casppbrg110.gif (6447 bytes)

Missouri Volunteer Resource Mothers

Maltreatment and Adolescent Pregnancy and Parenting Program

Adolescent Mother Journaling Program


Human Development and Family Studies Extension

Human Environmental Sciences Extension



This article was orginally published in Family Relations, vol. 48, no. 3, 1999.

Why Abstinent Adolescents Report
They Have Not Had Sex:
Understanding Sexually Resilient Youth

by Lynn Blinn Pike

The sample in this study consisted of 697 students from 20 schools in Missouri who indicated on a survey of sexual attitudes and behaviors that they had not had sex. The subjects completed the 18-item Reasons for Abstinence Scale and identified those items that were reasons why they had not had sex. The most frequent reasons for not having sex were related to fears of pregnancy and disease (including HIV/AIDS). The least frequent reasons were related to problems concerning the cost and availability of birth control and protection. Principal components factor analysis revealed three factors that were labeled "fear-based postponement," "emotionality and confusion," and "conservative values." Factor scores differed by gender, grades, alcohol consumption, family structure, father's education, and urbanicity. The discussion centers on the need to design different prevention strategies to build protective factors that result in "sexual resilience" in target groups of adolescents.

The debate over what messages to give adolescents about the prevention of sexually transmitted diseases (including HIV/AIDS) and pregnancy has proven to be confusing to youth and has proven conflictual and polarizing in many communities. The diverse messages that U.S. adolescents receive can be summarized as: a) remain abstinent until marriage, b) remain abstinent until emotionally and developmentally ready to become sexually active, c) remain abstinent but, if not able to, have accurate information about birth control and protection, and d) have accurate and factual information on how to use birth control and protection effectively because abstinence is not a realistic expectation. While many adults feel strongly about promoting one of the above positions with youth, it is not well understood how adolescents understand and incorporate these disparate admonitions into personal behavior patterns.

In the battle over the most appropriate approaches to reduce adolescent high risk sexual behaviors in the United States, more emphasis has been placed on understanding adolescents who report they have had sex than on understanding adolescents who report they have remained abstinent. For example, the Centers for Disease Control administered the Youth Risk Behavior Survey (YRBS) to a national sample of more than 16,000 high school students in 1997 and reported that 48% of the students had engaged in sexual intercourse. This figure was a significant decline from 54% in 1991 (MMWR, 1998) and generated great interest on the part of politicians, researchers and practitioners. However, there were little empirical data on which to explain this decline. Both supporters of conservative abstinence-only and more liberal comprehensive sex education programs claimed responsibility. No research was found that asked the adolescents who have not had sex why they have remained abstinent. Therefore, the present study used a protective factor model of resiliency to address why diverse groups of adolescents report they have not had sex and to examine how to support youth who make the decision to remain abstinent to become "sexually resilient."

Literature Review

While much research and discussion have occurred in the last decade about fostering both sexually abstinent youth and resilient youth, few researchers have attempted to merge the two areas to provide direction for adolescent HIV, sexually transmitted disease and pregnancy prevention. Although none of them specifically examined abstinence, Brooks-Gunn and Paikoff (1993), Small and Luster (1994), and Perkins, Luster and Villarruel (1998) were unique in applying the concept of resilience, originally developed in the field of developmental psychopathology, to understanding adolescent sexual well-being. Brooks-Gunn and Paikoff (1993) explored the roles of cultural (moral standards, gender, culture, and media), individual (biology and social cognition), and environmental (peers, family, and school) factors in understanding how to promote sexual well-being among adolescents. They stated that most of the interventions to promote healthy adolescent sexuality, while promoting principles of positive behavioral change, have not been explicitly linked to the developmental literature on risk and protective factors. Small and Luster (1994) and Perkins, Luster and Villarruel (1998) used an ecological risk-factor model to examine adolescents' sexual activity in relation to ethnicity, history of physical abuse, neighborhood monitoring, and attachment to schools.


Significant numbers of adolescents in the U.S. are putting themselves at risk for HIV, sexually transmitted diseases, and pregnancy at younger ages (MMWR, 1996, 1998). However, not all individuals become sexually active before adulthood and little attention has been paid to the group of adolescents who can be labeled as sexually resilient in the face of peer and media messages that make early sexual behaviors appear attractive and normal. Abstinence was catapulted into the public sphere when, on August 22, 1996, Congress appropriated $50M in the controversial Personal Responsibility and Welfare Reform legislation (Public Law 104-93) for the promotion of abstinence education each year from 1998-2002. The portion of this funding allocated to each state was determined by the proportion of the number of low income children in that state compared to the number of low income children nationally. In this legislation, the term "abstinence education" was defined as encompassing eight tenets: 1) there are social, psychological and health gains from abstaining from sexual activity, 2) abstinence from sexual activity before marriage is the expected standard for all school age children, 3) abstinence from sexual activity is the only certain way to prevent out-of-wedlock pregnancy, sexually transmitted diseases, and other associated health problems, 4) a mutually faithful and monogamous relationship in the context of marriage is the expected standard of human sexual activity, 5) sexual activity outside of marriage is likely to have harmful psychological and physical effects, 6) bearing children out-of-wedlock is likely to have harmful consequences for the child, the parents and society, 7) young people need to learn how to reject sexual advances and how alcohol and drug use increases vulnerability to sexual advances, and 8) young people need to attain self-sufficiency before engaging in sexual activity.

Organizations such as the National Abstinence Clearinghouse, Project Reality, and the National Coalition for Abstinence Education heralded this as much needed legislation. On the other hand, organizations such as the Sexuality Information and Education Council of the United States, the National Coalition to Support Sexuality Education, Advocates for Youth, the Alan Guttmacher Institute, and the National Commission on Adolescent Sexual Health disagreed and supported the provision of funding for "abstinence-based" rather than "abstinence-only" programs. The former include information on contraception and the later do not. It has yet to be determined what impact this abstinence-only legislation will have on adolescent sexual decision making in the future.


According to Masten, Best and Garmezy (1990, p. 425), resiliency refers to "the process of, capacity for, or the outcome of successful adaptation despite challenging or threatening circumstances." Remaining abstinent or making responsible sexual decisions during the adolescent years may be one of the most significant challenges facing youth today. The number of sexual messages that adolescents receive via the print and mass media each day makes it a challenging circumstance to be sexually resilient. According to Perry, Kelder and Komro (1993), during the 1960s the adolescent issues dealt with on television were rather innocuous and included dates, blemishes, after-school jobs, and cars. During the 1990s, adolescent issues portrayed on television included suicide, pregnancy, HIV/AIDS, sexual harassment, and sexual abuse. Lowry and Towles (1989a) found that in heterosexual relationships in daytime dramas, the ratio of unmarried to married partners was 24 to 1, reinforcing the message that sexual behavior is more likely to occur outside of marriage. They also reported that adolescents watch 11 sexual behaviors per hour during prime time television (Lowry & Towles, 1989b). Brown, Childers and Waszak (1990) reported that the average teenager watches almost 2,000 hours of sexual references on television each year and that references to birth control or to sexually transmitted diseases are almost nonexistent. Add the impact of films, music, videos, the world wide web, and magazines to that of television and it becomes evident that adolescents in the United States today learn about sexuality through almost unlimited exposure to sexual scenes where protection and responsibility are absent (Perry, Kelder & Komro, 1993). 

Bogenschneider, Small and Riley (1992) described how the study of resiliency has moved from an epidemiological risk-focused approach in the 1960s and 1970s to the present protective factor etiological perspective. The later perspective asks not only what is wrong with children, but what is right with children. What protects them? How can we enhance the child's ability to resist stressful life events and promote positive adaptation and competence? In addition, Bogenshneider (1996) moved the field a step further ahead by proposing an ecological risk/protective model of resilience based on Bronfenbrenner's (1976, 1986) ecological theory of human development and Lerner's (1995) developmental contextualism. Bogenshneider's model contends that human development is shaped by multiple processes that must be identified in multiple levels of human ecology, and that these processes are shaped by the dynamic and reciprocal nature of development.

Risk factors are defined as individual or environmental hazards that increase an individual's vulnerability to negative developmental outcomes. Risk factors do not necessarily guarantee negative consequences, but may increase the likelihood that problem behaviors will occur (Werner & Smith, 1990). As the number of risk factors increases, the probability of problem behaviors increases. The following are examples of risk factors for youth that can develop into persistent behavioral patterns: involvement in alcohol or other drugs, sexuality, depression/suicide, anti-social behaviors, poor academic standing, and eating disorders ( (Rutter, 1979).

Protective factors are defined as individual or environmental safeguards that enhance an individual's ability to resist stressful life events while adapting to the situation and developing competency in dealing with it (Garmezy, 1983; Werner & Smith, 1990). In the protective model of resilience, protective factors, such as positive adult role models, good schools, and community involvement, are thought to buffer the impact of risk factors by improving coping, adaptation and competence building. In the protective model, it is considered important to focus on those factors that foster health-promoting behaviors and competence in children. Protective factors are considered to work in interaction with risk factors to promote resiliency by moderating the effects of social or environmental risks so that more positive adaptation can take place than if they were not present.

Benson, Blyth, Deville, and Wachs (1997) contributed to the protective factor literature when they identified 40 developmental assets of resilient youth and divided them into those that were internal and external assets. The internal assets fall into three categories: commitment to learning, positive values and social competencies. The external assets fall into four categories: support, empowerment, boundaries and expectations, and constructive use of time.

There have been several longitudinal studies of resiliency that have provided valuable insight into how individuals fare after long term exposure to poverty, dysfunction, mental illness, and physical disability and these have consistently pointed to specific positive traits of resilient children (see Garmezy, Masten & Tellegen 1984; Rutter 1979; Werner & Smith, 1992). The positive traits that have emerged from these studies can be grouped into three categories. The first category involves their ability to seek out and cultivate positive relationships because they : a) are attractive and popular with other people, b) have more positive social relationships with friends and teachers, and c) have a well-developed sense of humor. The second category involves their better cognitive and intellectual abilities that result in them : a) being perceived by themselves and others as being competent in at least some areas, b) having positive school experiences, and c) having better verbal skills.

The third category involves their positive world view and outlook on life, exhibited as having: a) better impulse control, b) higher self-esteem and internal locus of control, c) the ability to delay gratification, and d) the ability to maintain a positive future orientation. The characteristics listed above act as protective shields that assist children in avoiding, regulating, or coping with aversive environmental or developmental conditions. In turn, they are better able to modify the negative impact of stressors and experience less damaging consequences.

However, the growing body of research on resilience has not been adequately applied to understanding adolescent abstinence, sexual decision making and risk taking behaviors. The following are unanswered questions concerning external protective factors and adolescent sexuality: Are the same external factors that have been identified as fostering resilience in areas such as drug prevention and delinquent behavior applicable in the prevention of early sexual activity (see Block, Block & Keyes, 1988)? What community and family charactersitics predict positive resistance to sexual pressure? What cultural and environmental assets help youth become proactive about securing and using birth control and protection? And what support systems factors assist in overcoming feelings of embarrassment when attempting to communication with adults and peers about sexual feelings and behaviors? The following are unanswered questions concerning internal protective factors. Are resilient youth: a) more creative in refusing unwanted sexual advances, b) more adept at building relationships with caring adults who can serve as confidants concerning sexual relationships, c) more competent in negotiating the use of condoms, d) more skilled in manipulating social situations so that they do not find themselves in high risk environments, and/or e) better able to synthesize conflicting messages into a coherent set of personal standards?

Research Questions

The research questions addressed in this study were as follows. First, what reasons do abstinent adolescents give for not becoming sexually active? Second, what are the underlying dimensions of adolescents' reasons for not being sexually active. And third, how do adolescents differ in their reasons for being abstinent based on individual (alcohol use, school grades, age),
cultural (race, gender) and environmental (family structure, father's education and urbanicity) factors related to resiliency?



The sample in this study consisted of 697 early adolescents (8 through 10th grade) attending 20 schools across the state of Missouri. Out of a larger sample of 1,112 subjects, the 697 students made up the 65% of the sample who responded that they had not had sex. Table 1 shows that the sample was approximately equally divided on gender (59% female). The highest percentages of subjects were from nonurban areas (67%), in ninth grade (43%), White (74%), had  fathers with high school educations or less (42%), and lived with both parents (73%). The students who were classified as urban lived in two designated metropolitan statistical areas (MSA): Kansas City and St. Louis. A MSA is a large population nucleus, together with adjacent communities that have a high degree of economic and social integration with that nucleus. Each metropolitan area must contain either a place with a minimum population of 50,000 or a Census Bureau defined urbanized area and a total metropolitan population of at least 100,000 (U.S. Census Bureau, 1993). The students not living in these two official metropolitan areas were classified as nonurban.

Table 1.
Demographic Data on Subjects (N =697)

Age (Mean/SD ) 14.56/.88



282 (41%)


414 (59%)



227 (32%)


468 (67%)
Father's Education

High School or less

278 (42%)

Some college

114 (17%)

4 or more years of college

120 (18%)

Not sure

147 (22%)


145 (21%)


516 (74%)


52 (5%)
Live With

Two parents

512 (73%)

Mother or stepmother

127 (18%)

Father or stepfather

12 (2%)


44 (6%)
Grades (self-reported)


373 (57%)


228 (35%)


55 (8%)


The results reported here are based on the 18-item Reasons For Abstinence Scale (RAS) that was developed for this study and part of a larger instrument (137 questions) administered to 1,112 students in fall 1998. The RAS was administered by trained classroom teachers in intact classrooms. The RAS questions were derived from an extensive review of the literature on adolescent sexual activity and field tested with 45 seventh through ninth grade students in one urban and two rural schools. Care was taken to insure that data collection was consistent across the 20 sites. Prior to the administration of the survey, the involved teachers attended a one-day workshop in which instructions given for securing parental consent and gathering data in a systematic and ethical manner.

In addition, it was considered important that the students understand the
vocabulary used in the survey. Prior to administration of the survey, the teachers wrote the following definition on the board and read it to the students: "Having sex means having intercourse or going all the way." The instructions for the RAS read: "Below are 18 reasons for not having sex. If you have never had sex, read each statement and mark yes' if it is one of reasons why you have not had sex. Mark no' if it is not one of the reasons why you have not had sex. If you have had sex, leave the questions blank." Examples of the 18 questions included: "I have never had sex because I would be too embarrassed," "I have never had sex because I am not ready," and "I have never had sex because I do not want to get AIDS." As a reliability check, the student was asked if he or she had had sex at five points in the 137-item survey. All of the 697 students responded five times that they had never had sex.


The first step in the analyses was to determine the frequencies of the reasons for not having sex. The second step was to determine the underlying nature of the adolescents' perceptions of why they had not had sex using principal components factor analysis. The number of factors retained for rotation was determined by several criteria: a) examination of the eigenvalue magnitudes using Kaiser's (1974) normalization, b) application of Cattell's (1952) scree test, and c) examination of the variance explained. Factors were rotated using varimax procedures (Kaiser, 1958). Factors were interpreted by examining questions with both positive and negative loadings above .40.

In the third analysis, unweighted, summed scales were calculated based on the factor results. Univariate analysis of variance (ANOVA) was used as test for significant main and interaction effects by subjects' family structure (two-parent family or other situation), grades in school (A-B, B-C, C-D), alcohol usage (never drink, less than one drink per month, more than one drink per month), ages (13,14,15,16), gender, fathers' education (high school or less, some college, four or more years of college), urbanicity (urban and nonurban), and race (African-American and White). When adolescents with missing data were eliminated, the sample size for the ANOVA procedures was 596.


The items with means below 1.20 or the most frequent reasons for not having sex were (1 = yes, 2 = no) fear of AIDS ( = 1.18/ SD = .39), fear of becoming pregnant or getting someone pregnant (1.18/.39), and fear of getting a disease (1.19/.39). The moderately frequent reasons were: believing it is wrong to have sex before marriage (1.50/.50) and waiting until marriage (1.50/.50). The least frequent reasons were not knowing where to get birth control or protection (1.90/.29), being embarrassed to use birth control or protection (1.90/.28), and not having enough money to buy birth control or protection (1.93/.24).

In the second analysis, the nature of adolescents' reasons for remaining abstinence were explored. The results of principal components factor analysis on the 18 questions revealed three factors that explained 48% of the variance (25%, 13% and 10%, respectively). All of the items, but one, loaded on one of three factors using the .40 criteria. The loading for the item "I do not have a partner at this time" was .32 and was dropped from further analysis. Reliability of the factor solution was assessed using theta. According to Carmines and Zeller (1979) the advantage of theta is that it provides a single coefficient for estimating reliability. Theta for the 17 items was .84.

The three factors were labeled: a) fear-based postponement (6 items), b)
emotionality and confusion (8 items), and c) conservative values (3 items). Table 2 shows a summary of the items and factor results. The label "fear-based postponement" was used to describe an adolescent who had considered the consequences and did not feel ready for sex because he or she believed it would be unwise due to the risk of pregnancy or disease and reprisal from parents.

The label "emotionality and confusion" was given to the items in the second
factor because of fears surrounding pain, embarrassment, lack of money for birth control or protection, and peer or partner disapproval. This person appeared to allow a mixed set of environmental or contextual factors to determine his or her level of sexual activity, as opposed to making a personal decision.

The label "conservative values" was given to the items that dealt with religion and waiting until marriage to have sex. The KR-20 reliability for each scale was "fear-based postponement" (.81), "emotionality and confusion" (.78) and "conservative values" (.84).

Table 2.
Summary of means and factor loadings
for the Reasons For Abstinence scale (N=697)

Factor Mean/SD Loading


Fear-based Postponement   
1.27/.42 .61 not ready for sex
.55 waiting for the right person
.66 waiting until I am older
  .77 fear of pregnancy or getting someone pregnant
  .84 fear of AIDS
  .43 parents would be upset
Emotionality and confusion   
1.85/.26 .47 friends think it wrong before
     .57 too embarrassed
  .50 partner does not want to have sex
  .67 do not know where to get birth control or protection
  .56 afraid it might hurt
  .78 embarrassed to used birth control or protection
  .65 not enough money to buy birth control/protection
  .73 embarrassed to buy condoms
Conservative values 1.53/.44 .87 wrong to have sex before
  .81 religion says it is wrong to have sex before marriage
  .84 waiting until I get married

In the third analysis, the factors identified above were compared based on the age, grades, family structure, race, fathers' education, urbanicity, and gender. There were insufficient numbers of observations in some cells to test for interaction effects.

Scores on factor 1 (fear-based postponement) were significantly different by gender and alcohol use. Females had lower (more affirmative) scores on this factor than males ( = 1.19/SD = .24 versus 1.37/.33); F (1,595) =53.10, p < .001). Scheffe follow-up procedures showed that adolescents who did not drink were significantly more likely than adolescents who drank less than once a month or more than once a month to give factor 1 as a reason for not having sex ( = 1.22/ SD = .27, 1.28/.31, and 1.36/.32, respectively) (F (2,592) = 11.55, p <.001.)

Scores on factor 2 (emotionality and confusion) showed that there were significant differences due to alcohol usage. Scheffe follow-up procedures showed that adolescents who did not drink at all ( = 1.83/ SD = .20) were significantly more likely than the adolescents who drank less than once a month ( = 1.90/ SD = .16) or the adolescents who drank more than once a month (1.89/.16) to affirm that factor 2 was a reason for remaining abstinent ( F (2,592) = 6.94, p < .001.)

Factor 3 (conservative values) revealed differences based on alcohol use, father's education, urbanicity, grades, and family structure. Table 3 shows the means and significance levels for these groups. Adolescents who did not use alcohol, had better educated fathers, were from nonurban areas, had higher grades, and lived with both parents were more likely to agree that factor three (conservative values) represented a reason for not having sex.

Table 3.
Significant Variables Related to Factor 3 (N=596)

  Group ( Mean/SD) F
Alcohol use    44.08 p<.001
no drinking 1.41/.41  
less than once a month 1.64/.40  
more than once a month 1.78/.35  
Father's education   8.63 p<.01
four or more years of college 1.45/.44  
some college 1.47/.44  
high school or less 1.59/.41  
Urbanicity   7.18 p<.01
nonurban 1.51/.43  
urban 1.61/.40  
Grades   8.68 p<.01
A-B 1.34/.44  
B-C 1.63/.40  
C-D 1.67/.33  
Family structure   4.59 p<.05
two parents 1.51/.43  
other 1.62/.40  

Scheffe follow-up procedures on factor 3 revealed that: a) all three pairwise comparisons were significant for alcohol use, b) adolescents who had fathers with high school educations or less were significantly different from either the students who had fathers with some college or four or more years of college, and c) adolescents who reported their grades were As and Bs were significantly different from the students who had either Bs and Cs or Cs and Ds.


The present study was unique in that it asked abstinent adolescents why they had not had sex. Abstinent adolescents can be labeled as "sexually resilient" because they face the same opportunities and pressures to have sex as their sexually active peers. The present results showed that the lack of an available partner did not appear to be a strong reason for remaining abstinent. Five hundred and four (71.5%) of the subjects stated that lack of a partner was not a personal reason for remaining abstinent. The adolescents in this study were young ( = 14 years) and, as can be expected, many of them had not yet initiated sexual activity. Nationally, only 7% of adolescents report having sexual intercourse before 13 years of age (MMWR, 1998). However, it is noteworthy that only approximately one third reported making a conscious decision to delay sexual activity. This is evident in the following results: did not feel ready for sex ( 32% or 229), waiting for the right person (25 % or 178), or waiting until they are older (35% or 251). Fears of pregnancy and AIDS were the most frequent reasons for not having sex. Issues related to the use of birth control and protection, such as cost, embarrassment and lack of availability, did not surface as important reasons for abstinence.

The adolescents showed that they had absorbed multiple and diverse messages about sex and synthesized them into three distinct abstinence-related factors, each of which was explained by either gender, alcohol use, family structure, grades, father's education, and/or urbanicity. Age may not have been a significant independent variable because of the limited age range of the sample.

The present research did not provide evidence of differences between Whites and African-Americans on the three identified factors. On one hand, it might be hypothesized that there would be racial differences on the first factor because previous research has shown that Black adolescents feel more vulnerable to health risks such as cancer, pregnancy and AIDS than do White adolescents (Eisen, Zellman, & McAlister, 1985; Price, et. al., 1988). On the other hand, it might be hypothesized that there would be racial differences on the third factor due to differences in religiosity (Billy et. Al, 1994; Day, 1992). Perkins, Luster and Villarruel (1998) conducted an ecological risk factor examination of individual, extrafamilial and familial risks related to sexual experience in over 15,000 Black, White and Latino adolescents. They reported little support for the hypothesis that risk factors for sexual activity differed significantly among ethnic groups. Low religiosity was the only risk factor that varied by ethnicity. Low religiosity was a significant predictor of sexual activity for White and Latino males and White and African American females. It remains to be seen if racial differences are revealed on these factors in future research. Future research in this area may need to include a specific measure of religiosity.

As the results revealed, drinking alcohol was a significant independent variable across all three factors. This is consistent with previous research that has pointed to alcohol consumption as a strong predictor of early sexual activity (Flick, 1986; Perkins, Luster & Villarruel, 1998; Small & Luster, 1994).

Factor 1
The first factor had the lowest of the three means indicating it contained items that were some of the most frequents reasons for not having sex. It was the called "fear-based postponement" because the author imagined an adolescent saying, "Having sex is stupid and I am going to wait because you can get pregnant or die, and besides my parents would kill me." Females were more likely to make such a statement. This subscale appears to be the most cognitively-based of the three that were identified here. Factor 1 represents primarily females who have weighed the consequences and decided not have sex yet. The fact that females were more likely to express this position is consistent with the resiliency literature that has pointed to gender differences in relation to cognitive ability and adjustment. Werner (1993) found that internal protective factors (e.g., temperament and cognitive skills) tended to have a greater impact on the adult adaptation of high risk females than of high risk males. The opposite was true for males who benefitted more from external support from family and community.

Much research exists about gender differences in risk factors, but very little exists about gender differences in protective factors (Clark, 1995). According to the Centers For Disease Control (MMWR, 1998), the decline in adolescent sexual activity in the U.S. from 1991 to 1997 was significant for males but not females. There is little research on the charactersitics of effective gender-based messages to help adolescents become sexually resilient.

Factor 2
The second factor had the highest mean indicating that it contained items that were the least likely to be selected as reasons for not having sex. It represented "emotionality and confusion" because adolescents who were low on this factor would be more likely to say, "I haven't had sex because it might hurt and it is embarrassing and hard to get the stuff you need to protect yourself." This was especially true for adolescents who did not drink alcohol. The items in this scale point to an individual who has not clearly delineated why he or she has not had sex and is operating with a mix of loosely related contextual factors. This individual may be at high risk for early initiation of sexual behavior because of a lack of a specific reasoning process that can act as a protective factor.

Factor 3
The third factor revealed a more conservative value stance on the part of adolescents who lived in nonurban areas, lived with their parents, did not drink, had better grades and had better educated fathers. Previous research has identified these variables as significant in the prevention of risky sexual activity (e.g., Costa et. al., 1995; Flick, 1986; Small & Luster, 1994). However, no research was found that examined the relationship between adolescents' reasons for being abstinent and protective factors such as: refraining from alcohol use, having better educated parents, having a higher grade point average, and living in a two-parent household. An adolescent whose views were represented by this factor might say, "I am waiting until marriage to have sex because my religion says that is what I am supposed to do."

In terms of urbanicity, it was not surprising that the nonurban adolescents held more conservative values. Previous research has shown that, compared to their urban counterparts, rural residents consider religion and the role of the church to be very important in community life, hold more traditional moral values, expect greater conformity to community norms, and are less tolerant of diversity (Rounds, 1988). However, such values may not impact behavior because few differences have been found between urban and rural adolescents in their ages of initiation of sexual activity, extent of contraceptive use or rates of adolescent pregnancy (Skatrund, Bennett, & Loda, 1998; McManus & Newacheck, 1985; Yawn & Yawn, 1987, 1993).

Implications For Prevention

While the resilience literature contains various related terms such as protective factors, risk factors, assets, and positive traits, it coalesces around an ecological model that stresses the roles of the individual, culture and environment in protecting youth from engaging in high risk behaviors and experiencing negative consequences from exposure to environmental stressors. The results reported here point to the significant roles that specific individual (alcohol consumption, fear of pregnancy and disease), culture (gender), and environmental (father's education, urbanicity, family structure) play in the adolescent's decision about when to begin having sex.

The "sexually resilient" youth has received little attention in the prevention literature. Based on the results reported above, two messages need to be given to early adolescents to build resiliency and to support behaviors that delay the initiation of sexual activity. The first message is that abstinence is the only 100% effective way of preventing unwanted pregnancy and disease. Some educators may choose to combine this with information on birth control and condoms., depending on local and cultural values. This first message is recommended because early adolescents have been shown to a) be sporadic users of contraception (Ford, Zelnick & Kantner, 1981; Herz & Reis, 1987), and b) misunderstand facts about pregnancy, contraceptive techniques, and the consequences of unintended pregnancy and parenthood (Herz & Reis, 1987). It appears that such a concrete message will be particularly effective with early adolescent females who fear pregnancy and disease.

While educational programs and media that attempt to reduce barriers to birth control and condoms (accessibility, cost, embarrassment) are important, they may not get the attention of early adolescents because the barriers are not perceived as salient. They may not have had experiences where they attempted to acquire or use birth control or protection and are not sensitized to the issues involved.

And second, given that alcohol consumption is a risk factor that is strongly related to early sexual activity, anti-drinking messages need to target adolescents in the age group studied here, below 15 years of age. Magazines and television shows that are popular with young females could emphasize the security and freedom from fear that comes with abstinence. While media outlets that are popular with both genders could place more emphasis in the anti-alcohol message. Finally, communities need to work together to foster the messages stated above, as well as to support the kinds of programs and services that are conducive to healthy adolescent development. These could include mentoring, academic tutoring, faith-based recreational activities, and broad-based family support. Parents, family life educators and teachers can use the RAS as an informal needs assessment tool and then target messages to build resiliency around the feelings of particular groups of adolescents.


Although the RAS needs research and documentation concerning its psychometric properties with different groups of adolescents, the results reported here point to three broad reasons why adolescents perceive they have not had sex: fear, confusion and religion. Sexually abstinent adolescents have not received the same amount of attention as their sexually active peers. Increasing our understanding of why some adolescents choose not to have sex is in keeping with the trend toward identifying protective rather than risk factors that contribute to resiliency. The pertinent question is: What have non-sexually active adolescents done right rather than what have their sexually active peers done wrong? Practitioners and researchers need to recognize that the diverse messages that have been given to American adolescents concerning sexuality have resulted in multiple reasons for some of them remaining abstinent. Some sexual resiliency in adolescents may be attributable to individual, cultural and/or environmental characteristics. No one message, no one philosophical base or value stance, and no one approach to resilience is going to be equally effective with all adolescents.

Perkins, Luster and Villarruel (1998, p. 663) stated, "There are different pathways to sexual activity...the risk factors that are important may differ from individual to individual." In light of the results presented here, the statement could be rewritten as, "There are many different pathways to sexual abstinence...the protective factors that are important may differ from
individual to individual and from group to group."


Benson, P.L., Blyth, D.A., Deville, C., & Wachs, J. (1997). Developmental assets among Seattle Youth. Minneapolis: Search Institute.

Billy, J.O.G., Brewster, K.L., & Grady, W.R. (1994). Contextual effects of sexual behavior of adolescent women. Journal of Marriage and the Family, 56, 387-404.

Bogenschneider, K., Small, S., & Riley, D. (1992). An ecological, risk-focused approach for addressing youth-at-risk issues. Chevy Chase, MD: National 4-H Center.

Bogenshneider, K. (1996). An ecological risk/protective theory for building prevention programs, policies, and community capacity to support youth. Family Relations, 45, 127-138.

Block, J., Block, J.H., & Keyes, S. (1988). Longitudinally foretelling drug usage in adolescence: Early childhood personality and environmental precursors. Child Development, 59, 336-355

Bronfenbrenner, U. (1976). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press.

Bronfenbrenner, U. (1986). Ecology of the family as a context for human development: Research perspectives. Developmental Psychology, 22, 723-742.

Brooks-Gunn, J., & Paikoff, R.L. (1993). "Sex is a gamble, kissing is a game": Adolescent sexuality and health promotion. In S.G. Millstein, A.C. Petersen, & E.O. Nightingale (Eds.), Promoting the health of adolescents: New directions for the twenty-first century (p.180-209). New York: Oxford University Press.

Brown, J.D., Childers, K.W., & Waszak, C.S. (1990). Television and adolescent sexuality. Journal of Adolescent Health Care, 11, 62-71.

Carmines, E., & Zeller, R. (1979). Reliability and validity assessment. Beverly Hills, CA: Sage.

Cattell, R.B. (1952). Factor analysis. New York: Harper & Row.

Clark, P. (1995). Risk and resiliency in adolescence: The current status of research on gender differences. Equity Issues, 1, 1-13.

Costa, F.M., Jessor, R., Donovan, J.E., & Fortenberry, J.D. (1995). Early initiation of sexual intercourse: The influence of psychosocial unconventionality. Journal of Research on Adolescence, 5, 93-121.

Day, R.D. (1992). The transition to first intercourse among racially and culturally diverse youth. Journal of Marriage and the Family, 54, 749-762.

Eisen, M., Zellman, G.L., & McAllister, A.L. (1985). A health belief model approach to adolescent fertility control: Some pilot program findings. Health Education Quarterly, 12, 185-210.

Flick, L.H. (1986). Paths to adolescent parenthood: Implications for prevention. Public Health Reports, 101, 132-147.

Ford, K., Zelnick, M., & Kantner, J. (1981). Sexual behavior and contraceptive use among socioeconomic groups of young women in the United States. Journal of Biosocial Science, 13, 31-45.

Garmezy, N. (1983). Stressors of childhood. In N. Garmezy & M. Rutter (Eds.), Stress, coping and development in children (pp. 43-84). New York: McGraw-Hill.

Garmezy N., Masten, A., & Tellegen, A. (1984). The study of stress and competence in children: A building block for developmental psychopathology. Child Development, 55, 97-111.

Herz, E.J., & Reis, J.S. (1987). Family life education for young inner-city teens: Identifying needs. Journal of Youth and Adolescence, 16, 361-377.

Kaiser, H.F. (1958). The varimax criterion for analytic rotation in factor analysis. Psychometrika, 23, 187-200.

Kaiser, H.F. (1974). An index to factorial simplicity. Psychometrika, 89, 31-36.

Lerner, R.M. (1995). America's youth in crisis: Challenges and choices for programs and policies. Thousand Oaks, CA: Sage.

Lowry, D.T., & Towles, D.E. (1989a). Soap opera portrayals of sex, contraception, and sexually transmitted diseases. Journal of Communication, 39, 76-83.

Lowry, D.T., & Towles, D.E. (1989b). Prime time TV portrayals of sex, contraception and venereal diseases. Journalism Quarterly, 66, 347-352.

Masten, A. S., Best, K. M., & Garmezy, N. (1990). Resiliency and development: Contributions from the study of children who overcome adversity. Development and Psychopathology, 2, 425-444.

McManus, M.A., & Newacheck P.W. (1985). Rural maternal, child, and adolescent health. Health Services Research, 23, 807-848.

MMWR. (1996). Youth Risk Behavior Surveillance, 45 (SS-4), Atlanta: Centers for Disease Control, 1-86.

MMWR. (1998). Youth Risk Behavior Surveillance, 47 (SS-3), Atlanta: Centers for Disease Control, 1-86.

Perkins, D.E., Luster, T., & Villarruel, F.A. (1998). An ecological, risk-factor examination of adolescents' sexual activity in three ethnic groups. Journal of Marriage and the Family, 60, 600-673.

Perry, C.L., Kelder, S.H., & Komro, K.A. (1993). The social world of adolescents: Family peers, schools and the community. In S. Millstein, A. Petersen, & E. Nightingale (Eds.). Promoting the health of adolescents: New directions for the twenty-first century (pp. 73-97). New York: Oxford University Press.

Price, R.H., Desmond, S.M., Wallace, M., Smith, D., & Stewart, P.M. (1988). Differences in Black and White adolescents' perceptions about cancer. Journal of School Health, 58, 66-70.

Rounds, K.A. (1988). AIDS in rural areas: Challenges to providing care. Social Work, 33, 257-261.

Rutter, M. (1979). Protective factors in children's responses to stress and disadvantage. In M.W. Kent & J.E. Rolf (Eds.), Primary prevention in psychopathology. Vol. 3: Social competence in children (pp. 49-74). Hanover, NH: University Press of New England.

Skatrud, J.D., Bennett, T.A., & Loda, F.A. (1998). An overview of adolescent pregnancy in rural areas. Journal of Rural Health, 14, 17-26.

Small, S.A., & Luster, T. (1994). Adolescent sexual activity: An ecological, risk-factor approach. Journal of Marriage and the Family, 56, 181-192.

U.S. Census Bureau. (1993). Census of Population and Housing: Summary Tape File 4 Technical Documentation, Washington, DC.

Werner, E.E. (1993). Risk, resilience, and recovery: Perspectives from the Kauai Longitudinal Study. Development and Psychopathology, 5, 503-515.

Werner, E.E., & Smith, R.S. (1990). Protective factors and individual resilience. In S.J. Meisels & J.P. Shonkoff (Eds.). Handbook of early childhood intervention (pp. 97-116). Cambridge, England: Cambridge University Press.

Werner, E.E., & Smith, R.S. (1992). Vulnerable but invincible: A study of resilient children. New York: McGraw-Hill.

Yawn, B.P., & Yawn, R.A. (1987). Teenage sexual activity in rural Minnesota. Minnesota Medicine, 70. 38-39.

Yawn, B.P., & Yawn, R.A. (1993). Adolescent pregnancies in rural America: A review of the literature and strategies for primary prevention. Family and Community Health 16 (1), 36-45.

*The research described here was supported by grants from the Pharmacia-Upjohn Corporation and Grant APR 000971-01-0 with the Office of Population Affairs, Department of Health and Human Services.

Key Words: Sexual abstinence, adolescents, resilience




For additional information, contact:

Kim Allen

Center on Adolescent Sexuality, Pregnancy and Parenting
Associate State Specialist University of MO Extension
303 Gentry Hall Columbia, MO 65211
Phone: 573-884-0644
Fax: 573.884.4878


University Outreach & Extension-University of Missouri

HES Extension Site Manager