Developmental Psychology © 2009 American Psychological Association

2009, Vol. 45, No. 6, 1695–1707 0012-1649/09/$12.00 DOI: 10.1037/a0016121

Mothers’ and Fathers’ Personality and Parenting:

The Mediating Role of Sense of Competence

Amaranta D. de Haan, Peter Prinzie, and Maja Deković

Utrecht University

This prospective longitudinal study addressed 3 key questions regarding the processes of parenting in a
large community sample of mothers (n ⫽ 589) and fathers (n ⫽ 518). First, the collective impact of
parental Big Five personality dimensions on overreactive and warm parenting, assessed 6 years later by
adolescents, was examined. Second, mediation of these associations by sense of competence in the
parenting role was addressed. Third, it was explored to what extent associations were similar for mothers
and fathers. Agreeableness and Extraversion were related to lower levels of overreactivity and higher
levels of warmth. Sense of competence completely mediated relations between personality and overre-
activity and partially mediated relations between personality and warmth. The associations were found
to be similar for mothers and fathers. Overall, sense of competence was shown to be an important
mechanism that can explain the link between personality and parenting.

Keywords: Big Five, parenting, sense of competence, fathers and mothers, adolescence

Linkages between parenting and children’s developmental out- resources, children’s characteristics, and contextual sources of
comes have been widely studied. Research has provided ample stress and support. Empirical research supports his notion that
evidence of detrimental effects of negative parenting, such as children’s characteristics (e.g., temperament) and contextual
rejection or overreactivity, on both externalizing (Dishion & sources of stress and support (e.g., work, social relations) impact
Patterson, 2006; Snyder, Cramer, Afrank, & Patterson, 2005) and parenting (for an overview, see Belsky & Jaffee, 2006). Parental
internalizing behaviors (Wood, McLeod, Sigman, Hwang, & Chu, personality has, however, been relatively neglected in research
2003). Conversely, positive parenting, which includes warmth and addressing the question of “why parents parent the way they do”
involvement, has been found to relate negatively to externalizing (Belsky, 1984, p. 83; Belsky & Barends, 2002; Caspi, Roberts, &
(Denham et al., 2001) and internalizing behaviors (Barrera et al., Shiner, 2005). This is surprising, as Belsky (1984) himself sug-
2002) and to promote children’s social competence (Brody, gested that parents’ personality should be considered the most
McBride-Murry, Kim, & Brown, 2002). In contrast to the well- important influence. Parental personality shapes parenting not only
established links between parenting and children’s development, directly but also indirectly by influencing the broader context in
less is known about individual antecedents of parenting (Belsky & which parent– child relationships exist.
Barends, 2002). With the current study, we aimed to increase the Existing work shows that personality shapes how people expe-
relatively limited knowledge base on the antecedents of parenting rience and respond to a wide variety of developmental tasks (Ozer
by examining the extent to which parental personality is longitu- & Benet-Martı́nez, 2006). Research on the relations between per-
dinally related to parenting behaviors, by determining the extent to sonality and parenting has, however, lacked a systematic approach
which these associations are mediated by parental sense of com- to the measurement of personality. Researchers have usually arbi-
petence, and by exploring whether these associations are similar trarily focused on one or two aspects of adult psychological
for mothers and fathers. functioning, especially adult psychopathology (e.g., depression),
and have examined links with parenting. To facilitate comparison
Parental Personality and Parenting Behaviors: of research findings, Belsky and Barends (2002) urged researchers
The Role of the Big Five to utilize a systematic, comprehensive approach to the measure-
ment of personality. The repeated identification of the Big Five
From previous theoretical and empirical work, Belsky (1984)
factors in personality ratings has led to the view that most person-
distilled three general sources of influence on parental functioning
ality traits can be captured by five broad content domains. The Big
in his process model of parenting: parents’ personal psychological
Five (or five-factor model) personality dimensions have tradition-
ally been numbered and labeled as follows: (I) Extraversion (Sur-
gency, Positive Affectivity); (II) Agreeableness (vs. Antagonism);
Amaranta D. de Haan, Peter Prinzie, and Maja Deković, Department of
(III) Conscientiousness (Constraint Dependability); (IV) Emo-
Child and Adolescent Studies, Utrecht University, Utrecht, the Nether-
tional Stability (vs. Neuroticism); and (V) Openness to Experience
lands.
Correspondence concerning this article should be addressed to Ama- (Autonomy, Intellect; Caspi, 1998; Goldberg, 1990; Hendriks,
ranta D. de Haan, Department of Child and Adolescent Studies, Utrecht Hofstee, & De Raad, 1999a, 1999b; McCrae & Costa, 1999).
University, P.O. Box 80.140, Utrecht 3508 TC, the Netherlands. E-mail: A recent meta-analysis (Prinzie, Stams, Deković, Reijntjes, &
a.d.dehaan@uu.nl Belsky, in press) examined relations among the Big Five dimen-

1695
1696 DE HAAN, PRINZIE, AND DEKOVIĆ

sions, considered separately, and three dimensions of parenting: nisms. Belsky et al. (1995), for example, found that daily hassles
warmth/supportiveness (vs. rejection), behavioral control/structure and transient mood (partly) mediated relations among Neuroti-
(vs. chaos), and autonomy support (vs. coercion). All Big Five cism, Agreeableness, and parenting. Belsky et al. argued that,
dimensions were found to be positively related to warm and because the direct relations between personality and parenting
structured parenting, and Agreeableness and Emotional Stability were not always and never were strongly mediated by mood and
were positively related to autonomy support. The associations hassles, other possible mediators should be explored.
were similar for mothers and fathers and were not moderated by A plausible possibility that has not yet been investigated is that
type of assessment of parenting (self-report or observations). Prin- parents’ personality impacts parental sense of competence and
zie et al. suggested that existing knowledge should be further parental sense of competence in turn impacts parenting. Parental
extended by examining (a) how much personality affects parenting sense of competence (i.e., the belief of parents that they can
when all Big Five dimensions are considered together, (b) pro- effectively manage parenting tasks) is believed to motivate and
spective rather than cross-sectional relations between personality shape behaviors and thus may be critical to understanding parent-
and parenting, and (c) why personality and parenting are related. ing behaviors (Coleman & Karraker, 1997; Goodnow & Collins,
It appears that some of the associations between the Big Five 1990; McGillicuddy-DeLisi & Sigel, 1995). Parental sense of
dimensions and parenting become redundant in the context of the competence is closely related to the construct of self-efficacy. The
other dimensions (Dunlap & Landis, 1998). Studies that took into two constructs are often used interchangeably, as is apparent from
account all Big Five dimensions found nonsignificant associations the definition of parental self-efficacy as “parents’ self-referent
between warmth, Emotional Stability (Smith et al., 2007), and estimations of competence in the parental role” (Coleman & Kar-
Openness (Clark, Kochanska, & Ready, 2000). When assessed raker, 2003, p. 128). Our overview of the literature includes both
individually, in contrast, these dimensions were significantly re- constructs, although we employ the term sense of competence for
lated to warmth (Belsky, Crnic, & Woodworth, 1995; Losoya, the sake of consistency. Furthermore, to our knowledge, no studies
Callor, Rowe, & Goldsmith, 1997). Nonsignificant cross-sectional have investigated either determinants or consequences of fathers’
associations have been found between overreactivity, Conscien- sense of competence; thus, this overview is limited to research
tiousness, and Emotional Stability (Prinzie et al., 2004, 2005; among mothers.
Smith et al., 2007). Significant associations between these person- Existing work shows that maternal sense of competence is an
ality dimensions and overreactivity were, however, found in re- important predictor of mothers’ capacity to provide an adaptive,
search that did not examine all Big Five dimensions simulta- stimulating, and nurturing child-rearing environment (Locke &
neously (Olsen, Martin, & Halverson, 1999). Although the Prinz, 2002). Sense of competence has been found to relate posi-
personality dimensions are assumed to be orthogonal (i.e., uncor- tively to maternal responsiveness (Donovan, Leavitt, & Walsh,
related), these diverging results suggest otherwise. 1990), and in an intervention study of the Home-Start parenting
Even among studies that did employ a longitudinal design, the support program, an increase in maternal sense of competence was
period of time over which the measures were assessed usually was accompanied by enhanced maternal sensitivity and consistency
not very long. Personality dimensions are, however, conceptual- (Asscher, Hermanns, & Deković, 2008). Conversely, sense of
ized as having enduring effects over time (Clark et al., 2000). competence has been reported to relate negatively to defensive
Empirical support for this conceptualization is offered by findings behaviors (Donovan et al., 1990) and to displays of negative affect
of Kochanska, Friesenborg, Lange, and Martel (2004), which and use of punitive disciplinary techniques (Bugental, Blue, &
showed that associations between maternal personality and the Cruzcosa, 1989; Bugental & Cortez, 1988). Drawing from self-
quality of the mother– child relationship seemed to grow stronger efficacy theory (Bandura, 1989; Coleman & Karraker, 1997; Jones
with the passage of time, as the dyadic relationship pattern coa- & Prinz, 2005), one can give several explanations of why sense of
lesced. Moreover, if the period of time between assessments is competence should be related to parenting. Parents who feel that
small, the ease of maintaining consistency between measures may they have little influence over their children’s behaviors may feel
bias correlations upward (Krathwohl, 1993). Assessing personality a sense of hopelessness. This feeling of hopelessness may lead
before assessing parenting also satisfies one of the three demands parents to derive less enjoyment from interactions with their chil-
for showing causality, that is, temporal precedence. Therefore, our dren, and this, in turn, would be related to lower levels of warmth
first aim in the current study was to examine prospective longitu- (Gondoli & Silverberg, 1997). In addition, individuals with a low
dinal associations between all parental Big Five dimensions and sense of competence tend to shy away from setting specific goals
parenting assessed 6 years later. In this way, it would be possible (Schunk, 1990). Personal goals provide a framework for behavior,
to assess the collective impact of the Big Five over time and the and the failure to develop clear goals is likely to be related to
relative impact of each of the dimensions in the context of the erratic, overreactive parenting (Coleman & Karraker, 1997). Sense
other dimensions. of competence also impacts the actual motivation to select or
engage in challenging tasks (Sexton & Tuckman, 1991). In the
Why Is Personality Linked to Parenting Behaviors: The parental domain, avoidance of challenging tasks may translate into
Mediating Role of Sense of Competence avoidance of more effortful disciplinary techniques, with parents
who feel less competent opting instead to control child behaviors
It is, as Belsky and Barends (2002, p. 431) signaled, one with overreactive tactics (Coleman & Karraker, 1997).
thing to observe that “parents’ personality is predictive of their The one study, to our knowledge, that has examined cross-
parenting . . . . [and] quite another to understand the mechanisms sectional relations between personality traits (Emotional Stability,
responsible for this relation.” The influence of personality on Openness) and maternal sense of competence found that Emo-
parenting may operate through multiple psychological mecha- tional Stability was positively related to maternal sense of com-
 PERSONALITY, SENSE OF COMPETENCE, AND PARENTING 1697

petence (Bornstein et al., 2003). Within vocational psychology, do not necessarily translate into differences in associations. Results
positive relations have been found between the Big Five dimen- from several studies support the possibility that associations be-
sions and individuals’ career sense of competence (Nauta, 2004) tween personality and parenting are similar for mothers and fa-
and sense of competence in the leadership role (Chan & Drasgow, thers, but the knowledge is limited and is not entirely consistent.
2001; Hendricks & Payne, 2007). Social cognitive career theory Belsky et al. (1995), for example, found that Extraversion was
(SCCT) provides vocational psychologists a framework that links more strongly related to fathers’ parenting and that Agreeableness
personality with sense of competence (Lent, Brown, & Hackett, was more strongly related to mothers’ parenting. Kochanska et al.
1994); from this framework, we can draw some cautious expecta- (2004) showed that parental Agreeableness and Openness had
tions. According to SCCT, personality plays a role in one’s sense differential effects on maternal and paternal positive affect, al-
of competence by impacting the kinds of learning experiences that though they emphasized that the differences were small and
people acquire. Parents high in Extraversion, with its connotation needed to be replicated. Other work has found that relations
of optimism and enjoying social interactions, may be more in- between personality and several parenting behaviors (warmth,
clined to emphasize positive experiences with their children and nurturance, restrictiveness) were similar for mothers and fathers
may thereby feel more competent as parents. Individuals high in (Metsäpelto & Pulkkinen, 2003; Verhoeven, Junger, Van Aken,
Agreeableness, with its emphasis on motives to maintain positive Deković, & Van Aken, 2007). Therefore, a third aim in this study
social relations, may feel more competent as parents because their was to explore the extent to which longitudinal associations among
easygoing interpersonal tendencies have resulted in (implicit) personality, sense of competence, and parenting are similar for
feedback from their children that they are doing a good job as mothers and fathers.
parents (Lent et al., 1994; Nauta, 2004). Parents high in Emotional On the basis of the results of a recent meta-analysis (Prinzie et
Stability, who are described as being calm, relaxed, and stable (as al., in press), we expected all Big Five dimensions would be
opposed to being anxious and insecure), likely emphasize those positively related to warmth and negatively related to overreactiv-
instances when their parenting led to the desired outcome and ity. On the basis of work done in the field of vocational psychology
thereby feel more efficacious as parents. Highly conscientious and of SCCT, we expected that all Big Five dimensions would
individuals, who are more organized and practical, may feel less relate to a higher level of perceived competence. Furthermore, we
overwhelmed by the tasks at hand and thereby feel more compe- expected that maternal sense of competence would mediate rela-
tent as a parent. Parents scoring high on Openness, which com- tions between maternal personality and parenting (see Figure 1).
prises being creative and imaginative, may feel they have a wider Maternal sense of competence was expected to relate positively to
range of options to guide their child in the desired direction and warm parenting and negatively to overreactive parenting. The
may, through this perceived choice of options, feel more compe- dearth of research examining sense of competence among fathers
tent. The extent to which all Big Five dimensions are longitudi- precluded us from formulating expectations about associations
nally related to parenting sense of competence is a question that among fathers’ sense of competence, personality, and parenting.
remains to be addressed (Bornstein et al., 2003; Coleman & Because of the limited and inconsistent results regarding possible
Karraker, 1997). moderation of the associations by parental gender, our examination
Sense of competence has previously been identified as a medi- of the extent to which associations are similar for mothers and
ator among adaptive and dysfunctional parenting and several his- fathers is exploratory.
torically recognized correlates of parenting. Maternal competence
beliefs have been found to mediate cross-sectional relations among Method
maternal depression, perceptions of infant difficulty, social-marital
support, and observed warm/stimulating parenting (Teti & Gel- Participants and Procedure
fand, 1991) and among emotional distress and parental acceptance This study is part of the longitudinal Flemish Study on Parent-
and psychological autonomy promotion (Gondoli & Silverberg, ing, Personality, and Development, which started in 1999 (for a
1997). Gerdes et al. (2007) found, among mothers of children with description of the recruitment of participants, see, e.g., Prinzie et
attention-deficit/hyperactivity disorder, that sense of competence al., 2004, 2005). If our results replicate findings of previous work
mediated cross-sectional relations between depressive symptom-
atology and overreactivity.
To our knowledge, no research to date has examined the medi- Sense of
ating role of sense of competence in longitudinal relations among Competence
Path a Path b
all Big Five dimensions and parenting. This is surprising, because
personality, sense of competence, and parenting have each been
found to be interrelated in meaningful ways. Therefore, a second Parental Path c Parenting
aim in this study was to examine the mediating role of parental Personality Behaviors
sense of competence in longitudinal relations among parental Big
Five dimensions and parenting. Parental
Furthermore, empirical research on the relations between per-
Gender
sonality and parenting has largely neglected fathers, despite the
growing acknowledgment that fathers play an important role in Figure 1. Hypothesized model on the relations between parenting behav-
child development. Although mothers and fathers have been found iors and parental sense of competence. Path a ⫽ first-stage effect; Path b ⫽
to differ in parenting behaviors (e.g., Kendler, Sham, & MacLean, second-stage effect; Path c ⫽ direct effect. Dashed arrows indicate mod-
1997; Metsäpelto & Pulkkinen, 2003), differences in mean levels eration by parental gender.
1698 DE HAAN, PRINZIE, AND DEKOVIĆ

on associations between personality and parenting among Ameri- (FFPI; Hendriks et al., 1999a, 1999b; Hendriks, Hofstee, & De
can mothers, this will provide further support for the generaliz- Raad, 2002). The FFPI comprises 100 brief nondispositional items
ability of these findings (Belsky & Barends, 2002). Parents rated assessing individual differences in behavior. Parents rated the
their own personality at Time 1 (2001) and their sense of parental items on a five-point scale ranging from 1 ⫽ not at all applicable
competence at Time 2, 6 years later. Adolescents rated their to 5 ⫽ entirely applicable. In the normal population, the FFPI
mothers’ and fathers’ warmth/involvement and overreactivity at scale and factor scores show high internal consistencies, substan-
Time 2. Use of adolescents’ ratings of parenting is consistent with tial stabilities, and good construct validity (Hendriks et al., 1999a,
the view that the impact of parenting on adolescent adjustment is 1999b, 2002). Factor weights, established in a large (N ⫽ 2,494)
mediated by how adolescents perceive their parents’ behaviors Dutch normative sample (Hendriks et al., 2002), were used to
(Neiderhiser, Pike, Hetherington, & Reiss, 1998; Powers, Welsh, produce uncorrelated factor scores. In this study, core items for
& Wright, 1994) and eliminates the risk of shared method vari- each of the personality dimensions (Hendriks, 1997; Hendriks et
ance. al., 1999b, 2002) and Cronbach’s alphas for mothers and fathers,
At Time 1, participants were 599 families with a child in respectively, were as follows: Extraversion, “Loves to chat” (␣s ⫽
elementary school (301 girls and 298 boys). The mean age of the .90, .91); Agreeableness, “Respects others’ feelings” (␣s ⫽ .89,
children was 7 years 6 months (range 5 years 0 months to 10 years .89); Conscientiousness, “Does things according to a plan” (␣s ⫽
10 months). Of these 599 families, 541 families (90.1%) were a .89, .89); Emotional Stability, “Can take his/her mind off his/her
two-parent household and 50 families were a one-parent house- problems” (␣s ⫽ .90, .88); and Autonomy, “Wants to form his/her
hold. Eight families (1.7%) did not supply information about their own opinions” (␣s ⫽ .85, .87). The labeling and content of the
household composition. Regardless of household composition, for fifth factor differ somewhat across measures. In the FFPI, the
545 families (92.7%), both parents provided data. For 39 children fifth factor, (Intellectual) Autonomy, describes an individual’s
only the mother (n ⫽ 594) completed the questionnaires, and for capability to take independent decisions, maintain an indepen-
5 children only the father (n ⫽ 550) completed the questionnaires. dent opinion, and not be influenced by social pressures to
All parents were of Belgian nationality. The mean age of the conform (Hendriks et al., 1999b, 2002; Hofstee, De Raad, &
mothers was 36 years 11 months (range 27 years 1 month to 52 Goldberg, 1992). Autonomy is related to but different from
years 0 months) and of the fathers was 39 years (range 27 years 11
Openness, which describes an individual’s unconventionality in
months to 61 years 10 months). The respective percentages of
addition to nonconformity (Ashton, Lee, & Goldberg, 2004),
mothers and fathers with various educational levels were as fol-
and from Intellect (De Fruyt, McCrae, Szirmak, & Nagy, 2004),
lows: elementary school (0.9%, 3.0%), secondary education
which includes unconventionality, behavioral flexibility, and
(41.1%, 43.3%), nonuniversity higher education (45.2%, 34.4%),
imagination. We refer to the fifth factor by the label used in the
and university (12.8%, 19.2%).
FFPI, that is, Autonomy.
At Time 2, participants were 480 families. Of these 480 fami-
Overreactivity. Adolescents rated their mothers’ and fathers’
lies, 452 (94.2%) were two-parent households and 26 (5.4%) were
overreactivity using the Parenting Scale (Arnold, O’Leary, Wolff,
one-parent households; in the latter group, all of the parents were
& Acker, 1993; Prinzie, Onghena, & Hellinckx, 2007). Overreac-
divorced. Two families did not supply information about their
tivity consists of nine items and measures parents’ tendency to
household composition. For the families that indicated that the
other parent was no longer present at Time 2 (n ⫽ 5 and n ⫽ 38 respond with anger, frustration, meanness, and irritation, impa-
for mothers and fathers, respectively), the respective parent was tiently and aversively, to problematic behavior of their children.
excluded from subsequent analyses. Regardless of household com- Items present discipline encounters followed by two options
position, for 414 families (86.3%), both parents provided data. For that act as opposite anchor points for a seven-point scale, on
52 families only the mother (n ⫽ 466) and for 12 families only the which 1 indicates a high probability of using an effective
father (n ⫽ 426) completed the questionnaires. For 430 families, discipline strategy (e.g., “When I misbehave my mother speaks
adolescents rated the parenting behaviors of both their parents. For to me calmly”) and 7 indicates a high probability of making a
34 families adolescents rated only their mother’s parenting (n ⫽ discipline mistake (“When I misbehave my mother raises her
464), and for 3 families adolescents rated only their father’s voice or yells”). The instrument has adequate test–retest reli-
parenting (n ⫽ 433). ability, distinguishes clinical from nonclinical samples, and has
For both mothers and fathers, missing values were missing been validated against behavioral observations of parenting
completely at random: Little’s missing completely at random test (Arnold et al., 1993; Locke & Prinz, 2002). In this study,
for mothers, ␹2(18, N ⫽ 589) ⫽ 22.08, p ⫽ .23; for fathers, ␹2(18, Cronbach’s alphas for adolescents’ ratings of their mothers’ and
N ⫽ 518) ⫽ 21.65, p ⫽ .25. To maximize sample size, we imputed fathers’ overreactivity were .81 and .80, respectively.
missing values for mothers and fathers separately, with the Warmth/involvement. Adolescents rated their parents’
expected-maximization algorithm (Schafer & Graham, 2002). In warmth using the Parenting Practices Questionnaire (Robinson,
this way, Time 1 data from all mothers and fathers for scores on Mandleco, Olsen, & Hart, 1995). This scale comprises 11 items
the personality dimensions could be included in the analyses (n ⫽ and assesses the extent to which parents exhibit warm parenting
589 and n ⫽ 518, respectively). and are involved in their children’s lives (e.g., “Expresses
affection by hugging, kissing, and holding me”). Adolescents
rated the items on a five-point scale ranging from 1 ⫽ never to
Measures
5 ⫽ always. In this study, Cronbach’s alphas for adolescents’
Parental personality. Mothers and fathers described their per- ratings of their mothers’ and fathers’ warmth were .87 and .87,
sonality characteristics using the Five-Factor Personality Inventory respectively.
 PERSONALITY, SENSE OF COMPETENCE, AND PARENTING 1699

Sense of competence. Parents assessed their sense of compe- Y ⫽ b0 ⫹ bX1 ⫹ bM ⫹ bZ ⫹ bX1Z ⫹ bMZ
tence in the parenting role using the Sense of Competence scale of
the Parenting Stress Index (Abidin, 1990; Jones & Prinz, 2005). ⫹ e 共Edwards & Lambert, 2007, p. 10兲,
This scale consists of 13 items and measures parents’ perceptions
where Y denotes the outcome, X the predictor, M the mediator, and
of their competence in terms of generally handling difficulties,
Z the moderator. The paths of the mediated model are as follows.
coping with daily demands, and exercising control over child
Relations between predictors and mediator are first-stage effects
behavior (e.g., “I feel that I am not very good at being a parent”).
(path a), relations between mediator and outcome are second-stage
Parents rated the items on a six-point scale ranging from 1 ⫽
effects (path b), and direct effects entail relations between predic-
disagree completely to 6 ⫽ agree completely. Cronbach’s alphas
tor and outcome (path c). The approach of Edwards and Lambert
for fathers’ and mothers’ sense of competence in this study were
overcomes several limitations of other procedures for combining
.71 and .72, respectively.
moderation and mediation. It produces statistical tests for moder-
ation of each of the three individual paths of the mediated model
Overview of Analyses (paths a, b, and c). It also gives estimates of indirect effects (a ⫻
First, descriptive statistics of the measures and bivariate b) and total effects (a ⫻ b ⫹ c) for different values of the
relationships among the measures were examined. Parents from moderator (gender).
the same family are assumed to resemble each other more than
they do parents from different families. This resemblance can
Results
occur because partners were similar when they paired together,
because they influenced each other’s personality and behaviors Descriptive Statistics and Correlation Analyses
subsequently (Kenny, 1996), or because they parent the same
child. Hence, we addressed the first aim in this study, regarding The means and standard deviations of the measures and bivari-
the impact of the Big Five dimensions in the context of the other ate correlations between the measures are reported in Table 1.
dimensions, by conducting multilevel analyses for warmth and One-way analyses of variance indicated that mothers rated them-
overreactivity separately with MLwiN 2.0 (Rasbash et al., selves significantly higher than did fathers on Extraversion, F(1,
2000). These analyses also allowed for identification of relevant 916) ⫽ 18.85, p ⬍ .001, ␩2 ⫽ .02; Agreeableness, F(1, 916) ⫽
variables for subsequent analyses. The relevance of variables 44.93, p ⬍ .001, ␩2 ⫽ .03; and Conscientiousness, F(1, 916) ⫽
was based on the criteria of Baron and Kenny (1986): (a) the 8.30, p ⬍ .01, ␩2 ⫽ .01. Mothers rated themselves significantly
predictor should be significantly related to the outcome variable lower than did fathers on Emotional Stability, F(1, 916) ⫽ 20.76,
in the model without the mediator, (b) the mediator should be p ⬍ .001, ␩2 ⫽ .02, and Autonomy, F(1, 916) ⫽ 19.44, p ⬍ .001,
significantly related to the predictors, and (c) the mediator ␩2 ⫽ .02. Adolescents rated mothers as significantly more warm
should be significantly related to the outcome. than fathers, F(1, 916) ⫽ 190.82, p ⬍ .001, ␩2 ⫽ .14, and as
In subsequent analyses, we addressed the second research aim significantly less overreactive than fathers, F(1, 916) ⫽ 3.95, p ⬍
regarding the mediational role of sense of competence and the .05, ␩2 ⫽ .00. Differences between mothers and fathers are around
third aim regarding possible gender differences in the associations a quarter of a standard deviation (SD) for Extraversion (d ⫽ 0.25),
by using the framework for moderated mediation proposed by Emotional Stability (d ⫽ ⫺0.26), and Autonomy (d ⫽ ⫺0.26).
Edwards and Lambert (2007). In this framework, mediation is Mothers scored 0.37 SD higher than fathers on Agreeableness and
framed as a path model (see Figure 1), and relationships among 0.17 SD higher than fathers on Conscientiousness. Mothers were
variables are expressed with regression equations. The combined rated over three quarters of a SD higher than fathers on warmth
model for moderation and mediation in reduced form equation is (d ⫽ 0.82) and 0.11 SD lower than fathers on overreactivity.
represented as Correlations between maternal and paternal personality character-

Table 1
Correlations, Mean Scores, and Standard Deviations for Measures of Personality, Sense of Competence, and Parenting

Measure 1 2 3 4 5 6 7 8 9 10 Mf SDf
ⴱ ⴱ ⴱ
1. Age — .09 ⫺.05 ⫺.02 .06 ⫺.04 .11 ⫺.06 .08 ⫺.09 38.43 4.15
2. Educational level .12ⴱⴱ — ⫺.08 ⫺.17ⴱⴱⴱ ⫺.03 .00 .21ⴱⴱⴱ .04 ⫺.05 .04 3.85 1.35
3. Extraversion ⫺.12ⴱⴱ ⫺.13ⴱⴱ — ⫺.03 .00 .15ⴱⴱⴱ ⫺.24ⴱⴱⴱ .24ⴱⴱⴱ ⫺.10ⴱ .11ⴱ 0.85 1.08
4. Agreeableness .10ⴱ ⫺.02 ⫺.12ⴱⴱ — .14ⴱⴱⴱ .17ⴱⴱⴱ .03 .14ⴱⴱⴱ ⫺.10ⴱ .10ⴱ 1.58 1.11
5. Conscientiousness .01 .00 ⫺.04 .14ⴱⴱⴱ — .09ⴱ .16ⴱⴱⴱ .11ⴱ .01 ⫺.01 1.03 1.02
6. Emotional Stability ⫺.02 .05 .14ⴱⴱⴱ .07 ⫺.02 — ⫺.13ⴱⴱ .28ⴱⴱⴱ ⫺.09ⴱ .04 1.13 0.89
7. Autonomy .16ⴱⴱⴱ .15ⴱⴱ ⫺.10ⴱ .17ⴱⴱⴱ .01 ⫺.12ⴱⴱ — .08 ⫺.06 .03 0.66 1.13
8. Sense of competence ⫺.02 ⫺.04 .17ⴱⴱⴱ .15ⴱⴱⴱ .08 .34ⴱⴱⴱ .21ⴱⴱⴱ — ⫺.28ⴱⴱⴱ .18ⴱⴱⴱ 3.15 0.61
9. Overreactivity ⫺.03 ⫺.07 .02 ⫺.11ⴱⴱ .03 ⫺.14ⴱⴱⴱ ⫺.05 ⫺.31ⴱⴱⴱ — ⫺.40ⴱⴱⴱ 3.36 1.00
10. Warmth ⫺.01 .03 .15ⴱⴱⴱ .12ⴱⴱ .11 ⫺.04 .13ⴱⴱⴱ .24ⴱⴱⴱ ⫺.36ⴱⴱⴱ — 3.06 0.73
Mm 36.44 3.77 1.11 2.02 1.20 0.88 0.38 3.08 3.24 3.62 — —
SDm 3.61 1.16 0.97 1.11 1.04 0.96 1.02 0.64 0.92 0.63 — —

Note. Results regarding mothers (n ⫽ 589) are presented below the diagonal; results regarding fathers (n ⫽ 518) are presented above the diagonal.
ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01. ⴱⴱⴱ p ⬍ .001.
1700 DE HAAN, PRINZIE, AND DEKOVIĆ

istics were small to moderate (range between r ⫽ .16, p ⬍ .001, fathers display different levels of warmth and overreactivity, as-
and r ⫽ .29, p ⬍ .001).1 Maternal and paternal sense of compe- sociations between personality and parenting are similar for moth-
tence, too, were related only moderately (r ⫽ .36, p ⬍ .001). ers and fathers.
Correlations between adolescent-reported maternal and paternal We computed structure coefficients for each of the personality
parenting were moderate to high (warmth, r ⫽ .34, p ⬍ .001; dimensions (results can be obtained from Amaranta D. de Haan
overreactivity, r ⫽ .56, p ⬍ .001; Cohen, 1988). upon request) to examine whether the other dimensions also ex-
Intercorrelations indicate that for both mothers and fathers, plained variance in the parenting measures. From these analyses, it
Agreeableness, Emotional Stability, and sense of competence were appeared that Emotional Stability, too, was related to maternal
related to lower levels of overreactivity. For fathers but not for overreactivity. Its insignificant association with overreactivity in
mothers, Extraversion was also related to lower levels of overre- the multivariate analyses, however, suggests that the contribution
activity. Mothers’ and fathers’ Extraversion, Agreeableness, and of Emotional Stability is redundant in the context of the other
sense of competence were related to higher levels of warmth. For dimensions (Dunlap & Landis, 1998).
mothers but not for fathers, Autonomy was also related to higher To test whether the theoretically chosen mediator, sense of
levels of warmth. Both fathers’ and mothers’ Extraversion, Agree- competence, should be included in the models, we conducted
ableness, and Emotional Stability were related to a higher sense of multilevel analyses for the prediction of sense of competence from
competence. For mothers, Autonomy was also related to a higher personality (see Table 2) and for the prediction of overreactivity
sense of competence, and for fathers, Conscientiousness was also and warmth from sense of competence (see Table 3). Table 2
related to a higher sense of competence. A more stringent test of shows that more extraverted, agreeable, emotionally stable, and
the relative impact of each of the personality dimensions on sense autonomous parents had a higher sense of competence. Parental
of competence and parenting was provided by the multilevel personality explained 18% of the variance in sense of competence.
analyses, in which the contribution of each of the five personality Although gender was a significant predictor of sense of compe-
dimensions was taken into account. tence, the interactions of gender with the personality dimensions
were not significant. The associations thus appeared to be similar
Prediction of Overreactive and Warm Parenting From for mothers and fathers. Parents who had a higher sense of com-
Personality and Sense of Competence petence exhibited lower levels of overreactivity and higher levels
of warmth (see Table 3). Sense of competence explained 8% of the
The following analyses were employed to examine the relative variance in overreactivity and 4% of the variance in warmth. The
impact of the personality dimensions and to identify variables for interactions of sense of competence with parental gender for
the test of moderated mediation. In the first step, the contribution overreactivity and warmth were nonsignificant, and they explained
of the personality dimensions and that of sense of competence are 0% of the variance in either parenting behavior (see Table 3).
examined in predicting overreactivity and warmth separately. Each These results again indicate that the associations were similar for
multilevel analysis consists of four models. The first model con- mothers and fathers.
cerns an intercept-only model. Second, the background variables In summary, when all personality dimensions were taken into
(parental age and educational level) and the moderator (gender) are account, Extraversion and Agreeableness were related to sense of
included. Third, personality or sense of competence is added. competence, to overreactivity, and to warmth. Sense of compe-
Fourth, the interactions between gender and the predictors are tence, too, was related to overreactivity and to warmth. On the
included (see Table 2).2 If the analyses are conducted in several basis of these findings, Extraversion and Agreeableness were
steps, an estimate of the explained variance on the individual level
selected for the moderated mediation analyses.
can be obtained that can be interpreted like an ordinary R2. It
should, however, be kept in mind that the measure is conditioned
on the variance to be explained that is not already explained by Moderated Mediation Analyses
differences between families (Hayes, 2006). To reduce multicol-
linearity between the main effects and the interaction terms, we Because the model described by Edwards and Lambert (2007)
standardized the values of the continuous independent variables (z includes one predictor at a time, their model was extended. The
scores) before we created interaction variables (Aiken & West, reduced form of the equations for overreactivity or warmth, re-
1991). Collinearity diagnostics showed that all the variance infla- gressed on two personality dimensions, can be presented as
tion factors were less than 1.06 and all the tolerance values were
greater than .94. This suggests there were no multicollinearity Y ⫽ b0 ⫹ bX1 ⫹ bX2 ⫹ bM ⫹ bZ ⫹ bX1Z ⫹ bX2Z ⫹ bMZ ⫹ e.
problems (Stevens, 2002, pp. 92–93).
Table 2 shows that extraverted and agreeable parents showed 1
lower levels of overreactivity and higher levels of warmth. The Additional testing showed that the differences between the model in
which the intercorrelations between the measures were constrained to be
differences in size of coefficients between these two personality
equal for mothers and fathers and the model in which the intercorrelations
dimensions are negligible. Personality explained 2% of the
between measures were estimated freely for mothers and fathers yielded a
individual-level variance in overreactivity and 3% of the variance significant improvement in model fit, ⌬␹2(39) ⫽ 80.43, p ⬍ .001. These
in warmth. Although parental gender appeared as a significant results suggest that the intercorrelations between the measures were not
predictor of both parenting behaviors, none of the interactions with equal for mothers and fathers.
parental gender were significant for either behavior, and the inter- 2
For the sake of brevity, only the coefficients obtained in the last model
actions explained no variance in overreactivity or warmth (0% and are shown. Full results can be obtained from Amaranta D. de Haan upon
1%, respectively). These results suggest that, although mothers and request.
 PERSONALITY, SENSE OF COMPETENCE, AND PARENTING 1701

Table 2
Results of Multilevel Analyses Predicting Overreactive Parenting, Warm Parenting, and Sense of Competence From Personality

Overreactive parenting Warm parenting Sense of competence

Predictor B SE ␤ ⌬R2
B SE ␤ ⌬R 2
B SE ␤ ⌬R2

0. Intercept 3.08ⴱⴱⴱ 0.32 3.33ⴱⴱⴱ 0.22 3.37ⴱⴱⴱ 0.19

1. Background variables .07 .15 .01
Age 0.01 0.01 .05 ⫺0.01 0.01 ⫺.05 ⫺0.01 0.00 ⫺.04
Educational level ⫺0.05 0.03 ⫺.07 0.03 0.02 .05 0.02 0.01 .03
Gendera ⫺0.10 0.06 ⫺.05 0.50ⴱⴱⴱ 0.03 .34 ⫺0.07ⴱ 0.03 .06
2. Personality .02 .03 .18
Extraversion ⫺0.10ⴱ 0.04 ⫺.10 0.08ⴱⴱ 0.03 .10 0.16ⴱⴱⴱ 0.02 .25
Agreeableness ⫺0.10ⴱ 0.04 ⫺.10 0.08ⴱⴱ 0.03 .11 0.06ⴱ 0.03 .10
Conscientiousness 0.05 0.04 .05 ⫺0.04 0.03 .05 0.03 0.02 .05
Emotional Stability ⫺0.09 0.04 ⫺.09 0.00 0.03 .00 0.14ⴱⴱⴱ 0.03 .23
Autonomy ⫺0.07 0.04 ⫺.07 0.02 0.03 .02 0.08ⴱⴱⴱ 0.02 .13
3. Interactions .00 .01 .01
Extraversion ⫻ Gender 0.10 0.06 .07 0.02 0.04 .03 ⫺0.05 0.03 .06
Agreeableness ⫻ Gender 0.01 0.06 .01 ⫺0.03 0.04 .04 0.00 0.03 .00
Conscientiousness ⫻ Gender 0.01 0.06 .01 0.07 0.04 .09 0.00 0.03 .00
Emotional Stability ⫻ Gender ⫺0.05 0.06 ⫺.04 ⫺0.04 0.04 .06 0.06 0.03 .07
Autonomy ⫻ Gender 0.02 0.06 .02 0.04 0.04 .06 0.06 0.03 .07
a
Gender was coded as follows: fathers ⫽ 0, mothers ⫽ 1.
ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01. ⴱⴱⴱ p ⬍ .001.

The test of the model started from two series of regression We then used the coefficients obtained from these analyses to
analyses. In the first series, the mediator (sense of competence) compute path coefficients for both indirect and total effects sepa-
was regressed on the relevant predictors (personality dimensions), rately for mothers and fathers using a macro (available online at
the moderator (gender), and the interaction term of predictors and http://www.apa.org/journals/supplemental/met_12_1_1/MET_
moderator (see Table 4). The results show that, although the main edwards.zip). We used constrained nonlinear regression (for de-
effects of Extraversion, Agreeableness, and gender on sense of tails regarding this procedure, see Edwards & Lambert, 2007, pp.
competence were significant, the interaction terms between these 12–13) and coefficients from 1,000 bootstrap samples to test the
measures were not. significance of expressions involving products of coefficients (i.e.,
In the second series of analyses, the outcome variables (over- differences across levels of the moderator variable) with confi-
reactivity and warmth) were each regressed on the predictors dence intervals. In other words, tests of gender differences are
(Extraversion and Agreeableness), the mediator (sense of compe- equivalent to tests of interactions reported in Tables 4 and 5 but are
tence), the moderator (gender), the interaction of predictors and based on bias-corrected confidence intervals derived from 1,000
moderator, and the interaction of mediator and moderator (see bootstrap estimates.
Table 5). The main effects of Extraversion, Agreeableness, and Table 6 shows the estimated coefficients for fathers and mothers
sense of competence on warmth were significant, and the main separately, for each stage of the model and for direct, indirect, and
effect of gender on both overreactivity and warmth was significant. total effects. For illustrative purposes, the results for Extraversion
However, the interaction terms between gender and the other are depicted in Figure 2. As can be seen in Table 6, none of the
measures were not significant. direct associations between the personality dimensions and over-

Table 3
Results of Multilevel Analyses Predicting Overreactive and Warm/Involved Parenting From Sense of Competence

Overreactive parenting Warm parenting

Predictor B SE ␤ ⌬R2 B SE ␤ ⌬R2

0. Intercept 3.22ⴱⴱⴱ 0.31 3.30ⴱⴱⴱ 0.20

1. Background variables .01 .15
Age 0.01 0.01 .03 ⫺0.02 0.01 .04
Educational level ⫺0.04 0.02 .06 ⫺0.02 0.02 .03
Gendera ⫺0.13ⴱ 0.05 .07 0.56ⴱⴱⴱ 0.03 .38
2. Competence ⫺0.27ⴱⴱⴱ 0.04 .28 .08 0.12ⴱⴱⴱ 0.03 .16 .04
3. Competence ⫻ Gender 0.03 0.05 .02 .00 ⫺0.01 0.03 .01 .00
a
Gender was coded as follows: fathers ⫽ 0, mothers ⫽ 1.
ⴱ
p ⬍ .05. ⴱⴱⴱ p ⬍ .001.
1702 DE HAAN, PRINZIE, AND DEKOVIĆ

Table 4 mediated by sense of competence. For warmth, in contrast, mater-

Test of the Paths From Predictors to Sense of Competence nal Extraversion and paternal and maternal Agreeableness had
(Path A) direct associations above and beyond the indirect effects through
sense of competence. All associations were similar for mothers and
Sense of competence for fathers.
Predictor B SE ␤ ⌬R2
Discussion
Intercept 3.20ⴱⴱⴱ 0.03
Gendera ⫺0.15ⴱⴱⴱ 0.03 ⫺.12 .07
In this study, associations between parental personality, as-
Extraversion 0.16ⴱⴱⴱ 0.03 .21
Agreeableness 0.08ⴱⴱⴱ 0.03 .13 sessed by parents, and parenting, assessed 6 years later by adoles-
Extraversion ⫻ Gender ⫺0.04 0.04 ⫺.04 cents, were examined in a large community sample of European
Agreeableness ⫻ Gender 0.02 0.04 .02 mothers and fathers. Our measurement of personality is in accor-
dance with the comprehensive Big Five model of personality,
a
Gender was coded as follows: fathers ⫽ 0, mothers ⫽ 1.
ⴱⴱⴱ
p ⬍ .001. which facilitates comparison of research findings (Belsky &
Barends, 2002). The inclusion of all Big Five dimensions allows
for an examination of how personality relates to parenting over
reactivity were significant for mothers or for fathers when sense of time when all dimensions are taken into account (Prinzie et al., in
competence was included in the model. Apparently, relations press) and of the impact of each of the dimensions in the context
between maternal and paternal Extraversion, Agreeableness, and of the other dimensions. The extensive period of time between
overreactivity are transmitted through sense of competence. Com- assessments is based on the conceptualization of personality as a
parison of path coefficients indicates that all of the associations set of dispositions that have enduring effects over time (Clark et
were similar (i.e., not statistically different) for mothers and for al., 2000), and the use of independent assessments of personality
fathers. and parenting eliminates the risk of shared method variance.
Comparison of path coefficients for warm parenting of mothers Results suggest that more extraverted and agreeable parents
and fathers (see Table 6) indicates that for mothers, but not for show lower levels of overreactivity and higher levels of warmth.
fathers, the direct association of Extraversion remained significant The associations are in line with our expectations, and they are
when sense of competence was included in the model (see Figure consistent with previous research (Clark et al., 2000; Losoya et al.,
2), although the difference between mothers and fathers did not 1997; Prinzie et al., 2005; Smith et al., 2007). Within the five-
reach significance. In addition, both for mothers and for fathers, factor structure of personality, both Extraversion and Agreeable-
the direct association between Agreeableness and warmth re- ness concern an individual’s behavioral tendencies in social inter-
mained significant when sense of competence was included in the actions (Caspi et al., 2005). Results from this study show that these
model. Thus, maternal Extraversion and paternal and maternal two dimensions are also important for the interpersonal behaviors
Agreeableness were related to warmth above and beyond the of parents toward their children. Extraversion and Agreeableness
indirect association through sense of competence. Moreover, the have somewhat different connotations regarding social interactions
insignificance of the differences across parental gender suggests (Caspi et al., 2005; Goldberg, 1990; Hendriks et al., 1999a,
that these associations, too, were similar for mothers and for 1999b). Extraversion reflects an individual’s capacity to enjoy
fathers. social interactions and optimism and dominance in social relations.
In summary, Extraversion and Agreeableness were related both Agreeableness reflects an individual’s motives to maintain positive
to overreactivity and to warmth. These personality dimensions social relations and has been found to relate to regulation of
apparently were equally important in the prediction of levels of emotions during social interactions (Tobin, Graziano, Vanman, &
overreactivity and warmth 6 years later. Furthermore, the relations Tassinary, 2000). Therefore, the impact of these two dimensions of
between parental personality and overreactivity were completely parenting may operate in somewhat different ways. Nevertheless,

Table 5
Test of the Paths From Predictors and Mediator to Outcomes (Paths B and C)

Overreactive parenting Warm parenting

Predictor B SE ␤ ⌬R2 B SE ␤ ⌬R2

Intercept 3.35ⴱⴱⴱ 0.04 3.07ⴱⴱⴱ 0.03

Gendera ⫺0.12ⴱ 0.05 ⫺.06 .19 0.54ⴱⴱⴱ 0.03 .36 .09
Extraversion ⫺0.03 0.04 ⫺.04 0.05ⴱ 0.03 .06
Agreeableness ⫺0.06 0.04 ⫺.06 0.06ⴱ 0.03 .08
Competence ⫺0.26ⴱⴱⴱ 0.05 ⫺.27 0.10ⴱⴱⴱ 0.03 .14
Extraversion ⫻ Gender 0.07 0.06 .05 0.03 0.04 .03
Agreeableness ⫻ Gender 0.01 0.06 .01 ⫺0.01 0.04 ⫺.01
Competence ⫻ Gender 0.02 0.06 .02 ⫺0.01 0.03 .01
a
Gender was coded as follows: fathers ⫽ 0, mothers ⫽ 1.
ⴱ
p ⬍ .05. ⴱⴱⴱ p ⬍ .001.
 PERSONALITY, SENSE OF COMPETENCE, AND PARENTING 1703

Table 6
Analyses of Moderation by Parental Gender of the Mediated Models

Overreactive parenting Warm parenting

First-stage Second-stage First-stage Second-stage

effect effect Direct effect Indirect Total effect effect Direct effect Indirect Total
Predictor (path a) (path b) (path c) effect effect (path a) (path b) (path c) effect effect

Extraversion
Fathers 0.23ⴱⴱⴱ ⫺0.28ⴱⴱⴱ ⫺0.03 ⫺0.06ⴱⴱⴱ ⫺0.10ⴱ 0.23ⴱⴱⴱ 0.11ⴱⴱⴱ 0.05 0.03ⴱⴱⴱ 0.08ⴱⴱ
Mothers 0.21ⴱⴱⴱ ⫺0.28ⴱⴱⴱ 0.07 ⫺0.06ⴱⴱⴱ 0.01 0.21ⴱⴱⴱ 0.13ⴱⴱⴱ 0.09ⴱⴱⴱ 0.03ⴱⴱⴱ 0.11ⴱⴱⴱ
Difference ⫺0.02 0.00 0.10 0.00 0.11 ⫺0.02 0.02 0.03 0.00 0.03
Agreeableness
Fathers 0.15ⴱⴱⴱ ⫺0.28ⴱⴱⴱ ⫺0.07 ⫺0.04ⴱⴱⴱ ⫺0.11ⴱⴱ 0.15ⴱⴱⴱ 0.11ⴱⴱⴱ 0.06ⴱ 0.02ⴱⴱⴱ 0.08ⴱⴱ
Mothers 0.18ⴱⴱⴱ ⫺0.28ⴱⴱⴱ ⫺0.05 ⫺0.05ⴱⴱⴱ ⫺0.10ⴱⴱ 0.18ⴱⴱⴱ 0.13ⴱⴱⴱ 0.07ⴱ 0.02ⴱⴱⴱ 0.09ⴱⴱⴱ
Difference 0.04 0.00 0.02 ⫺0.01 0.01 0.04 0.02 0.01 0.01 0.01

Note. Effects involving products of coefficients (indirect effects, total effects) were tested using confidence intervals derived from bootstrapping.
ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01. ⴱⴱⴱ p ⬍ .001.

Extraversion and Agreeableness were shown to remain important are related. The finding that parental sense of competence is a
for the parenting of both mothers and fathers, above and beyond psychological mechanism that explains why personality is impor-
the impact of the other personality dimensions. The findings ex- tant for parenting is an important contribution of this study. In
tend the results of studies in which the links between personality particular, Extraversion and Agreeableness predispose individuals
and parenting were examined contemporaneously and in which to a higher sense of competence, which in turn is related to higher
both constructs were assessed by self-report. It was demonstrated levels of warmth and lower levels of overreactivity. We added to
that the links remained even across an extensive period of time and existing research by showing that sense of competence is an
were not due to shared method variance. important mediator not only between psychosocial and contextual
In this study, Conscientiousness, Emotional Stability, and Au- factors and parenting (e.g., Gondoli & Silverberg, 1997; Teti &
tonomy were not significantly related to parenting in the multivar- Gelfand, 1991) but also between dispositional factors and parent-
iate analyses, although the bivariate correlations between these ing. Moreover, this study is, to our knowledge, the first to examine
personality dimensions and parenting were significant and the the correlates of sense of competence among fathers.
structure coefficient of Emotional Stability with overreactive par- It should be noted that the direct associations among Extraver-
enting was considerable. These findings are consistent with pre- sion, Agreeableness, and overreactivity became insignificant when
vious research that also took all Big Five dimensions into account sense of competence was included in the model, whereas the direct
(e.g., Smith et al., 2007). Taken together, these results suggest that associations between these dimensions and warmth remained sig-
the impact of Conscientiousness, Emotional Stability, and Auton- nificant. Parents with a lower sense of competence, who feel they
omy on parenting is not different from the impact of Extraversion cannot change their child’s (negative, problematic) behaviors, may
and Agreeableness (Dunlap & Landis, 1998). avoid the challenging task of disciplining their child in an effortful
Although previous work has examined relations between paren- way (e.g., by induction). Instead, they are more likely to uninten-
tal personality and parenting, few studies have examined why they tionally use overreactive techniques (e.g., yelling, screaming) to

A. Overreactive Parenting: Mothers B. Overreactive Parenting: Fathers

COM COM
0.21*** -0.28*** 0.23*** -0.28***

EX OR EX OR
0.07 -0.03

C. Warm Parenting: Mothers D. Warm Parenting: Fathers

COM COM
0.21*** 0.13*** 0.23*** 0.11***

EX WR EX
0.09*** 0.05 WR

Figure 2. Mediated models showing simple effects for mothers and fathers. Panels A and B concern
overreactive parenting; panels C and D concern warm parenting. EX ⫽ Extraversion; COM ⫽ sense of
competence; OR ⫽ overreactive parenting; WR ⫽ warm parenting. ⴱⴱⴱ p ⬍ .001.
1704 DE HAAN, PRINZIE, AND DEKOVIĆ

discipline their child. Warmth, on the other hand, can be seen to be analyses suggest that there was more stability than change between
related to how parents think they should parent their children. 2001 and 2004 (results can be obtained from Amaranta D. de Haan
Parents differ on the amount of warmth they deem appropriate to upon request). Rank-order stability (the relative placement of each
show in interactions with their children. Both Extraversion and respondent compared with the other respondents) was moderate to
Agreeableness, with their connotations of enjoying social interac- high for mothers and moderate for fathers (mothers, rs ⬎ .5;
tions and wanting to maintain positive social relations, thus may fathers, .4 ⬍ rs ⱕ .5). It could be the case that although one’s
affect parents’ warmth, but not parents’ overreactivity, above and Emotional Stability measured at a single point in time does not
beyond the indirect effect through sense of competence. uniquely predict parenting, a (large) change in one’s Emotional
Another possibility for these differential outcomes comes from Stability score does. Therefore, we urge researchers to explore
behavioral– genetic research. Personality is substantially influ- how changes over time in the Big Five personality dimensions
enced by genetics, with heritability estimates ranging from .40 to affect parenting and vice versa.
.60 (Bouchard, 2004; Bouchard & Loehlin, 2001). With respect to Due to the large sample size, all the measures were assessed by
parenting, behavioral– genetic studies have suggested evidence for questionnaires. Fortunately, we were able to use the adolescent
the heritability of warmth but not of overreactivity (Kendler, 1996; ratings of parental behaviors; this eliminated the risk of shared
Kendler & Baker, 2007). The differential associations found for method variance and emphasized the importance of the percep-
overreactivity and warmth may thus have arisen because warmth is tions of adolescents of their parents’ behaviors for their own
affected by genetics, as is parental personality, whereas overreac- adjustment. Relations between adolescents’ assessments and ob-
tivity is less genetically influenced (Kendler & Baker, 2007; Prin- served parenting are, however, suggested to be moderate (Smet-
zie et al., in press). Given the design of this study the argument of ana, Campione-Barr, & Metzger, 2006), and future research should
genetic influence is, of course, speculative, and future research include both observed and self- or other-reported measures of
could assess this possibility with a genetically informed design. parenting.
Personality explained only a small portion of individual-level We analyzed a population-based sample, and thus our results
variance in overreactivity and warmth (2% and 3%, respectively) should not be extended to clinical populations. For example,
and a substantial portion of individual-level variance in parental among parents who show very low levels of Emotional Stability,
sense of competence (18%). Furthermore, standardized beta coef-
this dimension may still play an important role in their parenting,
ficients were in the small range. These results suggest that the
even when all Big Five dimensions are analyzed simultaneously.
effect sizes of personality are small; however, the impact of
Indeed, research specifically assessing Emotional Stability in clin-
personality on parenting behaviors should not be understated.
ical samples consistently shows linkages between this factor and
Because the effects of personality (as well as of parenting) accu-
parenting (Belsky & Barends, 2002). Notwithstanding these find-
mulate over a child’s lifetime, a focus on a single parenting
ings, our results show that in nonclinical samples and within the
behavior measured at a single point in time may underestimate the
framework of the Big Five, parental Emotional Stability does not
contribution of personality to parenting and to children’s develop-
uniquely impact parenting in the context of the other dimensions.
ment.
Future research should assess whether our results replicate in
Relations among personality, sense of competence, and parent-
clinical samples. Moreover, we suggest that researchers should
ing were found to be similar for mothers and fathers. This finding
is in line with those of other studies that employed a stringent assess the impact of all Big Five dimensions simultaneously when
(statistical) test to assess possible gender differences (e.g., Mets- examining the impact of personality on interpersonal relations in
äpelto & Pulkkinen, 2003; Verhoeven et al., 2007). Gender may both clinical and nonclinical samples.
become less important for parenting when the influence of other Parental sense of competence is thought both to impact parent-
individual factors is taken into account (i.e., personality, age, ing through a complex interplay among affective, motivational,
educational level). Likely then, there are gender differences in the cognitive, and behavioral pathways (Bandura, 1989; Coleman &
bivariate correlations, but these differences “wash out” in more Karraker, 1997; McGillicuddy-DeLisi & Sigel, 1995; Teti & Can-
complex models. This study showed that the similarities hold, too, daleria, 2002) and to have a direct impact on parenting. Future
for the mediating role of sense of competence. To our knowledge, research should extend and qualify the results of this study by
this finding has not been assessed in previous work. examining which of these (or other) pathways play a role in linking
parental sense of competence with parenting and whether these
pathways are similar or different for each of the Big Five dimen-
Limitations and Future Research sions.
Some limitations of the current study should be mentioned. Although no gender differences were found in the associations
First, sense of competence and parenting were assessed concur- among personality, sense of competence, and parenting, we did
rently, and conclusions about causal relationships thus cannot be find mean-level gender differences in these constructs. These
made. In other words, it is possible that parents’ parenting behav- results are consistent with existing work (Smetana et al., 2006).
iors lead to a higher or lower level of sense of competence, instead Differences between mothers’ and fathers’ parenting behaviors
of the other way around. have been shown to be associated with marital discord and poorer
Second, we predicted parenting from previous level of person- child functioning (Block, Block, & Morrison, 1981; Harvey,
ality. Although such a model provides a useful framework for 2000). An important issue that could be assessed by future research
predicting the outcome over an extensive period of time, it does regards the extent to which discrepancies between parents in their
not address changes in these constructs. Although stability of the personality are related to differences in parenting and how these
personality dimensions at Time 2 could not be assessed, additional discrepancies are related to interparental conflict.
 PERSONALITY, SENSE OF COMPETENCE, AND PARENTING 1705

In the current study, the focus was on intraindividual mecha- Belsky, J., & Barends, N. (2002). Personality and parenting. In M. H.
nisms that explain linkages between parental personality and par- Bornstein & B. Webber (Eds.), Handbook of parenting (2nd ed., pp.
enting, although the proposed model is not intended to be exhaus- 415– 438). Mahwah, NJ: Erlbaum.
tive. Indeed, Belsky (1984) theorized that parental personality also Belsky, J., Crnic, K., & Woodworth, S. (1995). Personality and parenting:
shapes parenting behaviors by influencing the broader context Exploring the mediating role of transient mood and daily hassles. Jour-
nal of Personality, 63, 905–929.
(e.g., work, social relations). The importance of mechanisms un-
Belsky, J., & Jaffee, S. R. (2006). The multiple determinants of parenting.
derlying the relations between personality and parenting should In D. J. Cohen & D. Cicchetti (Eds.), Developmental psychopathology:
not be understated, and, for a more complete understanding of the Vol. 3. Risk, disorder, and adaptation (pp. 38 – 85). Hoboken, NJ: Wiley.
associations between personality and parenting, other psycholog- Block, J., Block, J., & Morrison, A. (1981). Parental agreement–
ical and contextual mediators need to be examined. disagreement on child rearing orientations and gender-related personal-
ity correlates in children. Child Development, 52, 963–974.
General Conclusions and Implications Bornstein, M. H., Hendricks, C., Hahn, C.-S., Haynes, O. M., Painter,
K. M., & Tamis-LeMonda, C. S. (2003). Contributors to self-perceived
The present study made an important contribution to the knowl- competence, satisfaction, investment, and role balancing in maternal
edge base on “why parents parent the way they do” (Belsky, 1984, parenting: A multivariate ecological analysis. Parenting: Science and
p. 83). In addition to showing that maternal and paternal Extra- Practice, 3, 285–326.
version and Agreeableness were related to lower levels of overre- Bouchard, T. J. (2004). Genetic influence on human psychological traits.
activity and higher levels of warmth 6 years later, the study found Current Directions in Psychological Science, 13, 148 –151.
that sense of competence was an important mediating mechanism Bouchard, T. J., & Loehlin, J. C. (2001). Genes, evolution, and personality.
in these associations. Behavior Genetics, 31, 243–273.
Sense of competence was strongly related to lower levels of Brody, G. H., McBride-Murry, V., Kim, S., & Brown, A. C. (2002).
Longitudinal pathways to competence and psychological adjustment
overreactivity and to higher levels of warmth/involvement. Unlike
among African American children living in rural single-parent house-
personality, sense of competence is relatively changeable, and it
holds. Child Development, 73, 1505–1516.
can be targeted in intervention programs. Treatment programs that Bugental, D. B., Blue, J., & Cruzcosa, M. (1989). Perceived control over
address sense of competence should include a focus on why caregiving outcomes: Implications for child abuse. Developmental Psy-
parents feel they cannot effectively manage parenting tasks. One chology, 25, 532–539.
important factor to take into account is parental personality, al- Bugental, D. B., & Cortez, V. L. (1988). Physiological reactivity to
though other factors, such as child behavior and the (lack of) fit responsive and unresponsive children as moderated by perceived con-
between parental and child characteristics, should not be ne- trol. Child Development, 59, 686 – 693.
glected. Including sense of competence in treatment programs Caspi, A. (1998). Personality development across the life course. In W.
designed to diminish overreactive parenting and promote warm Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child
parenting may considerably increase the efficacy of such pro- psychology: Vol. 3. Social, emotional and personality development (5th
ed., pp. 311–388). New York: Wiley.
grams.
Caspi, A., Roberts, B. W., & Shiner, R. L. (2005). Personality develop-
ment: Stability and change. Annual Review of Psychology, 56, 453– 484.
References Chan, K.-Y., & Drasgow, F. (2001). Toward a theory of individual differ-
Abidin, R. R. (1990). Introduction to the special issue: The stresses of ences and leadership: Understanding the motivation to lead. Journal of
parenting. Journal of Clinical Child Psychology, 19, 298 –301. Applied Psychology, 86, 481– 498.
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and Clark, L. A., Kochanska, G., & Ready, R. (2000). Mothers’ personality and
interpreting interactions. London: Sage. its interaction with child temperament as predictors of parenting behav-
Arnold, D. S., O’Leary, S. G., Wolff, L. S., & Acker, M. M. (1993). The ior. Journal of Personality and Social Psychology, 79, 274 –285.
Parenting Scale: A measure of dysfunctional parenting in discipline Cohen, J. (1988). Statistical power analysis for the behavioral sciences
situations. Psychological Assessment, 5, 137–144. (2nd ed.). Mahwah, NJ: Erlbaum.
Ashton, M. C., Lee, K., & Goldberg, L. R. (2004). A hierarchical analysis Coleman, P. K., & Karraker, K. (1997). Self-efficacy and parenting quality:
of 1,710 English personality-descriptive adjectives. Journal of Person- Findings and future applications. Developmental Review, 18, 47– 85.
ality and Social Psychology, 87, 707–721. Coleman, P. K., & Karraker, K. (2003). Maternal self-efficacy beliefs,
Asscher, J. J., Hermanns, J. M. A., & Deković, M. (2008). Effectiveness of competence in parenting, and toddlers’ behavior and developmental
the Home-Start parenting support program: Behavioral outcomes for status. Infant Mental Health Journal, 24, 126 –148.
parents and children. Infant Mental Health Journal, 29, 95–113. De Fruyt, F., McCrae, R. R., Szirmak, A., & Nagy, J. (2004). The
Bandura, A. (1989). Social cognitive theory. Annals of Child Development, Five-Factor Personality Inventory as a measure of the five-factor model:
6, 1– 60. Belgian, American, and Hungarian comparisons with the NEO-PI–R.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable Assessment, 11, 207–215.
distinction in social psychological research: Conceptual, strategic, and Denham, S. A., Workman, E., Cole, P. M., Weissbrod, C., Kendziora,
statistical considerations. Journal of Personality and Social Psychology, K. T., & Zahn-Waxler, C. (2001). Prediction of externalizing behavior
51, 1173–1182. problems from early to middle childhood: The role of parental social-
Barrera, M., Jr., Prelow, H. M., Dumka, L. E., Gonzales, N. A., Knight, ization and emotion expression. Development and Psychopathology, 12,
G. P., Michaels, M. L., et al. (2002). Pathways from family economic 23– 45.
conditions to adolescents’ distress: Supportive parenting, stressors out- Dishion, T. J., & Patterson, G. R. (2006). The development and ecology of
side the family, and deviant peers. Journal of Community Psychology, antisocial behavior in children and adolescents. In D. J. Cohen & D.
30, 135–152. Cicchetti (Eds.), Developmental psychopathology: Vol. 3. Risk, disorder,
Belsky, J. (1984). The determinants of parenting: A process model. Child and adaptation (pp. 503–541). Hoboken, NJ: Wiley.
Development, 55, 83–96. Donovan, W. A., Leavitt, L. A., & Walsh, R. O. (1990). Maternal self-
1706 DE HAAN, PRINZIE, AND DEKOVIĆ

efficacy: Illusory control and its effect on susceptibility to learned Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a unifying social
helplessness. Child Development, 61, 1638 –1647. cognitive theory of career and academic interest, choice, and perfor-
Dunlap, W. P., & Landis, R. S. (1998). Interpretations of multiple regres- mance. Journal of Vocational Behavior, 45, 79 –122.
sion borrowed from factor analysis and canonical correlation. Journal of Locke, L. M., & Prinz, R. J. (2002). Measurement of parental discipline
General Psychology, 125, 397– 407. and nurturance. Clinical Psychology Review, 22, 895–929.
Edwards, J. R., & Lambert, L. S. (2007). Methods for integrating moder- Losoya, S. H., Callor, S., Rowe, D. C., & Goldsmith, H. H. (1997). Origins
ation and mediation: A general analytic framework using moderated of familial similarity in parenting: A study of twins and adoptive
path analysis. Psychological Methods, 12, 1–22. siblings. Developmental Psychology, 33, 1012–1023.
Gerdes, A. C., Hoza, B., Arnold, L. E., Pelham, W. E., Swanson, J. M., McCrae, R. R., & Costa, P. T., Jr. (1999). A five-factor theory of person-
Wigal, T., & Jensen, P. S. (2007). Maternal depressive symptomatology ality. In L. A. Pervin & O. P. John (Eds.), Handbook of personality:
and parenting behavior: Exploration of possible mediators. Journal of Theory and research (pp. 139 –153). New York: Guilford Press.
Abnormal Child Psychology, 35, 705–714. McGillicuddy-DeLisi, A. V., & Sigel, I. E. (1995). Parental beliefs. In
Goldberg, L. R. (1990). An alternative “description of personality”: The M. H. Bornstein (Ed.), Handbook of parenting (pp. 333–358). Mahwah,
Big Five factor structure. Journal of Personality and Social Psychology, NJ: Erlbaum.
59, 1216 –1229. Metsäpelto, R.-L., & Pulkkinen, L. (2003). Personality and parenting:
Gondoli, D. M., & Silverberg, S. B. (1997). Maternal emotional distress Neuroticism, Extraversion, and Openness to Experience as discrimina-
and diminished responsiveness: The mediating role of parenting efficacy tive factors. European Journal of Personality, 17, 59 –78.
and parental perspective taking. Developmental Psychology, 33, 861– Nauta, M. M. (2004). Self-efficacy as a mediator of the relationships
868. between personality factors and career interests. Journal of Career
Goodnow, J. J., & Collins, W. A. (1990). Development according to Assessment, 12, 381–394.
parents: The nature, sources, and consequences of parents’ ideas. Hill- Neiderhiser, J. M., Pike, A., Hetherington, E. M., & Reiss, D. (1998).
sdale, NJ: Erlbaum. Adolescent perceptions as mediators of parenting: Genetic and environ-
Harvey, E. (2000). Parenting similarity and children with attention-deficit/ mental contributions. Developmental Psychology, 34, 1459 –1469.
hyperactivity disorder. Child & Family Behavior Therapy, 22, 39 –54. Olsen, S. F., Martin, P., & Halverson, C. F., Jr. (1999). Personality, marital
Hayes, A. F. (2006). A primer on multilevel modeling. Human Commu- relationships, and parenting in two generations of mothers. International
nication Research, 32, 385– 410. Journal of Behavioral Development, 23, 457– 476.
Hendricks, J. W., & Payne, S. C. (2007). Beyond the Big Five: Leader goal Ozer, D., & Benet-Martı́nez, V. (2006). Personality and the prediction of
orientation as a predictor of leadership effectiveness. Human Perfor- consequential outcomes. Annual Review of Psychology, 57, 401– 421.
mance, 20, 317–343. Powers, S. I., Welsh, D. P., & Wright, V. (1994). Adolescents’ affective
Hendriks, A. A. J. (1997). The construction of the Five-Factor Personality experience of family behaviors: The role of subjective understanding.
Inventory. Unpublished doctoral dissertation, University of Groningen, Journal of Research on Adolescence, 4, 585– 600.
Groningen, the Netherlands. Prinzie, P., Onghena, P., & Hellinckx, W. (2007). Re-examining the
Hendriks, A. A. J., Hofstee, W. K. B., & De Raad, B. (1999a). De Five Parenting Scale: Reliability, factor structure, and concurrent validity of
Factor Personality Inventory. Lisse, the Netherlands: Swets Test Pub- a scale for assessing the discipline practices of mothers and fathers of
lishers. elementary-school-aged children. European Journal of Psychological
Hendriks, A. A. J., Hofstee, W. K. B., & De Raad, B. (1999b). The Assessment, 23, 24 –31.
Five-Factor Personality Inventory (FFPI). Personality and Individual Prinzie, P., Onghena, P., Hellinckx, W., Grietens, H., Ghesquière, P., &
Differences, 27, 307–325. Colpin, H. (2004). Parent and child personality characteristics as pre-
Hendriks, A. A. J., Hofstee, W. K. B., & De Raad, B. (2002). The dictors of negative discipline and externalizing problem behaviour in
Five-Factor Personality Inventory: Assessing the Big Five by means of children. European Journal of Personality, 18, 73–102.
brief and concise statements. In B. De Raad & M. Perugini (Eds.), Big Prinzie, P., Onghena, P., Hellinckx, W., Grietens, H., Ghesquière, P., &
Five assessment (pp. 79 –108). Seattle, WA: Hogrefe & Huber. Colpin, H. (2005). Direct and indirect relationships between parental
Hofstee, W. K. B., De Raad, B., & Goldberg, L. R. (1992). Integration of personality and externalising behaviour: The role of negative parenting.
the Big Five and circumflex approaches to trait structure. Journal of Psychologica Belgica, 45, 123–145.
Personality and Social Psychology, 63, 146 –163. Prinzie, P., Stams, G. J. J. M., Deković, M., Reijntjes, A. H. A., & Belsky,
Jones, T. L., & Prinz, R. J. (2005). Potential roles of parental self-efficacy J. (in press). The relation between parent personality and parenting: A
in parent and child adjustment: A review. Clinical Psychology Review, meta-analytic review. Journal of Personality and Social Psychology.
25, 341–363. Rasbash, J., Browne, W., Goldstein, H., Yang, M., Plewis, I., Healy, M., et
Kendler, K. S. (1996). Parenting: A genetic– epidemiological perspective. al. (2000). A user’s guide to MLwiN (2nd ed.). London: Institute of
American Journal of Psychiatry, 153, 11–20. Education.
Kendler, K. S., & Baker, J. H. (2007). Genetic influences on measures of Robinson, C. C., Mandleco, B., Olsen, S. F., & Hart, C. H. (1995).
the environment: A systematic review. Psychological Medicine, 37, Authoritative, authoritarian and permissive parenting practices: Devel-
615– 626. opment of a new measure. Psychological Reports, 77, 819 – 830.
Kendler, K. S., Sham, P. C., & MacLean, C. J. (1997). The deteminants of Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state
parenting: An epidemiological, multi-informant, retrospective study. of the art. Psychological Methods, 7, 147–177.
Psychological Medicine, 27, 549 –563. Schunk, D. H. (1990). Goal setting and self-efficacy during self-regulated
Kenny, D. A. (1996). Models of non-independence in dyadic research. learning. Educational Psychologist, 25, 71– 86.
Journal of Social and Personal Relationships, 13, 279 –294. Sexton, T. L., & Tuckman, B. W. (1991). Self-beliefs and behavior: The
Kochanska, G., Friesenborg, A. E., Lange, L. A., & Martel, M. M. (2004). role of self-efficacy and outcome expectations over time. Personality
Parents’ personality and infants’ temperament as contributors to their and Individual Differences, 12, 725–736.
emerging relationship. Journal of Personality and Social Psychology, Smetana, J. G., Campione-Barr, N., & Metzger, A. (2006). Adolescent
86, 744 –759. development in interpersonal and societal contexts. Annual Review of
Krathwohl, D. R. (1993). Methods of educational and social science Psychology, 57, 255–284.
research. New York: Longman. Smith, C. L., Spinrad, T. L., Eisenberg, N., Gaertner, B. M., Popp, T. K.,
 PERSONALITY, SENSE OF COMPETENCE, AND PARENTING 1707

& Maxon, E. (2007). Maternal personality: Longitudinal associations to Tobin, N. M., Graziano, W. G., Vanman, E. J., & Tassinary, L. G. (2000).
parenting behavior and maternal emotional expressiveness toward tod- Personality, emotional experience, and the effort to control emotions.
dlers. Parenting: Science and Practice, 7, 305–329. Journal of Personality and Social Psychology, 79, 656 – 669.
Snyder, J., Cramer, A., Afrank, J., & Patterson, G. R. (2005). The contri- Verhoeven, M., Junger, M., Van Aken, C., Deković, M., & Van Aken,
butions of ineffective discipline and parental hostile attributions of child M. A. G. (2007). Parenting during toddlerhood: Contributions of paren-
misbehavior to the development of conduct problems at home and tal, contextual and child characteristics. Journal of Family Issues, 28,
school. Developmental Psychology, 41, 30 – 41. 1663–1691.
Stevens, J. P. (2002). Applied multivariate statistics for the social sciences. Wood, J. J., McLeod, B. D., Sigman, M., Hwang, W.-C., & Chu, B. C.
Mahwah, NJ: Erlbaum. (2003). Parenting and childhood anxiety: Theory, empirical findings,
Teti, D. M., & Candelaria, M. (2002). Parenting competence. In M. H. and future directions. Journal of Child Psychology and Psychiatry, 44,
Bornstein & B. Webber (Eds.), Handbook of parenting (2nd ed., pp. 134 –151.
149 –180). Mahwah, NJ: Erlbaum.
Teti, D. M., & Gelfand, D. M. (1991). Behavioral competence among Received May 6, 2008
mothers of infants in the first year: The mediational role of maternal Revision received March 16, 2009
self-efficacy. Child Development, 62, 918 –929. Accepted March 24, 2009 䡲