University of Southern California Dissertations and Theses

Attachment as a predictor of self -monitoring and sexual behavior

(USC Thesis Other)

Attachment as a predictor of self -monitoring and sexual behavior

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. U M I films the
text directly from the original or copy submitted. Thus, some thesis and
dissertation copies are in typewriter face, while others may be from any type of
computer printer.
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, cotored or poor quality illustrations and
photographs, print bieedth rough, substandard margins, and improper alignment
can adversely affect reproduction.
In the unlikely event that the author did not send U M I a complete manuscript and
there are missing pages, these w ill be noted. Also, if unauthorized copyright
material had to be removed, a note w ill indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning
the original, beginning at the upper left-hand comer and continuing from left to
right in equal sections with small overlaps.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6" x 9” black and white photographic
prints are available for any photographs or illustrations appearing in this copy for
an additional charge. Contact U M I directly to order.
Bell & Howell Information and Learning
300 North Zeeb Road, Ann Arbor, M l 48106-1346 USA
®
UMI
800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ATTACHMENT AS A PREDICTOR OF SELF-MONITORING AND SEXUAL BEHAVIOR
By
Susan E. Dieterich
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment o f the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
Education (Counseling Psychology)
August 1999
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number 9955069
_ _ ®
UMI
UMI Microform9955069
Copyright 2000 by Bell & Howell Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
Bell & Howell Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
TH E GRADUATE SCHOOL
UNIVERSITY PARE
LOS ANGELES. CALIFORNIA 90007
This dissertation, written by
Susan E. Dieterich
under the direction of hsr. Dissertation
Committee, and approved by all its members,
has been presented to and accepted by The
Graduate School, in partial fulfillment of re
quirements for the degree of
DOCTOR OF PHILOSOPHY
Dean of Graduate Studies
D ate Junfi..3*..LSl9.9.
(ISSERTATION COMMITTEE
Chairperson
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ii
TABLE OF CONTENTS
CHAPTER PAGE
1. Conceptual and Empirical Foundations o f the Study....................................1
Introduction ........................................................................................... 1
Review o f the Literature: Self-Monitoring and Related Concepts...... 6
Interactions o f High and Low Self-M onitors..................................7
Research on Self-Monitoring and Sexual Behavior/Dating 12
Models to Explain the Development o f Self-Monitoring Style. 14
Evolutionary Psychology as an Explanation for Self-
Monitoring and Sexual Behavior................................................... 14
Attachment Theory Proposed as an Alternative Hypothesis for
Development o f Self-Monitoring Behavior................................... 17
Infhnt Attachment....................................................................17
Adult Attachments.................................................................. 20
Attachment Style in Adulthood............................................. 21
Attachment and Adult Romantic Relationships................... 24
Other Attachment Research....................................................29
Reactance.........................................................................................30
Implications for this Study.............................................................32
Research Questions.........................................................................34
2. Method..™.................................................................................................... 36
Introduction .............................................................................36
Participants............................................................................................36
Materials and Procedure....................................................................... 36
Self-Monitoring Scale.................................................................... 36
Collins and Read (1990) Attachment Scale.................................. 37
Dating Confidence..........................................................................38
Reactance.........................................................................................38
Procedures.......................................................................................38
3. Results...........................................................................................................40
Introduction...........................................................................................40
Descriptive Statistics.............................................................................40
Research Questions...............................................................................45
Research Questions: Entire Group................................................ 45
Research Questions: Analysis for Men; for Women.................... 46
Additional Analyses..............................................................................48
Research Questions: Analyses for African American,
Euro-American, and Latino Participants.......................................48
Adult Attachment as Predictors o f Sexual Behavior and
Dating Confidence..........................................................................50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iii
4. Discussion.......................................................................................................52
Introduction............................................................................................ 52
The Results in the Context o f Existing Literature............................... 52
Attachment and Self-Monitoring, and Sexual Behavior/
Dating Confidence...........................................................................55
Reactance and Self-M onitoring......................................................59
Ethnicity........................................................................................... 61
Lim itations............................................................................................. 62
Suggestions for Future Research.......................................................... 63
REFERENCES........................................................................................................... 65
APPENDICES
A. Revised Self-Monitoring Scale.......................................................67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iv
TABLES AND FIGURES
TABLE
1. Descriptive Statistics........................................................................41
2. Correlation Matrix: Fem ale.............................................................42
3. Correlation M atrix: Male................................................................. 42
4. Summary o f Statistical Findings: Variables by
Self-Monitoring Groups____________________________________44
5. Summary o f Gender-Specific Relationships Between
Self-Monitoring and Dependent V ariables......................................... 47
6. Summary o f Ethnic-Specific Relationships Between
Self-Monitoring and Dependent V ariables......................................... 50
FIGURE
1. Proposed Relationship Among Variables......................................... 6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER ONE
Conceptual and Empirical Foundations o f the Study
Individual differences hi human, behavior have been an enduring research focus o f
psychologists (see, e.g., Dawis, 1992). M uch o f this research has been concerned with (a)
the extent to which people possess particular attributes and (b) the predictors and
correlates o f those attributes. The attributes psychologists have studied generally foil into
the two broad categories o f ability (intelligence, aptitude) and personality. The focus o f
this study was on the personality construct o f self-monitoring (SM; Snyder, 1974).
Specifically, its purpose was to investigate the possibility that certain attachment styles
predict self-monitoring behavior and, in turn, the possibility that self-monitoring behavior
may predict certain social and sexual relationships (e.g., sexual behavior; dating behavior;
the propensity toward reactant traits). Investigating the potential causes and correlates o f
such characterological traits is important for achieving better understanding o f human
nature and for improving the treatment o f individuals in counseling settings.
Snyder (1974) proposed the construct o f self-monitoring to describe a trait that
moderates the type o f social behavior people will exhibit cro ss-situationally. It is
grounded in the assumption that people who are highly motivated to alter their behavior to
fit a perceived set o f social constraints would demonstrate a changeable social exterior
that responds to the particular situation they encounter. High self-monitors (HSM) are
individuals who realize that “what they do” and “what they believe” are not necessarily the
same (McCann & Hancock, 1983, p. 111). Such individuals become astute at interpreting
subtle social cues and adept at tailoring their responses to meet the anticipated social
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
requirements, effectively masking internal beliefs they might hold.
On the other end o f the self-monitoring continuum are individuals whose overt
actions are generally consistent with their internal states. These low self monitors (LSM)
are less likely to adjust their behavior to meet the apparent requirements o f a social
situation. LSMs are “rather principled individuals” (McCann & Hancock, 1983, p. 110)
who are less concerned about social appropriateness than retaining consistency between
their internal state and external expression. Relative to HSMs, they are less sensitive to the
nuances o f social signals.
The development o f a scale to measure a particular construct is a necessary step in
operationalizing it and, thus, permitting research into its nature (see, e.g., McFall &
Townsend, 1998). In the case o f self-monitoring, Snyder described the first scale in his
seminal article (Snyder, 1974) on the self-monitoring concept. It was a 25 item
questionnaire that was based on the assumption that the goals o f self-monitoring are:
(a) to communicate accurately one’s true emotional state by means o f an
intensified expressive presentation; (b) to communicate accurately an arbitrary
emotional state which need not be congruent with actual emotional experience; (c)
to conceal adaptively an inappropriate emotional state and appear unresponsive
and unexpressive; (d) to conceal adaptively an inappropriate emotional state and
appear to be experiencing an appropriate one; (e) to appear to be experiencing
some emotion when one experiences nothing and a nonresponsive is inappropriate.
(Snyder, 1974, p. 527)
Later, after other researchers (see, e.g., Briggs et a l, 1980; Lennox & Wolfe,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1984) bad found the SM scale to be correlated with other characterological traits as well
as to show a three factor structure, Snyder and Gangstead (1986) responded with then:
own re-examination o f the scale. They attributed the negative changes in validity to
revisions made by those researchers to the original scale. According to Sndyer(1974), the
original SM scale reflected a general self-monitoring factor. Upon their re-evaluation o f
the scale, Snyder and Gangstead (1986) continue to support the theory and structure o f
the scale, but their analyses indicated that a few items discriminated high from low self
monitors poorly. Those items therefore were dropped, to create a new, 18-item version.
Further investigation has addressed whether SM is best represented on a
continuous dimension along which people can be arrayed from low to high. Miller and
Thayer (1988), for example, found that adjustment related to self-monitoring in an
inverted-U fashion, with middle scorers appearing to have more positive adjustment and
stable identity than those at either SM extreme. They did, however, find a significant
association o f social sources o f identity with SM scores as would have been predicted by
Snyder’s theory (Le., HSMs rated the importance o f their social identity more highly than
didLSM s).
This last finding is consistent with that o f Stewart et al. who found that the HSMs,
however unaware o f their true inner state, find it more important than LSMs to present
what they perceive as a particular (acceptable) social persona. Miller and Thayer’s (1988)
use o f Likert scaling (versus Snyder’s original dichotomous response to items) was
effective in increasing the scale’s variance and therefore in identifying a group o f “middle
self-monitors” who apparently function quite differently from those who score at one end
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o f the scale or the other. Although there could be some response bias in the manner in
which individuals answer Likert type scales, the findings suggest that HSMs and LSMs
share a common tendency toward maladjustment This characterological similarity has
implications from an evolutionary and psychodynamic perspective which will be addressed
later in this chapter. Similarly, Ickes and Barnes (1977) used Likert scaling to group high
middle and low self-monitors and create dyads to observe interaction between participants
differing in SM style. Their findings, too, suggested that those who are “moderate” in self
monitoring may have the “optimal [style] for minimizing disruptions in the flow o f
conversation” (p. 328).
An additional note on the phenomenon o f self-monitoring as a construct also is
warranted in light o f possible evaluative deductions that may be made. The research does
not suggest that any particular self-monitoring style is more desirable than another. It is
likely, however, that readers who value a certain set o f behaviors may conclude that a
person on one end o f the SM scale is inherently “better” than a person scoring in the
opposite direction.
Although the question o f motivators and precursors to SM behavior have not been
answered in the literature, many studies have investigated the correlation o f SM with such
other behaviors as initiation o f social interactions, empathic response and ability to
understand nonverbal emotional cues, and sexual behavior. These studies have attempted
to distinguish patterns o f behavior that are related to an individual’s propensity toward
self-monitoring. Such correlational data offers clues to a broader picture o f what would
constitute a HSM versus LSM and would be predictive o f personality types and specific
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
strengths and weaknesses h i interpersonal ability. It does not, though, begin to answer the
question o f what underlies the phenomenon o f SM behavior.
In this study, attachment is proposed as a possible antecedent to self-monitoring
and other social behaviors. A review o f the literature on self-monitoring, attachment, and
their impact on adult relationship styles and sexual behavior follows. First, a broad
overview o f the research on self-monitoring will be presented. Second will be a more
specifically focused section concerning research on self-monitoring and sexual behavior.
Next, a review o f the attachment literature follows, including more recent research on
adult relationship styles as they relate to early attachment styles. Finally, the construct o f
reactance will be reviewed and its possible relation to self-monitoring will be considered.
Figure 1 depicts the proposed relationship among constructs presented in the literature
review and investigated in this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 1
Proposed Relationship Among Variables
6
i i
Age
Gender
Sexual
Behaviors'
Dating
Confidence
Self-monitoring
Reactance
Attachment
* Sexual behavior variables are: 1) number o f sexual partners 2) number o f steady
relationships 3) age at first intercourse and 4) frequency o f intercourse per month
Review o f the Literature: Self-Monitoring and Related Concepts
The following review o f literature on SM addresses the several topics most salient to this
study. In choosing articles for review, broad inclusion rules were used in order to gamer
as much information as possible on the correlates o f SM.
Much o f the research cited on the topic o f self>monitoring was done in the 1980s,
soon after development o f the 18-item scale (Snyder & Gangstead, 1986). A literature
search, however, showed the se lf monitoring measure continues to be used in dissertations
and journal articles. For instance, Henderson-King and Henderson-King (1997) used the
se lf monitoring measure in a study o f the effects o f media images on women’s selfesteem .
Results suggested that women with a high s e lf monitoring style were differently affected
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
by ideal media images o f women (feeling m ote positive about their body image) than
participants who were LSMs. Havercamp (1994), in a study o f the use o f assessment in
counseling supervision, used the self-monitoring scale to investigate individual differences
in counselor approach to and interpretation o f assessment. Findings suggested that HSMs
and LSMs have different strengths and weaknesses in their clinical evaluation and
intervention. A number o f dissertation studies have focused on the relationship o f self
monitoring and counselor empathy (Treland, 1994; Squitieri, 1994) as well as that
between self-monitoring and social and communication skills (Walker, 1991; Barrow,
1991).
Interactions o f high and low self-monitors
Several studies have investigated the interactions ofH SM s and LSMs. Ludwig,
Franco, and Malloy (1986) studied the responses o f high and low self-monitors when
paired with a confederate who disclosed personal information at a high or low level. They
found that LSMs would reciprocate at the same level as the confederate whereas the HSM
would self-disclose at a high level regardless o f the confederate’s disclosing behavior. This
fits with the assumption that HSMs consistently will present an extraverted, friendly
appearance across social situations.
Shaffer, Ogden, and Wu (1987) took this scenario a step further and looked at
differences in the HSM versus LSM self-disclosure with (a) a confederate o f either high or
low self-disclosure and (b) one with whom participants believed they either would or
would not have further interaction in the fixture. Both high and low self-monitors
reciprocated the level o f intimacy and the emotional expression o f the confederate partner
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
when future interaction was not anticipated. Only the HSM reciprocated the more self-
disclosive material, however, when future interaction was expected.
The researchers interpreted this finding as endorsing the HSMs* hypothesized
tendency to become even more sensitive to social cues when they are deciding what
private information to disclose. This is in contrast to the LSM, who does not depend on
these situational cues to determine what she or he will express. Shaffer, Smith and
Tomarelli (1982) also reported findings consistent w ith those o f the previous researchers.
Their results indicated that the HSM was more likely to reciprocate the intimacy, detail,
and emotional display o f a confederate who presented private information than was the
LSM. Interestingly, the HSM and LSM post-experimentally indicated similar perceptions
o f the level o f intimacy their partner had expressed, though the HSM and LSM appeared
to be differently motivated to respond to the gestures. Shaffer et aL (1982) speculated that
HSMs may be inclined to adjust their responses in such situations as a way to “maximize
personal outcomes” (p. 174).
Danheiser and Graziano (1982) found that the prospect o f future interaction with a
partner increased cooperation on the part o f the high, but not the low SM. These
researchers suggest that these findings offer support that the HSM is differently motivated
than the LSM. That is, that HSMs have an intrinsic need to present themselves in a
seemingly positive light and thus make potential personal gains.
In their study o f interaction between high and low self-monitors, Ickes and Barnes
(1977) obtained three findings. First, they found that the HSM was consistently more
likely to initiate conversation. Second, they found that interactions o f the HSM-LSM
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
dyad were rated as particularly difficult, which the researchers interpreted as a result o f
insensitivity o f both high and low self-monitors to the other’s expression (contrary to
theory which would predict for the H SM to be keenly aware o f the other’s state). Third,
Ickes and Barnes found that HSMs demonstrated a greater tendency to control their
expressive gestures based on the quality o f the interaction they encountered. The first o f
these three findings is consistent with those o f Ellis (1988) whose data indicated that high
self-monitoring was associated with leadership emergence for male participants in group
situations.
It appears that HSMs are differently motivated than LSMs in how they will present
themselves in social situations. Snyder and Gangstead (1982) also investigated the types o f
social situations HSMs and LSMs would choose. They concluded that the HSM sought
situations in which their role would be clearly defined and they would thus “know what
type o f character one ought to be” (p. 134). LSMs, however, tended to seek social
situations which would fit their own already-existing dispositions. The researchers state,
“from this perspective, the strategic choice o f social situations may enable both high self
monitoring and low self-monitoring individuals to regularly and consistently enact their
characteristic behavioral orientations... [which] ensures the perpetuation o f their self
monitoring propensities” (p. 134).
Snyder and Monson (1975) conducted another study that focused on situational
variables o f participants. They found that HSM and subjects who were low in neuroticism
were more inclined to conform socially to discussions than were the LSMs and subjects
who were high in neuroticism. Subjects in the latter two categories were less likely to alter
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
their behavior to fit the social context. Snyder and Monson suggested that HSMs are
particularly attentive to social cues and apt to comply whereas people high in neuroticism
may ignore social cues as they may overgeneralize sim ilarities they encounter in social
situations. Additionally, self and acquaintance ratings predicted greater cross-situational
variability for the HSM than the LSM. Thus self-monitoring seems to be a social-
psychological process with people high in SM “demonstrat[mg] considerable situation-to-
situation discriminitiveness in their behavior” (p. 643) and the behavior o f those low in SM
“best be[ing] predicted from knowledge o f personal characteristics and measures o f inner
states” (p. 643).
Other behavioral correlates o f self-monitoring style have been studied. Wright
(1984), for example, studied the ability o f high and low self-monitors to decode
emotion/intention from audiotaped statements and to express empathy in
dyadic/therapeutic situations. They found that the HSMs were better able to decode verbal
cues and to interpret the vocal expression o f others than were the LSMs. The LSMs,
however, demonstrated what appeared to be more genuine empathy in the dyadic
situations than the HSMs. The researcher concluded that HSM s are adept at cognitive
type o f empathy and are thus able to analyze and interpret the vocal quality o f expression
whereas the LSMs are more skilled at experiencing and expressing affective empathy. This
is consistent with the theorized tendency ofLSM s to act in accordance with their internal
states.
Also consistent with this theory is Azjen, Timko, and W hite’s (1982) finding that
LSMs exhibited stronger attitude-behavior congruence than did HSMs. These researchers
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
concluded that although, both types o f Individuals were aware o f their attitudes at the time
they expressed them, the HSMs, were more influenced by situation demands and therefore
were “less likely to carry out their previously formed intentions than the LSMs, who
[were] more attentive to internal cues” (p. 426) when formulating then intentions.
Tunnell (1980) found that HSMs tended to rate themselves more highly on
personality dimensions such as achievement and affiliation than did their acquaintances.
Self and other ratings o f LSMs, on the other hand, were relatively congruent. Tunnel
(1980) speculated that HSMs would infate their self-ratings as a result o f apprehension
about being judged.
To investigate how the differential social inclinations o f HSMs and LSMs might
affect their encoding o f information, Lassiter, Stone and Wiegold (1988) examined
differences in memory for information about a scene they observed. Specifically, they
aimed to determine if high and low self-monitors responded differently when given leading
questions about a particular event they had observed. Although HSMs were hypothesized
to be more accurate in their memory o f details due to their sensitivity to social situations,
Lassiter et al. found them to be more susceptible to leading questions than LSMs.
Therefore Lassiter et al. concluded that the memories o f HSMs may be more likely than
LSMs to be influenced when certain social pressures are in place.
Overall, research on the different interpersonal styles ofHSM s and LSMs
shows that those high in self-monitoring tend to gauge social situations and adjust their
behavior to fit what they sense is appropriate and desirable behavior, seemingly in order to
gain social acceptance o r attention. Those low in self-monitoring, on the other hand,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
appear to behave in accordance with their internal states and beliefs, apparently less
influenced by desires to win social approval. How these different styles impact social
acceptance and social position is implied by the research which suggests, for
example, that HSMs are more likely to become leaders and LSMs are judged as more
genuinely empathetic in a given situation.
Research on Self-Monitoring and Sexual Behavior/Dating Another area o f self
monitoring research has concerned the dating and sexual behavior o f high and low self-
monitors. This is, after all, a specific instance o f social interaction. Snyder and Simpson
(1984) found that HSMs individuals were more likely than LSMs to terminate current
dating relationships in favor o f attractive alternative partners. HSMs also reported having
dated a greater number o f people in the preceding year than LSMs whereas LSMs
reported having had considerably longer dating relationships than those o f HSMs. Snyder
and Simpson correspondingly found that whereas HSMs appear to develop intimacy in
relationships more quickly, the LSMs tend to establish greater intimacy over a longer
period o f time. Thus, the HSM exhibits an “uncommitted” approach to dating
relationships whereas the LSM develops a “committed” approach. These authors
hypothesized that HSM individuals may invest less in one relationship to minimize the
distress and disappointment o f possible termination whereas the LSM would be more
willing to invest and risk possible dissolution o f the relationship.
Another hypothesis o f Snyder and Simpson (1984) was that LSMs meet potential
mates in the kinds o f environments in which they are normally comfortable and to which
they are accustomed and therefore are more likely to choose mates on the basis o f more
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
“stable dispositional characteristics” (p. 1290). The HSM, altering his o r her behavior to
meet perceived social demands, may not enter a relationship on the basis o f such stable
notions o f the other individual. A partner who appeared compatible under one
circumstance may prove inappropriate in another, thus resulting in a series o f more brief
and less committed relationships.
Yet another o f Snyder and Simpson (1984) hypotheses was that marriages o f high
and low self-monitors may be founded on fundamentally different ground. I f HSMs are
concerned primarily with their social success, it would follow that they may choose mates
who fit with their social activities. Thus, they would develop marriages based on
partnership and “doing” activities. LSMs, concerned with matching the fit between their
inner variables and the environment, would choose mates based on fit o f personal
attributes, thus basing the marriage relationship on ability to simply “be together”.
More research has investigated the differential approach to choosing dating
partners o f HSms and LSMs. For example, Snyder, Bersheid, and Glick (1985) asked
participants to choose between dating partners. To make this choice, they would have to
sacrifice one type o f desirable attribute (physical attractiveness) for another (personality
characteristics). They found that LSMs placed more emphasis on the personal attributes
information whereas the HSMs placed more importance on the physical attributes when
making their dating choice. Snyder et al. concluded that high and low self-monitoring
individuals represent “two broad classes o f people who systematically differ in their
approach to the selection o f relationship partners in certain settings” (p.1436).
In a similar study, Glick, DeMorest, and Hotze (1988), presented participants with
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
photographs and descriptions o f men and women to match for dates. They found that
HSMs tended to create couples who were more similar in ratings o f physical appearance
whereas LSMs create couples more similar in personality traits and interests. The
researchers speculated that the tendency ofhigh and low self-monitors to match couples
this way likely reflected the way they themselves might choose partners.
So far, however, no research has examined the relationship between self-
monitoring style and feh-competence in dating situations. That was, then, one purpose o f
this study.
Generally, research has shown that HSMs and LSMs in addition to differing in
how they interact with others also differ in their approach to dating situations. The data
suggests that HSMs tend to use physical attractiveness and social activity as a basis for
choosing mates, whereas LSMs appear to choose mates more on the basis o f character
traits and similar interests.
Models to Explain the Development o f Self-Monitoring Stvle
A reasonable question then, is how SM styles develops. Two primary models seem
useful for this purpose: evolutionary psychology; and, attachment theory. Each will be
considered in turn.
E volutionary Psychology as an Explanation for Self-Monitoring and Sexual
Behavior. Evolutionary psychology has been invoked to explain motivation for developing
a particular SM style in sexual situations. Based on the tenets o f evolutionary psychology,
the different patterns o f sexual contact demonstrated by high and low self-monitors may
constitute differing procreative styles.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 5
Snyder, Simpson and Gangestad (1986) surveyed the sexual behaviors o f college
students, and found that, controlling for effects o f gender, HSMs expected to have more
partners in the next 5 years, reported having had more one night stands, had more sexual
partners, would desire to have m ore current partners, and were more amenable to having
“casual sex” than LSMs. Snyder et aL (1986) termed this casual style o f sexual behavior o f
the HSM the “unrestricted” approach versus the “restricted” approach o f the LSM. High
self-monitors adopted a mating approach which is less discriminating than their LSM
counterparts. The HSM, therefore, will be more willing to “risk” procreation for
recreational purposes, to engage in sexual encounters with others with whom they have no
relationship or emotional investment, and to seek a wide variety of sexual partners to
enhance their procreative opportunity. The LSM takes a very different strategy
approaches the procreative task by making careful o r “low risk” choices in sexual partners.
In acquiring fewer sexual partners with whom they have an emotional investment or
ongoing relationship, LSMs limit their potential for genetic variety in offspring in order to
gain the certain support and investment o f the partner in the care of that offspring.
Not all sexual encounters are intended to result in pregnancy. The evolutionary
view, however, is that procreation is the root o f all sexual encounters, regardless o f the
social context o r the methods to prevent fertilization. The theory Snyder et al. (1986)
proposed to explain the differences in sexual behavior o f low versus high self-monitors is
akin to Buss’ theory (1985) o f the differences in sexual behavior of men and women. Buss
postulated that it makes more evolutionary sense for men to develop a strategy o f low
investment, high output whereas women would be most successful in genetic survival by
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
adopting a high investment, low output approach based on their physiological Imitations
to bear children. Due to women’s limited ability to reproduce relative to men, they
demonstrate greater investment in the production o f a child and adopt a more selective
approach to sexual intimacy. Thus, a woman would selectively look for cues in a potential
mate o f physical health, ability to provide for her and the child, and an inclination to be
nurturing and generous.
According to this theory, men, having the ability to produce a virtually unlimited
number o f offspring with relatively little investment (the procreative drive does not take
into consideration current legal and economic constraints on fathers), would try to
reproduce as much as possible. This would require the availability (and supposed
cooperation) o f many women. Therefore, men develop strategies to gain the attention and
admiration o f women including actual or feigned status, wealth, health, and sensitivity.
Though he may have little influence over the quality o f life or development o f the child, a
man best ensures genetic survival by producing as many offspring as possible with varying
genetic contribution.
That Snyder et al. (1986) obtained significant between-sex differences on all
dependent measures related to sexual experiences fits with Buss’ proposal. When the
effect o f participant gender was removed, a clear pattern o f “unrestricted” sexual behavior
and attitude was linked with high self-monitoring whereas a “restricted” approach was the
pattern for those low in SM.
These differences in individual SM regardless o f gender contrasts with the basic
proposal set forth by Buss in his theory that there is a primary gender basis to differences
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
in mating strategy. According to Buss’ approach m ost mates would logically fit the high
self-monitoring approach and females the low self-monitoring approach. Significant
gender differences in self-monitoring, were consistently not cited in the research (see, e.g.,
Glick, DeM orest & Hotze, 1988; and, Snyder & Simpson, 1984). However, Stewart and
Carley (1984) found a correlation o f .24 between self-monitoring and gender, with women
being overrepresented among low se lf monitors and men being overrepresented among
high se lf monitors.
According to the evolutionary psychology theory, it is logical that men would be
HSMs and women, LSMs. If sex differences do exist in se lf monitoring and more m en are
HSMs compared to women, then a study by Ickes and Barnes (1977), which found that
HSMs were less likely to establish deep affective relationships than LSMs, would imply
that there may be more to the differences in self-monitoring and sexual behaviors than the
evolutionary approach addresses.
Based on this literature, between-gender differences in self-monitoring was
examined in the present study.
Attachment Theory Proposed as an Alternative Hypothesis for Development o f S e lf
Monitoring Behavior
An alternative explanation for self-monitoring style differences is attachment
theory. Researchers such as Hazan & Shaver (1987) have investigated this theory with
respect to the establishment o f adult romantic relationships. Although they did not make
specific inquiry into sexual behaviors, these are logically related to romantic relationships.
I have focused particularly on attachment and adult relationships. In the section that
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
follows, I briefly review some o f the more salient literature on attachment theory.
Infant attachment. Attachment theory is rooted in psychodynamic thought and
based largely on the ideas o f John Bowlby (1969,1988) and subsequent research by his
student Mary Ainsworth (1978, 1989) who systematically observed the bonding behaviors
o f mothers and children in Africa and in the U.S. The basic premise o f this theory is that
the quality o f the primary relationship o f the infant to the parent (the mother) affects
subsequent behavior, such as comportment in school, self-esteem, interactions with peers,
interactions with adults, and competence. This early relationship thus shapes who we are
as developing adults and determines the kind o f social and emotional style we manifest.
In her observations o f children and their mothers, Ainsworth established clear
differences in the reaction o f children when left by their mother w ith a stranger and
reunited with her after a brief interval alone with the stranger. Known as the “Strange
Situation,” this intervention purportedly identified children as having “secure” or
“insecure” attachments and was deemed diagnostic o f the child’s attachment style: secure,
avoidant, o r anxious-ambivalent.
The secure child in this circumstance would be somewhat cautious but would
retain appropriate investigative curiosity. This child would maintain a safe boundary with
the stranger while having an enjoyable interaction. Secure children would be happy to see
the mother upon return and have little difficulty with reunion. Avoidant children, though,
were observed to be overly independent and to exhibit inappropriate autonomy and
distance from the mother when in an unfamiliar place. The child would appear undisturbed
by the presence o f a stranger and would not interact with the stranger or behave with any
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
noticeable distress while the m other was absent Upon the mother’s return, the child
would show minimal if any signs o f celebration for the recovery o f the mother.
The anxious-ambivalent child would appear nervous and distressed throughout the
events o f the situation. This child would be distrustful, concerned about his or her safety,
and exhibit unease (e.g., an inability to enjoy exploration without consistent referencing to
m other and tearful, disorganized behavior when left alone with the stranger). Upon her
return, this child would express anger toward the mother for having left while clinging to
her as if profoundly relieved that she had indeed not abandoned him o r her to the stranger.
Although the strange situation and attachment theory were criticized by
behaviorists and by others who were skeptical o f such a pschodynamic viewpoint (see,
e.g., Kagan, Reznick, Clarke, Snidman, & Garcia-KolL, 1984), many researchers saw this
as a potential answer to many questions o f human development. Attachment research
blossomed with investigations into parent-child relationships, competence, and school
behavior.
Stroufe and Waters (1977) found that securely attached children displayed more
task persistence and prosocial behaviors in the classroom, whereas anxious-ambivalent
children easily became fussy and displayed less flexibility in their interactions with the
environment. Avoidant children were observed to have less fantasy play, though what they
did express was characterized by irresolvable conflict. Avoidant children tended to be
oppositional and to victimize children who had been identified as having insecure
attachments. Stroufe et al. also discovered that teachers responded differently to these
groups o f children. Although they treated secure children in age-appropriate ways, they
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
20
tended to infantalize the anxious-ambivalent children and to behave in controlling and
harsh ways with the avoidant children.
Adult attachments. M ary Main, a student o f Ainsworth, believed that the way
parents recall and conceptualize their own childhood experiences would predict the
attachment style o f their own children (see Main, Kaplan, & Cassidy,1985). She observed
four distinct attachment patterns in her interviews with parents. She called the first
“autonomous”, characterized by individuals who were self-reliant and objective regarding
their childhood experiences and were able to consider both painfiil and happy memories as
parts o f their development. Main believed that people in this group had been securely
attached as children. Peopleexhibiting the second pattern seemed relatively indifferent to
deep emotional feelings in their most important relationships. They tended to idealize their
own childhood bonds, though they described events suggestive o f neglect and rejection
from their parents. Main called this group “dismissive o f early attachments” and concluded
that they had manifest avoidant attachments as children.
People exhibiting the third pattern, which she labeled “preoccupied with early
attachments,” was composed o f parents who expressed confused and indistinct feelings
about their childhood relationships. They often had overwhelming feelings o f anger about
relationships and described high dependency needs. Main thought this group displayed an
anxio us-ambivalent attachment style. The fourth style was termed “disorganized” and was
maintained by parents with a history o f abuse who had unresolved feelings o f conflict over
their past traumas. The finding most relevant to the following review o f literature on
adults and attachment was that Main’s categorization o f a sample o f adults’ attachment 75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
to 80 percent o f the tune corresponded to Ainsworth’s original classification found in a
sample o f children. Approximately 15% o f both groups were categorized as
anxious/ambivalent, about 23% were classified as avoidant and the remaining
approximately 62% were identified as securely attached (Campos, Barrett, Lamb,
Goldsmith, & Sternberg, 1 9 8 3 ). This suggests a stable pattern o f attachment across the
lifetime detectable and identifiable by careful clinical investigation.
Attachment Style in Adulthood. In a meta-analysis o f the research on Main’s adult
attachment interview (AAI), van IJzendoom (1995) discussed the nature o f the AAI and
how the demeanor o f the client and expressive quality o f the responses in the AAI are
what is rated rather than the content o f the responses given, van IJzendoom cited
numerous studies which support the “remarkable reliability and discriminant validity” o f
the measure (1995, p. 388). Namely, the attachment style as measured by the AAI should
correspond to the observed attachment o f that parent’s child in the Strange Situation test.
Rosenthal (1991) found this relationship to explain 22% o f the variance in attachment
style between the parent’s and child’s attachment style, van IJzendoom noted, however,
that the age o f the child being tested was an important predictor o f this parent-child
correspondence in attachment style. As the child being observed for study was older, the
predictive power o f the parent attachment decreased.
Van IJzendoom concluded that about half o f infant attachment could be attributed
to parental attachment whereas the other half would be based on chance. Such elements as
significant life changes (Le. loss o f parent, economic changes affecting the family) were
also demonstrated to alter a child’s attachment style, although initially secure children
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
were found to be more resilient to negative changes and more responsive to positive ones.
As Bowlby first theorized, the foundation for attachment style is laid in the first few years
(especially the first year) and is resistant to change thereafter. This would explain the high
correlation in numbers o f children and adults in each o f the three attachment groups.
In his review, van IJzendoom attem pts to explain the mechanism by which
attachment is transmitted from parent to child with such consistency. He cited “sensitive
responsiveness” on the part o f the parent as one source o f transmission. It appears that
parents express their own attachment in the way in which they respond behaviorally to
their children in a free play situation. Another source o f parental influence suggested by
van IJzendoom concerns such other forms o f communication as facial expression. Because
much o f the interaction that is rated in the attachment research is based on overt behaviors
and verbal communication, he hypothesized that more subtle interactions may account for
some o f the transmission o f attachment style from parent to child.
Although van IJzendoom included temperament as another possible factor
affecting transmission, he was more conservative in his consideration o f temperament as
part o f the transmission process. He based this speculation on what he considered
confounds in the research on personality characteristics and parental attachment. In a
commentary on the van IJzendoom article, Fox (1995) argued that there are multiple
factors affecting the establishment o f attachment style such as temperament, personality,
and psychological state. He preferred to view that adult attachment is the product o f
“early individual differences in temperament” (Fox, 1995, p. 408) controlling for
disruptive changes in the environment. It is suggested that the temperament o f the parent
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
and child create a unique attachment relationship.
In their 1998 study, Levy, Shaver, and B latt examined the college students’
parental representations hi an exercise hi which they gave written descriptions o f their
parents. Descriptions were based on the 4 attachment categories described by Barthlomew
and Horowitz (1991) which included 2 types o f avoidants; one dismissive o f attachment to
others, the other fearful o f attachment. Levy et al. made hypotheses about the written
outcomes based on their theory that cognitive and emotional aspects o f mental
representations become “increasingly accurate, articulated, and conceptually complex over
time” and “develop epigenetically” (Levy et al, 1998, p. 409).
An especially relevant point o f their study was the disproportionate number o f men
in the dismissive-avoidant group and the disproportionate number o f women grouped as
fearfiil-avo idant. The hypothesis they propose for this distinction is that socialization
processes may influence men to develop a strategy whereby their avoidance is used as a
defense and they de-emphasize the importance o f intimate relationships. Women’s
socialization may lead them to shrink away from what seem to be dangerous attachments
rather than detaching or re-organizing their experience in a self-serving manner like their
male counterpart avoidants. The authors noted that the responses o f these fearfiil-avoidant
women were unexpected in that they most closely resembled those responses o f the secure
individuals.
Based on this finding, Levy et al (1998) suggested that many o f these fearfiil-
avoidant women were “earned secure.” These individuals are able to organize their early
experiences with their parents in a cohesive manner like the “continuously secure”
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
individuals, yet maintain certain characteristics o f the insecure groups, mainly depressive
symptomatology. Noteworthy o f these avoidant individuals (women in this case) is th at
they are also similar to continuously secure individuals in their parenting style which seems
to promote secure attachment in their children (Pearson, Cohn, Cowen & Cowen, 1994
as cited in Levy et al, 1998, p. 416).
Attachment and Adult R om antic R elationships. In what has become a landmark
article investigating the association o f patterns o f adult romantic relationships to an earlier
attachment styles, Hazan and Shaver (1987) developed a love-experience questionnaire to
gain information on participant’s emotional experiences o f their relationships. Research
findings confirmed their hypotheses. Results indicated that the three attachment styles
found in the adult population corresponded to the frequencies o f these attachment styles
originally found in the child samples. They predicted that people in these three categories
would experience qualitatively different relationships as adults. Namely, securely attached
individuals described relationships as generally “happy, trusting, friendly” whereas
avoidant subjects emphasized a ‘ Tear o f closeness” and those in the anxious-ambivalent
group reported their relationships to have “ jealousy, emotional ups and downs, and a
desire for more reciprocation [from the partner]”.
As predicted, fundamental differences in mental models o f self and relationships
emerged, consistent with the separate attachment styles. Secure individuals reported th e
waxing and waning o f intense romantic love whereas those in the avoidant group believed
the “head-over-heels” love depicted in romance stories does not exist. The anxious-
ambivalent people also reported that romance waxes and wanes, but differed from the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
secure group in th at they felt vulnerable to falling in love frequently. Like the avoidants,
anxious-ambivalent people believed they rarefy found “real love.” The way that individuals
in the three different groups characterized their parental relationships was investigated. As
was suspected, the secure individuals described warm relations while both the avoidants
and anxious-ambivalents depicted those early relationships as cold, rejecting and unfair.
The researchers expected to find more emotional difficulties, specifically problems
with loneliness, in the insecure groups. Findings suggested that the anxious-ambivalent
individuals experiences trait loneliness though it was difficult to assess avoidants’
loneliness due to possible denial o f symptoms resulting from defensive type o f responses.
Hazan and Shaver acknowledged conceptual problems in comparing adult romantic
attachments, which typically include sexual involvement and reciprocity, with infant-parent
attachments though they believe there are fundamental similarities worthy o f exploration.
Although the above has focused on the social aspects o f attachment, Hazan and
Shaver in their conclusion theorized that attachment is a biological as well as a social
process. They offer the evolutionary psychology approach as a way to explain the
phenomenon o f attachment. From their perspective, adult romantic love may be a
biological process that encourages attachment between sexual partners who will likely
produce offspring that need their nurturance. Because it is virtually impossible to study
the integrity o f the evolutionary drive to either establish attachment behavior or to self
monitor, the present research investigated the social aspects o f such processes, while
keeping in mind the possible impact ofbiology and the evolutionary drive for survival.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Collins and Read (1990) factor analyzed the attachment scale devised by Hagan
and Shaver and found three dimensions: an individual’s comfort w ith closeness (Close), a
sense that he or she can depend on others (Depend), and how anxious or fearful the
person is about being rejected or abandoned by others (Anxiety). They found these
dimensions related to individuals’ measures o f self-esteem, expressiveness, instrumentality,
trust in others, beliefs about human nature, and styles o f loving. In attempting to match a
subject’s scores on these dimensional measures to a discrete attachment category, C ollins
and Read found some variability in the classification o f participants to one o f three
attachment groups. They suggest that classifying individuals to groups “may result in
groups whose members share some features but differ considerably on other dimensions”
(p.652).
In this case, Collins and Read noted that the classification on the anxious group in
particular was poor, with some secures shifting to the anxious group on analysis and some
individuals from the anxious group recategorized as avoidant. Therefore, they suggest that
there may be two groups o f anxious-ambivalent people.
Overall, their findings suggest that people with secure attachment styles tend to
have high scores on Close and Depend while scoring low on Anxiety. Avoidants tend to
have low scores for all three dimensions. People with anxious-ambivalent attachments
seem to have highest scores on Anxiety.
Lastly, they investigated how these dimensions may predict partner matching in
adult romantic relationships. Collins and Read found evidence that partners were most
alike for the care giving style o f the subject’s opposite-sex parent, whom they suggested
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
had served as a role model o f heterosexual attachments. For women, the ability for their
partner to feel close (Close) was the best predictor o f relationship satisfaction, whereas for
men, their partners’ anxiety level (Anxious) was the best predictor o f dissatisfaction.
Feeney and Noller (1990) examined the relationships between the Hendrick and
Hendrick (1986) love styles and the three attachment styles. As was predicted, the two
insecure groups differed significantly from the secure group members who generally
reported higher self-confidence and greater trusting in their relationships. The two
insecure groups differed from one another in that the avoidant individuals endorsed more
Ludus (game-playing) type o f love while scoring low on Romantic and Ideal love
measures. The anxious-ambivalent participants’ scores reflected high dependency needs
and greater desire for commitment from a partner. Their scores corresponded more with
the neurotic love styles o f Mania, Limerence and Love Addiction.
An important element o f the Feeney and Noller (1990) study which has impact on
further discussion o f attachment as a foundation o f self-monitoring is that they found no
gender differences in the frequency o f subjects who endorsed the three attachment styles.
This finding would counter the expectation, according to the evolutionary perspective,
that men would be more likely than women to exhibit behavior consistent with an avoidant
style.
Bartholomew and Horowitz (1991) proposed a four-group model o f attachment
styles based on their interviews o f adult participants. Their attachment ratings ofboth self
and friend reports were associated with four distinct patterns o f interpersonal functioning
termed secure, preoccupied, dismissing, and fearful. The previously discussed style o f
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
avoidance was found to be split between those seen as dismissing o f and lacking
apprehension about attachments and those who are avoidant o f attachments but have an
underlying fear o f intimacy. The four groups were based on the individual’s evaluation o f
him or herself (self model) and his o r her evaluation o f others in general (other model).
Those deemed secure have positive models o f both self and other while those foiling into
the preoccupied (akin to the anxious-ambivalent) category express negative evaluations o f
themselves and view others positively. The dismissing-avoidant individuals see themselves
in a positive light while evaluating others negatively, thus having a relatively detached
attitude towards developing relationships. People in the fearfiil-avoidant group had
negative views o f self as well as negative views o f others that contribute to their distancing
behaviors.
Another study examining behavior which evolutionary psychologists also have
examined in relation to gender concerned the relationship between jealousy and adult
romantic attachment. Sharpsteen and Kirkpatrick (1997) found distinctive behavioral
responses o f participants in each o f the three attachment groups. The avoidant subjects
were most likely to turn their anger at the “interloper” whereas the anxious-ambivalent
individuals were most likely to refrain from any overt expression o f anger. The secure
participants were most likely to turn their anger toward their partner and to maintain the
relationship. These responses are consistent with the findings o f previously cited studies
investigating the behavioral and attitudinal differences among attachment groups. The
avoidant subjects seem to make every attempt to use their anger in a way that fits with
maintaining themselves as right and invulnerable. The anxious-ambivalents stifle their
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
anger due to feelings o f inferiority and fear ofdisafifecting their mate. The secure
individuals feel justified in accusing their partner for transgressions against trust in their
relationship.
Sharpsteen and Kirkpatrick noted that Buss e t al. (1992) also had investigated
romantic jealously from the evolutionary viewpoint. In that frame, jealousy appeared again
to separate the sexes as men experienced more jealousy with “sexual” infidelity whereas
women’s jealousy was ignited with “emotional” infidelity on the part o f their mate.
Implications for this difference are again based on factors o f genetic investment and
promotion. This research indicates that jealousy may be a product o f the integration o f
two systems, the biological and the attachment system (Sharpsteen et al 1997).
Other Attachment Research. Other investigations o f the association o f attachment
style with personal attributes includes the finding that self-esteem was demonstrated
differently by secure and avoidant individuals. Although secure subjects endorsed more
self-liking items, dismissing-avoidance was associated with self competence (Brennan &
Morris, 1997). Thus, people with an avoidant attachment have found a successful way to
cope with an insecure base. Blysma et al (1997) repeated these findings, showing that
secure and dismissing-avo idant individuals reported greater global self-esteem and greater
self-competence than did either the preoccupied or fearfiil-avoidant participants. These
researchers point out that the attachment process may influence self-concept which in turn
affects an adult’s romantic attachment.
Another recent study focusing on adult attachment is that o f Mikulincer (1998)
who studied the experiences o f anger among people with the different attachment styles.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
He found that secure individuals did not become overwhelmed by then anger. Secure
subjects were able to use anger as a signal that something was wrong and took positive
action to rectify the situation. Anxious-ambivalent individuals tended to ruminate on their
anger which overwhelmed their cognitive system, thus impeding then taking action. The
author noted that anxious-ambivalent people’s patterns o f anger resembled the experiences
o f people prone to shame. This, he concluded, was due to maladaptive responses to
feelings o f anger. The avoidant group seemed to dissociate from their anger, perhaps as a
way to distance themselves from painful experiences or in an effort to control their self
presentation. These individuals demonstrated a discrepancy between their reported anger
and physiological measures o f arousal (heart rate).
Reactance
The research on SM style indicates a clear connection to the type o f overt social
behavior a person will display. Another construct that has been proposed to predict the
social behavior o f an individual is what Brehm and Brehm (1981) describe as reactance.
They defined reactance as a motivational force that impels individuals to attempt to gain
or regain lost or threatened freedoms. Whereas Brehm and Brehm considered reactance to
be situation-specific, others have regarded it as a characterological trait. Dowd,
Wallbrown, Sanders, and Yesenosky (1994) concluded from their data that reactant
individuals are less concerned about making a good impression and seem to “march to
their own drummer.”
Dowd and Trutt (1988) postulated that psychological reactance comes about as a
defiant response to a perceived threat. Thus, the client high in psychological reactance in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
the counseling setting would be likely to resist perceived manipulations by the therapist.
They hypothesized that highly reactant clients therefore would profit most from
paradoxical interventions.
Dowd, Milne, and Wise (1991) developed the Therapeutic Reactance Scale (TRS)
to assess reactance as a trait. Their data seemed to indicate that psychological reactance is
“in part an individual difference variable that is relatively stable over time and across
situations” (p. 544). They noted that TRS accounts for only about 22% o f the variance
with respect to the criteria o f attitudes toward verbal and behavioral interactions with
others, so situational aspects remain important when considering a person’s psychological
reactance.
Dowd and Wallbrown (1993) attempted to determine other personality
characteristics o f the high reactant versus low reactant individual. Consistent with Brehm
and Brehm’s original concept, they found the highly reactant participant to be “aggressive,
dominant, defensive and quick to take offense, and autonomous. They also tend not to
affiliate with others and neither seek support from others nor support them” (p. 537).
These people also did not describe themselves favorably in general or to others. Dowd and
Wallbrown considered this type o f individual to be a loner; individualistic and dominant.
Dowd and Wallbrown suggested that people high in reactance would seek control o f
situations and would be strong leaders. Perhaps consistently, Loucka (1990) has shown
men to be more reactant than women.
Based on the conceptions o f reactance as a characterological trait, to investigate a
relationship between SM and reactance scores seemed warranted. Both traits concern
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
responses to social situations. It is reasonable to wonder how these two attributes might
relate to each other, if at alL
Implications for this Study
The above review o f literature on the construct o f self-monitoring and its
purported influence on social behavior as well as the overview o f the identification and
correlates o f attachment styles, offer descriptions and predictions about individuals
categorized on these dim ensions as well as postulations about the origins o f these
behaviors.
Although it is reasonable to expect certain SM behaviors to change as a function o f
maturity and social changes experienced in the aging process, this has not been assessed.
As there have been no investigations into the impact o f developmental stages on self
monitoring behavior, this study will take the opportunity to use the age differences within
the sample to answer this question. Given the popular notion that individuals become less
self-conscious with age, the expectation is that older participants will have lower self
monitoring scores overall than the younger participants.
The evolutionary perspective is offered as a possible reason for the development o f
self-monitoring behavior (Snyder et al. 1984). This would imply that, based on the
differential sexual behaviors o f men and women, men would exhibit more high self
monitoring behaviors. These researchers suggested that people, regardless o f gender,
adopted different yet successful mating strategies to survive genetically, thus the high self
monitoring style would not be a predominantly male characteristic (as was suggested by
their data). Sexual behavior questions and questions about dating confidence will address
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
these issues in this sample.
Attachment theorists attribute differences in romantic relationships to a profound
difference among individuals in their sense o f self and others which is based on their
primary attachment. These researchers described the characteristic patterns o f behavior
among people with secure, anxious-ambivalent, and avoidant styles o f interaction.
I f self-monitoring can predict distinct patterns o f sexual behavior, could it not only be
attributable to differing mating strategies but also to an attachment process? I f so, how
would the individuals with insecure attachments differ both in self monitoring and sexual
behavior from those with secure attachments?
Following Thayer et aL’s (1988) foldings that individuals with either very high or
very low SM scores appeared m ost “neurotic” and with higher scores on maladjustment,
one would predict the secure participants to score in the middle range on the self
monitoring scale with the insecures taking position at the extreme ends o f the scale. That
is, it may be that self-monitoring is not a linear construct. For example, it is possible that
more extreme behaviors o f one type or another are manifest by people in the middle range
(versus the two extremes). Therefore for some o f the research questions, the sample was
broken into three groups (high, medium, low).
It is predicted that high and low se lf monitors will report similar patterns o f sexual
interaction as have been demonstrated in previous studies. Namely, HSMs will have more
frequent sex and more sex partners than LSMs. In this study, dating confidence and age at
first intercourse were additional dependent variables and, consistent with the sexual
behavior variables, it is predicted that HSMs will have higher dating confidence and be
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
younger at first intercourse. Since the variable o f “steady relationships” is more subjective
and was not defined by the parameters o f tune o r intensity, it was difficult to anticipate
how the high versus low self-monitor would interpret this question. Still, it would be
predicted that because HSMs would likely have a greater number o f partners, they also
may consider that they have had more “steady” relationships.
The construct o f reactance is examined as it relates to SM and attachment. As
reactance is similar to SM in that it purports to measure an individual’s social behaviors hi
relation to others, specifically one’s propensity to respond strongly to a perceived threat to
autonomy, it was predicted that those high in reactance would also be high in SM.
Research Questions
In light o f the above summary, the following research questions were developed to
drive this study:
1. Does age predict level o f self-monitoring style?
2. Does the attachment dimensions predict level o f self-monitoring style operationalized in
this study using the measure devised by Collins & Read, 1990, which identifies the
attachment dimensions o f Close, Depend, and Anxiety?
3. Does gender predict level o f self-monitoring style?
4. Does gender predict differences in the relationships among variables?
5. Does level o f self-monitoring style (high; medium; low) predict sexual behavior?
Specifically, does it predict:
a. age at first intercourse
b. frequency o f sexual intercourse during the past month
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
c. number o f sexual partners
d. number o f steady relationships
6. Does an individual’s level o f self monitoring style (high; medium; low) predict
reactance level?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
CHAPTER 2
Methods
This chapter provides demographic information about the participants in this study.
Specific details o f the instruments used and any relevant facts about then employment in
the study are described. Additionally, procedures used in development and administration
o f the instruments and in the handling o f missing data are presented.
Participants
Participants were 357 adolescents (224 females; 133 males) with an average age o f
17.39 (SD=2.58). O f these, 54 were African-American, 271 were Hispanic, and 32 were
White, non-Hispanic. Sites at which they were recruited included public health clinics,
public schools, and WIC (Women, Infant, and Children) nutrition centers. Participants had
an average educational level o f 5.72 years (SD=1.45) and reported an average family
income o f between $15,000 and $22,500.
Materials and Procedure
The data employed in this study were gathered as part o f a larger, funded project.
This particular study employed participants’ responses to demographic questions and to
questions about their sexual histories and behavior. These latter questions included:
frequency o f intercourse during the past month, frequency o f intercourse during the past
week, number o f lifetime sexual partners, and number o f lifetime steady relationships.
In addition, the following five measures were used:
Self-Monitoring Scale. Snyder & Gangstead’s (1986) 18-item Self-monitoring
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
scale was used after being revised in two ways. First, the wording o f a number of items
was changed to fit with current language standards among young adults, especially those
in a younger, community sample, rather than, university students for whom the original
measure was designed. Second, the item format was changed to employ a 5-point, Likert-
type scale f Strongly agree. Agree. NeutraL Disagree. Strongly disagreed rather than the
original dichotomous, True-False format (Appendix: A).
To establish the validity o f the revised measure, the original (Snyder & Gangstead,
1986) and the revised versions were administered to 40 students in an undergraduate
psychology class at a local community college. The two measures correlated at .73, which
is a satisfactory level o f correspondence. In this sample, reliability (alpha) o f the revised
measure was .73 (n=479).
Collins and Read 119901 Attachment Scale. The 18 item Collins and Read (1990)
attachment scale contains 6 statements for each o f 3 factor-analytically derived scales:
Close; Depend; and, Anxiety. Items are in a 5-point Likert-type format where 1= not at all
characteristic o f me and 5= very characteristic o f me. Scores on each scale can range from
6-30, with high scores indicating more o f that attachment dimension. High scorers on the
Close scale endorse feeling comfortable getting emotionally close to others. High scorers
on Depend are those who feel, when in need, that they can depend on others while high
scorers on Anxiety are those who are anxious or fearful o f being rejected or abandoned by
others.
Collins and Read noted that Close and Depend are moderately correlated (.41),
suggesting that people who feel they can depend on others are also more likely to feel
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
comfortable getting close. Anxiety was weakly correlated with Depend (.18) and not at all
with Close (.01).
A subset o f the Collins and Read sample completed the attachment scale again
about 2 months later. Test-retest reliabilities for Close, Depend, and Anxiety were .68,
.71, and.52 respectively. The researchers suggested that items on the Anxiety factor may
have been more closely tied to a particular relationship than were the Close and Depend
items. Overall, though, scores were considered fairly stable over a 2-month period.
Dating C onfidence. Dating confidence was measured by a five-item subscale o f the
Social Competency Scale o f Levenson and Gottman (1978). Sample items were “able to
maintain a long conversation with someone you are attracted to;” and “be able to sense
accurately how a member o f the opposite sex feels about you.” Items were in a four-point,
Likert format (scaled from never to always!. The scale had a reliability (alpha) o f .84 for
this sample.
Reactance. The 28-item Dowd, Milne, and Wise (1991) scale purports to assess an
individuals motivation to attem pt to gain or regain lost or threatened freedoms. Thus,
individuals scoring high on the scale likely would resist outside influences they perceive as
limiting their autonomy. Highly reactant individuals may be more aggressive and
dominant. Dowd et aL reported that test-retest reliability ranged from .57 to .76; internal
consistency ranged from .75 to .84. In this study, a truncated 5-item version with a
reliability (alpha) o f .52 was used.
Procedures
Gatekeepers (e.g., teachers, public health educators, etc.) in social service,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
educational, and public health settings in Los Angeles county were enlisted to help recruit
participants for a funded project concerning teen pregnancy. Therefore, most o f the
participants either were teens who were pregnant or had been within the past three months
(87.3% o f the females) o r were men who had been responsible for one or m ore teen
pregnancies (80.1% o f the men). The comparison group consisted o f men and women who
had not given birth or impregnated a teen, but who otherwise were similar w ith respect to
demographic and other characteristics.
In several o f the participating agencies, flyers were posted announcing the project.
Participation was based on informed consent. They were told that they would be asked to
complete a questionnaire that would take them about an hour; they also were told that
their participation was voluntary and that their responses would be anonymous. Each
participant received monetary compensation for participating.
The questionnaires described above were translated to Spanish, then back-
translated to ensure equivalence between English and Spanish forms. This follows the
procedures recommended by Marin and Marin (1991). Approximately 10 percent o f
participants used the Spanish form o f the questionnaire.
Missing values are a problem, especially with a data set that has multiple measures.
Therefore, to address this issue, data in this data set were broken out by ethnic group and
by gender (i.e., so that there were six different data sets). Then, for each missing value, the
mean value for that particular item was substituted for missing values. Finally, the three
data sets were reaggregated into a single set.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER THREE
Results
This chapter presents the results o f this study. The first section provides
descriptive information concerning the results obtained on the measures used, including
the number o f subjects, means, and standard deviations. The second section summarizes
inferential statistical tests o f each o f the research questions. In a final section, results o f
several additional analyses will be summarized.
Descriptive Statistics
The first step in data analysis was to examine relationships between and among the
variables. Table 1 summarizes between-variable correlations for the entire data set. It also
provides the mean and standard deviation for each measure. It shows, for example, that
self-monitoring correlated most highly with dating confidence (r =.22), assertiveness (r
= 20), and reactance (r =.24). With respect to the several attachment measures, it
correlated most highly with close attachment (r = 13).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
T ab let
Descriptive Statistics
1 2 3 4 5 6 7 8 9 10 11
1 Gender
2 Age
3 Anxious Attachment .06 .04
4 Close Attachment -.09 -.02 -.08
5 Dependent Attachment -.05 .00 -.44“ .18”
6 S elf Monitoring -.14” -.01 -.08 .13* .11*
7 Reactance -.05 .06 -.03 .06 .01 2 4 "
8 Dating Confidence -.08 .07 2 1 " .13* -2 0 " 2 2 " .02
9 Age a t first intercourse -.06 .12* .09 .02 -.07 -.04 .04 .10*
10 Frequency o f sex per month .09 .16“ .02 -.01 .08 .06 .02 .05 -.07
II Number o f sexual partners -26” .2 3 " -.04 .00 .04 -.01 -.10* .03 -.1 7 " 2 2 "
12 Number o f steady relats. .18” .1 3 " .03 .06 .03 -.02 .01 .04 -.11* .09 .1 6 "
* = £ < .05
**=£<.01
It is reasonable to assume that correlations between self-monitoring and the other
variables might differ by gender. Below, Table 2 summarizes between-variable correlations
for female participants only and Table 3 summarizes these relationships for males only.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2
Correlation Matrix: Female
42
1 2 3 4 5 6 7 8 9 10
1 Age
2 Anxious Attachment .04
3 C lose Attachment .03 -.08
4 Dependent Attachment -.06 <
*
0 0
I
2 0 “
5 SdFM omtoring .03 -.16“ .07 .1 3 '
6 Reactance .02 -.05 .1 3 ' -.01 .16*
7 D ating Confidence .10 .18“ .07 -.1 8 " 2 3 " .04
8 Age a t first intercourse 2 2 “ .1 4 ' .09 -.10 .07 .04 .17"
9 Frequency o f sex per month .1 8 " .03 -.01 .07 .11 .02 .10 .00
10 Number o f sexual partners .11 .01 .00 -.09 .02
0 0
o
»
.1 3 ' -2 6 " .06
II Number o f steady relats. .01 -.08 .02 .04 -.06 .01 .06 -.08 .09 2 6 "
*=£><.05
* * = j> < .0 1
Table 3
Correlation Matrix: Male
I 2 3 4 5 6 7 8 9 10
I Age
2 Anxious Attachment .04
3 Close Attachment .00 -.05
4 Dependent Attachment .09 -3 6 " .11
5 S elf Monitoring .05 .10 .1 9 ' .06
6 Reactance .14 .03 -.08 .05 .39**
7 D ating Confidence .11 2 8 " 2 2 “ -2 7 " .18* -.02
8 Age a t first intercourse .11 .05 -.06 -.05 -.16 .03 .02
9 Frequency o f sex per month .14 -.01 .03 .13 .03 .01 .00 -.12
10 Number o f sexual partners .1 8 ' -.12 .05 .13 .04 -.14 .05 -.1 7 ' 2 9 "
II Number o f steady relats. .13 .14 .1 7 ' .06 .09 .02 .05 -.11 .05 .10
* = jj< .05
* * = £><.01
In comparing Tables 2 and 3, it is apparent that different between-variable relations
occurred between the genders. For example, it is interesting to note that for females, a
significant correlation was obtained between Anxiety and Depend with self-monitoring but
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
not with Close; for males, however, a significant correlation was only found between
Close and self-monitoring. For females and males, there were significant correlations
between dating confidence with Anxiety and Depend though males also had a significant
correlation for dating confidence with Close whereas females did not. Age at first
intercourse correlated with Anxiety for females but not for males. Reactance correlated
with Close for females but not for males. Number o f sexual partners correlated with dating
confidence for females but not for males. Number o f steady relationships correlated with
Close for males but not for females. Also, number o f steady relationships correlated w ith
number o f sexual partners for females, but not for males.
The next set o f analyses were designed to examine whether there might be a
curvilinear relationship between self-monitoring and any o f the other variables. This
followed the finding by (Miller et al., 1988) that for some variables (namely, adjustment),
people who exhibited moderate levels o f self-monitoring exhibited higher levels o f those
variables than did people who were either high or low in se lf monitoring. To test this
possibility, participants were first grouped into those who scored high (N=138), medium
(N=142), or low (N=l 19) on the self-monitoring scale. Table 4 summarizes descriptive
information for participants according to self-monitoring status (high, medium, low) and
gender for each o f the several dependent measures.
Next, low and high self-monitors were collapsed together so that there were only
two groups (Medium SMs; High & Low SMs). In a series o f ANOVAs in which each o f
the other variables listed in Table 4 was treated as a dependent variable, no statistically
significant findings were obtained. Therefore, it was reasonable to assume that no
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
curvilinear relationship was present between self-monitoring and these several variables.
All subsequent analyses were based then on the assumption that self-monitoring was a
linear variable.
Table 4
Summary o f Statistical Findings: V ariables b v Self-M onitoring G roups
r n m
Low S elf M onitors Medium S elf Monitors High S elf M onitors
Gender Female Male Total Female Male Total Female Male Total
N = 6 S N = 54 AT=//9 N=87 N=55 A 1=142 N =I03 N = 3 5 U=138
Age
M
16.68 1837 17.54 16.82 17.84 1731 1632 1837 17.44
SD 1.6 2.79 2.41 1.43 3 3 2.49 138 4 3 6 2.63
Anxious Attachment
M 18.01 1737 17.67 16.76 18.1 1738 16.18 16.87 1635
SD 5.48 5.06 5.29 4.03 3.67 35)4 4.97 5.12 5
Close Attachment
M
18.88 17.65 ISJ3 18.7 1932 18.9 1939 19.46 19.85
SD 2.72 38 3 2.82 3.09 3.01 3.06 3 3 7 3.17 332
Dependent Attachment
M 1733 1738 1731 18.11 1739 173 1 18.44 183 18.46
SD 3.45 3 6 7 3.11 2 3 6 2 3 5 2.81 3.12 3.06 3.1
Dating Confidence
M 1231 1235 1238 1385 1235 12.65 14.07 12.82 13.75
SD 3.79 4.14 3.93 4.44 3.83 431 3.91 4 3 6 4.11
Intercourse Frequency
M
8 3 7 11.87 95)4 7 3 3 8 3 5 7.62 9 3 9 1135 10.38
SD 8.94 10.45 9.76 8.79 6.04 7.86 9.06 17.14 11.55
Reactance
M 1334 1332 13.44 14.04 133 13.99 1532 1534 1538
SD 3.41 3.45 3.42 3.1 2.68 2.93 3 3 6 3.05 3.74
Number o f Steady Reis.
M 2.11 3 4 6 237 2 3 5 333 2.63 1.86 3 7 5 2.09
SD 139 13 1.95 133 3 4 8 23 134 2.72 1.77
Number o f Sexual Partners
M 2.77 7 3 2 5.1 36 1 6 3 7 4.14 3 4 8 8 33 3.96
SD 3 3 9 14.01 10.03 2.75 132 8.09 3 1 4 1731 9.13
Age at First Intercourse
M 14.03 14.8 14.4 14.6 14 1435 14.65 13.45 14.34
J D _ 2 3 8 2 3 2.3 2 3 5 4.17 334 136 3.08 2.1
Note. The analyses for age and for the attachment scales were conducted via multiple regression, with self
monitoring treated as the criterion (continuous) variable.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
Research Questions
The research questions were first analyzed for the entire group and then analyzed
separately by gender and then by ethnicity.
Research Questions: Entire Group
First the questions “D oes a g e p red ict self-m on itorin g sty le? ” and “D oes
a ttach m en t d im en sion p red ic t self-m o n ito rin g style?” were examined by running a
step-wise multiple regression with self-monitoring as the criterion variable and the three
attachment dimensions and age were the four predictor variables.
The attachment dimension Close was a statistically significant predictor o f self
monitoring style for the entire group o f subjects, F (1, 369)=6.451, p< O il. None o f the
other variables; Anxiety, Depend, and age, were statistically significant in predicting
subjects’ self-monitoring level.
The next research question addressed for the entire data set was “D oes self
m on itorin g sty le p red ict sex u a l b eh a v io r an d d atin g co n fid en ce? ” To test this, a 3
(levels o f self-monitoring: high; medium; low) X 2 (gender) MANOVA was conducted,
with the four sexual history questions and dating confidence scores as the dependent
variables.
A significant effect was obtained for self-monitoring and frequency o f sexual
intercourse per month, F(2,613.662)=3.191, p< 042. However, no significance was
obtained for dating confidence F(2,77.563)=2.326, p< 099, number o f steady relationships
F(2,16.017)=2.082, p<. 126, number o f sex partners F(2,48.527)=.312, g<..732, or age at
first intercourse F(2,6.933)=.533, p< 587.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
Post hoc tests (Tukey) were run with alpha set at .05 to determine where the
significant between-group differences were for the three self-monitoring levels and
frequency o f sex per month. Results show that high self-monitors were significantly
different from medium self-monitors, with mean number o f sex acts o f 10.375 and 7.619
respectively.
There was a significant efiect for gender and number o f sex partners F (l,
2255.25), p<000 and number o f steady relationships F (l, 52.022), gc.000. Males had
more sex partners (M= 7.4) than females (M= 2.6) and more steady relationships (M= 2.8)
than females (M= 2.0). There was an interaction effect for self-monitoring and gender with
age at first intercourse, F(2, 51.837)=3.988, p<.019.
The last question for the group as a whole was “Does an individual’s self
m onitoring style predict reactance?” This was tested by performing an ANOVA with
self-monitoring (high; medium; low) as the independent variable. A significant effect was
found F(2,261.851)=11.501, p<.000.
Post hoc tests (Tukey) indicated that high self-monitors differ in reactance style
(M= 15.38) from both low (M= 13.44) and medium self-monitors (M = 13.99). Low and
medium self-monitors did not differ from one another statistically.
Research Questions: Analysis for Men: for Women
Next, the same questions were addressed separately by gender in order to answer
the fourth research question, “Does gender predict different relationships am ong the
variables?”. The questions concerned whether age and the attachment dimensions predict
self-monitoring. For females, a significant effect was found for Anxiety, F (l, 351.396)=
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
6 .8 5 7 , g<.009. T-tests indicated no significance for Close, Depend, and age w ith this
group. For males, a significant relationship was found for Close, F (l, 339.870)= 5.486,
£<.021. T-tests indicated no significance with this group for Anxiety, Depend, and age.
Table 5 summarizes the significant differences between male and female participants.
Table 5
Summary o f Gender- Specific Relationships Between Self-Monitoring and D ependent
Variables.
Female Male
Age
Anxiety *
Close ♦
Depend
Reactance
Dating
confidence
Age at first
intercourse
♦
Frequency per
month
if o f
partners
♦
ff o f
steady rels.
♦
Self
monitoring
♦
* = statistically significant relationship
Another research question not yet answered above was, “A re there gender
differences in self-m onitoring style?” For this, an ANOVA was run with gender as the
independent variable and self-monitoring as the dependent continuous variable. Results,
F (l,1 8 1 )= 8 .9 9 , £<.003, were statistically significant with women scoring higher (M=
4 7 .8 1 ) in self-monitoring than men (M= 45.64).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
Because male participants were approximately 2 years older than female
participants, an ANCOVA then was performed with age as the covariate, to control for
age. Gender continued to demonstrate a significant effect when age was controlled,
F(l,458.)=7.968, p<005. There was no statistically significant interaction effect for age
and gender, F(l,2.412)=.042, p<837.
Additional A nalyses
Several additional analyses were run. It was hoped that the results might provide
useful information for interpreting the primary findings that had been obtained.
R esearch Q uestions: A nalyses for African-Am erican, Euro-Am erican. and Latino
Participants
The main research questions that drove this study were addressed separately by
ethnic group (African-American; Euro-American; and, Latino). Whereas most research on
self-monitoring has been conducted on primarily-white college students, this sample was
multi-ethnic and primarily Latino. Therefore, it was reasonable to wonder whether
ethnicity affected the pattern o f relationships among the several variables o f interest in this
study.
The first two questions concerned the usefulness o f adult attachment scores and
age in predicting self-monitoring. For the African-American group, a significant overall
effect was found, F (4 ,869.608)= 5.391, p<.001. Follow-up t-tests indicated that age,
t=3.540, p<.001, and Depend, t=2.479, p<016, were both significant predictors o f self
monitoring for this ethnic group.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
For the Latino participants, the same regression indicated an overall significant
effect for age and the attachment dimensions with self-monitoring, F (4 ,559.137)= 2.688,
g<.032. Follow-up t-tests indicated that Anxiety, t=-2.089, p<038 was the only predictor
o f self-monitoring for this ethnic group.
The regression for the last group, Euro-American, did not show any statistically
significant relationships between the attachment dimensions, age, and self-monitoring.
For the research questions concerning the possible relationships between self
monitoring and the variables o f sexual behaviors, dating confidence and reactance,
separate regressions again were run for each o f the three ethnic groups.
For the African-American group, overall significance was found, F(6, 769.159)=
2.917, £<.016. Follow-up t-tests for this group showed significant relationships with self-
monitoring for dating confidence, t=3.145, £<.003, and for frequency o f sex per month,
fr=2.674, £<.010.
For the Latino group, overall significance was found, F(6, 834.334)= 2.709,
£<.014. Follow-up t-tests for this group indicated a significant relationship with self
monitoring for dating confidence, t=3.086, £<.002, and for reactance, t=2.518, £<.012.
For the Euro-American group, overall significance was found, F (6 ,263.868)=
5.264, £<.001. Follow-up t-tests indicated significant relationships only for reactance,
fr=3.730, £<.001.
Table 6 summarizes the significant between-group differences as described above.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
Table 6
Summary o f Ethnic-Specific Relationships Between Self-Monitoring and Dependent
Variables.
African-
American
Euro-
American
Latino
Age *
Anxiety *
Close
Depend ♦
Reactance * *
Dating
confidence
* *
Age at first
intercourse
Frequency
per month
*
# o f
partners
# o f
steady reis.
* = statistically significant relationship
Adult Attachment as Predictors o f Sexual Behavior and Dating Confidence
The first o f these analyses examined whether attachment dimension might predict
sexual behavior and dating confidence. This was addressed by conducting five separate
regression analyses, one each for the sexual behavior and dating confidence variables. In
each case, the three attachment scores were the predictors.
The regression analysis for dating confidence was statistically significant
F(3,577.508) =12.240, p<.000. Anxiety t=2.679, p<.008, and Depend t=-3.094, p<.002
each proved to be a useful predictor. Therefore, individuals who feel able to be close with
others have higher dating confidence scores. Those who have higher scores on Anxiety
also have higher scores for dating confidence. Converse^, individuals who scored high for
their feelings o f being able to depend on others scored lower on dating confidence.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 1
With respect to the other regression analyses, no statistical significance was
obtained for age at first intercourse F(3,27.887)= 1 .426, p<235, frequency o f intercourse
per month F(3,411.962}=L411, p<.239, number o f sexual partners F(3,77.142)=.312,
2<.817, o r for number o f steady relationships_F(3,10.314)=863, p<.460.
Gender differences were found for three o f the sexual behavior questions when a 2
(gender) by 3 (self-monitoring) MANCOVA was conducted with age controlled for. For
age at first intercourse, a significant effect was obtained, F=(l,25.67)=4.02,p<.045 with
men (M = 14.15) being younger on average than women (M= 14.46). A significant
relationship was found for number o f sexual partners F(l,1300.53)=17.13,p<.000 with
men having more partners (M = 7.4, versus 2.6). Men also had significantly m ore steady
relationships, F(l,32.26)=8.44,p<.003 (Ms. respectively, o f 2.8 and 2.0). An interaction
effect was found for the categorical variable o f se lf monitoring with gender on the
dependent variable o f age at first intercourse F(2, 51.66)= 4.05, p< 018.
Because this was a multi-ethnic sample, it was reasonable to wonder whether
ethnicity was related to se lf monitoring style. This was addressed by running an ANOVA
for three ethnic groups (African-American, Latino, and Caucasian) with the continuous
self-monitoring variable. No statistically significant between-group differences were
obtained, F(2,152)=2.725, p<.067.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
CHAPTER FOUR
Discussion
This chapter will consider the implications o f the findings and relate them to the
existing literature. It also will address the study’s limitations, then make suggestions for
further research.
The Results in the Context o f Existing Literature
This was the first study to examine adult attachment as a predictor o f self
monitoring style. I found that participants who scored higher in the attachment dimension
o f Close (Le., they endorse greater feelings o f being able to feel close to others) exhibited
higher levels o f self-monitoring.
HSMs are described in the literature as possessing more pro-social behaviors,
being more gregarious, and oriented to pleasing others. The motivation for this behavior,
however, has not been adequately explained. There are any number o f possible motivators,
including feelings o f security, social ambition, narcissism, or even social anxiety. In this
case, the high self-monitoring behavior o f the participants appears to be connected to
feelings o f being able to be close to others; a quality associated with secure attachment.
This does not rule out the possible influence o f other factors. However, it does provide
important new theoretical information.
There are several ways to explain this finding. One possibility is that people with
higher SM scores are most like people with secure attachments. People with lower scores
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
on the SM spectrum would be expected then to have most difficulty with insecure
attachment.
Another explanation is based on Miller and Thayer’s (1988) study on the
adjustment o f high, medium, and low self-monitors that indicated those a t the extreme
ends o f the SM spectrum have the lower adjustment scores. Based on the findings o f
Miller et al. wherein SM scores were trichotomized, it would be expected that HSMs, who
feel most comfortable in their ability to be close to others, might be experiencing an
inflated sense o f confidence about their social connections. Although this is a likely
explanation, the concept o f a middle self-monitoring group was not borne out by the
finding s o f this study. Therefore, all further conclusions are based on Snyder’s (1974)
original idea that the scale discriminates high from low self-monitors.
According to other research (Le, Tunnel, 1980), HSMs tend to rate themselves
higher on positive personality dimensions than do their acquaintances. This may predict
that in feeling especially close to others, HSMs may have an exaggerated sense o f
affiliation whereas others around them may rate their experience o f closeness with these
HSMs as being less close. So, this may mean that people lower in SM have a more
accurate estimation o f closeness than those higher in SM.
However, when the data set was split for gender, Anxiety was a significant
predictor o f self-monitoring for females, bot not for males. That is, only Anxiety predicted
self-monitoring scores for females and the lower Anxiety scores, were associated with
higher self-monitoring scores. In this split analysis, only Close continued to be predictive
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
54
o f self-monitoring for males. This would mean higher Close scores were associated with
higher SM scores for males.
Depend did not predict self-monitoring style. Based on the C ollins and Read
composites, secures would be expected to score high on both Close and Depend whereas
scores for anxious-ambivalents would be expected to be uniquely high for Anxiety and
scores for avoidants to be low on all three dimensions. A t this point, based on the C ollins
and Read findings, it would seem that the high self-monitor is most like the secure
individual. Because Anxiety and Depend did not predict self-monitoring style for the entire
group, it is unclear which attachment style is most similar to those higher and lower in
SM. In fact, if higher scores on Anxiety and Depend in combination with high Close
scores were determined by Collins and Read to be characteristic o f the anxious-ambivalent
style o f attachment.
Interestingly, it was noted by Collins and Read (1991) that partner satisfaction was
best predicted, for females, tty the extent to which their partner was able to be close (high
Close scores). For males, the best predictor o f their dissatisfaction with the relationship
was their partner’s level o f anxiety about being abandoned (high Anxiety scores). Because
high self-monitoring was found in this study to be associated with high Close scores for
males and lower Anxiety scores for females, it would appear that HSMs possess more
desirable traits to the opposite sex in terms o f predicting satisfaction in a relationship.
Perhaps this is more evidence that those higher in SM have more secure
attachment styles. What may be more appealing to the opposite sex in creating satisfying
relationships appears to be different for men and women. It is possible, then, that women
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 5
who are less anxious about abandonment are m ore secure and display higher SM more
secure attachment styles and display higher SM behaviors, behaviors. Similarly, men who
have the ability to feel closer to others, may have a more secure attachment style and
higher SM behaviors.
Attachment, self-monitoring. and sexual behavior/dating confidence
To better clarify the attachment-SM relationship, exploratory analyses were
conducted for relationships among sexual and dating behavior, SM scores, and attachment
scores. The literature on the topics o f sexual and adult relationships with SM and
attachment predict HSMs would have what was term ed an “unrestricted” approach to
sexual interactions (Snyder et aL, 1986); having more sex partners and more frequent
sexual contacts, and a greater number o f steady relationships than those lower in self
monitoring.
In this study, none o f these sexual behaviors (number o f sexual partners, number o f
steady relationships, age at first intercourse, and frequency o f sex per month) were
predicted by any o f the attachment dimensions (Close, Depend, and Anxiety). However,
the attachment dimensions o f Close and Anxiety did predict greater dating confidence
whereas Depend predicted less dating confidence.
It is intuitively understandable that feeling a greater ability to be close to others
would predict dating confidence. Perhaps greater anxiety, if not overwhelming, may
motivate people to master their insecurities about social connections. In analytic theory,
anxiety is not only seen as a neurotic defense, it is also viewed as necessary to survival
under certain circumstances. That the dimension o f Anxiety would predict dating
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
confidence may be akin to the way in which anxiety can be useful in giving impetus to
achieve one’s goals; hence, the exceptional presentation o f an anxious performing artist
(Fraiberg, 1959).
It is difficult to explain why higher Depend scores would predict lower dating
confidence. According to Collins and Read’s (1991) findings, Depend is associated with
Close in comprising “Function 1” which differentiates avoidant from secure and anxious-
ambivalent attachment types. Anxiety comprises “Function 2”, differentiating anxious-
ambivalent from secure ami avoidant types. Because o f this, it is possible that Depend, in
isolation, would not be a factor necessarily associated with a confidence in dating.
Although feeling one can depend on others may seem as if it would predict greater dating
confidence, it may mean that this dimension on its own identifies a person’s ability to feel
vulnerable, and hence, a negative relationship with the dating confidence measure used in
this study.
There was a significant relationship between self-monitoring and frequency o f
sexual intercourse per month. People scoring higher in self-monitoring reported more
frequent sexual contacts per month. These findings reinforce previous research (e.g.,
Snyder et al., 1984) that suggests people higher in SM are more sexually active in general
than those lower on the SM spectrum.
How these results together correspond with the existing literature on attachment
and relationship styles, and with how self-monitoring and the attachment dimensions are
related is complex. First, it appears that attachment style is unrelated to a person’s sexual
behavior. Yet it is related to one’s feelings o f being able to be attractive and confident in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
dating situations. Conversely, one’s self-monitoring score predicts the amount o f sexual
contact a person has in a month, but does not predict how successful that person sees him
or herself to be hi dating situations. Therefore, the relationships found in this study
between attachment, SM, and sexual behavior suggest that attachment may influence
one’s feelings about dating while self-monitoring would be a mediator for sexual behavior.
So, whether one has attachment dimensions indicating security (high Close and Depend,
low Anxiety) or insecurity (high Anxiety or low Close and Depend) in one’s attachment,
sexual behavior would be about equal except if one is also a HSM, then frequency o f sex
per month is increased. As well, those people who are highest in their ability to feel close
to others, tend to be higher in self-monitoring. People higher in this ability to feel close to
others, according to the research, also tend to be the most secure (Collins & Read, 1990).
Is it possible, then, that HSMs are most secure, have higher dating confidence and more
frequent sexual contacts than do lower self-monitors?
Consider that Anxiety also predicted dating confidence whereas Depend, which
logically would seem to fit with dating confidence, had a negative relationship with that
construct. Although for females, higher Anxiety would predict lower SM scores, it may be
that some amount o f anxiety, if not excessive, may raise and individual’s SM score. This
fits with Collins and Read’s findings that some people from the secure group had high
Anxiety scores in addition to high Close and Depend scores. This made them appear more
anxious-ambivalent while some from the anxious-ambivalent group had low Close and
Depend scores with high Anxiety scores making them appear more avoidant.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 8
In summary, then, at least some aspects o f attachment do predict self-monitoring.
What was expected, that people w ith attachment dimensions that fit avoidant styles
(especially the dismissive-avoidant style) would adopt high self monitoring behavior was
not evident in these findings. This hypothesis was based on research which portrayed the
avoidant individual as one who is not particularly close to others, is not as affected
emotionally by the loss o f relationships, but is often successful in their own domain; not
relying on or allowing relationships to affect their individual accomplishments. In this way,
the idea was that such an individual might be adept at social interactions to get what they
need in order to succeed themselves.
What does fit the hypothesis is that the attachment groups don’t appear to be
discrete in their categorization with respect to self-monitoring. As both Collins and Read
(1990) and Bartholomew and Horowitz (1991) noted, there seems to be a group classified
as “insecure” that would be considered “anxious-avoidant”. The original hypothesis o f this
study was that two groups o f anxious-ambivalents would emerge. The idea was that some
o f the anxio us-ambivalents may demonstrate HSM behaviors as a way to compensate for
their insecurities. They may look like the people that were categorized as both secure but
had Anxiety scores like the anxious-ambivalent group. They would have high scores for all
three attachment dimensions. I f they were anxious but too preoccupied to exhibit HSM
behavior, they would appear like an anxious-avoidant (also termed fearfiil-avoidant), with
relatively lower Close, Depend, and SM scores.
What was found, that Close was the best overall predictor (based on the entire
data set) suggests that people high in SM have traits associated with a secure o r anxious-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 9
ambivalent attachment. For females, lower Anxiety scores being more predictive o f HSM
supports the idea that it is not an anxious-ambivalent type o f attachment (which also
characteristically has high Close scores) that is predictive o f HSM for that gender.
Reactance and self-monitoring
This study also demonstrated a significant relationship between the constructs o f
reactance and self-monitoring- The tw o constructs have been described in ways that would
suggest a logical relationship based on the kind o f social qualities they each measure (see,
e.g., Dowd, Milne, & Wise, 1991; Snyder & Gangstead, 1986). In this study, I found that
HSMs scored higher in reactance than did either MSMs or LSMs. This confirms the
hypothesis that HSMs would be more reactive based on their propensity to seek individual
success and independence.
Although this may seem counter-intuitive because HMSs also seem to be the most
likely to want to please others and “fit in” socially, those that are highest in SM on the
continuum as described above, probably have some element o f anxiety and desire a
position o f social advantage or to have a social edge. For the people lower in SM on the
continuum, the threat o f losing freedom or independence is not so great. It makes sense
that those who are so concerned about being close to others and influencing others would
also be most worried that someone else could take away what they feel defines them.
Attachment was not found to predict reactance. Therefore it appears that SM may be a
mediator for the attachment and reactance variables.
Age and gender
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
In the analyses for the entire data set, I did not find participant age to predict self-
monitoring level. Gender, though, did predict responses to three o f the sexual behavior
variables. The pattern was for men to have had their first sexual experience at a younger
age, to have a greater number o f sexual partners, and more steady relationships. Men did
not, however, report having significantly more frequent sex per month than the women.
Interestingly, I found women to score significantly higher on self-monitoring than
men, even after the age correction. This finding contradicts that o f other general SM
research that found no gender differences with the SM measure. It also contradicts the
finding o f Stewart and Carley (1984) who in their study, found that men scored
significantly higher in SM in their group o f participants than did the women. This
difference may be due to the population represented in this study. Rather than a group o f
undergraduate college students recruited through a university, this sample had quite a
different constitution. Participants in this study were mainly minority (Hispanic) with an
average education less than a high school graduate. The majority o f the women
participating were adolescents who were currently or recently pregnant. Although no
significant effect was found here for SM status and ethnic heritage, educational level and
pregnancy status may contribute to the different result found in this study.
Ickes and Barnes (1977), had suggested that certain SM behaviors might be
mediated by sex role. That is, expression o f some o f the behaviors described on the SM
scale may seem inappropriate for men to endorse and vice versa for women. They also
suggested that the higher in self-monitoring the individual, the more likely he or she was to
inhibit these seemingly gender-inappropriate responses. Therefore, in this study, where the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
original SM scale was modified and the sample was unique for age, ethnicity, and
education level, an inflated effect due to sex role stereotyped response styles may also be
the reason for this difference in gender results for SM style.
Ethnicitv
When the data set was split for ethnicity (African-American, Euro-American,
Latino), some different patterns in the relationships among variables w ere found. In the
African-American group, the older the person and their sense that they can depend on
others were significant predictors on self-monitoring. This is interesting in that neither the
entire data set nor the analysis by gender found significance for age or Depend. For this
ethnic group, a person’s ability to see others as dependable is associated w ith HSM
behaviors.
The Latino group had a significant relationship between Anxiety and self
monitoring, in that the lower Anxiety scores predicted higher SM scores. This type o f
relationship was also found for female participants (regardless o f ethnicity). For this ethnic
group then, being less anxious about a relationship was associated with people who
demonstrate HSM behaviors.
Because no significant relationships were found for attachment and age with the
Euro-American group, it appears that SM behavior for this group is not predicted by
attachment or age.
Ethnic-specific relationships were found between self-monitoring and several other
variables. Dating confidence was a significant predictor o f self-monitoring for both Latinos
and African-Americans. Among the African-American group, frequency o f sex per month
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
was increased with higher SM scores. B oth the Latino and Euro-American groups had a
significant relationship between reactance and self-monitoring scores. No significant
relationships were found with the sexual behavior/dating confidence variables for the
Euro-American group.
One conclusion is that SM is a culturally linked phenomenon. People from
different ethnic backgrounds may exhibit sim ilar SM behaviors (whether low o r high), but
they are brought about by different cultural influences.
Limitations
A particular strength o f this study was its use o f a community sample. This is an
important extension o f previous self-monitoring research, which has relied primarily on
samples o f college students. At the same time, this difference (in age; educational level;
income level; ethnic diversity) raises important cautions about linking these findings with
research that has relied on those college samples. For example, it may be that over time
and life experiences, people’s self-monitoring behaviors change. Snyder’s (1974) finding
that theater actors scored higher and psychiatric patients scored lower than university
college students on self-monitoring supports the need for caution in generalizing across
samples.
Another limitation o f this study was that it relied on a nonrandom sample. The
possibility o f systematic sampling bias was reduced by our strategy o f recruiting
participants at multiple sites that varied in type and in location throughout Los Angeles
County. Nevertheless, this is a caution that needs to be considered.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 3
Studies that rely on self-report risk the possibility that attitudes and behaviors
might be misreported. There was a particular risk in this study that participants might have
exaggerated their numbers o f pregnancies (e.g., as a self-presentation strategy to increase
their sense o r perception o f “manhood”). The fact that the data were gathered
anonymously would reduce that likelihood, but it still is possible that this effect was
present to some unknown degree.
This idea o f self-monitoring style changing over time would also fit the Snyder
and Simpson (1984) hypothesis that se lf monitoring is a reproductive strategy. This seems
especially salient with this study, for fertility and reproduction are central issues with this
sample.
The modified version o f the 18-item se lf monitoring scale developed for this study
may have affected the results. Although it correlated .73 with Snyder’s revised scale, small
differences in meaning may have affected outcome in undetermined ways.
Finally, the ethnic constitution o f this group was primarily Latino and therefore
different from the (mainly Caucasian) samples used in the other SM literature. The results
o f the separate analyses by ethnicity suggest that this was a salient concern for this sample.
Suggestions for future research
First, in keeping with the limitations set forth in the previous section, a study o f
attachment predicting self-monitoring using a more heterogeneous group in terms o f age
and ethnicity may help to better distinguish whether there are patterns o f self-monitoring
behavior across age groups and ethnic groups. Because Main et al. (1985) have noted that
attachment style is generally stable over time, it would be interesting to find out if SM
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
level changed as a person matured, thereby indicating a process whereby other
characterological behaviors related to attachment may change while attachment remains
intact.
Additionally, further investigation into the gender differences in self-monitoring
may assist in clarifying the nature o f gender differences with regard to self-monitoring
behaviors. For example, was there something about this particular sample (e.g., age range;
education level; pregnancy status) that would cause women to score higher than men? Or
was this simply an anomolous finding?
Recent dissertation research has focused on the counselor-in-training and self-
monitoring behavior to see if there are differences in therapeutic effectiveness and
empathy (see e.g., Squitieri,1994; Ireland, 1994). An investigation into the effect
attachment style has on counselor-trainees may reveal different strengths and weaknesses
o f trainees with differing attachment styles.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
References
Ainsworth, M. D. S., Blehar, M. C., W aters, E., & Wall, S. (1978). Patterns o f
attachment: A psychological study o f the strange situation. Hillsdale, NJ:
Erlbaum.
Ainsworth, M. D. S. (1989). Attachments beyond infancy. A m erican Psychologist. 44.
709-716.
Ajzen, I., Timko, C. & White, J.B. (1 9 8 2 ). Self-monitoring and the attitude-behavior
relation. Journal o f Personality and Social Psychology. 4 2 . (3 ). 426-435.
Bartholomew, K. & Horowitz, L.M. (1991). Attachment styles among young adults: a
test o f a four-category model. Journal o f Personality and Social Psychology. 61.
(2). 226-244.
Brennan, K.A., & Morris, K.A. (1997). Attachment styles, selfesteem , and patterns o f
seeking feedback from romantic partners. Personality and Social Psychology
Bulletin. 23. ( 11.23-31.
Bowlby, J. (1969) Attachment and loss: Attachment. New York: Basic Books.
Bowlby, J. (1988). A secure base: Clinical applications o f attachment theory. London:
Routledge.
Briggs, S.R., Cheek, J.M ., & Buss, A.H. (1980). An analysis o f the self-monitoring
scale. Journal o f Personality and Social Psychology. 38. (41.679-686.
Buss, D. M. (1985). Human Mate Selection. Am erican Scientist. 7 3 .47-51.
Buss, D. M. (1994). The Evolution o f Desire. New York: Basic Books.
Bylsma, W.H., Cozzarelli, & Sumer, N. (1997). Relation between adult attachment
styles and global self-esteem. Basic and Applied Social Psychology. 19 . (11.1-
16.
Cassidy, J., & Berlin, L. (1994). The insecure/ambivalent pattern o f attachment: theory
and research. Child Development. 6 5 .971-991.
Collins, N.L., & Read, S.J. (1990). Adult attachment, working models, and relationship
quality in dating couples. Journal o f Personality and Social Psychology. 5 8 .4,
644-663.
Danheiser, P.R., & Graziano, W. G. (1982). Self-monitoring and cooperation as a self-
presentational strategy. Journal o f Personality and Social Psychology. 42. (3).
497-505.
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
66
Dawis, R. V. (1992). The individual differences tradition in counseling psychology.
Journal o f C ounseling Psychology. 39. (1). 7-19.
Dowd, T.E., & Wallbrown, F. (1993). M otivational components o f client reactance.
Journal o f C ounseling & Development. 7 1 .533-538.
Dowd, T.E., Milne, C.R., & Wise, S. L. (1991). The therapeutic reactance scale: a
measure o f psychological reactance. Journal o f C ounseling & Development.
69, 541-545.
Dowd, T. E., Trutt, SD ., & Watkins, C D .Jr. (1992). Interpretation style and reactance
in counselor’s social influence. Psychological Reports. 7 0 .247-254.
Ellis, R.E. (1988). Self-monitoring and leadership emergence ingroups. Personality and
Social Psychology Bulletin. 14. (4). 681-693.
Feeney, J.A., & Noller, P. (1990). Attachment style as a predictor o f adult romantic
relationships. Journal o f P ersonality and Social Psychology. 58. (2), 281-291.
Fox, N.A. (1995). O f the way we were: adult memories about attachment experiences
and their role in determ ining infant-parent relationships: a commentary on van
Ijzendoom (1995). Psychological B u lletin . 117. (3). 404-410.
Fraiberg, S. E L (1959). The Magic Years. New York: Fireside.
Gang stead, S. & Snyder, M. (1985). ‘T o carve nature at its joints”: on the existence o f
discreet classes in personality. Psychological Review. 92. (31.317-349.
Garbin, C.P., Robertson, E.A., & Bernstein, I.H . (1986). An analysis o f the internal
structure o f Levenson and Gottman’s social competency scale and suggestions
for its use. Journal o f Consulting and Clinical Psychology. 54. (31.406-407.
Glick, P., DeMorest, J.A., & Hotze, C.A. (1988). Self-monitoring and beliefs about
partner compatibility in romantic relationships. Personality and Social
Psychology Bulletin. 14. (3). 485-494.
Hamish, R. J., Abbey, A., & DeBono, K.G. (1990). Toward an understanding o f “the
sex game”: the effects o f gender and self-monitoring on perceptions o f sexuality
and likability in initial interactions. Journal o f Applied Social Psychology. 20.
(161. 1333-1344.
Hazaan, C., & Shaver, P. (1 9 8 7 ). Romantic love conceptualized as an attachment
process. Journal o f P ersonality and Social Psychology. 52. (31. 5 1 1 -5 2 4 .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
Hendrick, C., & Hendrick, S. (1986). A theory and method o f love. Journal o f
Personality and Social Psychology. 50. (21.392-402.
Howells, G.N. (1993). Self-monitoring and personality: would the real high s e lf
monitor please stand up? Journal o f Social Behavior and Personality. 8, (11. 59-
72.
Ickes, W., & Barnes, R.D. (1977). The role o f sex and self-monitoring in unstructured
dyadic interactions. Journal o f Personality and Social Psychology. 35 . (51. 3 1 5 -
330.
Kagan, J., Reznick, S., Clarice, C., Snidman, N., & Garcia-Koll, C. (1984). Behavioral
inhibition to the unfamiliar. Child Development. 5 5 .2212-2225.
Kelly, A , & Nauta, M.M. (1997). Reactance and thought suppression. Personality and
Social Psychology Bulletin. 23. (111. 1123-1132.
Lassiter, G.D., Stone, JJ ., & W eigold, M.F. (1988). Effect o f leading questions on the
self-monitoring/memory correlation Personality and Social Psychology
B ulletin 1J, (4), 537-545.
Lennox, R.D., & Wolfe, R.N. (1984). Revision o f the self-monitoring scale. Journal o f
Personality and Social Psychology. 46. (61.1349-1364.
Levenson, R.W., & Gottman, J.M . (1978). Toward the assessment o f social
competence. Journal o f Consulting and Clinical Psychology. 46. (31.453-462.
Levy, K.N., Blatt, S.J., & Shaver, P.R. (1998). Attachment styles and parental
representations. Journal o f Personality and Social Psychology. 74. (21.407-419.
Ludwig, D., Franco, J.N., & Malloy, T.E. (1986). Effects o f reciprocity and self
monitoring on self-disclosure with a new acquaintance. Journal o f Personality
and Social Psychology. 50. (61. 1077-1082.
Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood, and
adulthood: A move to the level o f representation. Monographs o f the Society for
Research in Child Development. 50. (1 & 2), 66-104.
Marin, G. & Marin, B.V. (1991). Research with H ispanic populations. Newbury Park,
CA; Sage Publications.
McCann, C.D., & Hancock, R.D. (1983). Self-monitoring in communicative
interactions: social cognitive consequences o f goal-directed message
modification. Journal o f Experimental Social Psychology. 19. 109-121.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
68
McFall, R.M., & Townsend, J.T.(1998). Foundations o f psychological assessment:
implications for cognitive assessment in clinical science. Psychological
Assessment. 10. (41.316-330.
Mill, J. (1984). High and low self-monitoring individuals: their decoding skills and
empathic expression. Journal o f Personality, 52.14'). 372-388.
Miller, M X ., & Thayer, J.F. (1988). On the nature o f self-monitoring: relationships with
adjustment and identity. Personality and Social Psychology Bulletin. 14. 131.
544-553.
Mikulincer, M. (1998). Adult attachment style and individual differences in functional
versus dysfunctional experiences o f anger.Journal o f Personality and Social
Psychology, 74, (2), 513-524.
Newcomb, M.D. (1985). Sexual experience among men and women: associations
within three independent samples. Psychological Reports. 5 6 .603-614.
Shaffer, D.R., Ogden, J.K., & Wu, C. (1987). Effects o f self-monitoring and prospect o f
future interaction an self-disclosure reciprocity during the acquaintance process.
Journal o f Personality. 55. 111. 75-96.
Shaffer, D.R., Smith, J.E ., & Tomarelli, M. (1982). Self-monitoring as a determinant o f
self-disclosure reciprocity during the acquaintance process. Journal o f
Personality and Social Psychology. 43. (1 1 .163-175.
Sharpsteen, D.J., & Kirkpatrick, L.A. (1997). Romantic jealousy and adult romantic
attachment. Journal o f Personality and Social Psychology. 72. (31.627-640.
Simpson, J.A. (1990). Influence o f attachment styles on romantic relationships. Journal
o f Personality and Social Psychology. 59. (51.971-980.
Snyder, M. (1974). Self-monitoring o f expressive behavior. Journal o f Personality and
Social Psychology. 30. (41.526-537.
Snyder, M. (1983). The influence o f individuals on situations: implications for
understanding the links between personality and social behavior. Journal o f
Personality. 5 1 . (31. 497-515.
Snyder, M., Berscheid, E ., & Glick, P. (1985). Focusing on the exterior and the interior:
two investigations o f the initiation o f personal relationships. Journal o f
Personality and Social Psychology. 48. (61.1427-1439.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
Snyder, M., & Gangstead, S. (1982). Choosing social situations: two investigations o f
self-monitoring processes. Journal o f Personality and Social Psychology. 4 3 . (11.
123-135.
Snyder, M., & Gangstead, S. (1986). On the nature o f self-monitoring: matters o f
assessment, matters o f validity. Journal o f Personality and Social Psychology.
5L (1), 125-139.
Snyder, M ., & Monson, T.C. (1975). Persons, situations, and the control o f social
behavior. Journal o f Personality and Social Psychology. 32. (41.637-644.
Snyder, M., & Simpson, J. (1984). Self-monitoring and dating relationships. Journal o f
Personality and Social Psychology. 47. (61.1281-1291.
Snyder, M., Simpson, J.A., & Gangstead, S. (1986). Personality and sexual relations.
Journal o f Personality and Social Psychology. 51. (11. 181-190.
Stewart, A., & Carley, L. (1984). Personality characteristics o f extreme scorers on the
self-monitoring scale. Perceptual and M otor Skills. 58. 199-205.
Stroufe, L.A., & W aters, E. (1977). Attachment as an organizational construct Child
Development. 4 8 .1184-1192.
Tunnell, G. (1980). Intraindividual consistency in personality assessment: the effect o f
self-monitoring. Journal o f Personality. 48. (21.221-232.
Turner, RG. (1980). Self-monitoring and humor production. Journal o f Personality. 48.
£2), 163-172.
van IJzendoom, M.H. (1995). Adult attachment representations, parental
responsiveness, and infant attachment: a meta-analysis on the predictive validity
o f the adult attachment interview. Psychological Bulletin. 117. (31.387-403.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7 0
A PPE N D IX A
Self-M onitoring S ca le (R evised )
1. I am not good at mutating people.
Strongly agree Agree Neutral____D isagree___Strongly disagree
2. At parties and social gatherings, I find m yself doing o r saying the kinds o f
things that other people will like.
Strongly agree Agree Neutral____D isagree___Strongly disagree
3. I can only argue for ideas that I already believe in.
Strongly agree Agree Neutral_ _ _ D isagree___Strongly disagree
4. I can stand up and talk about things that I don’t know much about
Strongly agree Agree Neutral_ _ _ D isagree___ Strongly disagree
5. I often find m yself putting on a show to impress o r entertain others.
Strongly agree Agree Neutral____D isagree___Strongly disagree
6. I think I would make a good actor.
Strongly agree Agree Neutral____ D isagree_Strongly disagree
7. When I am in a group o f people, I am usually not the center o f attention.
Strongly agree___ Agree Neutral____D isagree___ Strongly disagree
8. I find that I act like a different person depending on the situation and who is
around me.
Strongly agree Agree Neutral Disagree Strongly disagree
9. I am not very good at making people like me.
Strongly agree Agree Neutral_ _ _ D isagree___ Strongly disagree
10. I am not always thinking and feeling what other people think I am.
Strongly agree___ Agree Neutral_ _ _ D isagree___ Strongly disagree
11. I sometimes find that I change my opinions (or the way I do things) in order
to please someone or get their help.
Strongly agree Agree Neutral____Disagree_Strongly disagree
12. I have thought about being an entertainer as a hobby o r career.
Strongly agree Agree Neutral_ _ _ Disagree___ Strongly disagree
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
13. I have trouble changing my behavior to please different people and different
situations.
Strongly agree Agree Neutral___ Disagree Strongly disagree
14. When I am at a party I am usually quiet and let other people keep the jokes
and stories going.
Strongly agree Agree Neutral___ Disagree Strongly disagree
15. I can look anyone in the eye and tell a lie with a straight face (if it is fo ra
good reason).
Strongly agree Agree Neutral D isagree Strongly disagree
16. I may fool people by being friendly when I really don’t like them.
Strongly agree Agree Neutral___ Disagree___Strongly disagree
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Child sexual abuse in a sample of male and female Hispanic and White nonclinical adolescents: Extending the reliability and validity of the Trauma Symptom Inventory (TSI)

PDF

Behavioral and demographic predictors of breast cancer stage at diagnosis

PDF

A comparison of values among counseling psychologists characterized as scientist-practitioners, practitioners, or scientists

PDF

A correlational study of feminist/womanist identity development and depression in women

PDF

Cognitive predictors of intention -behavior consistency in safer sex practices

PDF

Gender differences in motivation for sexual intercourse: Implications for risky sexual behavior and substance use in a university and community sample

PDF

Characteristics of college males who are likely to sexually harass women: A test of a mediated model.

PDF

Differences in the eating behaviors among elite, recreational, and non-athletic women on four comparative measures: Dependency, self-criticism, self-efficacy, and alexithymia

PDF

A model for developing congruence, empathy, level of regard, and unconditional positive regard in counselor trainees

PDF

An investigation of the psychological well-being of unaccompanied Taiwanese minors/parachute kids in the United States

PDF

Cognitive and non-cognitive factors as predictors of retention among academically at-risk college students: A structural equation modelling approach

PDF

Integration of science and practice: A collective case study of scientist -practitioner programs in counseling psychology

PDF

A cross-cultural comparison of marital power and dyadic adjustment among American, Indo-American, and East Indian dual-career and single-career couples

PDF

Assessment of racial identity and self -esteem in an Armenian American population

PDF

Counselor Trainees' Perceptions Of And Attitudes Toward Sexual Harassment In Relation To Feminist Beliefs

PDF

A comparison of analogical and quantitative methods for teaching mechanical functions in science education

PDF

Supervisor self -disclosure as a predictor of supervisee self-disclosure, satisfaction with supervision, and perceptions of supervisor credibility

PDF

Application of the graded response model to the revised Tennessee Self-Concept Scale: Unidimensionality, parameter invariance, and differential item functioning

PDF

Attachment style, interpersonal guilt, parental alcoholism, parental divorce and eating disordered symptomatology in college women

PDF

Sexual harassment on campus: Refinement of a measure of students' perceptions

Asset Metadata

Creator Dieterich, Susan Ellen (author)

Core Title Attachment as a predictor of self -monitoring and sexual behavior

Degree Doctor of Philosophy

Degree Program Education (Counseling Psychology)

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Education, Guidance and Counseling,OAI-PMH Harvest,psychology, behavioral,psychology, personality,psychology, social

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Goodyear, Rod (committee chair), [illegible] (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-552493

Unique identifier UC11354632

Identifier 9955069.pdf (filename),usctheses-c17-552493 (legacy record id)

Legacy Identifier 9955069.pdf

Dmrecord 552493

Document Type Dissertation

Rights Dieterich, Susan Ellen

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA