Clueless: Parental Knowledge of Risk Behaviors of Middle School Students | Adolescent Medicine | JAMA Pediatrics | JAMA Network
[Skip to Navigation]
Sign In
November 1998

Clueless: Parental Knowledge of Risk Behaviors of Middle School Students

Author Affiliations

From the University of Kentucky Medical Center, Departments of Pediatrics (Dr Young) and Behavioral Medicine (Dr Zimmerman), Kentucky Clinic, Lexington.

Arch Pediatr Adolesc Med. 1998;152(11):1137-1139. doi:10.1001/archpedi.152.11.1137

Objectives  To determine parental knowledge of risk behaviors of their middle-school-aged children and to compare that knowledge with behaviors reported by the students.

Design  Confidential surveys were administered to 194 poor and middle-class middle school students and their parents.

Intervention  None.

Main Outcome Measure  Compare parental awareness and student admission of risk behaviors.

Results  Students and parents agree on the prevalence of some risk behaviors, including use of a seat belt, use of a bicycle helmet, arrest by police, use of diet pills, and attempts to lose weight through dieting and exercise. Small differences in perception, which were not statistically significant but which could be defined as trends, were found relating to the prevalence of dieting, exercising, or vomiting to lose weight. Statistically significant differences were found in the perceptions of the prevalence of the following risk behaviors: carrying a weapon to school (P<.001), LSD (lysergic acid diethylamide) or cocaine use (P=.02), suicide attempt (P<.001), sexual intercourse (P<.001), alcohol use (P<.001), tobacco use (P<.001), and marijuana use (P<.001). In many cases, parents dramatically underestimated the prevalence of these behaviors in students.

Conclusions  Parents are largely ignorant of the extent to which their adolescents are involved in major risk behaviors. Educating parents in this area encourages them to support comprehensive health education curricula and other preventive programs in schools and communities.

MANY HEALTH problems are associated with poor lifestyle choices.1,2 In adolescents, we monitor these poor choices by measuring risk behaviors. Some of the major risk behaviors are tobacco use, alcohol and other drug use, unprotected sexual activity, failure to use seat belts, and suicide attempts. While national surveys of risk behaviors have been conducted for years, particularly the Centers for Disease Control and Prevention's Youth Risk Behavior Survey,3 few have measured parent perception of adolescent risk behaviors. Parent and community awareness of adolescent risk behaviors has an effect on efforts to reduce these risk behaviors. Suprisingly. there are no local or national policies to address unprotected sexual activity, use of alcohol and other drugs, suicide risk, and other risk behaviors in adolescents. If parents are truly unaware of the kind of risk behaviors that their children are involved in, informing them of their lack of knowledge may stimulate their interest in efforts to reduce the prevalence of these behaviors.

As the school system is largely responsible for educating its students, it is responsible for making efforts to reduce the prevalence of the many risk behaviors that have a detrimental effect on the learning process. Parental support of school intervention programs, such as the Comprehensive School Health Education (CSHE) program, is essential. To gain this support, we need to know if parents accurately perceive the extent to which their child is involved in these behaviors. Informing parents that these perceptions may be inaccurate will help win their support for the CSHE program. Few studies have been done that compare student risk behaviors with parents' perception of these behaviors. Deffenbaugh et al4 conducted a survey of 2125 students (grades 4-12) and their parents on student risk behaviors. Significant differences were found between risk behaviors as reported by the students and as perceived by the parents relating to tobacco, alcohol, marijuana, and inhalant use. However, parents were not asked to report specifically on their perception of their own child's risk behaviors as they were in this study.

If parents are ignorant of the risk behaviors that their children engage in, they may show apathy toward programs designed to prevent the risk behaviors. This study was designed to measure the differences between students' reports and parents' perceptions of the students' risk behaviors.

Participants and methods

Survey design

A survey was designed as part of a school health needs assessment for a Community Access for Child Health grant from the American Academy of Pediatrics, Elk Grove Village, Ill. This survey assessed youth risk behaviors, student and parent concerns about different health problems, and their interest in many school-based health services. The Centers for Disease Control and Prevention Youth Risk Behavior Survey (1995)3 and a previous Community Access for Child Health needs assessment survey developed by Touloukian5 were used as models for this survey. Pilot surveys confirmed that students and parents understood the questions. Passive consent was used.

Survey population

As part of a schoolwide needs assessment, students enrolled in mandatory health classes in middle and high schools in one central Kentucky school district were surveyed. Health classes were chosen because they are mandatory for all students. Parents of the students at one middle school were surveyed after a broader, nonmatched survey revealed large differences in reported student risk behaviors. The surveys were completed in 1995. Parents were instructed to answer based on what they believed were their child's risk behaviors. Surveys were anonymous for both students and parents. Students and parents could not be matched on an individual basis. One hundred forty students and 140 parents participated in the survey. Participation was voluntary for both students and parents. Students completed the survey during class and parents completed their surveys at home and returned them to the classroom teachers. Parents had a 69% (96/140) survey completion rate and students had a 64% (89/140) completion rate. χ2 Tests of independence were conducted to assess similarity or difference in prevalence rates of risk behaviors reported by students and perceived by parents.


The students were in the seventh and eighth grades; girls represented 45%, and boys, 55%. Races included white (55%), African American (39%), Asian (1%), Hispanic (2%), and other (3%). Risk behaviors can be categorized into 3 groups: those that students and parents agreed on, those that they disagreed on slightly (not statistically significant), and those that they disagreed on significantly. Table 1 lists the health areas and risk behaviors that the parents and students estimated similarly. These risk behaviors and health risks tended to be the ones that can be easily observed by parents.

Table 1. 
Student and Parent Agreement
Student and Parent Agreement

In the second group of risk behaviors, there were some differences in perception between students and parents, but they were not statistically significant, as given in Table 2 (P<.10, but P>.05). Having had a physical fight in school almost reached statistical significance (P=.06).

Table 2. 
Trends but Not Statistically Significant Differences*
Trends but Not Statistically Significant Differences*

In the third group of risk behaviors, the differences between student and parent perceptions were not only statistically significant, but also obvious and large. Parents were unaware that their adolescents carried weapons to school, used illegal drugs including alcohol and marijuana, smoked tobacco, or were sexually active. Parents even failed to report that their child had attempted suicide. Table 3 gives the parent and student responses and the level of significance. The prevalence of these behaviors was 3 to 29 times higher according to students. For 5 of the 7 behaviors, students reported rates at least 10 times higher than the rates their parents reported.

Table 3. 
Large and Statistically Significant Differences*
Large and Statistically Significant Differences*


This study was designed to examine the differences between student risk behaviors and parents' perception of these risk behaviors. While parents showed an awareness of some risk behaviors, when it came to recognizing significant risk behaviors in their adolescents, they were "clueless." Risk behaviors including carrying weapons to school, LSD (lysergic acid diethylamide) and cocaine use, suicide attempts, sexual intercourse, alcohol use, tobacco use, and marijuana use were all significantly underestimated by parents. This is unfortunate, because if parents deny or fail to recognize risky behaviors of their adolescents, they are less likely to provide or support appropriate prevention and intervention strategies. Classes that teach parents how to recognize risk behaviors and how to respond to them are needed.

The lack of parental and community awareness of the extent of adolescent risk behaviors impedes interventions that could address the behavior problems. Sharing this information with parents—not to frighten them, but to enlighten them—can build their support for services needed to reduce these behaviors. Successful prevention strategies will require the support of pediatricians, parents, and the community.

We need parental support to establish CSHE programs in schools and to implement other effective interventions to counter problem behaviors in adolescents. While many school and parent organizations, including the National School Board Association6 and the National PTA (Chicago, Ill) support CSHE, grassroot support by local school boards and parents is needed to make CSHE a reality. Peditricians and child advocates can use the information in this article to educate parents and provide them with strategies to prevent risk behaviors. We need to think in terms of "herd mentality," in that all adolescents can benefit from effective intervention programs through "reverse" peer pressure. Even if parents and communities are driven to help only their own children, they should recognize that by helping other adolescents avoid risk behaviors, their own children will benefit. Replication studies with larger sample sizes would be useful to generate more general results relating to these issues.

We should also promote developmental assets for the optimal growth of adolescents. The Search Institute has shown that students' risk behaviors are inversely related to the number of developmental assets that the students have.7 Thinking in terms of building developmental assets is an optimistic and inclusive way to address the risk behavior problems of adolescents. It encourages positive solutions and focuses on prevention. Perhaps by recognizing that as parents and physicians we are largely clueless about the behavior of adolescents, we can begin to work together as a community to solve this problem.

Accepted for publication June 9, 1998.

Editor's Note: Is this a case of ostrich syndrome? Is the percentage of middle school "children" who are sexually active really that high? Is this a spurious finding in a relatively small sample? I'd say, "Yes, yes, and no."—Catherine D. DeAngelis, MD

Reprints: Thomas L. Young MD, University of Kentucky Medical Center Department of Pediatrics, Kentucky Clinic, Lexington, KY 40536.

Cunnane  SC Childhood origins of lifestyle-related risk factors for coronary health disease in adulthood.  Nutr Health. 1993;9107- 115Google ScholarCrossref
Twisk  JWKemper  HCMellenbergh  GJVan Mechelen  WPost  GB Relationship between the longitudinal development of lipoprotein levels and lifestyle parameters during adolescence and young adulthood.  Ann Epidemiol. 1996;6246- 256Google ScholarCrossref
Centers for Disease Control and Prevention, Youth risk behavior survey.  MMWR CDC Surveill Summ. 1995;44Google Scholar
Deffenbaugh  KBHutchinson  RLBlankschen  MP Substance use among youth (grades 4-12) in rural Indiana: students' reported use vs parents' perceptions of students use.  J Alcohol Drug Educ. 1993;3919- 33Google Scholar
Touloukian  C American Academy of Pediatrics CATCH Grant Report.  Elk Grove Village, Ill American Academy of Pediatrics1994;
National School Board Association, Comprehensive School Health Project.  Alexandria, Va National School Board Association1993;
Leffert  NBenson  PL Roehlkepartain: Starting Out Right: Developmental Assets for Children.  Search Institute Minneapolis, Minn1997;