Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Procedia Soc Behav Sci. Author manuscript; available in PMC 2011 January 1.
Published in final edited form as:
Procedia Soc Behav Sci. 2010; 2(2): 1620–1624.
doi:  10.1016/j.sbspro.2010.03.247
PMCID: PMC2997470

School as a risk factor for psychoactive substance use by middle school students


For the majority of Polish students school is a source of negative experiences and therefore may increase the risk of adolescent problem behaviors. The results of the study conducted in Warsaw middle schools (N=2244, 54% girls) indicated that changes for worse (between 7 and 8 grade) in students' behavior increase the risk of drug use. However, changes for better in students' perception of school value and school achievements are risk factors, too (even when family and peer risk factors are controlled).

Keywords: school, adolescents, alcohol, cigarettes, cannabis, risk factors

1. Introduction

School, together with family and peers, is one of the most important socialization factors. According to primary socialization theory (Oettting & Donnemeyer, 1998) explaining adolescents drug use and deviance, school and family are usually primary sources of positive, pro-social values, while peer influences lead teenagers toward breaking of social rules. Several studies have proved that this assumption holds true for the schools that enhance students' engagement and are perceived by adolescents as having “good climate” (e.g.: Hawkins, Catalano, Miller, 1992; McBride et al., 1995; Coker & Borders, 2001; LaRusso, Romer, Selman, 2008). Besides negative school climate (and students' disengagement, often indicated by truancy), well known school related factors enhancing the risk of using psychoactive substances include low educational aspirations and low school achievements (e.g.: Newcomb & Felix-Ortiz, 1992; Jessor, Van Den Bos, Vanderryn, Costa, Turbin, 1995; Dekovic, 1999; Epstein, Botvin, Griffin, Diaz, 2001).

In Poland as a rule, schools have been a source of negative experiences for most students (Dabrowska-Bak, 1987). In general, they declare feeling of alienation from school, lack of support from teachers and peers, teachers' unfairness and indifference (Karolczak-Biernacka, 2000). Among Polish adolescents compared to adolescents from other EU countries, lower rates of those who like their school and higher rates of those who not only dislike school, but also feel loaded with school work were observed (Woynarowska, 2005).

The low level of various indicators of students well-being at school suggests the questioning of the thesis that schools in general enhance positive behaviors (as it is assumed in primary socialization theory). The key question is whether school experiences lead Polish students to pro-social, healthy behaviors or to breaking of social rules?

The aim of this study was to assess the influence of changes in school related factors, such as grades, behaviors (truancy), perception of school climate and importance of school in general on psychoactive substance use. It was hypothesized that change for the worse increases the risk of substance use, while improvement in school related factors decreases the likelihood of substance use. It was also expected, that the influence of school related factors on adolescent problem behaviors is significant even when family and peer related risk factors are controlled.

2. Method

2.1. Respondents selection and sample attrition

This paper is based on the self-report data taken in the school year 2006/2007 (Wave 1) and 2007/2008 (Wave 2) from the representative sample of Warsaw middle school 7th (N=3165) and 8th (N=3141) grade students. Students' average age in wave 1 was 13,5 years.

A class was the unit of randomization and the final sample of 158 classes was selected from the sample frame of 600 seventh grade classes from all public and non-public schools. Once classes were selected and consents for the study were obtained from schools principals, parents and youths, the students participated in the anonymous survey conducted in class by specially trained research assistants.

The data attrition in both measures was due to schools principals or parents' refusals to participate in the study and students absenteeism during the data collection period. Therefore, the Wave 1 data base included 3103 students, and the Wave 2 – 3087 students which constituted in both waves about 82% of the original sample.

The questionnaires of 157 students were excluded from data files because of extensive missing data or answers indicating the youth did not take a study seriously (jokes, drawings and inconsistent answers). After matching Wave 1 and Wave 2 questionnaires (each student had a special ID code to track students in longitudinal study) the analyzed data included 2244 adolescents which constituted about 72% response rate from Wave 1 and Wave 2. Girls accounted for 54% of the sample.

2.2. Measures

Most of the measures used in the study (except the questions concerning school grades, which had to be adequate to the Polish school system), were adapted from the American - the Flint Adolescent Study (Zimmerman & Schmeelk-Cone, 2003).

2.2.1. Dependent variables

Dependent variables were assessed according to the data from Wave 2 and they concerned various substance use: cigarettes smoking and alcohol use in the past 30 days and getting drunk and cannabis use in the past 12 months. Because of high skewness of all dependent variables, the answers were classified as never or at least once and logistic regression method was chosen to test hypotheses.

2.2.2. Independent variables

Family variables included in the analysis concerned risk factors with well documented influence on adolescents' substance use, such as: family composition, low level of parental education, bad budget situation, hostile climate and family alcohol problems (e.g.: Garmezy, 1985; Luthar & Zigler, 1991; Zucker et al., 2003). To assess the relative meaning of peer related factors two indicators were chosen: association with negative friends (e.g.: Newcomb & Felix-Ortiz, 1992; Jessor et al., 1995; Epstein et al., 2001) and support from friends (higher support is associated with higher risk for alcohol and drug use, as demonstrated in Ostaszewski, 2009). All family and peer factor measures were taken from Wave 1.

Family variables were recoded to have binary values: family composition (two parents family versus other), family budget situation perceived by students (good versus at most mean), mother/father education (primary/vocational vs. secondary/university degree) and alcohol problems (measured by a single question: Do conflicts related to alcohol use by a member of your family take place in your home?, coded as: no or yes). The continuous measure of one family factor – hostile climate - was used. The scale consisted of 5 items (with 4-point answers scale each) concerning the frequency of quarrels, criticism, physical aggression, anger and losing temper in the family. It had good reliability (Cronbach's alpha=0,824) and normal distribution.

To assess negative peer influences the scale consisting of 17 items, asking: How many of your close friends …. e.g. skip classes, have left school, drink alcohol, etc. (answers ranged from any to all of them) was used. The scale had good reliability (alpha=0,888) but was highly skewed (2,869) and flat (kurtoza=13,475), so nominal coding was used (light, moderate and strong bad influences were distinguished). The continuous measure of friends' support was based on the 4-item's scale (alpha=0,865, normal distribution) asking whether it is true or false that a respondent can count on friends' emotional support and help to deal with problems and vice versa.

School variables were constructed by the combination of Wave 1 and Wave 2 measures to indicate the changes between grade 7 and grade 8, in students: grade point average (GPA) and evaluation of school behavior in the last semester; truancy (measured by two separate questions concerning skipping classes and skipping whole school days in the past four weeks); perception of the importance of school (4-items scale, sample item: School grades are very important for me, Wave 1 alpha=0,852 and Wave 2 - 0,873) and perception of school climate (7-items scale, sample item: I like my teachers, Wave 1 alpha=0,817 and Wave 2 - 0,819). All indicators of school related factors were coded in the same way into 3-cathegoris: increase/improvement, decrease/worsening or no change.

3. Results

Alcohol use in the past 30 days was reported by 33% of adolescents and cigarette use by 16%. Nearly one fourth of adolescents got drunk (23%), but only 6% used cannabis in the past year.

As shown in Table 1, the majority of study participants lived in two-parents families, without alcohol related problems and described their parents as well educated. More than a half of the sample admitted that the budget situation of their families is rather good. Nearly three in four adolescents experienced at least moderate negative peer influences. The rates of the students who between 7 and 8 grade became more skeptical about school's climate and importance, as well as the rates of students' truancy and lower achievements were much higher than the rates of those who improved their behaviors, grades and started to perceive school in more positive way.

Table 1
Sample characteristics

Cross-gender comparisons revealed only some significant differences: cannabis use was more prevalent among boys (7.4%) than girls (4.1%; χ2= 10.520, p=0.001); boys perceived their family situation as better than girls, which was indicated by their evaluation of family budget (χ2= 14.536, p=0.000), maternal education (χ2= 4.601, p=0.033) and less hostile climate (mean value for boys = 7.98 and for girls = 8.80; t=-6.515, p=0.000); girls assessed the relationship with friends as more supportive than boys (mean value for boys = 13.37 and for girls = 16.54; t=-21.613, p=0.000); boys more often than girls skipped some school classes (18.7% do 14.9%, χ2= 5.675, p=0.019).

To answer the study question, binominal logistic regression analyses were conducted separately for each of the dependent variables (cigarette, alcohol and cannabis use and getting drunk). Predictors were entered with forward conditional option in three steps: gender and family factors (step 1), negative peer influences (step 2) and school factors (step 3). Table 2 shows the results of the final step of each equation.

Table 2
Logistic regression of psychoactive substance use (Wave 2) on gender, family factors (Wave 1), negative peer influences (Wave 1) and changes in school functioning (between Wave 1 and 2)

Among family factors, the most important predictor of adolescents' substance use was the hostile climate (it became insignificant only for cannabis use having added school factors). Inclusion of school variables diminished the significance of family composition impact on cigarette use, too (this factor remained significant as a predictor of getting drunk and cannabis use).

In all equations inclusion the of peer influences (step 2) and school related variables (step3) improved predictions significantly. The likelihood of using psychoactive substances increased with the number of peers presenting risky behaviors. Moreover, the odds of getting drunk increased with the level of friends' support.

The probability of psychoactive substance use was higher for those students who in grade 8 compared to grade 7 skipped more school days (significant effect for alcohol use, smoking cigarettes and getting drunk) and skipped more classes (cannabis and alcohol use). Additionally, the odds of drinking alcohol and smoking cigarettes for adolescents who received lower grades for their behavior in school were higher than for adolescents whose grades for behavior did not change. The relationships between changes in grade point average (GPA) and alcohol use and abuse are more complicated. The study results suggest that the improvement of school grades is associated with drunkenness and that any changes (for better, as well as, for worse) may increase the risk of drinking alcohol. Similar two-sided effect is visible for changes in students perception of the importance of school as a risk factor for using and abusing alcohol.

4. Conclusions

This study results showed that moving from the 7th to the 8th grade is related to negative rather than positive changes, in terms of adolescent behaviors, achievements and attitudes toward school. As expected, the increase in truancy rates is associated with heavier alcohol, cigarette and cannabis use. Lower evaluation of school behavior in grade 8 than in grade 7, is a risk factor for alcohol and cigarette use. More surprising are the results suggesting that any change in the perception of the importance of school, for better, as well as for worse, increases the risk of using and abusing alcohol. Moreover, the lowest rates of alcohol use are observed for students who during one school year did not change their average school grades and the risk of getting drunk is the highest for those whose GPA became higher in grade 8 than in grade 7.

These results may be interpreted from the developmental perspective, as one of adolescents' steps toward adulthood. Adolescence is a period of testing new (often socially accepted for adults only) behaviors and attitudes, including functioning in school and psychoactive substance use. Perhaps 13-14 year olds who do not change their perception of school and whose school achievements are still at the same level are not yet in the period of experimenting in other life areas? And this is maybe the reason why they drink less alcohol?

The other hypothetical explanation links better GPA with high social attractiveness (which in this study was indicated by friends' support) which, on the other hand, is usually associated with alcohol use in peers' company (e.g.: Jarvinen & Gundelach, 2007; Bogren, 2006).

Even in spite of study limits (high data attrition between Wave 1 and Wave 2, measures based on self-reported and therefore not fully reliable information) this study provides interesting results suggesting a closer look at the changes in students' functioning in middle school. In fact the studies exploring the relationship between school achievements and heavy alcohol use are needed.


This research project has been supported by a grant from the Fogarty International Center, U.S. National Institute of Health, grant No 5R01TW007647


  • Bogren A. The competent drinker, the authentic person and strong person: lines of reasoning in Swedish young people's discussions about alcohol. Journal of Youth Studies. 2006;9(5):515–538.
  • Coker JK, Borders LD. An analysis of environmental and social factors affecting adolescent problem drinking. Journal of Counseling and Development. 2001 Spring;79(2):200–208.
  • Dabrowska-Bak M. Przemoc w szkole. Kwartalnik Pedagogiczny. 1987;2
  • Epstein JA, Botvin GJ, Griffin KW, Diaz T. Protective factors buffer effects of risk factors on alcohol use among inner-city youth. Journal of Child and Adolescent Substance Abuse. 2001;11(1):77–90.
  • Garmezy N. Stres-Resistant Children: The Search for Protective Factors. In: Stevenson J, editor. Recent Research in Developmnetal Psychopatology. Oxford – New York – Toronto – Sydney – Paris – Frankfurt: Pergamon Press; 1985. pp. 213–234.
  • Hawkins J, Catalano R, Miller J. Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for substance abuse prevention. Psychological Bulletin. 1992;112(1):64–105. [PubMed]
  • Jarvinen M, Gundelach P. Teenage drinking, symbolic capital and distinction. Journal of Youth Studies. 2007;10(1):55–71.
  • Jessor R, Van Den Bos J, Vanderryn J, Costa FM, Turbin MS. Protective factors in adolescent problem behavior: Moderator effects and developmental change. Developmental Psychology. 1995;31(6):923–933.
  • Karolczak-Biernacka B. Cechy uczniow a zrodlo poczucia zagrozenia. In: Czabala C, editor. Zdrowie psychiczne. Zagrożenia i promocja. Warszawa: Instytut Psychiatrii i Neurologii; 2000. pp. 14–39.
  • LaRusso MD, Romer D, Selman RL. Teachers as builders of respectful school climates: Implications for adolescent drug use norms and depressive symptoms in high school. Journal of Youth and Adolescence. 2008;37:386–398.
  • Luthar SS, Zigler E. Vulnerability and Competence: A Review of Research on Resilience in Childhood. American Journal of Orthopsychiatry. 1991;61(1):6–22. [PubMed]
  • McBride CM, Curry SJ, Cheadle A, Anderman C, Wagner EH, Diehr P, Psaty B. School-level application of a social bonding model to adolescent risk-taking behavior. The Journal of School Health. 1995;65(2):63–68. [PubMed]
  • Newcomb MD, Felix-Ortiz M. Multiple Protective factors for drug use and abuse: Cross-sectional and prospective Findings. Journal of Personality and Social Psychology. 1992;63(2):280–296. [PubMed]
  • Oetting ER, Donnermeyer JF. Primary Socialization Theory: The etiology of drug use and deviance. I. Substance Use and Misuse. 1998;33(4):995–1026. [PubMed]
  • Ostaszewski K. Determinants of health and health behaviours in Polish adolescents. Warszawa: Institute of Mother and Child; 2009. Risk/protective factors in salient adolescent problem behaviors. The Warsaw Adolescent Study; pp. 33–57.
  • Woynarowska B. Szkola w opinii mlodziezy w Polsce i w krajach Unii Europejskiej. Remedium. 2005;6:14–15.
  • Zimmerman MA, Schmeelk-Cone KH. A longitudinal analysis of adolescent substance use and school motivation among African American youth. Journal of Research on Adolescence. 2003;13(2):185–210.
  • Zucker RA, Wong MM, Puttler LI, Fitzgerald HE. Resilience and vulnerability among sons of alcoholics. Relationship to developmental outcomes between early childhood and adolescence. In: Luthar SS, editor. Resilience and Vulnerability. Cambridge University Press; 2003. pp. 76–103.