Statistical and descriptive Literature review on 4 empirical articles relating to sexual socialization through family peers and societal implementations of gender roles.
Literature Review: DACA and Mental Health of Immigrant Young Adults in The U.S
In the United States, a continuously growing population of 11.7 million undocumented
immigrants exist within the rest of U.S society. About 5 million are children and young adults
under the age of 30 that although live like U.S citizens– attend school, are employed to jobs,
contribute to Taxes– face serious stressors and anxieties of deportation and limitations which
include restrictions to job opportunities, social integration, access to healthcare, and financial aid
to pursue an education. These limitations cause a damaging psychological toll on this large
population of young adults. Deferred Action for Childhood Arrivals (DACA), a program
established through executive order by the Obama administration in 2012, provided some young
adults exclusion of deportation and temporary work authorization for a period of two years. This
temporary fix gives many immigrant young adults a relief and sense of security they lacked
before. The mental health wellbeing of this population is found to better with the adoption of the
DACA program as stressors are alleviated, but do not remove the fears of deportation
completely. Research examines potential pathways that DACA may influence health and mental
well-being both negatively and positively among this population in order to better understand
this group of individuals who undergo such an significant stage of development full of life
transitions and development of self-image. In gaining a better understanding, social aid and
support can better target and address health concerns for the undocumented young adult
population.
Because immigrants from Latin America make up over 90% of the population qualified
for DACA, most research is conducted in regards to the Latino population, even though the
DACA applicants are of many different ethnicities. Few studies have explored the impacts of
legal status on psychological well-being because the DACA program is fairly new and recently
has been terminated by the Trump administration. Some prior research on DACAs short-term
impacts indicated that it has increased young adults’ access to opportunities and therefore
contributed positively to their mental health well-being. Most recent research, suggests that the
social program although improves the mental health well-being of recipients, does not terminate
completely the ongoing fears and anxieties of being deported and having limited opportunities in
the United States. In addition, different ethnic-groups have been explored in order to extend
knowledge beyond the Latino population but also the Asian and Pacific Islander ethnic group.
Research conducted has been mostly qualitative but studies below show how quantitative
research provides an alternative view. Studies below range from the Summer of 2013 to March
of 2016 and have all been conducted in California which is home to 2.6 million and growing
undocumented immigrants. The following studies analyze the psychological stressors that come
with being a Latino, Asian, or Pacific Islander young adult immigrant in California and the effect
DACA has– whether positive or negative– on this population.
Psychological stressors of immigrant young adults. All available research displays
the life of an immigrant young-adult to be one containing an array of difficulties and
socioeconomic limitations that cause particular stress, anxiety, and an ongoing paranoia. Some of
the most common psychological effects due to their social status found in studies are depression,
distress, vulnerability, anxiety, negative emotions, isolation, and a perpetual fear of deportation.
Siemons’ study of 61 latino DACA eligible young adults ages 18-31 found that discrimination,
loss of social status, marginalization, reduced social integration, challenges self-identity, suicidal
thoughts, feelings of isolation, and living in a constant state of survival mode are some of the
psychological factors that came with being an undocumented immigrant in the United States
(2012). Similarly, a study conducted on 32 undocumented Asian and Pacific Islander (API)
participants in California also ages 18-31 found fear of deportation, uncertainty of the future, and
the financial burdens also as common stressors through a set of in-depth interviews and focus
group discussions (2017). Notably, participants described additional stressors in relation with
their ethnicity that Latino based research did not point out. Study shows that API undocumented
young adults “hold a prevailing sense of shame and stigma associated with undocumented
status” (Sudhinaraset, 2017, p. 742) due to a lack of open communication about documentation
status in their community in addition to the common stressors reported undocumented
immigrants face. Patler in 2016 through a statistical analysis pointed out that “socioeconomic
factors were the largest and most consistent predictor”(2) of past psychological well-being of
Latino undocumented immigrants prior to their DACA status. This study recognizes the three
specialized outcomes related to immigrants’ psychological well-being as distress, negative
emotions, and worry about deportation of self or family. Patler, unlike previous studies, also
makes a connection between structural racism and social stress of this population, pointing out
that through structural racism “exclusionary immigration policies can hinder immigrant’s access
to health services and contribute to a discriminatory climate”(2016, 2) which in turns has
negative effects on the mental health well-being of young undocumented immigrants. Similarly,
Raymond-Flesch in a study conducted in 2013 on Latino DACA-eligible young adults’ health
status, needs, and access to health care identifies the many barriers this population faces when
attempting to access this aid. Barriers include cost, limited health care literacy, difficulty
navigating health care logistics among the many listed. In turn, this population experiences a
detriment to their psychological well-being. Although little research has been done on the
psychological effect of undocumentation in the United States, findings are coherent and all point
to the same factors this population faces.
DACA’s positive effect of mental health and wellbeing. Research indicates a
positive effect in mental well being on DACA recipients that alleviates the stressors of being
undocumented. Siemons’ research in 2013 analyses through qualitative methods three mental
health wellbeing specialized outcomes of the DACA program: increased social integration,
greater peer support, and improved sense of self. Studies found that Latino DACA recipients
stated that DACA soothed their integration into US society as well as provided them with
increased access to resources, greater increased autonomy, improved sense of belonging, hope
for the future. They also experienced an improved sense of comfort in disclosing status (545).
Similarly, Research on Asian and Pacific Islander recipients of DACA stated the program helped
to “alleviate the financial barriers that are common for undocumented young
adults” (Sudhinaraset, 2017, 744). For many participants, these changes “profoundly empowered
them to envision their future life trajectories” (Sudhinaraset, 2017, p. 744). The work permit that
is included with DACA lessened the feelings of vulnerability and exploitation recipients felt in
their “under the table” jobs. Participants found this change “liberating” and in turned relieved
them from financial and personal burdens(2017). An improved sense of self was found to be a
result of the DACA program as well. Siemons states that DACA recipients’ “new sense of
belonging that improved with their self-esteem, impacting MHWB positively” (2017, p. 547).
Similarly, Sudhinaraset found that DACA benefits significantly identity and self-perception in
API young undocumented adults in stating that “DACA directly improved their sense of well-
being after having to cope with stressors in their daily lives” (2017, p. 744). Correspondingly,
statistics done in Patler’s (2017) research analyzing the psychological effects before recipients
had DACA and after they qualified for DACA displays the “likelihood of Psychological distress
was lower for DACA recipients than those without DACA” (4). This study also included a series
of controlled variables such as gender and age, which shows for the first time in comparison to
females, “males had significantly lower odds of… negative emotions” (Patler, 2017, p. 4). The
magnitude of change in psychological well-being by DACA status was also calculated and
found “the predicted probability of distress and negative emotions in the past 30 days for DACA
recipients dropped to under 20%, whereas probabilities were 40% or greater for those without
DACA” (Patler, 2017,4). In conclusion, this study found a “strong, positive, and significant
effect of legal status on psychological well-being. Receiving DACA reduced the odds of distress,
negative emotions, and worry about self-deportation by 76-87%, compared to respondents
without DACA” (Patler, 2017, 6). The mental health of this population shows to better with the
DACA program as in relieves the stressors that come with being undocumented.
DACA’s negative effect of mental health and wellbeing. Although positive effects
on the mental well being of undocumented immigrant young adults are found all across, research
shows that DACA does not completely terminate the psychological damage this population
experiences. The ongoing limitations and fear of deportation persists due to the temporary nature
of this program. Siemons’ research reveals most DACA recipients felt that trauma of growing up
undocumented still remained stressful and the fear of paranoia continues to exist (2017). Fear of
deportation does not go away. , actually, it has been shown to transfer or transform from personal
fear to fear of deportation of family members. Siemons’ discovers an “Increased emotional
responsibility and transfer of worry from their well-being to their families well being had mental
health consequences (2017, p. 547). Additionally, Patler’s cross-sectional telephone survey
research expresses “undocumented young adults’ experience emotional consequences due to
their parents’ legal status”(2017, p. 2). The unattained DACA status from family members
concerned young adults and in the case of API students’ in Sudhinaraset 2017 research. Unlike
the research done on the Latino population, Sudhinaraset finds some API DACA-eligible young
adults did not apply for the program due to strong feelings of shame because their family failed
to qualify. (2017, p.744). Research also reports increasing family responsibilities attached to the
DACA status. Feelings of overwhelment due to added responsibility contributed to their stress
and therefore affected their mental health. A DACA recipient in the a study done by Siemons in
2017 expresses that anxiety is still there, it has just shifted. DACA alleviates but does not
terminate stressors that affect the mental well being of this population.
The recent termination of DACA has limited research opportunities on the
undocumented immigrant young adult population. Future research can focus the psychological
effects the actual termination had on these young people. In regards to this research, it is evident
that as far as the Latino, Asian, and Pacific Islander population, DACA helped the psychological
well being of its recipients as it provided more opportunities and freedom than ever before
experienced. Results cannot be generalized to speak for the entire undocumented immigrant
population therefore further research on other ethnicities can benefit the overall understanding
and knowledge. All research mentioned was conducted in English which limits the contribution
of non-english speaking applicants.
Works Cited
Siemons, R., Raymond-Flesh, M., Auerswald, C.L. et al. J Immigrant Minority Health (2017).
Coming of Age on the Margins: Mental Health and Wellbeing Among Latino Immigrant
Young Adults Eligible for Deferred Action for Childhood Arrivals (DACA), 19: 543-551.
Caitlin Patler, Whitney Laster Pirtle, From undocumented to lawfully present: Do changes to
legal status impact psychological wellbeing among latino immigrant young adults?,
Social Science & Medicine (2017).
May Sudhinaraset, Tu My To, Irving Ling, Jason Melo, Josue Chavarin, The Influence of
Deferred Action for Childhood Arrivals on Undocumented Asian and Pacific Islander
Young Adults: Through a Social Determinants of Health Lens, In Journal of Adolescent
Health, Volume 60, Issue 6 (2017) Pages 741-746.
Marissa Raymond-Flesch, Rachel Siemons, Nadereh Pourat, Ken Jacobs, Claire D. Brindis,
“There Is No Help Out There and If There Is, It’s Really Hard to Find”: A Qualitative
Study of the Health Concerns and Health Care Access of Latino “DREAMers”, In Journal
of Adolescent Health, Volume 55, Issue 3, 2014, Pages 323-328.
Archives of Sexual Behavior, VoL 17, No. 4, 1988
Kinsey Revisited, Part II: Comparisons
of the Sexual Socialization and Sexual
Behavior of Black Women Over 33 Years
Gail Elizabeth Wyatt, Ph.D., 1’4 Stefanie Doyle Peters, Ph.D., 2
and Donald Guthrie, Ph.V. 1,»
Kinsey’s findings regarding the sexual behavior of black women were com.pared with data from a more recent study o f sexual socialization and experiences among women in Los Angeles County, Ca. The study examined
responses from two groups o f college-educated black women, ages 18 to 36,
196 women from the original Kinsey sample and 64 women from the new
sample. Log-linear analyses were used to control for differences between
the samples on age and marital status. Comparisons were conducted in the
areas of childhood family characteristics; sexual socialization and education;
sexual behavior in childhood, adolescence, and adulthood, contraceptive practices; and child sexual abuse. Results reflected changes that have taken place
in society and in patterns o f sexual behavior. Differences in sexual socialization pointed to the increased role of the media and the schools and to more
relaxed attitudes about nudity in the home. Shifts in sexual behavior were
particularly dramatic. As compared to women in the Kinsey sample, newer
subjects began intercourse earlier, were less likely to have a fiance or husband as their first partner, reported a higher number o f sexual partners, and
participated in a broader range o f sexual behaviors. Contraceptive practices
differed considerably, especially among never-married women. Women in
This research was supported by the Center for the Prevention and Control of Rape, NIMH,
MH 33603 and through a Research Scientist Career Development Award (MH 00269), to the
first author.
1Department of Psychiatry and Biobehavioral Sciences, 760 Westwood Plaza, University of
California, Los Angeles, California 90024.
2Barrington Psychiatric Center, 1990 South Bundy Drive, Suite 320, Los Angeles, California 90025.
aDivision of Biostatistics, School of Publ!c Health, University of California, Los Angeles.
4To whom correspondence should be addressed.
289
0004..0002/88/0800-0289506.00/0 © 1988PlenumPublishingCorporation
290
Wyatt, Peters, and Guthrie
this study were slightly more likely to report instances o f child sexual abuse.
Methdological and social factors contributing to the findings are discussed.
KEY WORDS: women’ssexualbehavior; femalesexuality;transgenerationalchanges.
INTRODUCTION
When Kinsey et aL, published Sexual Behavior in the Human Female
in 1953, an overwhelming interest in and concern abo ut the nature and prevalence of sexual behaviors emerged. The range of sexual experiences that were
reported, as weil as their age of onset and frequency, proved to violate previous conceptions regarding the adherence of American females to the JudeoChristian ethic. Consequently, the demand for more information to clarify
and enhance these findings grew. Kinsey and his colleagues collected data
from a wide variety of segments of the population and from more than one
ethnic group (Kinsey et aL, 1953). Thus, in 1958, Gebhard et aL published
Pregnancy, Birth andAbortion, which included data from a small and highly stratified sample of black women. Despite the limited generalizability of
the sample, until the 1960s, the results of the Kinsey study of these black
women’s sexual behavior and socialization were considered representative of
black women nationally. With the concentration of research on social policy and the human condition during the Kennedy era, studies concerning such
pertinent issues as pregnancy among unwed black adolescents and attitudes
towards and utilization of contraceptives increased. However, the research
of Kinsey and his associates served as an initial part of the documentation
that aspects of the sexual patterns of black women differed significantly from
their white peers. Indeed, with the publication of these two books, the inevitable occurred: comparisons between black and white women were made
(see Gebhard et al., 1958). Although some of the differences noted between
the two groups were influenced by income, education, marital status, and
the obvious sampling biases which severely limited interpretation of the findings, other differences in results appeared to be racial in origin. Unlike
research in areas other than sexuality, where more recent findings usually
become more prominent in the literature, use of the Kinsey data regarding
the sexual behavior of black women appeared to increase rather than decrease.
Several factors contribute to the continued use of these data: (i) Empirical studies that include adequate samples of black women remain few
in number; and (ii) the differences between black and white women apparently place some of the more controversial aspects of the sexual behavior
of white women in perspective. For example, akhough it was alarming to
learn about the prevalence of white women’s premarital coitus before age
20, data suggesting that black women became sexually active even earlier than
their white peers changed the focus of concern from white to black women.
And for the most part, that concern has been maintained in the last 20 years
Sexual Socialization and Sexual Behavior
291
of sex research without much progress in understanding the factors that contribute to the differences in black women’s sexual patterns.
While the significance of Kinsey and his colleague’s work must be acknowledged, there is a greater need to gather additional information which
will clarify and enhance knowledge about the sexual patterns of black
women within a sociocultural perspective. This paper examines differences
between the sexual behavior of black women in the Kinsey study and black
women in a more recent study conducted by Wyatt (1985). Each study is
summarized and the sexual socialization and sexual experiences of the two
samples of black women äre compared and contrasted.
Although many of the differences between the samples are controlled
for in the forthcoming analyses, the geographic diversity of the samples should
be noted. Kinsey and his associates interviewed black women in the Northeast and Midwest, whereas our sample was recruited from and is representative of Los Angeles County. Thus, differences that are reported may be as
much a function of regional differences in sexual patterns as they are the
result of generational changes of, of course, the limitations of the samples.
Another difference between the samples was the manner in which the
ethnicity of black women was defined. Kinsey included women who identified themselves as Negro or colored in his sample of black women. These
women were of African descent, but their socialization could have taken place
in a country other than the United States (e.g., the West Indies). Out sample
identified themselves as Afro-American or black. They were also asked how
many years before age 13 were spent in America, and those who spent more
than 6 years in other countries were deleted from the sample. Thus, an attempt was made to identify an Afro-American sample whose diverse cultural experience was an integration of African along with other cultural groups,
but whose early socialization occurred, for the most part, in the U.S. The
emphasis was placed upon African heritage as it interfaced with American
values, economics, language, and culture, rather than solely upon racial background (Bass et aL, 1982). Although some of the cultural experiences may
have differed between the two samples, it was assumed that they also shared
enough of the same values and customs, to the extent that their ethnic background was comparable. The term “black” is used to define the ethnic comparability of both groups of women.
METHODS
The Kinsey Study
Sampling
Detailed accounts of the sampling procedures used are provided in Sexual Behavior in the Human Female (Kinsey et al., 1953) and The Kinsey Data:
292
Wyatt, Peters, and Guthrie
Marginal Tabulations of the 1938-1963 Interviews Conducted by the Institute for Sex Research (Gebhard and Johnson, 1979). In summary, Kinsey’s
primary approach to sex research was atheoretical in its description of what,
rather than why, sexual behavior occurred, or its significance to the individual.
Instead of adopting random sampling techniques, Kinsey chose to increase
the size and diversity of his sample to offset sampling biases and other such
problems (Gebhard and Johnson, 1979). For example, Kinsey used quotas
to guide his search for subjects-that is, young, educated females. He sought
locations for this segment of the sample, such as colleges and sororities, and
interviewed any friends of the participants who volunteered. The types of
institutions or sources utilized are listed in 10 categories ranging from college students and faculty to those individuals known to belong to homose×ual organizations (see Gebhard and Johnson, 1979). Although more than
700 black women were interviewed, 353 were delinquent, meaning that they
had been convicted of a misdemeanor or felony and were or had been incarcerated, 4 86 had extensive homosexual experience, and 125 were nondelinquent, noncollege-educated women. Due to known sampling biases, these
women were deleted from this study. From the remaining 223 black collegeeducated women, a subsample of 196 women aged 18 to 36 years was used
in these analyses.
The interview was structured chronologically and obtained information
about sexual behavior and its frequency both pre- and postpubertally. Questions were often reworded to accommodate the age and experience of the
subject. White male interviewers conducted face-to-face interviews and engaged in a conversation format with subjects before collecting data. Responses
were transcribed by hand during the interview, and those which were interdependent with others were cross-checked so that inconsistencies could be
clarified and rectified.
The Wyatt Study
Sampling
Multistage stratified probability sampling with quotas was used to
r e c r u i t c o m p a r a b l e s a m p l e s o f A f r o – A m e r i c a n a n d w h i t e w o m e n , 18 t o 36
4Specifically, women who were or had been incarcerated were selected from the indiana Woman’s Prison; the New York Woman’s Reformatory, the Wilmington, Delaware Kruse School
(for delinquent black girls), and several juvenile homes or settlement houses for young black
females who were, or threatened to be, delinquents. Gebhard et al. “s Pregnancy, Birth and
Abortion (1958) included a chapter on black prison women.
Sexual Socialization and Sexual Behavior
293
years of age, residing in Los Angeles County. Details of the procedures used
to obtain a representative sample are described elsewhere (Wyatt, 1985). Quotas used in the study were based on education, marital status, and the presence
of children among Afro-American women.
Subjects were recruited by random-digit dialing of telephone prefixes
in Los Angeles County, combined with four randomly generated numbers.
Among households in which a woman resided, the estimated refusal rate was
33%. (This calculation is based on those women who were eligible to participate but refused and an estimated portion of those who terminated contact
before their eligibility could be determined.) Although the rate of 33% is
higher than desired, it should be noted that this is the only multiethnic community sample in research on sexuality or sexual abuse with comparable
demographic characteristics for both ethnic groups. The first 248 women
meeting the desired quotas were interviewed: 126 of these were Afro-American
women.
The Procedure
Each participant was interviewed by a highly trained and experienced
woman of the same ethnicity as the subject. Prior to data collection, interviewers underwent an intensive 3-month training program. Subjects were interviewed at the location of their choice, and were reimbursed $20.00 for
their time and up to $2.50 for baby-sitting or transportation costs. Interviews were usually conducted in two sessions, and the total length ranged
from 3 to 8 hr. At the completion of the interview, referrals for mental health
services were provided upon request.
Instrumentation
The Wyatt Sex History Questionnaire (WSHQ), a 478-item structured
interview, was used to obtain both retrospective and current data regarding
women’s sexual histories from childhood to adulthood, including data on
child sexual abuse. The WSHQ was developed from two pilot studies prior
to its use in this research (Wyatt, 1982). Reliability established on closedended items of the questionnaire was high (Pearson r = 0.90). when 10 audiotapes were selected at random, only two responses out of 4780 items
reviewed suggested errors in the interviewer’s written transcription. Reliability established on a weekly basis among the interviewers averaged 0.90.
Comparisons of retrospective information regarding women’s demographic characteristics obtained both as part of telephone recruitment and
during the interview, with 1 month to 2 years between the two data collec-
294
Wyatt, Peters, and Guthrie
-~, õ o . ~ , ~
.~
ù~
r~
õ
0
•- ~ ,
~o
õ
o
ü
o~”
.~ ~
Sexual Socialization and Sexual Behavior
‘~
c~ ~
o~
~’~
~,~.o
~
~
“~
•, ~ o
X
0
295
~
~
~
~.~
o
~ .~=
0
“~-
o
r~
<
,3
oa
0
0
0
o.~
~
~'-~
..~
~
"~-~
o~
o~
"0
.o
r~
0
00
c~
296
Wyatt~ Peters, and Guthrie
tion points, showed an average test-retest reliability of 0.84. These analyses
helped assure the reliability of the retrospective information obtained.
Comparison of the Kinsey and Wyatt Methods
Table I illustrates the basic methodologies of both studies and facilitates
their comparison. One noteworthy similarity between the studies is that both
require retrospective recall of information. The publication of Sexual Behavior
in the Human ,VIale (Kinsey et aL, 1984) brought forth concerns about
memory error performance from those who critiqued the methodology used
(Cochran et aL, 1954). The techniques developed to minimize memory performance errors are listed in Table I. An additional discussion of the mea,
sures used to elicit retrospective information and to maximize accurate recall
is provided in a companion paper (Wyatt et aL, 1988).
Although the two studies shared similarities in their approaches to using retrospective data, there are also important differences that should b e
highlighted. One concerns the conceptual framework that guided each :of the
studies. First, Kinsey was interested primarily in the occurrence or nonoccurrence of behaviors rather than in the individual's motivation or the meaning of the behavior. In contrast, we also collected data on the reasons,
circumstances, and affective reactions surrounding sexual behavior. A
second, important difference involves the framing of questions. Our subjects
were asked if they had engaged in a behavior, whereas Kinsey often asked
when the subject had engaged in it, placing the burden of denial on the subject. For certain of the comparisons presented, these methodological differences are critical to understanding the findings.
Current Demographic Characteristics
The demographic characteristics of both samples at the time of the interview are presented in Table II. Women were divided by age, 18 to 26 and
27 to 36, in order to examine generational differences.
Although the sample of black women in Kinsey's sample was markedly
biased, it tended to follow the same trend noted in previous research with
the sample overall: The women were predominantly young, single, and childless (Kinsey et al., 1953). The new sample tended to be older (27-36 years),
once married, and with at least one child. As mentioned earlier, only women with at least a college education were selected for this study, due to sampiing biases in Kinsey's non-college-educated sample (see Gebhard and
Johnson, 1979). As was evident in the sources of recruitment at universities,
over half of Kinsey's sample were students then currently in school. Similar
Sexual Socialization and Sexual Behavior
297
to the trend of the most common religious affiliations among black families
(Mendes, 1982), most women in both samples tended to be Protestant, with
the next largest groups being Catholic (Jacquet, 1986). It was interesting to
observe the increase in women who reported no religious affiliation at all
in the sample as compared to their counterparts in Kinsey's sample. However,
among those who did practice a religion, there was only a slight increase in
women in our sample who did not ever attend religious services. Thus, those
who were religious tended to attend services at least once a month.
The age of first marriage tended to shift slightly to 18 or older among
our sample, possibly due in part to sampling biases. Kinsey interviewed women in a rural area of Kansas, a few of whom were included in this sample
of college-educated women (personal communication with Paul Gebhard,
Kinsey Institute, Bloomington, Indiana, October 1984 and February 1986).
Their marital patterns, along with a shift towards a delay in the age of marriage, may have accounted for slightly more of Kinsey women's early age
of marfiage (Davis, 1972). The fact that both samples tended to marry only once
may not be as similar a finding as it seems. Women in our sample also tended to
have once been married or divorced, suggesting that these women may not
remarry, whereas Kinsey's sample may have had more single, ongoing marriages.
Occupational status showed a slight shift in the direction of upward
mobility, although the difference was not statistically significant. The proportion of women employed in managerial and professional positions went from
less than half (45%) in the Kinsey sample to slightly more than half (55%)
in the new sample.
Statistical Methods
As noted above, preliminary analyses of the two samples clearly showed
substantial differences in responses to items that are possible correlates of
responses to sexual behavior items. In order to compare the two samples,
it is therefore necessary to construct statistical adjustments for these differ=
ences. First, all data items were categorized; that is, responses that were numerical (e.g., age at first intercourse) were grouped to form categories. A
simple, unadjusted comparison of response across the two samples could thus
be represented as a two-way contingency table, and a chi-square test performed. This approach does not, however, adjust for other known correlates of sample membership. If, for example, the age range in one sample
were higher than in the other, and if the incidence of a particular behavior
were correlated with age, then a statisticaUy significant difference between
samples might be due to the difference in age.
To circumvent this problem, and to isolate the degree to which the
difference in response is uniquely due to sample membership, we performed
Wyatt, Peters, and Guthrie
298
,d
<
Z
<
Z
h¢l
II1
~t'%l
,II
k)
0
O
>
¢I,
Sexual Socialization and Sexual Behavior
299
o
zo
t
I
,n-
z
z<
~.
8
g
o
~3
"~
~
"~-
~
0
e~
g
0
0
o
300
Wyatt, Peters, and Guthrle
a series of log-linear analyses (Bishop et al., 1975) of the four-way contingency tables arising from cross-classification of response, sample, and the
categorized values of age and marital status. We followed the process suggested by Benedetti and Brown (1978). First, a log-linear model was constructed which admits dependence of the response variable on all
combinations of age and marital status, but not sample. A chi-square test
measured whether this model adequately represents the data; a significant
value of chi-square indicated that it does not, and that differences between
the samples are present in some form. We then successively added terrns to
the long-linear model until we encountered a nonsignificant chi-square test.
In some of our tests, no significant sample effect was found after adjustment for the other variables. In some cases, the addition of a "main effect"
term for sample differences provided a satisfactory statistical exptanation
of the table. In these cases, we concluded that after adjustment for the control variables, significant sample differences are present, and that they do
not interact with the levels of the control variables. In other cases, more complex models were necessary. For example, if it were found necessary to include a model term for both age and sample, it would be concluded that there
may not only be differences between the two samples but that they may also
differ in magnitude according to the age group of the respondents.
Most of the responses were analyzed by these methods. In some cases,
because of low rates of one level of response, it was necessary to combine
tables and disregard some of the control variables. These situations are noted as they occur.
RESULTS
Childhood Demographic and Environmental
Characteristics
Comparable data were available from the two samples on a variety of
items concerning the subject's life as a child. These included information on
environmental characteristics, parental demographics, family structure, and
religion (see Table III).
The two samples differed in terms of the area of the country in which
subjects had been raised. Over half the women in the new sample were raised
in the Pacific/Mountain region, whereas the majority of Kinsey's subjects
had grown up in either the South or the North Central region.
Parental demographic characteristics were quite similar across the two
samples. Controlling for the women's current demographic characteristics,
Sexua| Socialization and Sexual Behavior
301
log-linear analyses revealed no significant differences on mother's or father's
age, father's education, or mother's or father's occupation. The majority of
black women's parents were 55 or younger and had no college education.
For both mother's and father's occupation, there is a shift in the direction
of increasing white-collar employment among the new sample, but the differences were not statistically significant.
The vast majority of women in both samples reported their childhood
religious affiliation as Protestant, although the new sample included a higher
percentage of Catholic women. However, patterns of attendance have
changed over time. Over 90°70 of the Kinsey subjects attended church once
a week or more, as compared to 70070 of our sample.
Changes in family composition were also evident. The proportion of
women who lived with both patents throughout childhood was significantly
higher in the Kinsey sample as compared to our sample. Similarly, the percentage of women who lived with a stepfather at some time during childhood or adolescence was almost twice as high among women in our sample,
although the difference was not statistically significant. These findings are
consistent with trends evident in overall U.S. Census statistics (Rodgers-Rose,
1980).
Sexual Socialization and Education
Sexual socialization included the informal sources from which children
learned about human sexuality and the ways in which it should be expressed,
as weil as more formal sources such as the schools (see Table IV). The first
question in the area of sexual socialization had to do with the source from
which women first became aware of such aspects of sexuality as male-f emale anatomical differences, labels for genitalia, or sexual behaviors.
Although peers have been described as the most common source of sex-related
information (Herold and Goodwin, 1981; Kaats and Davis, 1970; Libby et
al., 1978; Mahoney, 1980; Mirande, 1968; Reiss and Miller, 1979; Shah and
Zelnik, 1981), our sample reported that their awareness was more a function of information from family members, sex education classes, the media,
or as a result of voluntary or abusive sexual experiences in childhood.
There did not appear to be much change in the age at which women
obtained specific knowledge about intercourse. Women in both samples
learned about coitus at age 12 or younger. However, similar to their source
of general sexual knowledge, our sample reported a wider variety of sources
of information about intercourse including family, formal education, and
the media, as weil as personal sexual contact.
Although family nudity appears not to be an acceptable type of informal sex education in the homes of black families (Stables, 1973), almost half
302
Wyatt,
Peters, and Guthr~e
0
e~
.(
Z
o
o
0
o
o
Z
Z
Z
Z
Z
~
0
$.,
e~
e~
"0
0
0
r~
"0
(W3
,,w
{J
lP
[,,W
0
~
~.~u~
Z0 Z0 ~~
0
Sexual Socializaüon and Sexual Behavior
303
;»-, ù o
ù
"C=
I.
oB m
0
O'3
Z
C~
~'.=,
"~ 8
U
e~
e~
~
m
i»t.~
~q
•-
o
~~=.-~
~~
~
~.R
8~
,~
~
~
304
Wyatt, Peters, and Gu~hr~e
~:
,-4
"
0
0
0
0
ell
o
0
~.
a
68
o
~
o
Sexual Socialization and Sexual Behavior
1
m
z
305
m
~
~Z
8
~D
~
°~
.,..r
~z
N~z
g~z
O0
Wyatt, Peters, and Guthrie
306
(47°70) of the new sample recaUed at least occasional nudity as compared to
Kinsey's sample. However, in spite of a 10% increase in women in the new
sample who as children recaUed their parents having coitus, the freedom to
learn about sex through observing parents seems to have been limited.
Comparison of the two samples on formal education regarding human
sexuality yielded striking increases in educational information in grade school
and particularly in high school in our sample. Apparently, there has either
been an overaU increase in some sort of sex education, or women in Los Angeles County reported such an increase.
Childhood Sexual Behavior
The two samples were compared on various aspects of childhood sexual behavior (summarized in Table V) pertaining to sexual play with other
children and masturbation.
Sexual Play With Other Children
Sex play with other children was reported by significantly more women in the Kinsey sample. Over 40% of the Kinsey subjects had engaged in
homosexual and/or heterosexual play during childhood as compared to only
about 20% of our subjects. Age of onset of these behaviors was also younger among the Kinsey women, although no statistical tests of these differences could be conducted due to the small numbers of subjects.
However, we believe that these findings are likely due in part to differences in question wording and content (see Wyatt et aL, 1988 for a discussion of this point).
Masturbation
There was a significant difference between the two samples in the
proportion of women who masturbated during childhood. Only 14%0 of the
Kinsey subjects reported masturbating as compared to 33% of our women.
Because the questions asked in the two studies were more similar for masturbation than they were for sex play involving other children, we are more inclined to view these findings as a true change in behavior patterns. Although
societal and religious proscriptions regarding masturbation have been noted
(Staples, 1973), parental attitudes may have become more tolerant over time.
As was the case for the white women sampled (see Wyatt et al., 1988), only
10%0 of the Afro-American women in our sample recalled having received
prohibitive communications about masturbation from their parents during
childhood.
Sexual Socialization and Sexual Behavior
307
Among women who masturbated, guilt was reported by a significantly greater proportion of women in the new sample. However, there
are several factors that might contribute to these findings. Differences between the two interviews may have led to an overestimation of guilt among
black women in our sample and underestimation among the Kinsey sample.
(For further elaboraüon of this point, see Wyatt et al., 1988.) AddiüonaUy, masturbatory behavior has been associated with other psychological
problems (Clower, 1975) and with child sexual abuse (Friedrich et al.,
1986).
Adolescent Sexual D e v e l o p m e n t and Behavior
Adolescent sexuality is an area that has been of great interest to sex
researchers. For this study, we were able to compare data on several topics:
sexual development, onset of sexual activity, and homosexual experience (see
Table VI).
Sexual Development
Two aspects of sexual development-age of breast development and
age of first menses-were examined, with the ages categorized as 12 or
younger versus 13 or older. The two samples did not differ significantly on
either variable. Around 70°70 of the women in both samples reported that
breast development began at 12 or younger, and over 50070 reported first
menses as occurring in that same period.
The lack of differences between samples was somewhat surprising, given
the general idea that puberty in females has been occurring at increasingly
earlier ages. However, this downward trend appears to be a very gradual
one. A survey of over 3000 women (of whom 469 were black) conducted
by the National Center for Health Statistics (MacMahon, 1973) found a
difference of only 3 months in the mean age at menarche for black women
aged 55 to 79 (.~ = 13.10) compared to those aged 18 to 34 (.~= 12.85).
A difference this small would not be expected to be evident if the data were
examined in broad categories, as they were for our comparisons.
First Intercourse
The two samples contained several pieces of comparable information
concerning the subject's first experience with intercourse-the age at which
it took place, the relationship to the partner, and the degree of enjoyment
(see Table VI). In evaluating the results, however, it is important to be aware
that the two studies differed somewhat in their approach to first coitus. Our
308
Wyatt, Peters, and Guthr|e
0
P~
~
z
z
0
O
0
r~
ù0
0
0
,~
-o
o
0
0
,.,
O
O
N
N
<
~ -~
c~ c~
Sexual Socialization and Sexual Behavior
309
~q
ù.~ " 0
u~
"~
o
ù.~ ,.ù
o
z
z
c~
ù.
"O
8
o
c~
0
ù~
c~
O
-a
ù~
O
o
e~
310
Wyatt, Peters, and Guthrie
A3
O~
co
e~ q~
0
q,)
.
c~ ~t
0
r~
"S
Z
r~
8
;$
8
0
~8
[.,.
8
'a
O
O
~88
<
~.-~
.=. 8
Sexual Socialization and Sexual Behavior
311
sample includes only voluntary experiences, whereas the Kinsey sample includes an unknown number of experiences that results from sexual abuse.
(For further discussion of this point, see Wyatt, in press, and Wyatt et al., 1988).
The two samples differed significantly on age of first intercourse. About
three-fourths of the new subjects had begun sexual activity by the age of
18 as compared to only half the women in the Kinsey sample. This is highly
consistent with other studies that have documented earlier onset of sexual
activity (Chilman, 1978; Hartley, 1975; Kantner and Zelnik, 1972; Sorenson, 1973; Vener and Stewart, 1974; Zelnlk and Kantner, 1977, 1980; Zelnik
et al., 1981). A finding of particular concern is that the proportion of
women who began intercourse by age 15 had increased from 11% (Kinsey)
to 24°70 (Wyatt). Examining the data by age groups (18 to 26, and 27 to 36),
the onset of sexual activity appears to have shifted downward withinthe Kinsey sample but not within our sample (see Table VII).
There was also a significant difference in the relationship with the first
partner. The percentage of women whose first experience occurred with their
fianc6 or husband dropped from 24070among the Kinsey women to 8o70among
our subjects. However, the breakdown by age groups (see Table VII) did
not reveal substantial shifts over time within each sample.
Finally, women in our sample had somewhat more positive feelings
about first coitus, even though their experiences occurred at younger ages
and in the context of less committed relationships. Forty-two percent of our
subjects reported at least some degree of enjoyment, compared with only
14070 of the Kinsey women. It should be noted, however, that first coitus
was not regarded as an enjoyable experience by the majority of women in
both samples.
Table Vll Aspects of First Intercourse by Age Group for Black Women
Kinsey
Wyatt
27-36 ~
%
18-26
%
27-36
%
18-26
%
Age of first intercourse
15 or younger
16-18
19 or older
13
30
57
10
48
42
28
46
26
17
56
26
Relationship to first partner
Boyfriend or acquaintance
Fiance or husband
78
22
75
25
90
10
96
4
~Age groups are arranged so that in reading across each row from left to
right, one encounters women who were born progressively later.
312
Wyatt, Peters, and Guthrie
Homosexual Experiences
Sexual activity with members of the same sex was another aspect of
adolescent sexual behavior that was assessed in both studies. Somewhat more
women in the new sample (9%) than in the Kinsey sample (2%) reported
having had sex between the ages of 13 and 17 with a girt or woman. However,
this difference could not be tested statistically due to the small numbers of
subjects.
Experiences Spanning Adolescence and Adulthood
Several aspects of sexual behavior encompassed the entire period of
sexual activity or the entire postpubertal period, rather than falling specifically into adolescence or adulthood (see Table VIII).
First Homosexual Experience
Both studies gathered information about sexual experiences with other
women. However, whereas Kinsey identified these as homosexual, we
described these sexual behaviors as occurring with the same sex. This was
done in order to avoid labeling exploratory incidents like some of these included here as homosexual, because women who engaged in these experiences
often did not identify themselves as homosexual.
Among those women who had an experience with someone of the same
sex in adolescence or adulthood, our subjects were somewhat more likely
to report their first experience as occurring during adolescence (45% as compared to 33% in the Kinsey sample). Because of the very small number of
women with data for this item, the difference could not be tested statistically. However, this is consistent with what appears to be a general pattern of
increased adolescent sexual activity in the new sample, as indicated in areas
such as earlier onset of intercourse.
We had planned to compare the two samples on other aspects of the
first experience with another woman, such as relationship to the partner and
type of sexual activity. However, Kinsey and his associates collected this information only for women with extensive homosexual experience, so further comparisons were not possible.
Number of Male Partners
The number of sexual partners often serves as an index of Che extent
of women's sexual activity. The two samples differed significantly in terms
Sexual
Socialization
and
Sexual
Behavior
313
~[-.ù
o
~~~.~
~ ~ ~~,~
0
z
ù~
o
o
<
ù~
0
ù~
<
ù« e.~
r~
ù~ ~
ù~
o
ù~
~,~
~
0
ù.ù
a
o
e.~
..-!
o
~~~
<
~e
Wyatt, Peters, and Guthrie
314
o f the total number of male coital partners, suggesting a substantial change
in sexual activity patterns. Only 26% o f the Kinsey women reported six or
more sexual partners as compared to 60% of the new subjects. Because the
distributions were so discrepant, we were forced to use only two categories
in the log-linear analyses to ensure adequate celi sizes. However, more specific
categorizations can be used on a descriptive basis to illustrate further the
extent of the difference between samples. The proportion of women who
had only one male partner dropped from 24% in the Kinsey sample to 3%
in our sample. Similarly, the proportion reporting 11 or more partners doubled, from 14% in the Kinsey sample to 29% in our sample.
Orgasmic Patterns
Comparisons were conducted on two aspects of orgasmic response: age
of first postpubertal orgasm and consistency of orgasm in postpubertal
masturbation. There was a significant difference between the samples on age
of first orgasm. A greater proportion of women in Kinsey's sample (38%)
as compared to ours (18%) reported having their first postpubertal orgasm
before age 15. Because this item did not differentiate the various possible
sources of the first orgasm, we cannot pinpoint the exact reason for this difference. Indeed, since women in the Kinsey sample tended to have sexual intercourse at slightly later ages and fewer women engaged in masturbation, it
is difficult to determine factors that might contribute to Kinsey women's earlier orgasmic behavior. One issue that deserves consideration is the way in
which "orgasm" was defined to women. Kinsey's definition of orgasm might
have led to women who responded to this item falling to have a consensus
on the various biological and emotional responses included in this behavior.
The two samples also differed on consistency of orgasm in postpubertal masturbation. Among women who masturbated, 64% o f the Kinsey sample were always orgasmic, compared to only 28% of our sample. This
difference may have resulted from differences in interview formats. It appears that the criteria used to categorize a woman as "always orgasmic" may
have been more stringent in our study (see Wyatt et al., 1988, for a more
detailed discussion).
Adult Sexual Behavior
The aspects of adult sexual behavior that were examined tended to fall
into two groups (see Table IX). The first group of items referred to preferences for particular sexual behaviors within the context of a given
relationship-specifically, cunnilingus, fellatio, and anal sex. The remain-
Sexual Socialization and Sexual Behavior
315
Z
g
r~
i1
e~
Z
316
Wyatt, Peters, and Guthrie
ing items - engaging in extramarital sex or prostitution- pertained to sexual
behavior outside of a primary relationship. For the preference items, it is
important to note that our questions were asked in the context of the woman's primary sexual relationship, whereas Kinsey's data represent a lifetime
measure of these behaviors.
Sexual Preferences
Both types of oral sex showed significant differences between samples.
Only 18% of Kinsey's subjects had engaged in cunnilingus as compared to
70% of ours. A similarly dramatic increase was found for fellatio, which
increased from 15% to 65°7o. The difference between samples on anal sex
was also statistically significant, though not as substantial as for oral sex.
The proportion of women who engaged in anal sex rose from 9°70 in the
Kinsey sample to 21% in ours. It should be noted that these figures do not
indicate the frequency of the particular behavior, only whether or not the
subject had ever engaged in it. Within the new sample, we found that frequency depends on the type of behavior involved. Women who engaged in
either or both forms of oral sex tended to do so on a regular basis, whereas
involvement in anal sex generally occurred on an experimental or very occasional basis. Nonetheless, these findings suggest two interpretations: (i)
Women have broadened their sexual repertoires to include a variety of behaviors besides vaginal intercourse; or (ii) women have broadened their sexual repertoires in their primary relationships to include behaviors other than
coitus. Hypothetically, long-term relationships are those in which trust is established and where both partners have the opportunity to learn to communicate their sexual needs and hope to explore their preferences for sexual
gratification with each other. Perhaps in the context of primary relationships
black women broadened their sexual repertoires.
The influence of marital status on sexual preferences was also an area
of interest because it had been focused on by Kinsey and his associates (see
Gebhard et al., 1958). We found a consistent pattern for all three preference
behaviors. Among the Kinsey sample, ever-married women were slightly more
likely than never-married women to have engaged in each of the behaviors.
In out sample, by contrast, marital status appeared to have little influence
on preferences. However, the effect was not powerful enough to require the
addition of marital status to the model. Thus, although the marital relationship may have provided opportunities for relationships to develop the qualities mentioned earlier, these findings may also suggest that trust and
communication about preferences for a variety of sexual practices may best
be conveyed in the "relationship in which [women] spent the most t i m e ' -
Sexual Socialization and Sexual Behavior
317
the way in which the primary partner was described to the new sample. The
primary partner was not necessarily the marital partner.
Sexual Behavior Outside of the Primary Relationship
Among ever-married women, there was only a slight difference between
samples on extramarital sexual activity. The proportion of women who had
engaged in extramarital sex was 31%0 in the Kinsey sample as compared to
40% in ours, a difference that was not statistically significant.
Prostitution was a very rare behavior for both samples. None of the
Kinsey women and only 2% of out subjects reported ever engaging in prostitution. Clearly, no statistical test could be performed due to the rarity of
affirmative responses.
Contraception and Pregnancy
Comparisons between the two samples were made for several methods
of contraception (see Table X). These were natural methods (e.g., rhythm,
withdrawal), chemical methods (e.g., jellies, suppositories, foam), the diaphragm, and condoms. Two cautions should be noted in considering these
findings. First, two of the methods most commonly used in the new samplethe birth control pill and the intra-uterine device, or IUD (used by 87% and
32% of women, respectively) were not available to women in the Kinsey sample. Thus, the data presented here give a complete picture of contraceptive
use for Kinsey's sample but not for ours. Second, we were able to compare
only the percentage of women who had ever used a given method. The findings do not address issues pertaining to the duration or consistency with which
each type of contraceptive was used.
The two samples differed significantly in overall contraceptive use.
Although the vast majority of women in both samples had used some method
at least once, there was a detectable increase from 89% among Kinsey's sample
to 98% among ours.
Looking first at specific methods dependent primarily on the behavior
of the male partner, there was a significant difference between samples in
the use of condoms. Of the Kinsey subjects, 80% reported their partners as
having used condoms as compared to 46% of the women in our sample. The
use of withdrawal also appears to have decreased somewhat, from 44% in
Kinsey's sample to 33% in ours. However, in the case of withdrawal, none
of the models tested provided an acceptable fit, possibly due to the presence
of higher-order interactions.
318
Wyatt, Peters, and Guthrie
0
<
o,
~J
Sexual Socialization and Sexual Behavior
319
=
O
o
320
Wyatt, Peters, and Guthrie
Among the methods used by the women themselves, significant differences between samples were evident for both natural and chemical methods.
The proportion of women who had used natural methods increased from
4% in Kinsey's sample to 22% in ours, and the percentage reporting use o f
chemical methods rose from 12% to 44070, respectively. (For the latter
method, although addition o f the "sample" term led to a highly significant
improvement in the fit o f the model, none of the models tested provided
a satisfactory fit.)
Overall, use of the diaphragm increased slightly, from 24% of Kinsey's
women to 32% among our subjects. However, marital status had a marked
influence on the direction o f change. Among never-rnarried women, use of
the diaphragm increased from 8% in Kinsey's sample to 35% in ours, whereas
diaphragm use among ever-married women dropped from 45% to 30%,
respectively. The decrease among ever-married women may represent a shift
to newer methods that are more convenient and effective, such as the pill
and the IUD. Increased diaphragm use among the never-married may reflect
the greater access that these women now have to medical methods of contraception. Overall, marital status appears to have had a significant influence
on diaphragm use among the Kinsey women as compared to almost no influence among the new subjects.
Looking across the various methods, two trends emerge. Reliance on
methods that depend primarily on the partner's b e h a v i o r - namely, condoms
and w i t h d r a w a l - h a s dropped. This is consistent with a general shift away
from these methods in recent years (Westoff and Ryder, 1977). Conversely,
there has been an increase in the use of methods that depend primarily on
the woman's own behavior and that presuppose her having access to at least
some information about contraception. The case of the d i a p h r a g m - the only
method compared that requires a medical p r e s c r i p t i o n - p r o v i ä e s an excellent illustration o f the dramatic changes that have occurred in the availability of contraception to nonmarried women (Pitpel, 1968).
Finally, among women who had been pregnant, we were able to compare the rate of pregnancy due to contraceptive failure. Although the percentage o f women reporting such a pregnancy was slightly higher among the
Kinsey women (29%) than it was among our subjects (20%), the difference
was not statistically significant. It appears that the rate of unwanted pregnancy stemming from contraceptive failure has not changed markedly, despite
the increased availability of more effective methods. These data raise questions as to whether contraceptive failure was more or less a function of
whether the method was being used appropriately.
We had also hoped to compare data on induced abortion across the
two samples, hut this was not possible due to lack of cornparability in the
way the data had been coded.
Sexual Socialization and Sexuai Behavior
321
Child Sexual A b u s e
The Kinsey study was one of the first to provide documentation of sexual encounters between female children and older males, experiences that
are now commonly regarded as sexual abuse. However, in evaluating the comparisons of findings from the two studies, it is important to be aware of differences in the approaches to data collection. In both studies, screening questions
were used to determine whether or not the woman had been sexually abused;
if she had been, a series of follow-up questions were asked to determine specifm aspects of the experience(s). In the Kinsey study, inquiry about sexual abuse
took place in the first half of the interview and included only a single screenlng question about an adult male approaching the subject for sex or having
sex with her. On the other hand, we placed the inquiry towards the end of the
interview, and used a series of eight screening questions, each focusing on
a specific type of abuse. Both studies, however, were similar in that the questions simply described a particular behavior rather than to label it as sexual.
These similarities and differences have important implications for understandmg the differences between findings from the two studies.
Comparing the results of the two studies also required that we first adjust for differences in the definitions used (see Wyatt et al., 1988 for a more
detailed discussion). To approximate a match with the Kinsey s t u d y - which
e×amined only prepubertal experiences with postpubertal m a l e s - o u r data
set was modified to exclude incidents that either occurred after the subject
had reached puberty or involved a perpetrator whose age was close to that
of the victim. Thus, the abuse experiences described here are only those that
corresponded to the Kinsey definition. Table XI presents the items pertaining to sexual abuse that were compared across the two samples.
Before discussing the results, two cautions should be noted. First, the
results should be regarded very tentatively because of the small numbers of
subjects involved. This is reflected in the statistical analyses by the lack of
significant differences between samples, even in cases where the distributions
of responses were quite discrepant. Second, the findings presented here pertain only to prepubertal abuse, because Kinsey and his associates did not
collect data on postpubertal experiences. There may be patterns of difference in adolescent sexual abuse that could not be identified in these comparisons.
P r e v a l e n ce
Prevalence refers to the proportion of women who reported at least
one incident of sexual abuse within the designated period (in this case, prior
322
Wyatt, Peters, an~ Guthrie
@
Z
~
Z
Z
Z
<
<
0
0
<
<
"0
Sexual Socialization and Sexual Behavior
=
O
Z
o,
O
Z
r-:
0
Z
323
,,~
O
Z
8
.o
t~
Wyatt, Peters, and Guthrle
324
to puberty). There was a nonsignificant trend toward higher prevalence in
our sample (39%) as compared to Kinsey's (29%).
This difference of 10 percentage points, although not statistically significant, raises the question of whether abuse has become more prevalent
over time. If this were the case, we would expect to see increasing rates for
successive cohorts of women within each sample. E×amining prevalence by
successive age groups, we find the following:
Kinsey,
Kinsey,
Wyatt,
Wyatt,
27-36:
18-26:
27-36:
18-26:
35%
25%
50%
21%
There is no clearly increasing trend evident in these figures. Nor is there evidence in other studies that the prevalence of abuse has risen over time (see
review by Peters et al., 1986).
If the prevalence of abuse has not truly increased, how can we account
for the small difference between samples? We suspect this difference is due
largely to the number and type of questions used to determine whether abuse
had occurred. As noted above, the Kinsey interview included only a single
screening question, whereas we used a series of eight questions. In reviewing
a large number of contemporary prevalence studies, we found that the use
of multiple specific questions consistently yielded higher prevalence rates than
the use of a single general question, such as Kinsey's (Peters et aL, 1986;
Wyatt and Peters, 1986). Therefore, we believe that methodological factors
play an important role in accounting for the small difference in prevalence
rates between the two samples. It is nevertheless important to keep in mind
that Kinsey did not inquire about postpubertal abuse, in contrast to most
contemporary prevalence studies (including ours) that do include some or
all adolescent experiences. The fact that Kinsey omitted such incidents from
his sample makes it more difficult to evaluate whether the present findings
point to an increase in the prevalenee of abuse.
Another factor which should be considered in connection with prevalence rates is the possibility of regional variation, particularly the question
of whether sexual abuse is more common in California than it is in other
areas. Available evidence on this point is complex and contradictory (see
Peters et aL, 1986, for a more detailed discussion). We compared women
raised in California with women raised in other stares and found that prevalence rates were identical for the two groups. However, another study involving a national telephone survey (Lewis, 1985) found a higher prevalence
rate for the Pacific region as compared to the rest of the county. It is important to keep in mind that the question of regional differences involved three
Sexual Socialization and Sexual Behavior
325
distinct issues: where victims are currently residing, where they were raised,
and where the abuse took place. Until more definitive evidence becomes available, the possiblity of regional variation in the prevalence or characteristics
of sexual abuse remains an open question.
Characteristics o f A b u s e lncidents
Among the women who had been abused, there was only one statistically significant difference between the samples. The proportion of women
who had two or more separate abusive relationships with different perpetrators was much higher in our sample (40070) than in Kinsey's (907o). Several
other aspects of abuse experiences showed interesting trends that did not
emerge as statistically significant, possibly because of the small numbers of
subjects included. Although the majority of women in both samples reported that the first perpetrator was 25 or older, more of out subjects than
Kinsey's reported him as being 24 or younger (40 rs. 18070, respectively). Indeed, there is evidence to suggest that the identification of adolescents as
perpetrators of sexual abuse has previously been ignored and neglected; that
due to greater sophistication in reporting procedures, the numbers of reported
adolescent sex offenders tend to be increasing (Knopp, 1982). Our subjects
were more likely to report being abused by someone known to them, either
an acquaintance or a relative, than were Kinsey women (64 rs. 4407o, respectively), and the former were more likely than the latter to report repeated
abuse by the same perpetrator (2407o as compared to 507o of Kinsey's).
Although this finding is a nonsignificant trend, there is research to suggest
that sexual abuse is perpetrated increasingly by persons known to the victim
(Finkelhor, 1979; Russell, 1983; Wyatt, 1985). Due to the nature of the relationship of the perpetrator to the victim, sexual abuse may tend to have been
repeated; relatives and acquaintances typically have more frequent opportunities for contact with the victim than do strangers. Finally, the proportion
of women who had a completely negative reaction to the experience was higher
among the new sample (88 vs 5807o among the Kinsey sample). The slight
shift in perpetrators known to the victim and repeated abuse incidents with
that perpetrator may have accounted for the slightly more negative reaction
among the new sample.
No differences between samples were evident on the remaining two
aspects of abuse experiences- age at which the abuse occurred and type of
sexual activity. Slightly more than half the women in both samples were first
abused at age 10 or older. They also reported that their first experience did
not involve physical contact with the perpetrator.
326
Wyatt, Peters, and Gmhrie
DISCUSSION
This study compared two samples of b-lack women, 18 to 36 years of age,
one collected mainly in the Northeast and Midwest and the other ha Los Angeles
County some 33 years later. The comparisons of these data provJde an empirical examination of the changes in women's sexual behavior. Although research
in the 1960s and 1970s idenüfied changes in childbearing and contraceptive practices, few studies had the scope of Kinsey's. Our study is similar to IOnsey's
in that a range of sexual socialization and behavioral patterns are exarnined
from childhood to adulthood. Consequently, we can compare changes in a
broader fange of sexual behaviors. At first glance, the comparison of a small
and highly stratified sample of black women with one that ranged in age, education, marital status, and the presence of children, and that was representative of women with comparable demographic characteristics in Los Angeles
County might be considered a capricious disregard for some of the tenets of
science (Meyers, 1982). The likelihood is great that many of the life experiences
of these women differ. However, much of sex research includes and compares
black and white females who are high school students (Reiss, 1967, 1971;
Roebuck and McGee, 1977), college students (Christopher and Cate, 1984;
Herold and Goodwin, 1981; Sack et al., 1984; Weis, 1983), or participants in
clinical samples (Castadot, 1975). Very little research has been conducted with
community samples (Hunt, 1974; Reiss, 1967, 1971). Despite the volume of
research reported, the findings are often limited in their generalizability due to
age or regional or sampling restrictions. On the one hand, when black samples
are included, subjects often tend to be indigent (Dixon, 1980; Gebhard et aL,
1958; Hammond and Ladner, 1969; Johnson, 1971; Rainwater, 1966; Roebuck
and McGee, 1977; Staples, 1967). On the other hand, information about demographic characteristics and the comparability of black and white samples is orten
unavailable in studies that report across ethnic-group comparisons (Belcastro,
1985; Reiss, 1967, 1971; Weiss, 1983). Studies that include more sophisticated statistical techniques (e.g., Sack et al., 1984; Strahle, 1983, Newcomb et
al., 1986; Weiss, 1983) or those that introduce theoretical perspectives based
upon previous research (D'Augelli and D'Augelli, 1977; Davidson and Leslie, 1977; Kelley, 1978; Mirande, 1968; Rook and Hammem 1977) of ten exclude ethnicity as a variable. The limited information available illustrates
the need to understand the strengths and limitations of existing research and
to build upon that information.
These studies were similar in that they exarnined epidemiological issues
in women's sexual socialization and sexual behavior; they collected retrospective as weil as current information; subjects were interviewed face-to-face
with a structured interview; and both samples were at least college educated.
However, Kinsey’s approach was to use puberty to divide data developmentally, and to label experiences with males and females as indicafive of sexual
preference. Although interviews were conducted within the framework of
Sexual Socialization and Sexual Behavior
327
a conversation in Kinsey’s study, white male interviewers were used to collect data. One may well wonder about the level of comfort and disclosure
of Kinsey’s black female sample under these circumstances (personal communication with Joanna Blackhawk, Los Angeles, California, 1980). Our
approach adhered more strictly to data collection within a developmental
framework, and relationships with the same sex were not considered indicative of sexual preference. Interviewers were clinically trained black females
who developed rapport with subjects prior to the interview and monitored
the subject’s reaction to affect-laden topics in order to adjust the pace of
the interview.
Kinsey obtained only a limited amount of information about women’s
feelings in relation to behaviors, whereas we examined the age of onset, frequency, and circumstances of each behavior, and obtained affective ratings
of the first time that the behavior was engaged in. Although a great deal
of time was expended in recoding items for purposes of comparability, these
factors influence the contextual framework of the questions used in this study
and, consequently, some of the similarities and differences found therein.
These findings are not representative of black women. One of the main
purposes of comparison of these inherently different samples is to place some
of Kinsey’s findings in perspective, acknowledging the sampling and methodological limitations described earlier, and to offer additional information on
a more representative sample of black women.
Although differences between the samples in age, marital status, and
the presence of children were controlled for where possible by use of loglinear analyses, there are, however, other differences between the samples
that were not controlled for, such as region or religion, that were of interest
in this study. It is important to assess differences in other developmental,
personal, and family variables that could also influence the findings. The
families of these samples of women were comparable across demographics.
Although some shifts in religious practices in America have been noted
(Jacquet, 1986), fairly consistent religious affiliation and attendance was
maintained over one-quarter of a century for these black families. However,
changes in the number of women in the new sample who lived with two parents proved to be a measure of the changing patterns in family constellations in the United States (U.S. Bureau of Census, 1980).
Sexual socialization appears to have more widespread resources, including the schools, the media, and voluntary and abusive sexual experiences that
women encounter. Although parents of our sample were included among
the resources of general sex-related information, they appeared no more willing to allow the observation of their lovemaking by their children than when
Kinsey’s subjects were growing up.
Kinsey’s sample reported more exploratory childhood sexual behaviors,
most likely due to the manner in which the questions asked elicited this information. However, less than half of Kinsey’s and only one-fifth of our
328
Wyatt, Peters, and Guthrie
sample engaged in such behaviors. Childhood masturbation appears to be
on the increase in spite of cultural and religious proscriptions (Staples, 1973).
Indeed, in the new sample, one in four women recalled masturbating at some
time at or before age 12. Although this does not appear to be a c o m m o n
sexual behavior among black women, some who engaged in it may have done
so for a variety o f reasons, including pleasure seeking or as a result of emotional and psychological problems. There is need for additional research,
not only to identify m a s t u r b a t o r y patterns in women who have not traditionally been active participants but also to understand better those factors
that influence the increase in masturbatory behaviors.
When the samples were compared, the proportion of women who experienced first coitus at or before age 18 rose from one in two to three in
four. Additionally, those becoming sexually active at or before 15 rose from
one in ten to one in four. However, when the two generations within that
sample were compared, the women in our sample do not appear to be experiencing first coitus earlier. The consistent pattern of sexual intercourse
in adolescence for black women is similar to other patterns, such as early
childbearing out of marriage (Washington, 1981), and may suggest that the
factors influencing early sexual activity in the past still appear to be operating today (Wyatt, in press).
In contrast to early sexual activity, twice as m a n y of Kinsey’s sample
reported reaching their first orgasm at or before age 15. Factors that may
account for this difference include some variation between the studies in what
women understood to be “orgasm,” or later orgasmic response in the new
sample. Perhaps, if the latter group experienced early sexual intercourse with
someone other than a husband or fiance, orgasmic response in first intercourse may have been delayed until the length or quality of relationships facilitated such a response. More research in this area is also needed.
Consistent with the increased span of time in which sexual activities
could occur, two in three of our sample reported si× or more partners as
compared to one in four of Kinsey’s. Additionally, the preference for other
sexual behaviors also increased in the new sample. However, while both types
of oral sex appear to have been added to black women’s sexual repertoire,
anal sex was not added for the vast majority of women. In contrast to changing sexual preferences of black women, the prevalence of sex outside of marriage remained at similar levels for both samples.
Although the use of more convenient forms of contraception increased
for black women in our sample, it should be noted that almost all of the
women had used some form of contraception at least once. Marital status
appears to have less influence upon the use of medical methods of cõmraception, perhaps beeause o f rnore availability and fewer proscriptions against
use by unmarried women. Indeed, as the proportion of nonmarried women
Sexual Socialization and Sexual Behavior
329
in the black population increases, it is encouraging to observe that marital
status does not continue to dictate the type of contraceptive available.
Finally, although there was no significant increase in the prevalence
of child sexual abuse by adult perpetrators, women in our sample appear
to have been more at risk for repeated incidences. One of the contributing
factors may have been that these women tended to live in more urban areas
than Kinsey’s sample, and urbanization has been identified as a risk factor
for child sexual abuse.
One in four black women in Kinsey’s sample experienced at least one
incidence of prepubertal sexual abuse with an adult perpetrator who tended
to be unrelated to them. It is ironic that more attention was paid to the prevalence of first coitus rather than those experiences that were abusive at or before age 12. Both early coitus and child sexual abuse could have had
potentially negative outcomes for these women and both merited serious consideration by researchers and clinicians.
It is difficult to interpret the difference found between samples on the
age of perpetrators because of the scant research that has been available in
this area. Recent research indicates that the methodologies developed to obtain current data are contributing to the higher prevalence of adolescent sex
offenders (Knopp, 1982; Neilson, 1984). It is likely that earlier discoveries
of sexual abuse by an adolescent may not have resulted in a report that came
to the attention of researchers (Becker and Abel, 1985). (Personal telephone
communication with Judith Becker, Columbia University, New York City,
New York, August, 1986).
Although these comparisons do attempt to offer insight into the sexual
experience for black women, more research with representative samples is needed in order to present a consistent and balanced view. Although many aspects
of these women’s experiences are not changing as rapidly as for other ethnic
groups, the sexual patterns noted do rend to be a function of the assimilation of black women into the American mainstream. Attention needs to be
focused on those patterns that appear to be changing more radically so that
information can be generated to understand better the factors that influence
the sexual socialization and behavior of American black women.
ACKNOWLEDGMENTS
The authors acknowledge the support and assistance of June Reinisch,
Ph.D., Director, Paul Gebhard, Ph.D., Former Director, and Tom Albright,
programmer, at the Kinsey Institute. We also thank Sondra Perdue, Dr. P.H.,
Earl Leonard, and Dolores Adams for statistical consultation and programming at the Neuropsychiatric Institute, UCLA.
330
Wyatt, Peters, and Guthrie
REFERENCES
Bass, B. A., Wyatt, G. E., and Powell, G. P. (1982). TheAfro-American Family: Assessment,
Treatment and Research Issues, Grune and Stratton, New York.
Becker, J. V., and Abel, G. G. (1985). Methodological and ethical issues in evaluating and treating
adolescent sexual offenders. In Otey, E. M., and Ryan, G. D. (eds.), Adolescent Sex
Offenders: Issues in Research and Treatment (DHHS Publication No. ADM 85-1396),
U.S. Government Printing Office, Washington, DC, pp. 109-129.
Belcastro, P. A. (1985). Sexual behavior differences between Black and White students. J. Sex
Res. 21:56-67
Benedetti, J. K., and Brown, M. B. (1978). Strategies for the selection of log linear models.
Biometrics, 34: 680-686.
Bishop, Y. M. M., Fienberg, S. E., and Holland, P. W. (1975)o DiscreteMultivariateAnalysis,
MIT Press, Cambridge, MA.
Castadot, R. G. (1975). Need for, difficulties and experience with sexual behavior quesfionnaires among teenagers. Maryland State Med. J. 24: 40-44.
Chilman, C. S. (1978). Adolescent Sexuality in a Changing American Society: Social and Psychological Perspectives, Public Health Service, DHEW, U.S. Government Printing Office,
Washington, D.C.
Christopher, F. S., and Cate, R. M. (1984). Factors involved in premarital sexual decision-making.
J. Sex Res. 20: 363-376.
Clower, V. L. (1975). Significance of masturbation in female sexual development and function. In Marcus, I. M., and Francis, J. J. (eds.), Masturbation From Infancy to Senescence, Internationnal Universities Press, New York, pp. 107-143.
Cochran, W. G., Mosteller, F., and Tukey, J. (1954). Statistical Problems of the Kinsey Report
on Sexual Behavior in the Human Male, American Statistical Association, Washington,
DC.
D’Augelli, J. E., and D’Augelli, A. R. (1977). Moral reasoning and premarital sexual behavior:
Toward reasoning about relationship. J. Soc. Issues 33: 46-66.
Davis, K. (1972). The American family in relation to demographic change. In Westoff, C. F.,
and Parke, R. (eds.), Demographic and SocialAspects ofPopulation Growth, Vol. 1,
U.S. Government Printing Office, Washington, DC, pp. 235-265.
Davidson, J. K., and Leslie, G. R. (1977). Premarital sexual intercourse: An application of
axiomatic theory construction. J. Marr. Fam. 39: 15-25.
Dixon, R. D. (1980). The absence of birtb order correlatiõns among unwed and married black
first-coneeptors. J. Sex Res. 16: 238-244.
Finkelhor, D. (1979). Sexually Victimized Children, Free Press, New York.
Friedrich, W., Uriquiza, A., and Beilke, R. (1986). Behavior problems in sexually abused children. J. Pediat. PsychoL 11: 47-57.
Gebhard, P. H., and Johnson, A. B. (1979). The Kinsey Data: Marginal Tabulations of the
1938-1963 Interviews Conducted by the Institute for Sex Research, W. B. Saunders,
Philadelphia.
Gebhard, P., Pomeroy, W., Martin, C., and Christenson, C. (1958). Pregnancy, Birth, and
Abortion, Harper & Row, New York.
Hammond, B., and Ladner, J. (1969). Socialization into sexual behavior in a Negro slum ghetto. In Broderick, C., and Bernard, J. (eds.), The lndividual, Sex and Society, John Hopkins Press, Baltimore.
Hartley, S. (1975). lllegitimacy, University of California Press, Berkeley.
Herold, E. S., and Goodwin, M. S. (1981). Adamant virgins, potential nonvirgins and nonvirgins. J. SexRes. 17: 97-113.
Hunt, M. (1974). Sexual Behavior in the 1970% Playboy Press, Chicago.
Jacquet, C. H., Jr. (ed.) (1986). Yearbook of American and Christian Churches, Abingdon,
Nashville.
Johnson, L. (1971). Sexual behavior of southern Blacks. In Staples, R. (ed.), The Black Family: Essays and Studies, Wadsworth, Belmont, CA.
Sexual Socialization and Sexual Behavior
331
Kaats, G. R., and Davis, K. E. (1970). Dynamics of sexual behavior of college students. J.
Marr. Fam. 32: 390-399.
Kantner, J. F., and Zelnik, M. (1972). Sexual experience of young unmarried women in the
United States. Fam. Plann. Perspect. 4: 9-18.
Kelley, J. (1978). Sexual permissiveness: Evidence for a theory. J. Marr. Fam. 40: 455-468.
Kinsey, A., Pomeroy, W., and Martin, C. (1948). Sexual Behavior in the Human Male, W.
B. Saunders, Philadelphia.
Kinsey, A., Pomeroy, W., Martin, C., and Gebhard, P. (1953). Sexual Behavior in the Human
Female, W. B. Saunders, Philadelphia.
Knopp, F. H. (1982). Remedial Intervention in Adolescent Sex Offenses: Nine Program Descriptions, Safer Society Press, Syracuse, NY.
Lewis, I. A. (1985). Los Angeles Times Roll #98. Unpublished raw data.
Libby, R. W., Gray, L., and White, M. (1978). A test and reformulation of reference group
and role correlates of premarital sexual permissiveness theory. J. Marr. Fam. 40: 79-92.
MacMahon, B. (1973). Age at Menarche. United Stares (DHEW Publication No. HRA 74-1615),
National Center for Heahh Statisties, Rockville, MD.
Mahoney, E. R. (1980). Religiosity and sexual behavior among heterosexual college students.
J. Sex Res. 16: 97-113.
Mendes, H. A. (1982). The role of religion in psychotherapy with Afro-Americans. In Bass,
B. A., Wyatt, G. E., and Powell, G. J. (eds.), TheAfro-American Family: Assessment,
Treatment, and Research Issues, Grune and Stratton, New York.
Meyers, H. F. (1982). Researeh on the Afro-American family: A critical review. In Bass, B.
A., Wyatt, G. E., and Powell, G. J. (eds.), The Afro-American Family: Assessment,
Treatment, and Research Issues, Grune and Stratton, New York.
Mirande, A. M. (1968). Reference group therapy and adolescent sexual behavior. Jr. Marr. Fam.
30: 572-577.
Neilson, G. (1984). Therapy and Treatment o f Adolescent Sex Offenders, Paper presented at
the Child Weil Fair Workshop, Salem, OR, April 4.
Newcomb, M. D., Huba, G. J., and Bentler, P. M. (1986). Determinants of sexual and dating
behaviors among adolescents. J. Pers. Soc. Psychol. 50: 428-438.
Peters, S. D., Wyatt, G. E., and Finkelhor, D. (1986). Prevalence. In Finkelhor, D. (ed.), A
Sourcebook on Child Sexual Abuse, Sage, Beverly Hills, CA.
Pilpel, H. F. (1968). The law, social welfare, and family planning. In Haseikorn, F. (eds.), Family Planning: The Role o f Social Work, Garden City, NY.
Rainwater, L. (1966). Some aspects of lower class sexual behavior. J. Soc. Issues 22: 96-108.
Reiss, I. L. (1967), The Socia• Context o f Premarital Sexual Permissiveness, Holt, Rinehart
and Winston, New York.
Reiss, I. L. (1971). Premarital sex codes: The old and the new. In Grummon, D., and Barclay,
A. (eds.), Sexuality: A Search f o r Perspective, Reinhold, New York.
Reiss, I. L., and Miller, B. C. (1979). Heterosexual permissiveness: A theoreticaI analysis. In
Burr, W. R., Hill, R., Nye, F. I., and Reiss, I. L. (eds.), Contemporary TheoriesAbout
the Family, Vol. 1, Free Press, New York.
Rodgers-Rose, L. F. (1980). Some demographic characterics of the black woman: 1940 to 1975.
In Rodgers-Rose, L. F. (eds.), The Black Woman, Sage, Beverly Hills, CA.
Roebuck, J., and McGee, M. G. (1977). Attitudes toward premarital sex and sexual behavior
among black high school girls. J. Sex Res. 13: 104-114.
Rook, K. S., and Hammen, C. L. (1977). A cognitive perspective on the experience of sexual
arousal. J. Soc. Issues 33: 7-29.
Russell, D. (1983). The incidence and prevalence of intrafamilial and extrafamilial sexual abuse
of female ehildren. Child Abuse Neglect. 7: 133-146.
Sack, A. R., Keller, J. F., and Hinkle, D. E. (1984). Premarital sexual intercourse: A test of
the effects of peer group, religiosity, and sexual guilt. J. Sex Res. 20: 168-185.
Shah, F., and Zeinik, M. (1981). Parent and peer influence on sexual behavior, contraceptive
use, and pregnancy experience of young women. J. Marr. Fam. 43: 339-348.
Sorenson, R. B. (1973). Adolescent Sexuality in Contemporary America: Personal Values and
Sexual Behavior, Ages Thirteen to Nineteen, World, New York.
332
Wyatt, Peters, and Guthrie
Staples, R. E. (1967). Sex Behavior of low-income Negroes. Sexology 34: 152-155.
Staples, R. E. (1973). The Black Woman in America, Nelson-Hall, Chicago.
Strahle, W. M. (1983). A model of premarital coitus and contraceptive behavior among female
adolescents. Arch. Sex. Behav. 12: 67-95.
U.S. Bureau of the Census. (1980). America’s Black Population: 1970 to 1982. A Statistical
View (Special Publication. Series plp/Pop-83), U.S. Government Printing Office,
Washington, DC.
Vener, A. M., and Stewart, C. S. (1974). Adolescent sexual behavior in rniddle America revisited: 1970-1973. J. Marr. Fam. 36: 728-735.
Washington, A. C. (1981). A cultural and historical perspective on pregnancy-related activity
among U.S. teenagers. J. Black PsychoL 9: 1-28.
Weis, D. L. (1983). Affective reactions of women to their initial experience of coitus. J. Sex
Res. 19: 209-237.
Westoff, C. F., and Ryder, N. B. (1977). The Contraceptive Revolution, Princeton University
Press, Princeton, NJ.
Wyatt, G. E. (1982). The sexual experienee of Afro-American women: A middle income sample. In Kirkpatrick, M. (ed.), Women’s Sexual Experience: Explorations o f the Dark Continent, Plenum Press, New York.
Wyatt, G. E. (t985). The sexual abuse of Afro-American and white women in childhood. Chi!d
Abuse Neglect. 9: 507-519.
Wyatt, G. E. (in press). Reexamining factors affecting Afro-American and white women’s first
coitus. Arch. Sex. Behav.
Wyatt, G. E., and Peters, S. D. (1986). Methodological considerations in research on the prevalence of child sexual abuse. Child Abuse Neglect. 10″ 241-251.
Wyatt, G. E., Peters, S. D., and Guthrie, D. (1988). Kinsey revisited, Part I: Comparisons of
the sexual socialization and sexual behavior of white women over 33 years. Arch. Sex.
Behav. 201-239.
Zelnik, M., and Kantner, J. F. (1977). Sexual contraceptive experience of young unmarried
women in the United States, 1976 and 1971. Farn. Plann. Perspect. 9: 55-71.
Zelnik, M., and Kantner, J. F. (1980). Sexual activity, contraceptive use and pregnancy among
metropolitan-area teenagers. 1971-1979. Fam. Plann. Perspect. 12: 230-23%
Zelnik, M., Kantner, J. F., and Ford, K. (1981). Sex and Pregnancy in Adolescence, Sage, Beverly
Hills, CA.
Psychology oj Women QwrterZy, 18 (1994), 203-219. Printed in the United States of America.
SEXUAL SOCIALIZATION
DIFFERENCES BY GENDER, GREEK
MEMBERSHIP, ETHNICITY, AND
RELIGIOUS BACKGROUND
llsa L. Lottes
University of Maryland, Baltimore County
Peter J. Kuriloff
University of Pennsylvania
Socialization theories have included parents and peers as important
determinants of the initial sexual standards and sexual behavior of teenagers and young adults. The purpose of the research reported here was
to examine how parental and peer sexual socialization influences are
related to gender, ethnicity, religious background, and college membership in a fraternity or sorority. A sample that included a majority of
Caucasian university students and about 13% Asian and 7% Black students completed questionnaires both as entering first-year students and
as seniors. Results indicated that compared to women, men continue to
experience a more permissive sexual socialization from both parents
and peers. Greek membership was associated with a more permissive
socialization from peers but not parents. Asian students reported a more
restrictive socialization than Blacks or Caucasians. Findings are discussed with respect to concerns of social scientists regarding the influence of fraternities and differential gender socialization.
Research for this paper was funded in part by grants from the University of Pennsylvania
Research Foundation and the National Institute of Justice (Grant NIJ89-IJ-CX-0048, Assessment and Evaluation of SMART and Related Programs, Robert Boruch, PI). Opinions expressed in it are those of the authors and do not necessarily represent the official views of the
University of Pennsylvania or NIJ. We thank Samuel C. Haffer for his help in data processing
associated with this research and two anonymous reviewers for helpful comments on an
earlier version of this manuscript.
Address correspondence and reprint requests to: Ilsa L. Lottes, Ph.D., University of Maryland Baltimore County, Department of Sociology and Anthropology, 5401 Wilkens Ave.,
Baltimore, MD 21228.
Published by Cambridge University Press 0361-6843194 $5.00
+ .00
203
204
LOlTES AND KURILOFF
DeLamater and MacCorquodale (1979) and Reiss (1967) have proposed
models of the social influences on heterosexual sexual expression. In these
models, the sexual standards of one’s parents are an important determinant of one’s initial sexual standards and behavior. A second socializing
agent in these models is one’s peers. Reiss (1967) has argued that as young
adults become more involved in dating and courtship, they are increasingly influenced by their friends and peer group. Supporting Reiss’s view
are several studies cited by DeLamater and MacCorquodale (1979) that
reported a correspondence between measures of peer and individual standards. Parental values are influenced, at least in part, by the family’s
ethnic and religious background. The purpose of this research was to
examine how parental and peer sexual socialization influences are related
to gender, ethnicity, religious background, and membership in a fraternity or sorority. The present research draws on data from a broader longitudinal study that is investigating the relationship among background
variables, residential and social affiliations, and the attitudes, values, and
sexual experiences of students at a highly selective eastern university.
Social scientists (e.g., Allgeier & McCormick, 1983; Reiss, 1967, 1986,
1990; Weis & Borges, 1975) have emphasized that traditional socialization
has encouraged opposite goals and roles for women and men in dating
and courtship. For example, men have been socialized to enjoy their sexuality and to take active roles in initiating sexual activity with women. In
contrast, women have been socialized to take passive or reactive roles in
sexual interactions and to engage in sexual activity only in relationships
characterized by romance, affection, love, and commitment. Until the
last decade, surveys of sexual behavior generally supported this differential
sexual socialization of women and men.
A substantial amount of recent research, however, has described dramatic changes in the sexual behavior of women since the 1960s. Ehrenreich, Hess, and Jacobs (1987) wrote:
Men changed their sexual behavior very little in the decades from the fifties
to the eighties. . . , Women, however, have gone from a pattern of virginity
before marriage and monogamy thereafter to a pattern that much more
resembles men’s . . . it is not only that women came to have more sex and
with a greater variety of partners but they were having it on their own
terms, and enjoying it as enthusiastically as men are said to. (p. 2)
They further emphasized that, although economic and political parity
with men remain elusive, sexual equality has become a definite possibility.
Reiss (1986) also concluded that in the last 20 years American women
have made substantial gains in sexual equality. Longitudinal studies (e.g.,
Bell & Coughey, 1980; Bishop & Lipsitz, 1991; Sherwin & Corbett, 1985)
and the results of national surveys reported by Reiss (1990) document the
dramatic increases in women’s premarital coital experiences, shifts toward
Sexual Socialization
205
more liberal sexual norms for women, a decrease in the age of first intercourse, and an increase in the number of their sex partners as well as their
frequency of intercourse.
Given the trend toward increased sexual experience of women, we
wanted to examine parental and peer sexual socialization influences on
women and men to determine if such influences were now similar or if
traditional differential sexual socialization still applied. Studies (e.g., Lottes, 1985; Smith, 1990; Tallichet & Willits, 1986; Thornton, Alwin, &
Camburn, 1983) have shown that egalitarian attitudes about gender roles
and permissive views about premarital sexuality have gained increasing
acceptance over the last two decades. Nevertheless, Reiss (1990) notes that
despite the trend toward overall gender equality, remnants of traditional
socialization and the double standard of sexuality still prevail. Thus, we
expected at least some tendency for men to report more permissive sexual
socialization influences from parents and peers than did women.
The term permissive here implies acceptance of nonmarital sexual interactions. A highly permissive view would be one that is accepting-for
both oneself and others-of single young adults in a variety of relationships (from casual acquaintances to partners in a stable relationship) for
varying degrees of affection between the participants (from not much
affection to love) engaging in a variety of sexual activities (from kissing to
intercourse). Of course, the degree of permissiveness varies according to
acceptance of specific sexual activities and the level of involvement of
and affection between participants in the activities. Therefore, we define
permissive influences as influences that would encourage sexual involvement in a wide range of circumstances, including casual and short-term
relationships. Nonpermissive influences would be those that discourage
casual sexual encounters and promote either abstinence for single adults
or sex for individuals only in relationships characterized by high degrees
of affection and commitment (see Reiss, 1967, and Reiss & Miller, 1979,
for a discussion of premarital sexual permissiveness).
Because the influence and value of fraternities and sororities (the Greek
system) has become a controversial issue on college campuses, we also
chose to investigate sexual socialization differences between Greeks and
non-Greeks. In the 1980s, over 60 colleges and universities undertook investigations to determine if the Greek system should be abolished on their
campuses (Grant, 1990). Much of this criticism involved a concern that
fraternities promote a social context that facilitates the sexual coercion of
women (Copenhaver & Grauerholz, 1991; Sanday, 1990). For example,
Martin and Hummer (1989) concluded that fraternities encourage their
members to value a narrow definition of masculinity that includes dominance, winning at all costs, sexual conquest and control of women, and
the use of exploitative and forceful strategies with women in order to
engage in sexual activities. Supporting these views, Boeringer, Shehan,
and Akers (1991), Garrett-Gooding and Senter (1987), and Kanin (1967)
206
LOTTES A N D
KURILOFF
found that membership in a fraternity was associated with a higher level
of sexual coercion than nonmembership. Thus, we expected fraternity
membership to be associated with a more permissive peer socialization
than nonfraternity membership. Previous research has focused on the possible negative peer socialization pressures on fraternity members. Neither
the peer sexual socialization of sorority and nonsorority members nor the
parental sexual socialization of Greek and non-Greek students have been
the focus of previous research. Therefore, no predictions were made regarding the relationships between peer socialization and sorority membership or parental sexual socialization and Greek membership.
Although religion has been cited as an important influence on sexual
standards and behavior, studies have found contradictory and inconclusive results regarding the effects of religious background. Ehrmann (1959)
and Bayer (1977) found tendencies for Jewish students to report higher
rates of sexual experience and more permissive views on premarital sex
than Catholics and Protestants. Similarly, Reiss (1990) concluded that, on
average, for many issues related to sexuality and gender, Jews (excluding
the Orthodox) tend to be most liberal and egalitarian, Protestants the
least, and Catholics somewhere in between. On the other hand, DeLamater and MacCorquodale’s (1979) study of mostly White middle-class
young adults did not find substantial differences in sexuality as a function
of religious affiliation. An early study of Reiss (1967) also found no major
differences in sexually permissive attitudes of Protestants, Catholics, and
Jews. Recently, Reiss (1990) has emphasized that those with higher education tend to belong to liberal rather than conservative congregations. Because the respondents in this study’s sample included mainly students from
highly educated families, the tendency found in some studies for Jews to
hold more permissive sexual norms may be attenuated. Thus, no predictions were made about the sexual socialization influences of parents and
peers by…