], [25], [29]. Fifth, several longitudinal studies showed that parental smoking behavior is moderately to strongly related to smoking onset among adolescents [13], [19]. A study of Flay et al. [18] indicated that parental smoking influences adolescents' smoking onset indirectly, through negative outcome expectation, perceived parental approval, and intention. In summary, there is some evidence that parental factors are related to smoking onset and therefore these factors might be a substantial contribution to the explanatory value of The theory of planned behavior.
The aim of the present study has been to explore an extension of The theory of planned behavior by including the quality of the parent–child relationship, psychological control, strict control, parental knowledge, and parental smoking behavior as distal factors in predicting smoking onset. We investigated to what extent parental factors influenced adolescents' smoking behavior indirectly through smoking-related cognitions (see Fig. 1). We used cross-sectional data to examine the associations among parental factors, smoking cognitions, and current smoking behavior. Furthermore, longitudinal data were used to examine whether these explanatory factors also predicted future smoking behavior. It should be mentioned that no assumptions are made about the magnitude of indirect effects of each of these parental factors on adolescents' smoking onset. Data on a sample of 1,070 early adolescents were conducted to test our model.
Section snippets
Participants
The present study was conducted in the Netherlands among six secondary schools in the area of Utrecht. From these schools, all first year students participated in this study. The students were approached in two waves: in the fall (November 2000) (T1) and spring (May 2001) (T2) of the first year of secondary education. A total of 1,173 secondary school children participated in both waves. However, 103 participants were excluded because of missing data on the variable “smoking behavior of their
Results
The input matrix of correlations used for Mplus is given in Table 1. To avoid complicated graphic presentations of the models, results of the analyses are given in two parts. The measurement part of the models (lambdas) and the significant correlations between latent variables are presented in Table 2. The structural parts of the models are given in Fig. 2, Fig. 3.
The latent variable “quality of parent–child relationship” was linked to three indicators (i.e., communication, trust, and
Discussion
In this study, we examined whether The theory of planned behavior [1] could be extended with parental factors (i.e., the quality of the parent–child relationship, psychological control, strict control, parental knowledge, and parental smoking behavior) as important distal factors in predicting adolescents' smoking onset.
First of all, the results of this study, like those of other studies on smoking initiation, are to a large extent in agreement with The theory of planned behavior [1]. The model
References (57)
The theory of planned behaviour
Organ Behav. Hum. Dec
(1991)- et al.
Measuring nicotine dependence among high-risk adolescent smokers
Addict. Behav
(1996) DSM-III-R tobacco dependence and quitting during late adolescence
Addict. Behav
(1995)- et al.
Contributions of the social context to the development of adolescent substance use: a multivariate latent growth modeling approach
Drug Alcohol Depend
(1998) - et al.
Psychosocial, school and parent factors associated with recent smoking among early-adolescent boys and girls
Prev. Med
(1999) - et al.
Emotional autonomy, psychosocial adjustment and parenting: interactions, moderating and mediating effects
J. Adolesc
(1999) - et al.
Social cognitive determinants of drinking in young adults: beyond the alcohol expectancies paradigm
Addict. Behav
(2001) - et al.
Antecedents of smoking cessation among adolescents: who is motivated to change?
Prev. Med
(1998) - et al.
Do as I say: parent smoking, anti smoking socialization and smoking onset among children
Addict. Behav
(1997) - et al.
Predicting the stages of smoking acquisition according to the theory of planned behavior
J. Adolesc. Health
(1997)