Is Hong Kong Experiencing Normalization of Adolescent Drug Use? Some Reflections on the Normalization Thesis

Abstract

The upsurge of consumption of party drugs among adolescents in recent years in Hong Kong has been part of the global trend of adolescent recreational use of drugs at rave parties, discos and similar party settings. Scholars in Western societies have recently proposed the thesis of “normalization of adolescent drug use” to describe such a trend. The normalization thesis points at three major aspects of the normalization phenomenon, namely, a rapid increase of the prevalence of drug use in young people, the widespread popularity of recreational drug use that is closely linked with the recent arrival of dance club culture, and a receptive attitude towards drug use as a normal part of leisure. This article aims to examine whether the normalization thesis can be applied to analyze the situation of adolescent drug use in Hong Kong. Data are drawn from official statistics and a recent survey conducted in 2002–2004 of drug use of Hong Kong marginal youths (N = 504). The case of Hong Kong only partially supports the thesis. Our findings show that the normalization of drug use among young people has occurred in Hong Kong, but the extent of normalization is smaller than those in Western societies like the United Kingdom. They also suggest that a recognition of possible cultural differences may be complementary to the normalization thesis. Limitations of the study are also noted.

Keywords:

• drug use
• normalization
• risk society
• social capital
• adolescence
• Hong Kong

Notes

^aThe journal's style utilizes the category substance abuse as a diagnostic category. Substances are used or misused; living organisms are and can be abused. Editor's note.

The argument concerning the cost–benefit drug decision-making in the normalization thesis has introduced the thesis of rational addiction (Vuchinich and Heather, 2003). It should be noted, however, that human judgments might have irrational underpinnings (Kahneman, Slovic, and Tverskey, 1982).

All the four student surveys enumerated large samples of Form 1 to Form 7 students from mainstream Chinese-speaking secondary schools in Hong Kong. The sample size of each student survey was 84,117 for the 1990 survey, 81,100 for the 1992 survey, 84,515 for the 1996 survey, 67,100 for the 2000 survey, and 66,386 for the 2004 survey. The sampling frame was a list of secondary schools provided by the Education Department. For each survey, a two-stage stratified cluster sampling method was employed. At stage one, schools were stratified according to the criteria of location (Hong Kong Island, Kowloon, and the New Territories) and by type of school (government, aided, and private). A sample of schools were then randomly selected for each of the strata. At stage two, classes of students from selected schools were randomly selected for participation in the survey.

It should be noted that the use of different data collection methods might have affected the comparability of the data of the student sample and the marginal youth sample. Self-administered questionnaires have the advantage of encouraging more candid responses on sensitive questions, but they have the weaknesses of more incomplete questionnaires and some of the questions being misunderstood. Although the weaknesses of self-administered questionnaires could be minimized by the use of structured interviews, structured interviews also have drawbacks, including reluctance of giving true responses on sensitive questions, interviewer bias, and social desirability effect. Despite having the possibility of affecting the comparability of the data, these two methods of data collection were adopted for some reasons. Self-report questionnaires were used for the student sample because it was difficult to conduct structured interviews when the size of the student sample was very large. In data collection of marginal youths, social workers of participating outreaching social work agencies were recruited to conduct structured interviews. Conducting interviews by social workers had an advantage, since the trust between social workers and their clients could enhance the reliability of responses given by marginal youths. Several briefing sessions were arranged for social workers, in order to help them to avoid interviewer bias and social desirability effect during interviews.

The response format for the statement measuring sense of uncertainty ranges from “Strongly agree” (scored 4), “Agree” (scored 3), “Disagree” (scored 2), to “Strongly disagree” (scored 1).

The ten-item general self-efficacy scale has been adapted and widely used in Chinese context (Zhang and Schwarzer, 1995). The response format for each item ranges from “Strongly agree” (scored 4), “Agree” (scored 3), “Disagree” (scored 2), to “Strongly disagree” (scored 1). The ten items are as follows: (a) I can always manage to solve difficult problems if I try hard enough; (b) if someone opposes me, I can find means and ways to get what I want; (c) it is easy for me to stick to my aims and accomplish my goals; (d) I am confident that I could deal efficiently with unexpected events; (e) thanks to my resourcefulness, I know how to handle unforeseen situations; (f) I can solve most problems if I invest the necessary effort; (g) I can remain calm when facing difficulties because I can rely on my coping abilities; (h) when I confront with a problem, I can usually find several solutions; (i) if I am in a bind, I can usually think of something to do; and (j) no matter what comes my way, I am usually able to handle it. The ratings are summed to form a general self-efficacy scale, with scores ranging from 10 to 40. The lower the score, the lower the level of self-efficacy. Cronbach alpha for the general self-efficacy scale was 0.8 for the marginal youth sample and 0.88 for the student sample.

The concept of social capital has been used for analyzing a wide range of social behaviors, such as educational and occupational attainment (Hagan et al., 1996; Teachman, Paasch, and Carver, 1997; Lin, 2001), health (Lomas, 1998; Ziersch, 2005), post-treatment drug use (Cheung and Cheung, 2003), and crime and delinquency (Hagan and McCarthy, 1997; McNulty and Bellair, 2003; Messner, Baumer, and Rosenfeld, 2004). Social capital refers to those resources embodied in the structure of social relations, including interpersonal ties and institutional linkages (e.g., family, schools, work, and community setting) that can facilitate social actions or achievement of goals (Coleman, 1988, 1990). In order to derive social capital, embeddedness in social relations is necessary.

Family social capital and school social capital capture the extent to which parents and school generate social capital for adolescents. The measurement of family social capital focuses on three dimensions—parental informal control, parental support, and parental positive labeling. The first dimension of direct parental informal control was measured by the question: “How often did you need to seek approval from parents when going out?” The response categories for this question were “Never” (scored 1), “Seldom” (scored 2), “Sometimes” (scored 3), and “Often” (scored 4). The second dimension of parental support was assessed with the item: “How often were your parents willing to try to help you solve problems when you need their help?” The response categories for this item were “Often” (scored 4), “Sometimes” (scored 3), “Seldom” (scored 2), and “Never” (scored 1). The third dimension of parental positive labeling was measured by the question: “How did your parents think of you as a son/daughter?” The response categories for this question were “Very good” (scored 4), “Good” (scored 3), “Poor” (scored 2), and “Very poor” (scored 1). We combined the scores of the three items measuring direct parental informal control, parental support, and parental positive labeling to construct a scale of family social capital, with scores ranging from 3 to 12. The higher the score, the more family social capital an adolescent can obtain. Cronbach alpha for the family social capital scale was 0.55 for the marginal youth sample and 0.53 for the student sample. For the measurement of school social capital, we propose three dimensions including direct school informal control, school support, and teachers' positive labeling. The first dimension of direct school informal control was assessed with this question: “How strict was your school's supervision on students?” The response items for this question were “Strict/very strict” (scored 4), “Fair” (scored 3), “Not strict” (scored 2), and “Not strict at all” (scored 1). The second dimension of school support was measured by this question: “Do you think that your teachers showed concern in students' non-academic matters, such as those concerning personal hobbies, emotions, friendship and family?” The response items were “Cared very much” (scored 4), “Somewhat cared” (scored 3), “Didn't care” (scored 2) and “Didn't care at all” (scored 1). The third dimension of teachers' positive labeling was assessed with this question: “How did your teachers think of you as a student?” The response items included “Very good” (scored 4), “Good” (scored 3), “Poor” (scored 2), and “Very poor” (scored 1). The scores of the three items measuring the three dimensions of school social capital were summed to form a school social capital scale, with scores ranging from 3 to 12. The higher the score, the more school social capital an adolescent can obtain. The school social capital scale had Cronbach alpha values of 0.53 for the marginal youth sample and of 0.58 for the student sample. It should be mentioned that, unfortunately, the measurements of the family social capital scale and the school social capital scale were less than satisfactory, as their alpha values were barely acceptable. However, they were quite close to the minimally accepted value of 0.6 and, as such, they were used in the present analysis. More research efforts are needed to improve the family social capital scale and the school social capital scale constructed in this study.

Socially disadvantageous experiences are operationalized with three variables, educational disadvantage, association with drug-using peers, and trouble with law. Two dimensions are used to represent educational disadvantage: educational underachievement and diminished educational effort (Hagan and McCarthy, 1997; Hagan and Parker, 1999). Educational underachievement was measured by asking the question: “How did you rate your academic performance in school?” The response categories for this question were “Very good” (scored 4), “Quite good” (scored 3); “Quite poor” (scored 2), and “Very poor” (scored 1). Diminished educational effort was measured with this item: “Did you think you are a hard-working student?” The response items for this question were “Very lazy” (scored 4), “Quite lazy” (scored 3), “Quite hard-working” (scored 2), and “Very hard-working” (scored 1). We constructed an educational disadvantage scale by combining the scores of the measures of the two dimensions. Scores of this scale were from 2 to 8. Cronbach alpha of this scale was 0.62 for the sample of marginal youths and 0.6 for the sample of students. The measurement of association with drug-using peers consisted of two items asking the respondent to report how many of his/her good friends had engaged in drug use in Hong Kong, Shenzhen, or other locations of mainland China. Two questions were asked: (a) “How many of your good friends have taken drugs in Hong Kong?”; and (b) “How many of your good friends have taken drugs in Shenzhen or other places of mainland China?” The response categories for each question were “Many” (scored 4), “Some” (scored 3), “Very few” (scored 2), and “No/don't know” (scored 1). A scale of association with drug-using peers were formed by summing the scores of the two questionnaire items, with scores ranging from 2 to 8. Cronbach alpha for this scale was 0.59 for the marginal youth sample and 0.67 for the student sample. To assess trouble with law, we asked the respondent to report how many times he had been put under police superintendent's discretionary scheme, probation orders or community service orders, or had been sent to penal institutions prior to the interview (or prior to the admission to the residential drug treatment center or penal institution, if the respondent was at the time of interview under the residential drug treatment, or if the respondent was still in penal institution). The higher the scores, the more the socially disadvantageous experiences an adolescent has encountered.