Journal of Social and Personal

Relationships
http://spr.sagepub.com/

Big Five personality traits and relationship quality: Self-reported,

observational, and physiological evidence
Ashley S. Holland and Glenn I. Roisman
Journal of Social and Personal Relationships 2008 25: 811
DOI: 10.1177/0265407508096697

The online version of this article can be found at:

http://spr.sagepub.com/content/25/5/811

Published by:

http://www.sagepublications.com

On behalf of:
International Association for Relationship Research

Additional services and information for Journal of Social and Personal Relationships can be found at:

Email Alerts: http://spr.sagepub.com/cgi/alerts

Subscriptions: http://spr.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

Citations: http://spr.sagepub.com/content/25/5/811.refs.html

>> Version of Record - Oct 14, 2008

What is This?

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

 Big Five personality traits and
relationship quality:
Self-reported, observational, and
physiological evidence
Ashley S. Holland & Glenn I. Roisman
University of Illinois at Urbana–Champaign

ABSTRACT
This report explored links between Big Five personality traits
and the quality of romantic relationships. Interpersonal func-
tioning was operationalized at three levels of analysis (self-
reported quality, observed emotional tone, and physiological
reactivity) in three samples (dating, engaged, and married
couples). Couples completed questionnaires about their own
and partners’ personalities and then discussed a disagreement
in their relationship while being physiologically monitored.
Two conceptual frameworks were examined; the first predicted
that personality would be consistently associated with all indi-
cators of functioning. The second framework predicted that
personality would be consistently linked to perceived quality
and trivially associated with observed emotional tone and
electrodermal/cardiac reactivity. Significant associations were
identified between personality and self-reported quality, and
only trivial associations with other indictors.

KEY WORDS: Big Five • dating couples • engaged couples • married

couples • observed quality • personality traits • physiological
reactivity • self-reported quality

This article is based on a Masters thesis completed by the first author in the Department of
Psychology at the University of Illinois at Urbana–Champaign. Data collection for the studies
involving engaged and married couples was supported by a grant from the Research Board
at the University of Illinois at Urbana–Champaign to the second author. The dating couples
study was sponsored by a grant by the National Science Foundation to R. Chris Fraley. All
correspondence relating to this article should be addressed to Glenn I. Roisman, Department
of Psychology, University of Illinois at Urbana–Champaign, 603 East Daniel Street, Cham-
paign, IL 61820, USA [e-mail: roisman@uiuc.edu]. Duncan Cramer served as Action Editor
for this manuscript.

Journal of Social and Personal Relationships Copyright © 2008 SAGE Publications

(www.sagepublications.com), Vol. 25(5): 811–829. DOI: 10.1177/0265407508096697

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

812 Journal of Social and Personal Relationships 25(5)

Social and personality psychologists have long been interested in under-

standing how personality traits influence the quality of adults’ romantic
relationships (Caspi, Roberts, & Shiner, 2005). Indeed, even a cursory review
of the literature reveals unambiguously that there is a reliable association
between self-reported personality traits and relationship quality and satis-
faction, as assessed through questionnaires (Karney & Bradbury, 1997;
Robins, Caspi, & Moffitt, 2000, 2002; Watson, Hubbard, & Wiese, 2000).
However, fewer studies have examined how personality traits may be
reflected in other indicators of interpersonal functioning, such as observed
quality (Donnellan, Conger, & Bryant, 2004; Donnellan, Larsen-Rife, &
Conger, 2005). In addition, no studies to our knowledge have examined the
association between personality traits and relationship quality as opera-
tionalized through patterns of physiological responding during dyadic
interactions, despite research that confirms that patterns of autonomic
response are intimately tied to interpersonal functioning (Gottman, Coan,
Carrere, & Swanson, 1998; Gottman & Levenson, 1988, 1992). In order to
address these gaps in the literature, this report explored links between the
Big Five personality traits and the quality of adults’ romantic relationships
utilizing a multi-method approach that operationalized interpersonal func-
tioning at three levels of analysis (self-reported quality, observed emotional
tone, and physiological reactivity during a dyadic interaction) in three
samples (dating, engaged, and married couples).

Measuring personality

A complicating issue involved with evaluating the previous literature in this

area is that researchers use different operationalizations and measures of
personality. Nonetheless, although different personality characteristics are
assessed across studies, there is some overlap. For example, Caspi et al.
(2005) state that there is increasing agreement among researchers about the
organization of higher-order traits (e.g., extraversion); however, there is less
agreement about lower-order traits that are included within the higher-order
dimensions.
The focus of the current analysis, therefore, was on higher-order trait
dimensions; namely, extraversion, neuroticism, conscientiousness, agreeable-
ness, and openness to experience. This Five Factor Model is a well-validated
approach to the measurement of personality traits (McCrae & Costa, 1997).
In this taxonomy, individuals high on extraversion or positive emotionality
are generally described as outgoing, animated, and lively. Individuals low on
extraversion or positive emotionality are quiet, submissive, and inhibited.
Highly neurotic individuals (i.e., individuals high on negative emotionality)
are anxious, moody, and insecure in relationships. Emotional stability and
flexibility describe individuals who score low on this trait dimension. Indi-
viduals who are highly conscientious are responsible, focused, and organ-
ized; individuals who are low on this trait are irresponsible, forgetful, and
unreliable. Agreeable individuals are thoughtful, kind, and supportive and

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

Holland & Roisman: Personality and relationship quality 813

disagreeable people are impolite, stubborn, and aggressive. The last trait,
openness to experience (sometimes called intellect), is the least-well under-
stood trait of the Big Five dimensions (Caspi et al., 2005) and reflects indi-
viduals who are inventive and creative.

Personality traits and relationship quality

A number of studies have examined which personality traits are most highly
correlated with relationship satisfaction. For instance, low levels of negative
personality traits such as neuroticism and negative emotionality have been
consistently associated with self-reports of relationship quality and satis-
faction (Caughlin, Huston, & Houts, 2000; Robins et al., 2002; Watson et al.,
2000). A meta-analysis (Heller, Watson, & Ilies, 2004) found that overall
neuroticism is the strongest personality correlate of marital satisfaction. In
addition, Karney and Bradbury (1997) have found that negative personality
traits such as neuroticism are associated with decreased satisfaction from
the onset of marriage. Kelly and Conley (1987) similarly found that neuroti-
cism in both husbands and wives was negatively associated with marital
satisfaction. Another study (Gattis, Berns, Simpson, & Christensen, 2004)
found that neuroticism was significantly higher in a sample of distressed
couples than in a sample of non-distressed couples. Finally, a study by
Caughlin et al. (2000) found that trait anxiety is associated with marital
satisfaction, and this association was largely explained by self-reports of
communication patterns between partners. Consequently, it seems clear
that the presence of highly neurotic individuals in relationships may have
important implications for self-reported relationship quality and satisfaction
(Robins et al., 2000).
Positive emotionality and other positive personality traits are also fre-
quently examined in conjunction with self-reported relationship quality.
Robins et al. (2002) found that positive emotionality was related to higher-
quality relationships and also low levels of negative relationship outcomes,
such as self-reported conflict and abuse. Likewise, Watson et al. (2000)
examined personality and relationship satisfaction in a sample that included
both married couples and a sample of dating couples. Specifically, they
found that extraversion was associated with satisfaction in married couples
and that conscientiousness and agreeableness were associated with satis-
faction in dating couples. In another study (Botwin, Buss, & Shackelford,
1997), conscientious wives had husbands who were more sexually satisfied
and conscientious husbands had wives who were more generally satisfied
with their relationships.
It is important to note that all of the results described above are based
on self-reported and informant-reported personality and quality/satisfaction
assessed using questionnaire methods. Recently, new research has emerged
that has utilized other indicators of relationship quality in addition to self-
reported quality, including observed quality (Donnellan et al., 2004, 2005).
In these path-finding studies, participants talked about agreements and

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

814 Journal of Social and Personal Relationships 25(5)

disagreements in their relationships during a laboratory interaction pro-

cedure. Trained observers rated both males and females on several types of
negative behaviors (e.g., hostility, angry coercion) in the context of their
interactions. Donnellan et al. (2004) focused on a sample of long-term
married couples. For the female participants, agreeableness, neuroticism, and
openness were associated with observed negative interactions. For the male
participants, only neuroticism was significantly associated with observed
quality. Another report drawn from the same study (Donnellan et al., 2005)
examined committed relationships, with a majority of married couples. In
this study, positive emotionality was negatively correlated, and negative
emotionality was positively associated, with negative interactions. Thus,
some evidence for an association between personality and observed quality
has emerged. However, surprisingly few researchers have incorporated
comparable measures of observed relationship quality into their research
designs. As such, these results await replication by other investigators.
In addition, while many current studies are now examining both partners
involved in romantic relationships, few researchers utilize partner-reports
of personality in their work (Botwin et al., 1997; Donnellan et al., 2004;
Kosek, 1996), despite the fact that a secondary informant can provide valu-
able validation of findings based entirely on self-report. A notable excep-
tion is Watson et al. (2000), who examined both self- and partner-reports
of personality. Their study assessed a sample of married and dating couples
and each individual was asked to rate his or her own personality and then
rate his or her partner’s personality. In addition, each individual rated their
own relationship satisfaction. In the dating couples sample, conscientious-
ness and agreeableness, irrespective of who provided the data on personal-
ity (self or partner), were associated with both the males’ and females’
satisfaction. Significant associations were also found for neuroticism and
extraversion. In the married couples sample, self-reports of extraversion,
agreeableness, and neuroticism were all associated with self-reports of satis-
faction, results that were corroborated by partners’ reports of personality.

The importance of other indicators of relationship quality

Although rarely examined in relation to personality traits, observed quality

and physiological responding are fundamental indicators of the quality of
adults’ romantic relationships. Observed quality is an important component
of relationship quality because it provides a window on romantic partners’
actual interactions. For example, Gottman and Levenson (1988) have
demonstrated that observed negative affect consistently differentiates satis-
fied from dissatisfied couples. Observational methods also provide a unique
opportunity to examine negative affect, positive affect, and conflict during
couples’ interactions. Observation may be especially important in terms of
assessing negative affect because it is possible that individuals in romantic
relationships may under-report negative behaviors in their relationship.
Researchers focused on romantic relationships have also utilized patterns
of physiological response as an additional component of interpersonal func-

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

Holland & Roisman: Personality and relationship quality 815

tioning and as a predictor of relationship stability (Gottman et al., 1998).

Gottman et al. (1998) proposed that the ability to physiologically self-
soothe and also soothe a partner during conflict interactions results in more
successful relationships. Gottman et al. (1998) found that in happy couples
(as opposed to unhappy or divorced couples) that the presence of humorous
wives was associated with decreases in their husbands’ heart rates during
interactions. In addition, it was also found that in happy couples, the
husbands’ own characteristics, such as de-escalating arguments related to
negative topics, validating their wives, and showing affection towards their
partners, were also successful in decreasing their own cardiac reactivity.
Furthermore, Fowles (1980, 1988) maintains that two antagonistic motiva-
tional systems whose activation can be inferred from unique physiological
correlates. Specifically, Fowles (1980) argued that engagement of the Be-
havioral Inhibition System, which has often been characterized as being
involved in the effort to inhibit behavior, is reflected in increases in electro-
dermal activity, whereas the Behavioral Activation System, which is involved
in approach-related (hyper)activation, is signaled by heart rate reactivity.
Using this theoretical framework we have recently shed light on the possible
developmental foundations of the patterns of autonomic response so
routinely observed among distressed couples in research on the psycho-
physiology of marriage and divorce (Gottman & Levenson, 1988) in that
adults insecure with respect to their attachment experiences appear to show
the largest increases in heart rate and skin conductance during their marital
and premarital interactions relative to resting conditions (Roisman, 2007).

Two conceptual frameworks

It seems plausible that personality traits influence the nature and quality of
adults’ romantic relationships. However, there are at least two different
ways of conceptualizing the nature of this association. In this study, two
different frameworks for understanding links between personality traits and
relationship quality were contrasted. As described earlier, robust associ-
ations between personality and self-reported relationship quality have been
found in previous studies and consequently, broad conclusions about the
nature of the association between personality and relationship quality have
been suggested. For example, Caspi et al (2005) state that there is “robust
evidence that early-emerging individual differences in personality shape
how individuals experience, interpret, and respond to the developmental
tasks they face across the life course” (p. 471), and more specifically that
“personality continues to be an important predictor of relationships in
adulthood” (p. 471). Consistent with this predominant view, the first frame-
work anticipates that personality will be broadly associated with interper-
sonal functioning. Thus, this framework would predict consistent effects of
personality on self-reported quality, observed emotional tone, and physio-
logical responding. A lack of previous literature limits the scope of predic-
tion in reference to specific physiological patterns of responding, although
it might be predicted that a negative trait, such as neuroticism, would create

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

816 Journal of Social and Personal Relationships 25(5)

stressors in the interaction which would be reflected in patterns of physio-

logical responding. In addition, a positive trait, such as agreeableness, may
calm or inhibit stressful encounters, which would result in less physiological
change from baseline. In this study, we specifically examine electrodermal
and heart rate reactivity, because both of these indicators have been consist-
ently associated with marital dissatisfaction and even divorce (Gottman et
al., 1998; Gottman & Levenson, 1992).
Although many studies investigating the association between personality
traits and relationship satisfaction have been conducted, the vast majority
examine only self-reports of satisfaction, only a few have focused on
observed quality (Donnellan et al., 2004, 2005), and none to our knowledge
have thus far examined autonomic reactivity during dyadic interactions. In
addition, the findings that exist with respect to observed quality are based
on results from a single sample (parents and their children involved in a
longitudinal project) and have not yet been replicated by other researchers.
Due to this limitation of the literature, it is worthwhile to explore a second
framework that suggests a more limited role of personality with respect to
relationship outcomes. This second framework predicts that personality
would be most strongly associated with self-reported quality and perhaps
only trivially associated with observed emotional tone and physiological
responding. Therefore, this second framework would predict that an indi-
vidual’s personality may primarily constrain his or her own evaluations of
the quality of his or her relationships, but may be less reflected in observed
interpersonal processes and physiological responding.
It seems reasonable that self-reported personality is commonly associated
with self-perceptions of one’s own relationship. In fact, it seems inevitable
that how an individual defines him or herself will affect how he evaluates
his or her relationships. However, it is important to investigate this associ-
ation across different levels of analysis (i.e., self-reported quality, observed
emotional tone, and physiological responding), which this study did. No
investigation has simultaneously examined all three of these indicators of
interpersonal functioning in a single study to more completely understand
the association between personality and relationship quality.
Toward that end, this study also utilized both self-reports of personality
and reports of personality completed by one individual about the other
partner (i.e., partner-reports). As described above, partner-reports have the
potential to provide corroboration of findings related to self-reported
personality. Independent of providing support for a single framework,
neuroticism was expected to emerge as an important personality trait in
predicting relationship quality in light of findings of previous literature. It
is also important to note that this is a secondary analysis that utilized data
from three different heterosexual couple types: exclusive dating, younger
engaged, and older married couples (Roisman, 2007; Roisman, Clausell,
Holland, Fortuna, & Elieff, 2008). Exploring associations across different
couple types may be important because previous research has primarily
focused on a single couple type. Although we believe it is possible that
systematic differences may exist between couple types, because of a lack of

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

Holland & Roisman: Personality and relationship quality 817

previous data we did not make specific predictions with respect to couple
type a priori.

Method

Participants
Three different samples of heterosexual couples were utilized for this study.
First, 109 dating couples (average length of relationship = 15.6 months,
range = 1 to 69 months) were recruited from a small town in the Midwest.
The men averaged 20.7 years of age (SD = 1.8, range = 18 to 25 years) and
the women averaged 20.1 years of age (SD = 1.5, range = 18 to 25 years).
Second, 50 engaged couples (between the ages of 18 and 30 years who were
not previously married; average length of relationship = 41 months, range
= 4 to 108 months) were also recruited from the same community. The
men averaged 22.7 years of age (SD = 2.7, range = 19 to 29 years) and the
women averaged 22.2 years of age (SD = 3.0, range = 18 to 30 years).
Finally, 40 married couples (50 years of age and older and married at least
15 years; average length of relationship = 379 months, range = 201 to 621
months) were again recruited from the same small town. The men averaged
57.9 years of age (SD = 7.6, range = 50 to 77 years) and the women averaged
55.8 years of age (SD = 6.2, range = 50 to 73 years). The three samples were
homogeneous in terms of ethnicity (dating = 75% Caucasian, engaged =
87% Caucasian, married = 97.5% Caucasian). The dating couples received
US$20.00 as compensation for their participation and both the engaged and
married couples received US$50.00 as compensation for their participation.

Procedure
Before arriving in the laboratory, all of the participants completed a packet
of self-report questionnaires about themselves, their significant other, and
their relationship in general, including a self- and partner-report on person-
ality. After providing informed consent, the participants were separated
and each was interviewed about their childhood experiences (see Roisman,
2007). Participants then completed several self-report measures, including
a questionnaire that listed 11 common relationship problem areas (e.g.,
money, communication, sex, etc.). Participants were informed that this
questionnaire would be the only questionnaire that their significant other
would see. Participants rated whether the problem area was currently a
problem in their relationship using a 10-point scale (where 1 = not a problem,
10 = is a serious problem).
After completing the problem area questionnaire, participants were re-
united to engage in a standard marital interaction task. Couples used the
problem area questionnaire to decide on a single problem area to discuss.
Participants were given 10 minutes to discuss and arrive at a solution to this
problem area in their relationship (i.e., the disagreement epoch). As a way
of “cooling down” following the disagreement discussion, couples then
spent five minutes discussing areas in their relationship they agreed about.

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

818 Journal of Social and Personal Relationships 25(5)

Physiological sensors measuring heart rate were attached to participants’

torsos and sensors measuring skin conductance levels were attached to
participants’ fingertips prior to the interaction. Physiological readings were
monitored second-by-second from an adjoining room during the inter-
action as well as during a three-minute rest period prior to the interaction,
which provided a baseline measure of responding. Just prior to the resting
baseline, a research assistant instructed participants to be silent and to
empty their minds of all thoughts, feelings, and emotions before beginning
the rest period. The interaction followed the rest period.

Apparatus

Physiological equipment. The acquisition system consisted of two Pentium

computers, Snapmaster software, and James Long Inc. bioamplifiers. This
system allowed continuous recording of physiological responses from both
participants during the interaction.

Video equipment. High-resolution color video cameras recorded the couples’

interactions. The video cameras were embedded within a bookshelf located
across the room from the participants who were seated on a couch. Micro-
phones were clipped to the participants’ clothing and were used to record
the conversation during the interactions.

Measures

Personality. Personality ratings were obtained using the NEO-Five Factor

Inventory (NEO-FFI), which was part of the pre-lab packet of question-
naires. The NEO-FFI measures the “Big Five” personality traits: neuroticism,
extraversion, openness, agreeableness, and conscientiousness. The NEO-
FFI contains 60 items and each personality trait was derived by computing
the mean of 12 different items. Participants completed one inventory about
their own personality characteristics (neuroticism: dating M = 2.7, SD = .72;
engaged M = 2.8, SD = .78; married M = 2.4, SD = .71; extraversion: dating
M = 3.6, SD = .55; engaged M = 3.5, SD = .54; married M = 3.5, SD = .54;
openness: dating M = 3.6, SD = .54; engaged M = 3.6, SD = .51; married M
= 3.4, SD = .49; agreeableness: dating M = 3.7, SD = .49; engaged M = 3.6,
SD = .53; married M = 4.0, SD = .43; conscientiousness: dating M = 3.7, SD
= .58; engaged M = 3.6, SD = .61; married M = 3.9, SD = .59) and a second
inventory about their partner and his or her personality. This second
personality inventory was created by re-wording the original NEO-FFI
items to probe about the partner’s personality (neuroticism: dating M = 2.6,
SD = .77; engaged M = 2.7, SD = .75; married M = 2.5, SD = .78; extraver-
sion: dating M = 3.6, SD = .62; engaged M = 3.5, SD = .52; married M = 3.4,
SD = .55; openness: dating M = 3.4, SD = .59; engaged M = 3.4, SD = .48;
married M = 3.2, SD = .54; agreeableness: dating M = 3.8, SD = .58; engaged
M = 3.8, SD = .59; married M = 3.9, SD = .58; conscientiousness: dating M
= 3.8, SD = .65; engaged M = 3.7, SD = .61; married M = 3.9, SD = .69). The

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

Holland & Roisman: Personality and relationship quality 819

NEO-FFI has well-documented reliability and validity. In this analysis, the

alphas for self-reports of personality ranged from .74 to .87 for the dating
couples, from .72 to .90 for the engaged couples, and from .69 to .90 for the
married couples. The alphas for the partner-reports ranged from .78 to .89
for the dating couples, from .65 to .89 for the engaged couples, and from .75
to .91 for the married couples.

Self-reported relationship quality. Both the engaged and married partici-

pants filled out the Emotional Tone Index (ETI; Berscheid, Snyder, &
Omoto, 1989) as a part of their pre-lab packet of self-report questionnaires.
Participants were given a list of 25 emotion words (e.g., angry, calm, and
joyful) and asked to rate on a scale of 1 (never) to 7 (almost always) how
often they experience each feeling in their relationship. Scores correspond-
ing to the positive and negative items were separately summed and then
the total negative item score was subtracted from the total positive item
score (engaged M = 4.0, SD = 1.2; married M = 3.6, SD = 1.3). The reliability
analysis produced an alpha of .89 for the positive emotions and .89 for the
negative emotions. Alphas for the married couples were comparable; .93
for the positive emotions and .87 for the negative emotions. In contrast, the
dating couples completed the Dyadic Adjustment Scale (DAS; Spanier,
1976). The DAS is a 32-item questionnaire that has been used with a wide
variety of couples to assess adjustment/satisfaction. All of the items of the
DAS were composited to create a total dyadic adjustment score (M = 119.8,
SD = 13.3). The reliability analysis produced an alpha of .87.

Observed relationship quality. The Interactional Dimensions Coding System

(Kline et al., 2005) was used to code the couples’ interactions. The Inter-
actional Dimensions Coding System consists of nine individual scales (e.g.,
communication, problem-solving skills) and five dyadic scales (e.g., negative
escalation, commitment) ranging in scores from 1 (extremely uncharacter-
istic) to 9 (extremely characteristic). The individual scales that were used in
this analysis are positive affect and negative affect. These two scales were
chosen because they most closely paralleled the positive and negative self-
reported emotional tone rating described above.
Positive affect was used to describe how positive the participant’s face,
voice, and body were while interacting with their partner. For example,
good eye contact is considered an important element of positive facial cues,
laughing would be considered a component of positive voice, and orien-
tation towards their partner would be included in the positive body
category. Negative affect also consisted of negative face, voice, and body.
Similarly, negative facial cues include lack of eye contact, a cold or angry
voice would be considered negative voice, and negative body would include
a tense or rigid posture. A composite score of observed emotional tone was
created by subtracting negative affect from positive affect (dating M = –.35,
SD = .3.1; engaged M = .92, SD = 2.2; married M = 43, SD = .2.0). Two
graduate students were trained to use the system. Intra-class correlations
between raters were as follows:

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

820 Journal of Social and Personal Relationships 25(5)

• positive affect: dating couples = .84, engaged couples = .78, married couples
= .72;
• negative affect: dating couples = .78, engaged couples = .69, married couples
= .70.

Physiological responding. Physiological responses were recorded from parti-

cipants’ cardiac system during a baseline period and the disagreement epoch
of the interaction. The specific cardiac measure obtained was heart rate
(HR). Electrode stickers were placed on opposite sides of each participant’s
torso and a ground lead was placed on the sternum. Cardiac inter-beat
intervals (IBI) were measured in milliseconds between successive R waves
of the electrocardiogram (EKG). Heart rate was calculated using the
standard formula: HR = (1/IBI)
60,000 ms (i.e., beats per minute, dating
M = 5.1, SD = 4.4; engaged M = 4.2, SD = 4.1; married M = 3.4, SD = 3.3).
Electrodermal response was measured by skin conductance level (SCL). A
constant-voltage device was used to pass a small voltage between elec-
trodes attached to the fingertips of the second and fourth fingers of the
non-dominant hand. SCL was measured in microsiemens (dating M = 4.1,
SD = 2.2; engaged M = 4.2, SD = 2.3; married M = 2.8, SD = 1.7). Changes
in physiological responses were calculated by subtracting mean levels of
physiological responses during the baseline from mean levels during the
disagreement epoch of the interaction. All physiological data were stan-
dardized for analyses.

Results

Analytic plan
HLM 6.0 (Bryk & Raudenbush, 1992) was used to examine the effects of
personality on each measure of relationship quality. Two sets of analyses
were conducted per relationship outcome in HLM. First, we examined the
self-reported personality traits in relation to each relationship quality indi-
cator. Next, partner-reports of personality were examined, in part to deter-
mine whether personality traits per se or a single informant (i.e., self-report)
accounted for expected associations between personality and self-reported
relationship quality. Thus, partner-report analyses were conducted as a
means of corroborating analyses based on self-reported personality traits.
Note that analyses were conducted by relationship type (dating, engaged,
and married) and only included actor effects (influence of an individual on
his or her own outcome) and not partner effects (influence of individual on
his or her partner’s outcome). We were not especially interested in partner
effects in this study, but rather in using partner-reports as a secondary infor-
mant to corroborate self-reported findings. Finally, additional analyses were
conducted in HLM to determine whether gender or duration of the rela-
tionship moderated any of the effects of personality traits on relationship
quality. More specifically, gender was examined as a potential moderator of
Level 1 personality effects (e.g., slopes) and duration was entered as a Level
2 (dyadic) predictor of each Level 1 personality slope.

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

Holland & Roisman: Personality and relationship quality 821

In order to examine the associations between personality and indicators

of relationship quality in the context of the HLM analyses, t-ratios were
converted to correlations to estimate the standardized magnitude of effects
(r = √t 2/[t 2 + df ], Rosenthal & Rosnow, 1984 as cited in Kurdek, 2004).
Pseudo R2 s were also calculated by comparing the error terms from the
restricted model to an unrestricted model (only containing a dependent
variable and intercept). The following formula was utilized for these calcu-
lations, 1 – ((level 1 restricted error + level 2 restricted error)/(level 1 un-
restricted error + level 2 unrestricted error)) (R-squared in a hierarchical
model, n.d.).

Preliminary analyses
Correlations examining the association between self-reported personality
and partner-reported personality were conducted (see Table 1) to gain an
understanding of how the self-reports and partner-reports converged as the
partner-reports were important in establishing corroboration of the self-
report findings. Overall, for the dating, engaged, and married couples there
was significant convergence between informants within personality trait,
thus providing evidence of the validity of using the partner-reports as a
corroborating measure.
Analyses were conducted in HLM (see Table 2) to investigate the over-
lap between the different measures of relationship quality. These analyses
were important to investigate how the different measures of relationship
quality converge to ensure that we are capturing different constructs. Signifi-
cant associations were found between self-reported quality and observed
emotional tone for the dating couples and for the married couples. Signifi-
cant correlations between self-reported quality and observed emotional
tone were not found for the engaged couples. For both the dating couples
and the married couples, skin conductance and heart rate were significantly
correlated. While some overlap exists between the dependent measures, the
fact that such overlap was modest in magnitude suggested that it was best
to examine these constructs separately in analyses.

HLM analyses

Dating couples. Self-reported conscientiousness was associated with greater

self-reported quality (see Table 3). In addition, self-reported agreeableness
was associated with greater quality as rated by observers. Partner-reported
neuroticism was negatively associated with self-reported quality and posi-
tively associated with conscientiousness. Thus, the self-reported conscientious-
ness finding was corroborated by partner-report. Finally, partner-reported
extraversion was associated with an increase in heart rate. Pseudo R2 s were
highest for self-report quality (range = .07 to .09) in comparison to observed
emotional tone and physiological responding (range = –.02 to .03). Only
one effect was moderated by gender; more specifically, female partner-
reported extraversion was more strongly associated with increases in heart
rate than was male partner-reported extraversion. No effects were moder-
ated by duration of relationship.

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

822 Journal of Social and Personal Relationships 25(5)

TABLE 1
Correlations between male and female self-reported personality and male and
female partner-reported personality

Partner-reported personality 1 2 3 4 5

Dating couples
Male self-reported personality
1. Neuroticism .37** –.24* –.00 –.21* –.08
2. Extraversion –.39** .62** .00 .23* .14
3. Openness .11 –.09 .44** .15 .06
4. Agreeableness –.12 .14 .21* .44** .02
5. Conscientiousness –.08 .13 –.15 .14 .40**
Female self-reported personality
1. Neuroticism .13 –.16 .03 .12 –.22*
2. Extraversion –.23* .42** –.07 .17 .25**
3. Openness .23* .00 .41** –.08 –.15
4. Agreeableness .01 .20* .04 .31* .10
5. Conscientiousness –.02 .21 –.15 .02 .36**

Engaged couples
Male self-reported personality
1. Neuroticism .59** –.35* –.01 –.24 –.15
2. Extraversion –.22 .65** –.25 .24 .11
3. Openness .11 –.21 .65** .05 –.12
4. Agreeableness –.42 .29 –.10 .47** .36**
5. Conscientiousness –.06 .26 –.19 .31* .71**
Female self-reported personality
1. Neuroticism .57** –.29* –.03 –.38** .05
2. Extraversion –.15 .69** .17 .38** –.04
3. Openness .27 .05 .62** .04 –.33*
4. Agreeableness –.23 .17 –.05 .64** .08
5. Conscientiousness .01 .11 .04 .06 .55**

Married couples
Male self-reported personality
1. Neuroticism .44** –.20 .02 –.46** –.39*
2. Extraversion –.15 .50** –.14 .13 .10
3. Openness –.16 –.03 .68** –.13 –.09
4. Agreeableness .01 .12 –.09 .53** .01
5. Conscientiousness –.07 .17 –.13 .15 .62**
Female self-reported personality
1. Neuroticism .65** –.22 .06 –.06 –.23
2. Extraversion –.39* .43** .17 .14 .10
3. Openness –.20 .27 .47** .16 .06
4. Agreeableness –.04 –.18 –.01 .14 –.06
5. Conscientiousness –.23 –.04 –.03 –.09 .51**

* p < .05; ** p < .01.

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

Holland & Roisman: Personality and relationship quality 823

TABLE 2
Standardized effects identified in HLM between self-reported quality,
observed emotional tone, and physiological responding for dating couples,
engaged couples, and married couples

Self-report Observed Heart rate SCL

Dating couples
Self-report
Observed .15*
Heart rate .05 .01
SCL .07 –.03 .17*

Engaged couples
Self-report
Observed .05
Heart rate –.12 .01
SCL –.08 –.04 –.13

Married couples
Self-report
Observed .29**
Heart rate –.16 .08
SCL .01 –.14 .38**

* p < .05; ** p < .01.

TABLE 3
HLM analyses predicting dating couples’ self-reported quality, observed
emotional tone, and physiological responding from self-reported personality
and partner-reported personality

Self-report Observed Heart rate SCL

Self-reported personality
Neuroticism .07 .05 .06 .08
Extraversion .08 .06 .06 .06
Openness .05 .09† .03 .04
Agreeableness .02 .11** .01 .05
Conscientiousness .12** .02 .07 .07
Pseudo-R2 .07 .03 –.02 –.02

Partner-reported personality
Neuroticism –.12** .06 .06 .07
Extraversion .02 .03 .10* .05
Openness .07 .06 .06 .05
Agreeableness .02 .05 .07 .02
Conscientiousness .13** .05 .08 .04
Pseudo-R2 .09 .01 .01 –.01

† p < .10; * p < .05; ** p < .01. Unless otherwise noted, coefficients above are standardized betas.

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

824 Journal of Social and Personal Relationships 25(5)

Engaged couples. Individuals who reported being high on extraversion (see

Table 4) and conscientiousness also reported having greater relationship
quality. Self-reported extraversion was associated with greater relationship
quality as rated by observers. Finally, conscientiousness was associated
with increases in skin conductance levels from baseline. With respect to
partner-reported personality, neuroticism was associated with reporting
lower relationship quality. Partner-reported openness, agreeableness, and
conscientiousness were all associated with greater self-reported quality.
Once again, partner-reported conscientiousness corroborated the associ-
ation identified between self-reported conscientiousness and self-reported
quality. Partner-reported neuroticism, extraversion, and agreeableness were
all associated with increases in heart rate during the disagreement epoch.
Consistent with the findings from the dating sample, pseudo R2s were highest
for self-report quality (range = .22 to .30) in comparison to observed emo-
tional tone and physiological responding (range = –.06 to .03). No effects
were moderated by either gender or duration of relationship.

Married couples. Individuals high on self-reported neuroticism also reported

higher levels of relationship quality (see Table 5). Self-reported extraversion
and agreeableness were associated with greater self-reported quality. With
respect to physiological outcomes, self-reported conscientiousness was
associated with increases in skin conductance levels. Only one unique effect
was found for partner reported personality: agreeableness was associated
with greater relationship quality as rated by observers. Once again, pseudo

TABLE 4
HLM analyses predicting engaged couples’ self-reported quality, observed
emotional tone, and physiological responding from self-reported personality
and partner-reported personality

Self-report Observed Heart rate SCL

Self-reported personality
Neuroticism –.14† .09 .14 .07
Extraversion .15* .14* .02 .05
Openness .06 .06 .05 .08
Agreeableness .13† .08 .01 .12
Conscientiousness .15* .13 .09 .18**
Pseudo-R2 .22 .02 –.06 .03

Partner-reported personality
Neuroticism –.15* .09 .16* .09
Extraversion .07 .09 .17* .08
Openness .17** .10 .07 .10
Agreeableness .18** .06 .16* .07
Conscientiousness .19** .09 .09 .11
Pseudo-R2 .30 .02 .00 –.05

† p < .10; * p < .05; ** p < .01. Unless otherwise noted, coefficients above are standardized betas.

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

Holland & Roisman: Personality and relationship quality 825

TABLE 5
HLM analyses predicting married couples’ self-reported quality, observed
emotional tone, and physiological responding from self-reported personality
and partner-reported personality

Self-report Observed Heart rate SCL

Self-reported personality
Neuroticism –.17* .13 .05 .06
Extraversion .18* .10 .08 .05
Openness .13 .07 .06 .15†
Agreeableness .16* .05 .09 .06
Conscientiousness .07 .14 .13 .18*
Pseudo-R2 .26 –.01 .04 –.02

Partner-reported personality
Neuroticism .11 .08 .10 .09
Extraversion .08 .13 .01 .08
Openness .15† .08 .13 .07
Agreeableness .16† .17* .14 .13
Conscientiousness .16† .08 .15† .11
Pseudo-R2 .22 .02 –.04 .03

† p < .10; * p < .05. Unless otherwise noted, coefficients above are standardized betas.

R2 s were highest for self-report quality (range = .22 to .26) in comparison

with observed emotional tone and physiological responding (range = –.04 to
.04). No effects were moderated by gender; however, one effect was moder-
ated by duration of relationship. More specifically, duration of relationship
moderated the effect of self-reported extraversion and self-reported rela-
tionship quality, such that the effect was larger for longer-term relationships.

Discussion

This study contrasted two frameworks designed to better understand the

nature of links between the Big Five personality traits and the quality of
adults’ romantic relationships. As described in the introduction, the first
framework predicted that personality traits would be coherently and reliably
associated with all of the relationship outcomes (i.e., self-reported quality,
observed emotional tone, and physiological responding) explored in this
study. In contrast, the second framework predicted that personality would
be most strongly associated with self-reported quality, and only weakly
associated with observations of relationship quality and physiological
response during a dyadic interaction. Significant associations were identi-
fied between personality (in aggregate) and self-reported quality. In contrast,
few significant associations were identified for both observed emotional
tone and physiological responding. Moreover, the overall magnitude of the
effect of personality traits in aggregate on non-self-report indicators of

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

826 Journal of Social and Personal Relationships 25(5)

relationship quality was trivial across dating, engaged, and married couples.
Thus, more support was found for the second framework.
An association between personality traits and self-reported relationship
quality has been consistently identified in the literature (Karney & Bradbury,
1997; Robins et al., 2000, 2002; Watson et al., 2000). As such, neuroticism was
expected to emerge as an important personality trait in predicting relation-
ship quality. However, while a few significant effects were found for each
of the personality traits, a consistent pattern across informants and indi-
cators of quality did not clearly emerge for any specific personality trait. In
addition, there was some heterogeneity in unique effects across couple
type. The most consistent unique effects were found in the domain of self-
reported quality – and conscientiousness was the most consistent predictor
of this dependent measure. For example, among dating and engaged couples,
individuals who were more conscientious reported higher quality relation-
ships. In addition, married couples reported greater conscientiousness and
also exhibited higher levels of skin conductance. That said, we consider this
latter finding counterintuitive in that increases in skin conductance are often
associated with efforts to inhibit behavior – which would not be indicative
of a high quality interaction with a significant other.
A major strength of this study is the incorporation of partner-reports of
personality in an effort to corroborate self-reports of personality. It is
important to note that few studies have incorporated partner-reports of
personality in their work (Botwin et al., 1997; Donnellan et al., 2004; Kosek,
1996; Watson et al., 2000). In this study, the partner-reports of personality
only provided modest corroboration for self-reported personality. For the
dating and engaged couples, both self-reported and partner-reported con-
scientiousness was associated with higher self-reported relationship quality.
However, no other partner-reports corroborated self-reported personality
findings. Perhaps the lack of significant partner-reported findings is related
to the partner’s own current satisfaction with the relationship. Watson and
Humrichouse (2006) hypothesized that when raters are unsure about how
to answer a question about their partner’s personality that they compen-
sate by referring to their current level of satisfaction with the relationship.
Their data showed that when participants were very satisfied with their
relationship, that they rated their partners as more positive than even their
partners self-reported. However, they judged their partners more nega-
tively when they were less satisfied with the relationship.
A major goal of this study was to more fully explore how personality and
relationship quality are associated by including a range of measures of
romantic relationship quality. This goal was complicated by the fact that
much of the existing literature does not explicitly acknowledge that the
association between personality and relationship quality is based on an
approach that primarily utilizes self-reports of personality and self-reports
of relationship quality. Thus, an additional strength of this study is the use
of a multi-method, multi-informant approach. This design allowed us to
examine questions previously not extensively explored in the existing litera-
ture related to the range of relationship-relevant outcomes associated with
personality traits.

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

Holland & Roisman: Personality and relationship quality 827

Limitations
This analysis focused on a specific operationalization of personality (i.e.,
self-reported and partner-reported personality). This operationalization was
used because is it is the most common method utilized when measuring
personality. Nonetheless, as a reviewer of this paper noted, both observed
emotional tone and physiological responding are behavioral assessments
of relationship quality. As such, if a more behaviorally-based measure of
personality was utilized, more evidence that personality has broad predic-
tive significance might have emerged in this analysis.
In addition, the sub-samples of this study were relatively small in size and
we chose not to aggregate them because different measures of self-reported
quality were used for the dating couples (the DAS) and the engaged and
married couples (the ETI). Re-scaling one of the measures seemed inappro-
priate conceptually, as the two measures differ in their focus and scope, even
though they both assess self-reported quality. In addition, it is important to
acknowledge that our methodologically intensive design limited our ability
to collect data from larger samples. Finally, a significant limitation is that this
study only examined personality and relationship quality at a single time
point. However, before conducting longitudinal studies, it seems appropriate
first to examine cross-sectional associations, which were not identified in this
study for observations and physiological indicators of relationship quality.
Nonetheless, future work still needs to incorporate multiple levels of
analysis longitudinally in order to fully examine how personality traits and
relationship quality are interrelated over time (Donnellan et al., 2005;
Robins et al., 2002; Watson et al., 2000). Researchers have in fact acknowl-
edged that individuals are likely to influence their partner over the duration
of the relationship; thus individuals may become increasingly dissimilar from
or similar to their partner as the relationship progresses (Gattis et al., 2004).
This study utilized samples of exclusive dating, engaged, and married
couples and, interestingly, the smallest effects of personality were found for
dating couples. One possible explanation for this finding is that different
measures were used for the dating couples versus the engaged and married
couples. That said, previous studies have not specifically investigated how
the association between personality and relationship quality differs as a
function of relationship type; based on these results perhaps associations in
this domain are stronger in more serious relationships, thus reflecting a
developmental shift in terms of the salience of personality traits in adults’
romantic relationships. In other words, it may be that dating relationships
are not yet fully developed enough to see the same kind of effects that were
found in more seriously committed couples. However, a related explanation
is that tasks of social living, such as having children and beginning a career
path may influence personality change (Roberts, Wood, & Smith, 2005).
Thus, younger couples, such as the dating and engaged couples who are
more likely beginning to approach these challenges, may be exhibiting more
personality change than the older, married couples. In fact, Roberts et al.
(2005) discuss how investing heavily in social relationships may help explain
increases in agreeableness and conscientiousness in young and middle
adulthood. Consequently, it is possible that increases in agreeableness and

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

828 Journal of Social and Personal Relationships 25(5)

conscientiousness would have the potential to positively affect romantic

relationships over time.
Finally, it is important to acknowledge that the lack of an association
between personality and observed emotional tone stands in contrast to
previous findings investigating the association between these two constructs
(Donnellan et al., 2004, 2005). However, the Donnellan et al. (2004, 2005)
results were based on a single sample of parents and children followed
longitudinally and have not been replicated by other researchers.

Conclusion
This study explored how the Big Five personality traits are associated with
relationship quality in the context of two different conceptual frameworks.
Among samples of dating, engaged, and married couples, the data were more
consistent with a framework that predicted that personality traits would be
most strongly associated with self-reported quality and only weakly associ-
ated with other indicators of relationship quality, such as observed quality and
physiological response during a dyadic interaction. The common approach
to addressing questions about the predictive significance of personality
traits typically uses self-reports of personality and self-reports of relation-
ship quality. However, this study clearly indicates that such a mono-method
approach is not sufficient. Instead, it is necessary to utilize multi-method,
multi-informant designs in order to fully clarify how personality is associ-
ated with the full range of relationship quality indicators routinely used by
interpersonal relationships scholars. Understanding the nature and differ-
ential magnitude of these associations across relationship quality outcomes
is a critical step in accurately portraying the predictive significance of person-
ality traits for the quality of adults’ interpersonal relationships.

REFERENCES

Berscheid, E., Snyder, M., & Omoto, A. M. (1989). The relationship closeness inventory:
Assessing the closeness of interpersonal relationships. Journal of Personality and Social
Psychology, 57, 792–807.
Botwin, M. D., Buss, D. M., & Shackelford, T. K. (1997). Personality and mate preferences:
Five factors in mate selection and marital satisfaction. Journal of Personality, 65, 107–136.
Bryk, A., & Raudenbush, S. W. (1992). Hierarchical linear models for social and behavioral
research: Applications and data analysis methods. Newbury Park, CA: SAGE.
Caspi, A., Roberts, R. W., & Shiner, R. L. (2005). Personality development: Stability and
change. Annual Reviews of Psychology, 56, 453–484.
Caughlin, J. P., Huston, T. L., & Houts, R. N. (2000). How does personality matter in marriage?
An examination of trait anxiety, interpersonal negativity, and marital satisfaction. Journal
of Personality and Social Psychology, 78, 326–336.
Donnellan, M. B., Conger, R. D., & Bryant, C. M. (2004). The Big Five and enduring marriages.
Journal of Research in Personality, 38, 481–504.
Donnellan, M. B., Larsen-Rife, D., & Conger, R. D. (2005). Personality, family history, and
competence in early adult romantic relationships. Journal of Personality and Social Psychol-
ogy, 88, 562–576.
Fowles, D. C. (1980). The three arousal model: Implications of Gray’s two-factor learning theory
for heart rate, electrodermal activity, and psychopathy. Psychophysiology, 17, 87–104.

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014

Holland & Roisman: Personality and relationship quality 829

Fowles, D. C. (1988). Psychophysiology and psychopathology: A motivational approach.

Psychophysiology, 25, 373–391.
Gattis, K. S., Berns, S., Simpson, L. E., & Christensen, A. (2004). Birds of a feather or strange
birds? Ties among personality dimensions, similarity, and marital quality. Journal of Family
Psychology, 18, 564–574.
Gottman, J. M., Coan, J., Carrere, S., & Swanson, C. (1998). Predicting marital happiness and
stability from newlywed interactions. Journal of Marriage and the Family, 60, 5–22.
Gottman, J. M., & Levenson, R. W. (1988). The social psychophysiology of marriage. In
P. Noller & M. A. Fitzpatrick (Eds.), Perspectives on marital interaction (pp. 182–200).
Philadelphia, PA: Multilingual Matters Ltd.
Gottman, J. M., & Levenson, R. W. (1992). Martial processes predictive of later dissolution:
Behavior, physiology, and health. Journal of Personality and Social Psychology, 63, 221–233.
Heller, D., Watson, D., & Ilies, R. (2004). The role of person versus situation in life satisfaction:
A critical examination. Psychological Bulletin, 130, 574–600.
Karney, B. R., & Bradbury, T. N. (1997). Neuroticism, marital interaction, and the trajectory
of marital satisfaction. Journal of Personality and Social Psychology, 72, 1075–1092.
Kelly, E. L., & Conley, J. J. (1987). Personality and compatibility: A prospective analysis of
marital stability and marital satisfaction. Journal of Personality and Social Psychology, 52,
27–40.
Kline, G. H., Julien, D., Baucom, B., Hartman, S., Gilbert, K., Gonzales, T., & Markman, H. J.
(2005). The Interactional Dimensions Coding System: A global system for couple inter-
actions. In P. Kerig & D. Baucom (Eds.), Couple observational coding systems. Mahwah,
NJ: Erlbaum.
Kosek, R. B. (1996). The quest for a perfect spouse: Spousal ratings and marital satisfaction.
Psychological Reports, 79, 731–735.
Kurdek, L. A. (2004). Are gay and lesbian cohabiting couples really different from hetero-
sexual married couples? Journal of Marriage and Family, 66, 880–900.
McCrae, R. R., & Costa, P. T. (1997). Personality trait structure as a human universal. American
Psychologist, 52, 509–516.
Roberts, B. W., Wood, D., & Smith, J. L. (2005). Evaluating five factory theory and social
investment perspectives on personality trait development. Journal of Research in Personal-
ity, 39, 166–184.
Robins, R. W., Caspi, A., & Moffitt, T. E. (2000). Two personalities, one relationship: Both
partners’ personality traits shape the quality of their relationship. Journal of Personality and
Social Psychology, 79, 251–259.
Robins, R. W., Caspi, A., & Moffitt, T. E. (2002). It’s not just who you’re with, it’s who you are:
Personality and relationship experiences across multiple relationships. Journal of Personal-
ity, 70, 925–964.
Roisman, G. I. (2007). The psychophysiology of adult attachment relationships: Autonomic
reactivity in marital and premarital interactions. Developmental Psychology, 43, 39–53.
Roisman, G. I., Clausell, E., Holland, A., Fortuna, K., & Elieff, C. (2008). Adult romantic
relationships as contexts of human development: A multimethod comparison of same-sex
couples with opposite-sex dating, engaged, and married dyads. Developmental Psychology,
44, 91–101.
R-squared in a hierarchial model. (n.d.). Retrieved June 18, 2007, from http://www.utexas.edu/
its-archive/rc/answers/hlm/hlm4.html
Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for assessing the quality of
marriage and similar dyads. Journal of Marriage and the Family, 38, 15–28.
Watson, D., Hubbard, B., & Wiese, D. (2000). General traits of personality and affectivity as
predictors of satisfaction in intimate relationships: Evidence from self- and partner-ratings.
Journal of Personality, 68, 413–449.
Watson, D., & Humrichouse, J. (2006). Personality development in emerging adulthood: Inte-
grating evidence from self-ratings and spouse ratings. Journal of Personality and Social
Psychology, 91, 959–974.

Downloaded from spr.sagepub.com at Dicle Ãœniversitesi on November 4, 2014