Skip to main content

Peer influences on adolescent alcohol consumption: evidence using an instrumental variables/fixed effect approach


In order to address empirical difficulties in research examining peer effects in alcohol consumption, I use instrumental variables/fixed effects methodology that compares students in different grades within the same school who face a different set of classmate decisions. Within this context, I suggest that alcohol availability in classmates’ homes and classmates’ parents’ alcohol abuse can be used as instruments. Results indicate that a 10% increase in the proportion of classmates who drink increases the likelihood an individual drinks by five percentage points. This paper also provides evidence of peer effects in problem drinking, such as binge drinking and drunkenness.

This is a preview of subscription content, access via your institution.


  1. See also the recent discussion between Cohen-Cole and Fletcher (2008a, 2008bb) and Christakis and Fowler (2007) and Fowler and Christakis (2008) for examinations of peer influences on weight status. See Cohen-Cole and Fletcher (2008c) for an example where one can estimate large “social effects” for outcomes where the true social effect is zero, such as the transmission of height. The authors argue that weak empirical models commonly employed in the literature can produce these results.

  2. The assumption of the relevant reference group is also an important difficulty with social interactions research. I follow the literature in assuming that classmates are a relevant reference group, although other researchers have assumed larger types of reference groups (e.g. city blocks by Case and Katz 1991).

  3. If all families choose schools based on time-invariant school characteristics, then controlling for school fixed effects controls for the main source of selection into schools.

  4. Hoxby (2000) and others have pointed out that contextual effects that are non-linear could imply distributional consequences of changing the composition of schools.

  5. Norton et al. (1998) focuses on younger students and defines peer groups at the neighborhood level rather than the school level. There is also a complementary research area examining peer influences on alcohol consumption of college students (Sacerdote 2001; Kremer and Lavy 2003; DeSimone 2007). See also Kooreman (2007).

  6. Gaviria and Raphael (2001) and others argue that this approach is valid because, “...students are less exposed to the family background of their school peers than they are exposed to the family background of peers residing in the same neighborhood.”

  7. The authors compared individuals who were recent movers versus those who were immobile to examine the potential bias from the endogeneity of school and found relatively large differences—a difference of nearly 16% in the estimated endogenous effect coefficient in the case of drinking.

  8. McEwan (2003) uses a similar approach with data from Chile on educational achievement for middle-schoolers.

  9. Manski (1995, p.136) states, “Of course one cannot simply specify a dynamic model and claim that the problem of inference on social effects has been resolved. Dynamic analysis is meaningful only if one has reason to believe that the transmission of social effects follows the assumed temporal pattern.” Clark and Loheac also note that they must also assume that behavior is not in a steady state equilibrium in order to use their approach.

  10. Clark and Loheac (2007) report that the correlation between the one-year lagged peer drinking measure and contemporaneous measure is 0.43.

  11. Increasing the proportion of male peer group members who drink by 25% is predicted to increase own-drinking participation by 4.5%.

  12. This paper is similar in spirit to Hanushek et al. (2003) and Hoxby (2000) who use administrative data from Texas to examine peer effects in educational achievement. Important differences include the use of non-administrative data that includes rich family-level information as well as the examination of risky behaviors decisions rather than academic achievement. Using survey data rather than administrative data allows me to have broad geographic coverage, increasing the generalizeabilty of the results, as well as the ability to examine risky behaviors that are not included in school administrative data. Bifulco et al. (2011) use the same data but focus on the effects of peer characteristics (e.g. racial composition) rather than peer behaviors.

  13. See Udry (2003) for full description of the Add Health data set. Also see for further information:

  14. In order to keep 3,000 students whose parent did not complete the separate parental survey, I impute family income and maternal education and create a dummy variable for missing parental data.

  15. The exact wording of the question was, “During the past 12 months, on how many days did you drink alcohol?”

  16. The Add Health Picture Vocabulary Test (AHPVT) is a computerized, abridged version of the Peabody Picture Vocabulary Test-Revised (PPVT-R). The AHPVT is a test of hearing vocabulary, designed for persons aged 2 1/2 to 40 years old who can see and hear reasonably well and who understand standard English to some degree. Each test included a set of practice, or pretest items, followed by a series of test items arranged in order of increasing difficulty. The respondent was asked to listen to the word spoken by the interviewer and to select the picture on the plate that he or she believed best illustrated the meaning of the stimulus word. Once the response was entered into the computer, the program indicated the next plate to use in the test. In addition, the computer program determined test results automatically. These test results were not made available to the interviewer or to the respondent.” The test scores are standardized by age. Some psychologists interpret PVT scores as a measure of verbal IQ. Information on the test is provided online at

  17. See Averett et al. (2011).

  18. The survey question is: “Is alcohol easily available to you in your home?” The responses are “yes” or “no”.

  19. Previous research has used variation in multiple cohorts’ peer characteristics within a school to examine the effects of peer characteristics (but not outcomes) on individual outcomes, where the coefficient of interest is δ rather than α (e.g. Hoxby 2000; Hanushek et al. 2003; Lavy and Schlosser 2007; Bifulco et al. 2011).

  20. One potential weakness of the grademates’ alcohol availability instrument is that, in principle, it could have direct effects on the consumption patterns of a student though providing direct alcohol access. For example, some grademates are also friends. However, the grademates’ parental alcohol status measure should not have a similar problem. The over identification tests shown below suggest similar results regardless of the instrument used, which reduces the concern with the availability instrument, However, this potential limitation with the instrument should be considered when evaluating the results.

  21. Results stratified by race and gender are available upon request.


  • Arcidiacono P, Nicholson S (2005) Peer effects in medical school. J Public Econ 89(2–3):327–350

    Article  Google Scholar 

  • Averett SL, Argys LM, Rees DI (2011) Older siblings and adolescent risky behavior: does parenting play a role? J Popul Econ. doi:10.1007/s00148-009-0276-1

    Google Scholar 

  • Bifulco R, Fletcher JM, Ross SL (2011) The effect of classmate characteristics on individual outcomes: evidence from the add health. Am Econ J: Econ Polic 3(1):25–53

    Article  Google Scholar 

  • Brock W, Durlauf S (2001) Discrete choice with social interactions. Rev Econ Stud 68(2):235–260

    Article  Google Scholar 

  • Carpenter C, Dobkin C (2008) The drinking age, alcohol consumption, and crime. Working Paper

  • Carpenter C, Dobkin C (2009) The effect of alcohol access on consumption and mortality: regression discontinuity evidence from the minimum drinking age. Am Econ J: Appl Econ 1(1):164–182

    Article  Google Scholar 

  • Case A, Katz L (1991) The company you keep: the effects of family and neighborhood on disadvantaged youth. NBER Working Paper 3705

  • Christakis N, Fowler J (2007) The spread of obesity in a large social network over 32 years. N Engl J Med 357(4):370–379

    Article  Google Scholar 

  • Clark A, Loheac Y (2007) ‘It wasn’t me, it was them!’ Social influence in risky behavior by adolescents. J Health Econ 26(4):763–784

    Article  Google Scholar 

  • Cohen-Cole E, Fletcher JM (2008a) Is obesity contagious? Social networks vs. environmental factors in the obesity epidemic. J Health Econ 27(5):1382–1387

    Article  Google Scholar 

  • Cohen-Cole E, Fletcher JM (2008b) Estimating peer effects in health outcomes: replies and corrections to fowler and christakis. SSRN Working Paper:

  • Cohen-Cole E, Fletcher JM (2008c) Detecting implausible social network effects in acne, height, and headaches: longitudinal analysis. Br Med J 337:a2533

    Article  Google Scholar 

  • Cook PJ, Ostermann J, Sloan F (2005) Are alcohol excise taxes good for us? Short and long term effects on mortality rates. NBER Working Paper 11138

  • DeSimone J (2007) Fraternity membership and binge drinking. J Health Econ 26(5):950–967

    Article  Google Scholar 

  • Finlay K, Magnusson LM (2009) Implementing weak instrument robust tests for a general class of instrumental variables models. Stata J 9(3):398–421

    Google Scholar 

  • Fletcher JM (2007) Social multipliers in the sexual initiation decisions among U.S. high school students. Demography 44(2):373–388

    Article  Google Scholar 

  • Fletcher JM (2010a) Social interactions and smoking decisions: evidence using multiple cohorts, instrumental variables, and school fixed effects. Health Econ 19(4):466–484

    Article  Google Scholar 

  • Fletcher JM (2010b) Social interaction and college enrollment: evidence from NELS. Yale University Working Paper

  • Fletcher JM, Tienda M (2009) High school classmates and college success. Sociol Educ 82(4):287–314

    Article  Google Scholar 

  • Fowler JH, Christakis NA (2008) Estimating peer effects on health in social networks: a response to Cohen-Cole and Fletcher; and Trogdon, Nonnemaker, and Pais. J Health Econ 27(5):1400–1405

    Article  Google Scholar 

  • Gaviria A, Raphael S (2001) School-based peer effects and Juvenile behavior. Rev Econ Stat 83(2):257–268

    Article  Google Scholar 

  • Hanushek EA, Kain JF, Markman JM, Rivkin SG (2003) Does peer ability affect student achievement? J Appl Econ 18(5):527–544

    Article  Google Scholar 

  • Hoxby C (2000) Peer effects in the classroom: learning from gender and race variation. NBER Working Paper 7867

  • Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE (2008) Monitoring the future national results on adolescent drug use: overview of key findings, 2007 (NIH Publication No. 08-6418). National Institute on Drug Abuse, Bethesda.

  • Kawaguchi D (2004) Peer effects on substance use among american teenagers. J Popul Econ 17(2):351–367

    Article  Google Scholar 

  • Keng SH, Huffman WE (2010) Binge drinking and labor market success: a longitudinal study on young people. J Popul Econ 23(1):303–322

    Article  Google Scholar 

  • Komro KA, Maldonado-Molina MM, Tobler AL, Bonds J, Muller KE (2007) Effects of home access and availability of alcohol on young adolescents’ alcohol use. Addiction 102(10):1597–1608

    Article  Google Scholar 

  • Kooreman P (2007) Time, money, peers, and parents: some data and theories on teenage behavior. J Popul Econ 20(1):9–33

    Article  Google Scholar 

  • Kremer M, Lavy V (2003) Peer effects and alcohol use among college students. NBER Working Paper 9876

  • Lavy V, Schlosser A (2007) Mechanisms and impacts of gender peer effects at school. NBER Working Paper 13292

  • Lundborg P (2006) Having the wrong friends? Peer effects in adolescent substance use. J Health Econ 25(2):214–233

    Article  Google Scholar 

  • Manski C (1993) Identification of endogenous social effects: the reflection problem. Rev Econ Stud 60(3):531–542

    Article  Google Scholar 

  • Manski C (1995) Identification problems in the social sciences. Harvard University Press, Cambridge

    Google Scholar 

  • Manski C (2000) Economic analysis of social interactions. J Econ Perspect 14(3):115–136

    Article  Google Scholar 

  • McEwan P (2003) Peer effects on student achievement: evidence from Chile. Econ Educ Rev 22(2):131–141

    Article  Google Scholar 

  • Moore MJ, Cook PJ (1995) Habit and heterogeneity in the youthful demand for alcohol. NBER Working Paper 5152

  • Norton EC, Lindrooth RC, Ennett ST (1998) Controlling for the endogeneity of peer substance use on adolescent alcohol and tobacco use. Health Econ 7(5):439–453

    Article  Google Scholar 

  • Powell L, Taurus J, Ross H (2005) The importance of peer effects, cigarette prices, and tobacco control policies for youth smoking behavior. J Health Econ 24:950–968

    Article  Google Scholar 

  • Rehm J, Gmel G, Sempos CT, Trevisan M (2003) Alcohol-related morbidity and mortality. Alcohol Res Health 27(1):39–51

    Google Scholar 

  • Rothstein J (2008) Teacher quality in educational production: tracking, decay, and student achievement. Working Paper.

  • Sacerdote B (2001) Peer effects with random assignments: results from Dartmouth roommates. Q J Econ 116(2):681–704

    Article  Google Scholar 

  • Summers A, Wolfe B (1977) Do schools make a difference? Am Econ Rev 67(4):639–652

    Google Scholar 

  • Udry JR (2003) The National Longitudinal Study of Adolescent Health (Add Health), Waves I and II, 1994–1996; Wave III, 2001–2002, Carolina Population Center, University of North Carolina, Chapel Hill, NC (2003) machine-readable data file and documentation

  • Vigdor J, Nechyba T (2004) Peer effects in elementary school: learning from “apparent” random assignment. Working Paper

  • Williams J (2005) Habit formation and college students’ demand for alcohol. Health Econ 14(2):119–134

    Article  Google Scholar 

  • Zimmerman D (2003) Peer effects in academic outcomes: evidence from a natural experiment. Rev Econ Stat 85(1):9–23

    Article  Google Scholar 

Download references


The author thanks two anonymous reviewers, Kitt Carpenter, Jeff DeSimone, Andrew Francis, David Frisvold, Dave Marcotte, Sara Markowitz, and seminar participants at Emory University, University of Maryland-Baltimore County, and the 2008 Southern Economic Association Meetings for very helpful comments.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Jason M. Fletcher.

Additional information

Responsible editor: Erdal Tekin

This research uses data from Add Health, a program project directed by Kathleen Mullan Harris and designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris at the University of North Carolina at Chapel Hill and funded by grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 23 other federal agencies and foundations. Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Information on how to obtain the Add Health data files is available on the Add Health website ( No direct support was received from grant P01-HD31921 for this analysis.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

(PDF 197 KB)

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Fletcher, J.M. Peer influences on adolescent alcohol consumption: evidence using an instrumental variables/fixed effect approach. J Popul Econ 25, 1265–1286 (2012).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Peer effects
  • Social interactions
  • Alcohol consumption

JEL Classification

  • I12
  • I20