“Abstinence” or “Comprehensive” Sex Education?
The Institute for Research and Evaluationa
Salt Lake City, Utah
June 8, 2007

The debate about “abstinence” vs. “comprehensive” sex education has been occurring for at
least three decades.b The common ground that drives these competing approaches is concern about
the negative consequences of adolescent sexual activity to the health and well-being of individuals
and society. This debate has been re-energized recently by the release of findings from a national
study by Mathematica Policy Research, Inc., in which four different abstinence education programs
were selected as subjects for a long-term evaluation. This study reported that teen participants in
these abstinence programs did not abstain from sexual activity more than non-participants, when
measured 2½ to 5½ years after the program ended.1

These results have caused some to conclude that the abstinence approach to preventing teen
sexual risk behavior does not work. As one advocate of the “comprehensive” approach stated,
“This report should serve as the final verdict on the failure of the abstinence-only industry in this
country.”2 Implicit in such statements is the corollary conclusion that the comprehensive sex
education approach does work. Neither of these conclusions is supported by the full body of
research evidence about “abstinence” and “comprehensive” sex education.

While new evidence can add to the debate, it should never be accepted without scrutiny, and
should be viewed alongside the broader base of evidence upon which important policy questions
must rely. We have reviewed that body of evidence pertaining to “abstinence” and
“comprehensive” approaches to education regarding teen sexual activity. Our institute has also
conducted more than 100 evaluations of abstinence education interventions in 30 states over the
past 15 years. We draw on this broad base of evidence to share the following observations.

• • Many serious problems are associated with adolescent sexual activity.
• • Condom interventions have serious limitations.
• • When held to the same criteria employed by the Mathematica evaluation, comprehensive
  sex education programs do not appear to work.
• • The Mathematica study, and the four programs it evaluated, cannot be generalized to
  represent the efficacy of abstinence programs in general.
• • Several well-designed evaluations of abstinence programs have found significant, longterm
  reductions in adolescent sexual activity.
• • Abstinence education offers benefits to adolescents and society that are not found in the
  comprehensive sex education approach.
• • Abstinence interventions are most effective if they incorporate what has been learned
  about how to reduce adolescent sexual risk behavior.

We have found that well-designed and well-implemented abstinence education programs
can reduce teen sexual activity by as much as one half for periods of one to two years. For a more
detailed discussion of each of our points above, see the pages that follow.a Stan E. Weed, Ph.D., Irene H. Ericksen, M.S., Paul J. Birch, M.S., Joseph M. White, Ph.D., Matthew T. Evans, Ph.D., Nicole E. Anderson, M.S.(Refer questions to This e-mail address is being protected from spambots. You need JavaScript enabled to view it .)
b “Comprehensive” interventions place primary emphasis on teaching prevention methods for sexually active teens, centered on the use of condoms, although some also present abstinence as the preferred option. “Abstinence” interventions place primary emphasis on teaching the postponement of sexual initiation or the discontinuation of sexual activity, and often include information about the limits of condoms in preventing STDs. For the purposes of this paper, we will also refer to comprehensive sex education as condom-based interventions.

I. Consequences of Teen Sex—In 2005, 63.1% of American adolescents had experienced sexual
intercourse by the end of high school.3 Many serious health and social problems in American
society are related to teen sexual activity. These include:


A. Teen Pregnancy: One in 13 high-school-age girls becomes pregnant each year in America.4
Adverse consequences associated with teen pregnancies include abortion, unwed teen
parenthood, father absence, poverty, welfare dependence, and the growth of drug abuse,
gang culture, and crime.5–9


B. STDs: STDs have emerged as a significant threat to adolescent health. The consequences
include chronic pelvic pain, genital lesions, lifetime infection, infertility, ectopic (tubal)
pregnancy, damage to unborn children, cancer, and in some cases death.10–13 Adolescent
STD rates are higher than rates for all other age groups. One quarter of sexually active
teens have an STD,1 and adolescent rates for most STDs are on the rise.11,14,15 The growing
STD problem has been called a hidden epidemic.11,16 The direct medical cost of 9 million
new cases of STDs that occurred among U.S. adolescents and young adults (15–24-year
olds) in the year 2000 was estimated at $6.5 billion (in year 2000 dollars).17


C. Poorer Emotional Health: There is a strong association between sexual activity and poor
emotional health for adolescents.


1. Sexually active teens are more than twice as likely as virgin teens to be depressed or
attempt suicide. Adolescents report a drop in self-esteem after initiating sexual
intercourse, and the majority express regret for becoming sexually active.18–21


2. Sexually experienced teens, especially girls, are much more likely to experience dating
violence than their virgin peers, and sexual exploitation (such as statutory rape) and
unwanted or forced intercourse/rape are not uncommon among sexually experienced
teen girls. In 2005, one out of eight 12th grade girls in the U.S. reported being
physically forced to have intercourse against her will.3,22,23


II. Condom Limitations—Condom use is advocated by many as the best protection for the
sexually active from both pregnancy and STD transmission. Yet many consequences of teen
sexual activity are not prevented by condom use.


A. Even with consistent and correct use (which is rare), condoms may diminish but do not
effectively prevent STDs that are spread through skin-to-skin or skin-to-sore contact.
These STDs are on the rise in the adolescent population.24–27


B. After 20-plus years of comprehensive sex education efforts in the U.S., adolescent rates of
consistent condom use are not high enough to eliminate the STDs for which condoms are
most preventive, such as HIV, let alone STDs for which condoms are least preventive.
Adolescents contract one fourth of all new HIV infections.15 Among sexually active U.S.
teens, only 47.8% of males and 27.5% of females report using condoms consistently over a
one-year period.28 Efforts to improve those rates have not proven successful.


C. Consistent condom use cannot prevent the negative emotional impact or the sexual
exploitation and sexual violence that are associated with teen sexual activity, as described
above.

III. Failure of Comprehensive Programs—When studies are held to the same criteria as the
Mathematica evaluation (random assignment, a follow-up period of 2½ to 5½ years, a high
level success criteria), there is ample evidence that condom-based sex education interventions
do not work. In the past 20 years, studies evaluating abstinence education programs have been
limited in number and in rigor, while during the same time period research on comprehensive
sex education has abounded.29–32 One recent and thorough summary of this research33
reviewed 50 well-designed evaluation studies of comprehensive sex education programs in the
United States, going back to 1990, and included these findings:


A. None of the programs increased the prevalence of consistent condom use (CCU)c among
adolescents for a period greater than one year. CCU is the only condom measure that
approaches the stringent standard of the abstinence measure. Only one program produced a
significant increase in the prevalence of CCU that was sustained for a period of one year.35


B. Thirteen control trials of comprehensive sex education found no increase in teen condom
use for any period of time.


C. Only two comprehensive sex education programs succeeded in improving less stringent
measures of teen condom use (not CCU) for a period longer than two years, and none lasted
beyond three years.


IV. Mathematica Study Limitations—The Mathematica study, and the four programs it
evaluated, cannot be generalized to represent the efficacy of abstinence education or of
comprehensive sex education.


A. The Mathematica Study Did Not Examine Comprehensive Sex Education Programs: The
interpretation some have ascribed to the Mathematica report is that “abstinence programs
don’t work, therefore we must provide “safer sex” programs to reduce the risks of early
sexual activity.” The Mathematica study did not draw this conclusion, did not examine
safer sex programs, nor suggest that they are the obvious default if abstinence programs are
not successful. A substantial number of studies have examined condom-based interventions
and can inform policy decisions. In summary, of 50 rigorous studies spanning the past 15
years, only one of them reports an improvement in consistent condom use after a period of at
least one year.35 This study showed that 58% of females visiting a health clinic for STDs
one year after the CCU intervention reported CCU while the control group reported 45%.
The other 49 studies either did not measure CCU (the best comparison with abstinent
behavior), or did not find a significant program effect of at least one year.33 This pattern of
evidence (1 success out of 49) does not provide a reasonable basis for replacing abstinence
education with a condom-based sex education policy.


B. Cross-contamination of Program Effects: The benefits of a random assignment research
design are best realized when the treatment and control group can be kept separate and their
integrity can be maintained. In this way, the treatment or “medicine” is not shared between
the groups. However, in field experiments, this requirement is difficult to achieve,
especially with teenagers, and particularly with an intervention that deals with a topic as
highly charged as sex. Students randomly assigned to the two groups don’t live in these
groups—they interact with friends, siblings, and dating partners in the other group. Any
new values or behaviors adopted by each group are shared across the groups, and the longer
that sharing lasts the more likely it is that the differences between the two groups will
disappear as their attitudes, values, beliefs and behaviors merge over time. This cross-group
contamination is likely to be a stronger intervention than a typical one-hour-per-day shortterm
intervention. With almost six years for this spillover effect to operate, this would
minimize the measurable differences between the groups, even if the program had
successfully reduced the participants’ sexual activity. The Mathematica study did not
address this problem, nor did it make exception for it in reporting its findings. This
limitation and those that follow demonstrate that it requires more than an initial random
assignment of participants to claim a “gold standard” study. It also illustrates how difficult
it is to do good field studies.


C. Non-Representative Study Sample: The high-risk population used in the study does not
represent the teen population in the U.S. (The majority of the sample were African
American youth from poor, single-parent households—see pp. 9 and 20.) The fact that these
programs produced no impact on this sample does not indicate whether these same programs
would have had an impact on a more representative group of teens.


D. Unusually Long Follow-up Timeframe: The follow-up time frame employed in this study—
2½ to 5½ years after the program end—is too long for any type of sex education
intervention to have a sustained effect on behavior without interim reinforcement of the
program message. We are not aware of any evaluations of comprehensive sex education
programs that have shown positive changes in teen condom use after three years, and are
aware of only two that have shown impact after two years, and these were using the lower
standard of success. A myriad of negative influences operate in adolescents’ lives to
overpower any initial program effect that may have occurred so far in the past. An outcome
evaluation with a 5½ year follow-up time period and no interim program reinforcement does
not provide a realistic indication of program effectiveness.


E. Inappropriate Timing of Program Dose: The age group for the interventions in the
Mathematica study was quite young—elementary and early middle school. Some were as
young as 4th and 5th grade. The interventions did not continue, follow-up with, or reinforce
the initial treatment during the key years (9th, 10th, 11th grade) when transition into sexual
activity typically occurs. Thus, the treatment was not delivered or reinforced when it was
most relevant and needed. As stated in the Mathematica report “the findings provide no
information on the effects programs might have if they were implemented for high school
youth or began at earlier ages but served youth through high school” (p. 61). At the outset
then, the evaluation started with programs that had little hope of impacting behavior in the
long run.


F. Inadequate Utilization of Mediator Variables: A major disappointment with the study was
the insufficient attention given to identifying and tracking the important causal factors that
mediate adolescent sexual risk behavior. The study’s generic logic model was not tailored
to the four specific programs, and therefore the specific theory of these programs was not tested.
Without understanding and monitoring these causal factors, success or failure cannot be understood
or explained, intervention modifications can not be made, and longer term
program potential cannot be identified. Of the mediating variables they did measure and
test, only two showed a significant relationship to the targeted behavior, but neither of them
showed significant pre-post change. A more appropriate evaluation model for new and
developing programs is one that would share interim data with programs to support their
evolution and improvement. In Mathematica’s case, data was not shared with the programs
until four or five years later. Had we taken that approach with some of our own program
evaluations, (e.g., Arkansas, Virginia, South Carolina, Georgia) we would likely have seen
the same result when measuring behavior five years later. Instead, these projects have
evolved and matured overtime, and are now realizing up to 50% reduction in initiation of
sexual activity.


V. Evidence of Abstinence Effectiveness—Several well-designed evaluations of abstinence
programs have found significant, long-term reductions in adolescent sexual activity, with both
moderate and high-risk populations.


A. A randomized controlled trial conducted by Jemmott et al. found that an abstinence-only
intervention significantly reduced sexual initiation among young African American
adolescents after a 24-month follow-up period, and did not reduce condom use for those
virgins who did become sexually active (p<.05).36


B. An abstinence curriculum that was taught in addition to an existing comprehensive sex
education program decreased sexual initiation by approximately 40% after 20 months for
program students versus comparison students in a high-risk population (p<.01).37


C. An evaluation of the Reasons of the Heart abstinence curriculum found that adolescent
program participants were approximately one half as likely as the matched comparison
group to initiate sexual activity after one year (p<.05). The program’s effect was as strong
for the African American subgroup in the sample as it was overall (p<.05).38


D. A study of the Heritage Keepers abstinence program found that one year after program
participation virgin middle school students were about one half as likely to initiate sexual
activity as the comparison group (p<.001). Roughly one half of the sample was African
American, for whom the program effect was equally strong (p<.001).39


E. The Sex Respect and Teen Aid abstinence-only programs reduced the rate of initiation of
sex by more than one third (p<.01) for the high-risk students in a Caucasian high school
sample after 12 months.40


VI. Benefits of Abstinence—Abstinence education offers benefits to adolescents and society that
are not found in the comprehensive sex education approach.


A. Abstinence provides 100% protection from the biological consequences of sex (pregnancy,
abortion, teen parenthood, the full spectrum of STDs).

B. Youth who abstain can avoid the negative emotional consequences related to teen sex—
lowered self-esteem, regret, depression, etc.—as well as reducing the likelihood of
experiencing sexual coercion and sexual violence.


C. Abstinence programs emphasize principles of self-restraint, self-esteem, future goals, longterm
commitment, and unselfishness in relationships, and teach healthy relationship skills,
all of which support the formation of strong marriages and healthy families.


D. Several studies have found that teaching abstinence does not reduce rates of condom use for
virgin teens who become sexually active.36,41


E. Abstinence education addresses the relationship of sexuality to the well-being of the whole
person, rather than treating sexual activity as an isolated and unrelated behavior.


VII. Programs That Work—Our research shows, not surprisingly, that some programs work and
some don’t. The important questions are “which ones do, and why?” Abstinence interventions
are most effective if they incorporate what has been learned about how to reduce adolescent
sexual risk behavior.


A. Well-designed programs target teen attitudes, values, efficacy, and goals regarding
abstinence, sexuality, and relationships, as key mediators of sexual behavior.


B. The classroom teacher plays a crucial role in the process of changing teen attitudes and
behaviors about sexuality through his/her personal example, mentoring, and teaching skills.


C. Successful programs utilize a variety of instructional methods that include interactive
participatory activities, role playing, skill-building, personal application, and commitment.


D. An initial program installment of 20 hours of instruction, repeated annually, and followed
by regular reinforcement of the abstinence message is the minimum dose recommended to
facilitate an increase in teen sexual abstinence.


E. Well-designed abstinence interventions will contain a strong parent component that
includes direct parent instruction and “homework” assignments to facilitate parent–teen
interaction about abstinence.

REFERENCES
1. Trenholm, C., Devaney, B., Fortson, K., Quay, L., Wheeler, J., & Clark, M. (2007). Impacts of Four Title V, Section
510 Abstinence Education Programs. Princeton, NJ: Mathematica Policy Research, Inc. April 2007.

2. SIECUS. (2007). Press Release: Federal abstinence-only-until-marriage programs not proven effective in delaying
sexual activity among young people. April 13, 2007. Available at: http://www.siecus.org/media/press/press0141.html.

3. Eaton, D.K., Kann, L., Kinchen, S., Ross, J., Hawkins, J., & Harris, W.A. (2006). Youth Risk Behavior
Surveillance—United States, 2005. MMWR Surveillance Summaries, June 9, 2006/55(SS05);1–108. Retrieved Jan. 12,
2007 from http://www.cdc.gov/mmwr/preview/mmwrhtml/ss5505a1.htm.

4. Guttmacher Institute. (2006). U.S. Pregnancy Statistics: National and State Trends and Trends by Race and Ethnicity.
New York: Author. September, 2006.

5. Hofferth, S.L., & Reid, L. (2002). Early childbearing and children’s achievement and behavior over time.
Perspectives on Sexual and Reproductive Health, 34(1):41–49.

6. Jaffee, S.R. (2002). Pathways to adversity in young adulthood among early childbearers. [Electronic version]
[Abstract]. Journal of Family Psychology, 16, 38–49.

7. Maynard, R.A. (Ed.) (1997). Kids having kids: Economic costs and social consequences of teen pregnancy.
Washington, DC: The Urban Institute. Cited in Emerging Answers: Research Findings on Programs to Reduce Teen
Pregnancy, Kirby, 2001.

. Miller, B.C., & Moore, K.A. (1990). Adolescent sexual behavior, pregnancy, & parenting: Research through the
1980s. Journal of Marriage & the Family, 52, 1025–1044.

9. Warren, C.W., Kann, L., Small, M.L., Santelli, J.S., Collins, J.L., & Kolbe, L.J. (1997). Age of Initiating Selected
Health-Risk Behaviors among High School Students in the United States. Journal of Adolescent Health, 21, 225–231.

10. American Social Health Association. (1998). Sexually Transmitted Diseases in America: How Many Cases and a
What Cost? Menlo Park, CA.: Kaiser Family Foundation. Retrieved October 6, 2003, from
http://www.kff.org/content/archive/1445/std_rep.html.

11. Centers for Disease Control and Prevention. (2001). Tracking the Hidden Epidemics 2000: Trends in STDs in the
United States. Atlanta, GA: U.S. Department of Health and Human Services, July, 2001. Retrieved March 12, 2003,
from http://www.cdc.gov/nchstp/od/news/RevBrochure1pdfintro.htm.

12. National Institute of Allergy and Infectious Diseases (NIAID). (2001). Workshop Summary: Scientific Evidence on
Condom Effectiveness for Sexually Transmitted Disease (STD) Prevention. July 20, 2001. Washington, DC: National
Institutes of Health, Department of Health and Human Services.

13. Sulack, P.J. (2003). Sexually transmitted diseases. Seminars in reproductive medicine, 21 (4): 399–413.

14. Centers for Disease Control and Prevention. (2003a). Sexually Transmitted Disease Surveillance, 2002. Atlanta,
GA: U.S. Department of Health and Human Services, September 2003. Retrieved February 9, 2004, from
http://www.cdc.gov/std/stats/tables/table12B.htm.
15. Centers for Disease Control and Prevention. (2003b). Fact Sheet—Young People at Risk: HIV/AIDS Among
America’s Youth. Division of HIV/AIDS Prevention. Atlanta, GA: U.S. Department of Health and Human Services.
Retrieved June 24, 2003, from http://www.cdc.gov/hiv/pubs/facts/youth.htm.
16. Fortenberry, J.D. (2002). Unveiling the hidden epidemic of sexually transmitted diseases [Electronic version].
Journal of the American Medical Association, 287, 768–769.
17. Chesson, H.W., Blandford, J.M., Gift, T.L., Tao, G., & Irwin, K.L. (2004). The estimated direct medical cost of
sexually transmitted diseases among American youth, 2000. Perspectives on Sexual and Reproductive Health, 2004,
36(1):11–19.
18. Hallfors, D.D., Waller, M.W., Ford, C.A., Halpern, C.T., Brodish, P.H., & Iritani, B. (2004). Adolescent depression
and suicide risk: association with sex and drug behaviors. American Journal of Preventive Medicine, 27:224–230.
19. Rector, R., Johnson, K., & Noyes, L. (2003). Sexually Active Teenagers are More Likely to be Depressed and to
Attempt Suicide. Heritage Foundation Center for Data Analysis 2003: Report #03-04. Available at
http://www.heritage.org/Research/Family/cda0304.cfm. Accessed June 5, 2003.
20. Bearman, P.J., & Bruckner, H. (2001). Promising the future: virginity pledges and the transition to first intercourse.
American Journal of Sociology, 106:859–912.
21. National Campaign to Prevent Teen Pregnancy. (2003). America’s Adults and Teens Sound Off About Teen
Pregnancy: An Annual National Survey. December 2003. Washington DC: Author. Retrieved January 5, 2004, from
http://www.teenpregnancy.org.
22. Moore, K., & Manlove, J. (2005) A demographic portrait of statutory rape. Presentation to Conference on Sexual
Exploitation of Teens. March 2005. Washington, DC: Child Trends. Accessed October 30, 2006 at: Statutory_rapefinal_
version_sent_to_OPA.ppt.
23. Silverman, J.G., Raj, A., & Clements, K. (2004). Dating violence and associated risk and pregnancy among adolescent girls in the United States. Pediatrics, 114(2), e220–225.
24. Weller, S., & Davis, K. (2002). Condom effectiveness in reducing heterosexual HIV transmission. Cochrane
Database Syst Rev, 1. [Abstract].
25. Winer, R.L., Hughes, J.P., Feng, Q., O’Reilly, S., Kiviat, N.B., Holmes, K.K., & Koutsky, L.A. (2006). Condom use
and the risk of genital human papillomavirus infection in young women. New England Journal of Medicine, 354:2645–
54.
26. Crosby, R.A., DiClemente, R.J., Wingood, G.M., Lang, D., & Harrington, K.F. (2003). Value of consistent condom
use: A study of sexually transmitted disease prevention among African American adolescent females. American Journal
of Public Health, 93:901–2. [In Holmes, et al., 2004—see Reference #18.]
27. Wald, A.; Langenberg, A.G.; Link, K.; Izu, A.E.; Ashley, R.; Warren, T.; Tyring, S.; Douglas, J.M. Jr., & Corey, L.
(2001). Effect of condoms on reducing the transmission of herpes simplex virus type 2 from men to women. Journal of
the American Medical Association, 285(24):3100–06.
28. Centers for Disease Control and Prevention. Teenagers in the United States: Sexual Activity, Contraceptive Use,
and Childbearing, 2002 (online). Vital Health Stat 23, Number 24. Hyattsville, Maryland: U.S. Department of Health
and Human Services, December, 2004. Available at: http://www.cdc.gov/nchs/data/series/sr_23/sr23_024.pdf. Accessed
December 16, 2004.
29. Thomas, M. (2000). Abstinence-based programs for prevention of adolescent pregnancies: A review. Journal of
Adolescent Health, 26, 5–17.
30. Kirby, D. (2002). Do Abstinence-Only Programs Delay the Initiation of Sex Among Young People and Reduce Teen
Pregnancy? Washington, DC: National Campaign to Prevent Teen Pregnancy.
31. Manlove, J.M., Terry-Humen, E., Papillo, A., Franzetta, K., Williams, S., & Ryan, S. (2002). Preventing teenage
pregnancy, childbearing, and sexually transmitted diseases: What the research shows. In Child Trends & John S. and
James L. Knight Foundation (Eds.), American teens: A special look at Awhat works@ in adolescent development (pp. 6–
23). Washington, DC: Child Trends.
32. Scher, L.S., Maynard, R.A., & Stagner, M. (2005). Interventions intended to reduce pregnancy-related outcomes
among adolescents. (Unpublished manuscript, June 30, 2005)
33. Kirby, D., Laris, B.A., & Rolleri, L. (2006). The impact of sex and HIV education programs on sexual behaviors of
youth in developing and developed countries. Youth Research Working Paper Series, No.2, Family Health International.
34. Holmes, K.K., Levine, R., & Weaver, M. (2004). Effectiveness of condoms in preventing sexually transmitted
infections. Bull World Health Organ, 82(6):454–461.
35. DiClemente, D.J., Wingood, G.M., Harrington, K.F., Lang, D.L., Davies, S.L., Hook III, E.W., et al. (2004).
Efficacy of an HIV prevention intervention for African American adolescent girls: a randomized controlled trial.
Journal of the American Medical Association, 292, 171–9.
36. Jemmott III, J.B., Jemmott, L.S., & Fong, G.T. (2006). Efficacy of an abstinence-only intervention over 24 months:
a randomized controlled trial with young adolescents. Presentation at XVI International AIDS Conference, Toronto,
Canada; Aug. 13–18, 2006.
37. Howard , M., & McCabe, J.B. (1990). Helping teenagers postpone sexual involvement. Family Planning
Perspectives, 22, 21–26.
38. Weed, S.E., Ericksen, I.H., Lewis, A., Grant, G.E., & Webberly, K.H. (2007). An Abstinence Program’s Impact on
Cognitive Mediators and Sexual Initiation. American Journal of Health Behavior (in press).
39. Weed, S.E., Ericksen, I.H., Birch, P.J. (2005). An evaluation of the Heritage Keepers Abstinence Education
program. In Golden A (Ed.) Evaluating Abstinence Education Programs: Improving Implementation and Assessing
Impact. Washington DC: Office of Population Affairs and the Administration for Children and Families, Department of
Health & Human Services, 2005:88–103.
40. Olsen, J.A., Weed, S.E., Daly, D., Jensen, L. (1992). The effect of abstinence sex education programs on virgin
versus nonvirgin students. Journal of Research and Development in Education, 25:69–75.
41. Borawski, E.A., Trapl, E.S., Lovegreen, L.D., Colabianchi, N., & Block, T. (2005). Effectiveness of abstinence-only
intervention in middle school teens. American Journal of Health Behavior, 29:423–434.