the construct validity of openness to experience in middle ... · the construct validity of...
TRANSCRIPT
The Construct Validity of Openness to Experience in Middle Childhood: Contributions from Personality and
Temperament
by
Kathrin Herzhoff
A thesis submitted in conformity with the requirements for the degree of Master of Arts
Department of Psychology University of Toronto
© Copyright by Kathrin Herzhoff 2011
ii
The Construct Validity of Openness to Experience in Middle
Childhood: Contributions from Temperament and Personality
Kathrin Herzhoff
Master of Arts
Department of Psychology
University of Toronto
2011
Abstract
Controversy exists over the validity of child Openness to Experience (OE), which is typically
considered a major trait in adult personality models. In an effort to establish construct validity
for child OE, data were collected for 346 children (51% girls) approximately 9–10 years of age
(M = 9.92, SD = 0.83). Parents completed questionnaires about their children’s personality,
temperament, and behavioral problems and competencies. Factor analyses of relevant
personality and temperament facets revealed a robust and measurable OE factor made up of
three facets: Intellect, Imagination, and Sensitivity. Evidence for convergent and discriminant
validity was established via associations with other higher-order personality traits, behavioral
problems, and behavioral competencies. The results underscore the importance of drawing from
both temperament and personality literatures in attempts to establish construct validity for child
trait domains as well as of moving beyond the higher-order domain and examining facet-level
associations between OE and child behavior.
iii
Acknowledgments
I would like to thank my supervisor, Dr. Jennifer L. Tackett, for her invaluable support and
encouragement throughout my years under her supervision. Furthermore, I would also like to
thank my subsidiary advisor, Dr. Marc A. Fournier, for his insightful comments and
recommendations on earlier drafts of this thesis. Finally, I would like to thank the families in the
study for their participation and the research assistants at the Personality Across Development
Lab for their help in the data collection. This research was supported by funding from the Social
Sciences and Humanities Research Council of Canada.
iv
Table of Contents
Acknowledgments ......................................................................................................................... iii
Table of Contents ............................................................................................................................ iv
List of Tables .................................................................................................................................. vi
List of Figures ................................................................................................................................vii
Chapter 1 .......................................................................................................................................... 1
1 Introduction ................................................................................................................................. 1
1.1 Defining Openness to Experience in Middle Childhood ..................................................... 2
1.2 Relationship between Openness to Experience and Personality in Childhood ................... 4
1.3 Relationship between Openness to Experience and Childhood Behavior ........................... 6
1.4 The Present Study ................................................................................................................ 7
Chapter 2 .......................................................................................................................................... 9
2 Method ........................................................................................................................................ 9
2.1 Participants .......................................................................................................................... 9
2.2 Measures .............................................................................................................................. 9
2.2.1 Personality ............................................................................................................... 9
2.2.2 Temperament ......................................................................................................... 10
2.2.3 Behavioral Problems and Competencies ............................................................... 10
2.3 Procedure ........................................................................................................................... 10
Chapter 3 ........................................................................................................................................ 12
3 Results ....................................................................................................................................... 12
3.1 The Structure of Openness to Experience in Middle Childhood ....................................... 12
3.2 Relationship between Openness to Experience and Personality in Middle Childhood ..... 13
3.3 Relationship between Openness to Experience and Behavior in Middle Childhood ........ 14
Chapter 4 ........................................................................................................................................ 15
v
4 Discussion ................................................................................................................................. 15
4.1 The Structure of Openness to Experience in Middle Childhood ....................................... 15
4.2 Relationship between Openness to Experience and Personality in Middle Childhood ..... 16
4.3 Relationship between Openness to Experience and Behavior in Middle Childhood ........ 17
4.4 Limitations and Future Directions ..................................................................................... 18
4.5 Conclusion ......................................................................................................................... 20
References...................................................................................................................................... 22
Tables ............................................................................................................................................. 27
Figures ........................................................................................................................................... 31
vi
List of Tables
Table 1 Item Loadings > .40 for Exploratory Factor Analysis of Mother-Reported Openness to
Experience Items .......................................................................................................................... 27
Table 2 Summary of Correlations between the Openness to Experience, ICID, and CBCL scales
Based on Averaged Mother and Father Report. .......................................................................... 28
Table 3 Hierarchical Multiple Regression Analyses Predicting the Lower-Order Openness to
Experience Facets from Mother- and Father-Reported Personality Traits. ............................... 29
Table 4 Multiple Regression Analyses Predicting CBCL Scales from Mother- and Father-
Reported Openness to Experience. .............................................................................................. 30
vii
List of Figures
Figure 1. Structure of Openness to Experience in middle childhood based on confirmatory factor
analysis of father report. .............................................................................................................. 31
1
Chapter 1
1 Introduction
When parents describe characteristics of their children, items from the fifth factor of the Five
Factor Model (FFM)—Openness to Experience (OE)—play a prominent role (Halverson et al.,
2003). When considering how best to characterize children, it is easy to apply terms such as
imaginative, curious, and inquisitive to individuals from an early age. Despite the natural ease
with which we conceptualize young children in these terms, researchers currently lack a clear
and cohesive understanding of when OE emerges in development and how best to conceptualize
the construct at an early age. Gaining a greater understanding of child OE is important, as
previous research has shown that it predicts numerous outcomes such as behavioral problems
(e.g., Gjerde & Cardilla, 2009; Prinzie et al., 2004) and behavioral competencies (e.g., Abe,
2005; Muris, Meesters, & Diederen, 2009).
The purpose of the present study is to advance our current knowledge about OE in childhood by
establishing evidence for construct validity of the trait in middle childhood. Construct validity is
a broad concept that refers to whether a test is measuring its target construct in addition to the
ongoing process of establishing a nomological network around a given construct (Cronbach &
Meehl, 1955). Attempts to establish construct validity frequently include structural
investigations and analyses examining convergent and discriminant validity. Factor analytic
approaches are often combined with rational item selection to examine the structural validity of
a measure, whereas tests of convergent and discriminant validity are typically conducted to
contribute to the nomological network around a construct. In the current study, we attempted to
establish construct validity for child OE by taking the novel approach of defining child OE in
terms of both personality and temperament constructs and then testing the nomological network
of child OE by examining convergent and discriminant validity via associations with other
personality traits, behavioral problems, and behavioral competencies.
2
1.1 Defining Openness to Experience in Middle Childhood
Controversy exists over whether an OE factor analogous to that found in adults even emerges in
childhood, resulting in a lack of understanding over how best to conceptualize child OE. Some
studies have found an analog of OE in childhood (Goldberg, 2001; Halverson et al., 2003)
whereas others suggest that it may not be a meaningful personality dimension prior to
adolescence (Lamb, Chuang, Wessels, Broberg, & Hwang, 2002) and again others have found
that OE is less differentiated in childhood than it is in adulthood and is primarily defined by
items measuring Conscientiousness (Mervielde, Buyst, & De Fruyt, 1995; Mervielde & De
Fruyt, 2000). This controversy is further mirrored in major temperament models of early
individual differences, which do not converge on a major OE trait (Shiner & Caspi, 2003).
Difficulty in establishing an OE trait in childhood has been exacerbated by measurement
limitations in this area of research (Tackett, 2006). For example, Lamb et al. (2002) used a
measure of OE with low internal consistency (Cronbach’s α = .33–.57). They concluded that
this was lack of evidence for a meaningful OE trait prior to adolescence, which may have been
preemptive given that others have reliably measured OE in childhood (e.g., Goldberg, 2001;
Halverson et al., 2003).
Even the psychometrically sound child personality measures that have emerged over the last
decade cover a much narrower OE content base than do typical adult measures. For example,
the OE analog in the Inventory of Child Individual Differences (ICID; Halverson et al., 2003) is
heavily defined by intellectual tendencies whereas the OE analog in the Hierarchical Personality
Inventory for Children (HiPIC; Mervielde & De Fruyt, 1999) is heavily defined by imaginative
tendencies. Similarly, in his re-analysis of Digman’s child personality data, Goldberg (2001)
primarily selected markers of the OE factor defined by intellectual and imaginative tendencies.
In some samples, he added a marker that focused on the child’s aesthetics. Adult OE measures,
on the other hand, often provide broader coverage of OE. For example, the NEO PI-R (Costa &
McCrae, 1992) measures the OE facets Fantasy (e.g., vivid imagination and fantasy life),
Aesthetics (e.g., appreciation of art, music, and poetry), Feelings (e.g., receptiveness to inner
emotional states and valuing emotional experience), Actions (e.g., trying new activities, visiting
3
new places, and trying new foods), Ideas (e.g., intellectually curious and open to new ideas), and
Values (e.g., re-examination of traditional social, religious, and political values). To date, it
remains unclear whether the content of child OE should be similarly broadened. The present
study will address this issue by integrating relevant characteristics from both temperament and
personality frameworks in our conceptualization of OE.
As previously noted, researchers working from a temperament framework do not include items
directly measuring OE in any of their major temperament models (Shiner & Caspi, 2003).
Furthermore, although researchers increasingly propose that personality constructs are
theoretically related to temperament constructs (e.g., Shiner & Caspi, 2003), researchers have
not shown much progress in empirically relating temperament constructs to OE in child
populations (Mervielde, De Fruyt, & Jarmuz, 1998). One study in a sample of Flemish children
found that a factor made up of the temperament constructs Perceptual Sensitivity and Low
Intensity Pleasure was moderately correlated with OE measured by the HiPIC (Mervielde & De
Fruyt, 2002) in early childhood (De Pauw, Mervielde, & Van Leeuwen, 2009). Perceptual
Sensitivity measures the detection of slight, low intensity stimuli from the external environment
whereas Low Intensity Pleasure measures pleasure or enjoyment related to situations involving
low stimulus intensity, rate, complexity, novelty, and incongruity (Simonds & Rothbart, 2004).
Perceptual Sensitivity seems conceptually similar to Goldberg’s (2001) marker of OE that
focused on the child’s aesthetics. Although Low Intensity Pleasure does not seem to
conceptually overlap with OE (e.g., McCrae, 1994), it has been related to characteristics of OE
in early childhood (De Pauw et al., 2009) and to the temperament construct Fantasy/Openness in
middle childhood (Simonds & Rothbart, 2004). Specifically, open children were found to have
higher levels of Low Intensity Pleasure. Given these findings, we selected this variable for the
present investigation. Broadening the content of child OE to include temperament constructs
might prove critical to the study of OE earlier in life given that temperament constructs such as
Perceptual Sensitivity can be reliably measured even in infants (Gartstein & Rothbart, 2003).
Thus, empirical integration of such constructs into a broad and robust child OE domain was a
4
primary goal of the current investigation in hopes of contributing to a more developmentally
generalizable understanding of OE.
1.2 Relationship between Openness to Experience and Personality in Childhood
In addition to better understanding content and structure, it is important to examine whether
child OE is clearly differentiated from—and additionally shows theoretically expected patterns
of convergence with—other personality traits. Some evidence does suggest that relations
between OE and other personality traits differ in adult and child populations. Specifically, OE
primarily covaries with Extraversion in adults (Markon, Krueger, & Watson, 2004; DeYoung,
Peterson, & Higgins, 2002) whereas it primarily covaries with Conscientiousness in children
(Goldberg, 2001; Mervielde & De Fruyt, 2000; Tackett et al., 2011) and to a lesser degree with
Extraversion (Goldberg, 2001). Furthermore, one study found that child OE predicted adult OE
only in females whereas it predicted adult Conscientiousness in males (Gjerde & Cardilla,
2009).
One reason for the closer relationship between child OE and Conscientiousness could be that
child OE measures focus more on Intellect content than adult measures do (Gjerde & Cardilla,
2009). For example, in Mervielde and colleagues’ (1995, 2000) studies, OE measured with a
focus on Intellect content was not clearly differentiated from Conscientiousness. When they
split their fifth factor into two separate factors, one measuring Intellect and the other measuring
Openness, they found that only the Intellect factor was related to Conscientiousness in 10-to-12-
year-old children (Mervielde et al., 1995). Intellect was also negatively related to
Agreeableness; a finding that other studies that measured a higher-order OE factor did not find
(Goldberg, 2001). Another reason for the closer relationship between OE and Conscientiousness
might be the relatively greater importance of academic achievement during childhood compared
to adulthood. Specifically, OE and Conscientiousness might seem more intertwined to reporters
of child personality because of the developmental salience of academic achievement, which
involves both Conscientiousness and OE (e.g., John, Caspi, Robins, Moffit, Stouthamer-Loeber,
1998; Mervielde et al., 1995). A better understanding of the relationship between Intellect and
5
Conscientiousness is predicated on broadening the OE construct to include aspects other than
Intellect. The present study will do so by introducing temperament constructs such as Perceptual
Sensitivity and Low Intensity Pleasure to the study of OE.
Similarly, Perceptual Sensitivity has been shown to load on both Effortful Control (which is
largely related to Conscientiousness) and Extraversion (Gartstein & Rothbart, 2003; Rothbart,
Ahadi, Hershey, & Fisher, 2001) in early childhood. However, in contrast to findings with OE,
Perceptual Sensitivity has also demonstrated some connections with Neuroticism and
Agreeableness characteristics. In later childhood, Perceptual Sensitivity also showed substantial
covariance with Sadness and Affiliation/Agreeableness (Simonds & Rothbart, 2004), and in
adolescence, Perceptual Sensitivity (and Low Intensity Pleasure) loaded on
Affiliation/Agreeableness and was related to depressive mood (Ellis & Rothbart, 2001). A
conceptually related construct to Perceptual Sensitivity, the trait of Absorption, also covaried
highly with Negative Emotionality/Neuroticism in middle childhood (Tackett, Krueger, Iacono,
& McGue, 2008). It could be that heightened sensitivity, potentially common to both Perceptual
Sensitivity and Absorption, shows relationships with Neuroticism beginning in middle
childhood.
In summary, different aspects of child OE are hypothesized to relate to different personality
traits in the present study. Specifically, Intellect aspects of OE are hypothesized to relate most
strongly to Conscientiousness. Aspects of OE such as Openness/Fantasy are hypothesized to
relate most strongly to Extraversion given that these aspects have been emphasized more in
adult OE measures than in child measures, and adult measures of OE have been found to be
most strongly related to Extraversion. Less robust findings for links between Intellect–low
Agreeableness, Perceptual Sensitivity–Neuroticism and Perceptual Sensitivity–Agreeableness
suggest additional tentative hypotheses.
6
1.3 Relationship between Openness to Experience and Childhood Behavior
Some of the important behavioral domains that child OE predicts both concurrently and
longitudinally are behavioral problems (e.g., Gjerde & Cardilla, 2009; Prinzie et al., 2004) and
behavioral competencies (e.g., Abe, 2005; Muris et al., 2009). To date, more research has been
conducted on the relationship between OE and behavioral problems than on OE and
competencies.
Behavioral problems in children are commonly classified across two major dimensions:
externalizing and internalizing problems (Achenbach, 1978). Externalizing problems in children
typically include acting-out behaviors such as conduct and oppositional problems; internalizing
problems in children typically include anxiety and mood symptoms as well as somatic
complaints. Research on the association between OE and externalizing problems in childhood
has generally revealed negative associations (e.g., Ehrler, Evans, and McGhee, 1999; John et al.,
1994; Victor, 1994) with some inconsistent findings demonstrating positive associations
(Prinzie et al., 2004).Research on the association between OE and internalizing problems has
also yielded somewhat inconsistent findings. Some researchers have found that OE is unrelated
to internalizing problems (John et al., 1994), others finding negative associations (Abe, 2005;
Ehrler et al., 1999; Victor, 1994), and another finding a positive association, but only in girls
(Gjerde & Cardilla, 2009).
Results across all these studies may be somewhat mixed due to use of different measures of OE
and behavior as well as investigations only at the OE domain level. Facet-level investigations,
however, may help clarify such inconsistencies. Investigations with adult samples have shown
that OE facets have divergent associations with psychological problems (e.g., Carrillo, Rojo,
Sanchez-Bernardos, & Avia, 2001; DeYoung, Peterson, Séguin, & Tremblay, 2008; Ring,
Ryder, Lavgine & Bagby, 2008). Specifically, one study found that cognitive aspects of OE,
which are emphasized in the Intellect facet, were negatively related to externalizing problems
whereas Extraversion aspects of OE were positively related to externalizing problems
(DeYoung et al., 2008). Similarly, another study found that the Intellect aspects of OE were
7
negatively related to schizotypal personality disorder whereas the Extraversion aspects of OE
were positively related to schizotypal personality disorder (Ring et al., 2008). Other work has
suggested that the Fantasy facet of OE is positively related to depression (Carrillo et al., 2001).
In sum, different aspects of OE are hypothesized to relate to behavioral problems in divergent
directions, which may result in the higher-order OE factor showing inconsistent relationships
with childhood behavioral problems.
A dominant definition of behavioral competencies refers to successful adaptation in age-specific
developmental tasks (Masten & Curtis, 2000). Common domains examined are school, social
relationships, and extracurricular activities (e.g., Achenbach & Rescorla, 2001). OE has been
found to positively predict school competencies in children and adolescents ranging in age from
4 to 17 years (Digman, 1989; Graziano & Ward, 1992; John et al., 1998; Lamb et al., 2002;
Mervielde et al., 1995; Muris et al., 2005; Shiner, 2000). Child OE has also been positively
related to social competencies (Abe, 2005; Muris et al., 2005). The relationship between OE and
competencies in extracurricular activities has so far been unexplored, representing an important
area for closer investigation. A more nuanced understanding of childhood OE at the facet level,
which this study will attempt to provide, may provide a more detailed understanding of these
relationships.
1.4 The Present Study
The present study examined evidence for the construct validity of OE in middle childhood.
Specifically, the current investigation aimed to (a) broaden the content of child OE to include
both personality and temperament constructs, (b) employ reliable measures of children’s
personality and temperament, (c) collect data from multiple informants and, (d) examine
relationships with a variety of other constructs. The following hypotheses were tested:
A robust and psychometrically sound analog of OE would be identified in middle
childhood and include Intellect, Fantasy/Openness, Perceptual Sensitivity, and Low
Intensity Pleasure content.
8
Intellect aspects of child OE would be most strongly related to Conscientiousness,
Fantasy/Openness aspects of child OE would be most strongly related to Extraversion,
and Perceptual Sensitivity aspects would be most strongly related with Agreeableness
and Neuroticism.
The higher-order OE factor would not be related to behavioral problems. Intellect
aspects of OE would be negatively related to externalizing problems and
Fantasy/Openness aspects of OE would be positively related to internalizing problems.
The higher-order OE factor would be positively related to school and social
competencies. Intellect aspects of OE would be positively related with school
competencies, above and beyond contributions from general OE. No hypotheses were
formulated regarding extracurricular activities, given a lack of previous research on this
topic.
9
Chapter 2
2 Method
2.1 Participants
Participants were 328 mothers and 226 fathers of 346 children (51% female) primarily 9-to-10-
years of age (M = 9.92 years, SD = 0.83 years). The child’s ethnicity was designated by the
parents: 70.9% European/White, 13.3% Multiracial, 9.8% Asian/Pacific Islander, 2.9%
African/American, 0.6% Hispanic, 0.6% Other, and 2% did not report ethnicity. Participants
were recruited through a database of interested parents maintained by the Psychology
Department at the University of Toronto, advertisements in newspapers, and flyers posted in the
Greater Toronto Area. All participants were fluent in English and children did not have previous
diagnoses of mental retardation, autism, or schizophrenia.
2.2 Measures
2.2.1 Personality
The Inventory of Child Individual Differences (Halverson et al., 2003) was completed by both
mothers and fathers. The ICID is an empirically derived bottom-up measure developed for use
in children and youth. It assesses five higher-order factors (analogous to the adult FFM) and 15
lower-order facets. It consists of 108 items which caregivers rated on a scale ranging from 1
(much less than the average child or not at all) to 7 (much more than the average child). The
current study utilized the lower-order facets Openness and Intellect as well as the higher-order
factors Extraversion, Agreeableness, Conscientiousness, and Neuroticism. Internal consistencies
for the higher-order factors ranged from α = .90 (Neuroticism) to α = .96 (Agreeableness) with a
mean of α = .94. Internal consistency for the Intellect facet was α = .95 and for the Openness
facet was α = .87 in the present sample.
10
2.2.2 Temperament
The Temperament in Middle Childhood Questionnaire (TMCQ; Simonds & Rothbart, 2004)
was also completed by both mothers and fathers. The TMCQ assesses 17 lower-order facets of
temperament. It consists of 157 items, which are rated on a scale ranging from 1 (Almost always
untrue) to 5 (Almost always true) with an additional option of “Not applicable”. Caregivers
were asked to indicate how well a statement described their child within the past 6 months. The
current study utilized the Fantasy/Openness (α = .80), Low Intensity Pleasure (α = .65), and
Perceptual Sensitivity (α = .81) dimensions.
2.2.3 Behavioral Problems and Competencies
The Child Behavior Checklist (CBCL; Achenbach & Rescorla, 2001) was completed by both
mothers and fathers. The CBCL assesses competencies, adaptive functioning, and behavioral,
emotional, and social problems. It consists of 118 problem behavior items, which are rated on a
scale ranging from 0 (not true) to 2 (often true of the child) and additional competency items,
which measure the amount and quality of the child’s involvement in social, academic, and
extracurricular activities. The problem behavior items are combined into the higher-order
problem domains Total Problems, Externalizing Behaviors, and Internalizing Behaviors, and the
competency items are combined into the higher-order competency domains Total Competencies,
School Competencies, Social Competencies, and Extracurricular Competencies. In our sample,
internal consistencies were α = .83 for Total Problems, α = .88 for Externalizing Behaviors, α =
.83 for Internalizing Behaviors, α = .81 for Total Competencies, α = .80 for School
Competencies, α = .63 for Social Competencies, and α = .69 for Extracurricular Competencies.
2.3 Procedure
Both parents independently completed questionnaires about their child either at home or at the
lab. One parent and the child came to the lab to complete the questionnaires as well as
interviews and behavioral tasks that were part of a larger study examining the development of
personality and behavioral outcomes. During this lab visit, questionnaires by the second parent
were returned. Data collection from a second caregiver (which was typically the father) was
11
added several months into the study, which accounts for a larger amount of missing data for
father participants than for mothers. Missing data were imputed using the maximum-likelihood-
based EM algorithm in SPSS. Families were compensated with CAD $40 and children received
two small gifts for full participation in the complete lab-based protocol which lasted
approximately 2–2.5 hours. The University of Toronto Ethics Board approved all study
protocols and materials.
12
Chapter 3
3 Results
3.1 The Structure of Openness to Experience in Middle Childhood
Given that data were collected from both mothers and fathers, mothers were used as the initial
exploratory sample with follow-up confirmatory analyses conducted with the fathers. Using
mother-report only, item-level exploratory factor analysis was conducted on the Openness and
Intellect facets of the ICID and the Fantasy/Openness, Perceptual Sensitivity, and Low Intensity
Pleasure facets of the TMCQ using Mplus (Muthén & Muthén, 1998–2006). One- through four-
factor structures were extracted using oblique rotation and a maximum likelihood estimator.
Results of the EFA supported a three-factor structure (RMSEA = .08) over a one-factor
(RMSEA = .10) and two-factor (RMSEA = .09) structure. Although the four-factor structure
(RMSEA = .07) provided a slightly better fit than the three-factor structure, clarity of extracted
factors was enhanced for the three-factor solution. Specifically, rather than measuring a separate
personality facet, the fourth factor seemed to measure an affinity to reading with the items “likes
reading or listening to make believe stories”, “enjoys looking at books”, and “likes quiet reading
time”. Thus, the three-factor solution was retained.
Based on content, we labeled the first factor “Intellect”, the second factor “Imagination”, and
the third factor “Sensitivity”. Items on the Intellect factor came from the ICID Intellect and
Openness scales, items on the Imagination factor came from the ICID Openness, the TMCQ
Openness/Fantasy and the TMCQ Low Intensity Pleasure scales, and items on the Sensitivity
factor came from the TMCQ Perceptual Sensitivity scale (for item loadings, see Table 1). Apart
from the item “likes the sound of poems” that loaded on the Imagination factor, none of the Low
Intensity Pleasure items loaded substantially on any of the factors.
The EFA was followed up with a confirmatory factor analysis on the father-report. For the CFA,
we eliminated items that did not load above 0.4 on any factor in the EFA. None of the remaining
13
items had secondary cross-loadings above 0.4. Following Little, Cunningham, Shahar, and
Widaman's (2002) recommendation to parcel items if the goal of a study is to examine the
nature of a construct, we randomly aggregated the items into clusters of four to five items within
factor. Results of the CFA also provided support for a three-factor structure (CFI= .99, RMSEA
= .06; see Figure 1). Based on the results of the CFA, we created indicators of the higher-order
OE factor and the lower-order facets Intellect, Imagination, and Sensitivity by summing the
items that were retained in the CFA. Internal consistencies were α = .93 for mother- and α = .91
for father-reported Intellect, α = .87 for mother- and α = .86 for father-reported Imagination, α =
.81 for mother- and α = .80 for father-reported Sensitivity, and α = .92 for the mother- and α =
.93 for the father-reported higher-order OE factor. Based on scores averaged across parent
informants, the Intellect and Imagination facet correlated with each other at 0.44 (p > .01), the
Intellect and Sensitivity facet correlated with each other at 0.28 (p > .01), and the Imagination
and Sensitivity facet correlated with each other at 0.39 (p > .01; see Table 2).
3.2 Relationship between Openness to Experience and Personality in Middle Childhood
Evidence for convergent and discriminant validity of the lower-order OE facets was investigated
via a series of hierarchical regressions. Scores from the higher-order ICID factors
Agreeableness, Conscientiousness, Extraversion, and Neuroticism were averaged across parent
informants (for correlations among those scales and the higher-order OE factor and its lower-
order facets, see Table 2). Each OE facet was designated the dependent variable in one
hierarchical regression, with the remaining two facets simultaneously entered in the first step
and the four higher-order ICID factors entered in the second step. This allowed for tests of
specificity among the three facets. Unique variance in the Intellect facet was most strongly
predicted by Conscientiousness (β = .49, p < .001; see Table 3). Unique variance in the
Imagination facet was most strongly predicted by Extraversion (β = 28, p < .001). Finally,
unique variance in the Sensitivity facet was most strongly predicted by Neuroticism (β = .52, p
< .001) and was also predicted by Agreeableness (β = .37, p < .001).
14
3.3 Relationship between Openness to Experience and Behavior in Middle Childhood
Evidence for both common and specific associations with behavioral problems and
competencies were investigated via separate hierarchical regressions for each of the three CBCL
problem behavior scales and the four behavioral competency scales. Regressions were
conducted separately with the higher-order OE factor as the independent variable in one
regression and its lower-order facets as the independent variables in another regression. Scores
from the CBCL were averaged across parents.
The results from these analyses revealed that Total Problems, Externalizing Behaviors, and
Internalizing Behaviors were not significantly predicted by the higher-order OE factor (see
Table 4). After controlling for the other OE facets, Total Problems was negatively predicted by
Intellect (β = -.27, p < .001) and positively predicted by Imagination (β = .16, p = .010).
Externalizing and Internalizing Behaviors were negatively predicted by unique variance in
Intellect. Total, School, Social, and Extracurricular Competencies were all positively predicted
by the higher-order OE factor (see Table 4). Total Competencies were also predicted by the
unique variance in Intellect (β = .36, p = .001) and Sensitivity (β = .11, p = .035). School
Competencies were predicted by the unique variance in Intellect (β = .47, p < .001). Social
Competencies were predicted by the unique variance in Intellect (β = .25, p < .001) and
Sensitivity (β = .11, p = .049). Extracurricular Competencies were predicted by the unique
variance in Intellect (β = .17, p = .004).
15
Chapter 4
4 Discussion
The present study is the first to provide a comprehensive investigation of the construct validity
of OE in childhood. A major strength of the study is the use of a multi-measure, multi-informant
approach to establishing construct validity for child OE, including both mother and father report
on reliable child personality and temperament measures. The results of the study move the field
of child personality research toward a clearer and more cohesive understanding of how to
conceptualize OE in middle childhood. These results also underscore the importance of drawing
from both temperament and personality literatures in establishing construct validity for child
trait domains.
4.1 The Structure of Openness to Experience in Middle Childhood
Our first hypothesis about the structure of OE was largely supported. Specifically, an analog of
OE—which included content from the domains of Intellect, Fantasy/Openness, and Perceptual
Sensitivity—was clearly identified in middle childhood, showing replication across informants.
The identification of a clear OE factor in middle childhood is consistent with previous work
(e.g., Halverson et al., 2003; Goldberg, 2001); however, distinct from previous work with both
child and adult populations, a new three-factor hierarchical structure for OE emerged. Both
personality and temperament items were important indicators of OE, underlining the importance
of including content from both domains (e.g., De Pauw et al., 2009). The present study
represents an important extension of previous work in child and adolescent samples by
empirically integrating these domains in a measure of OE for the first time. Given that
temperament constructs can be reliably measured from infancy (Gartstein & Rothbart, 2003),
these findings offer the opportunity for future work to take an even broader lifespan perspective
of the OE domain.
16
Contrary to our hypothesis, the OE factor did not include Low Intensity Pleasure content with
the exception of one item: “likes the sound of poems”, which loaded on the Imagination facet.
As noted previously, Low Intensity Pleasure is defined as the amount of pleasure or enjoyment
related to situations involving low level intensity, rate, complexity, novelty, and incongruity of a
stimulus (Simonds & Rothbart, 2004). Conceptually, these characteristics do not overlap with
OE (e.g., McCrae, 1994). Empirically, however, Low Intensity Pleasure has been related to
characteristics of OE in early childhood (De Pauw et al., 2009) and to the temperament
construct Fantasy/Openness in middle childhood (Simonds & Rothbart, 2004), which is why it
was selected for the present investigation. The findings regarding the relationship between Low
Intensity Pleasure and OE are too inconsistent to date to conclude whether the construct is
relevant to early OE, representing an important question to address in future research.
4.2 Relationship between Openness to Experience and Personality in Middle Childhood
With regard to the relationship between OE and personality in middle childhood, our hypotheses
were again largely supported. Specifically, our first hypothesis was supported: the Intellect facet
was most strongly predicted by Conscientiousness (e.g., Mervielde et al., 1995). Our second
hypothesis was also supported: the Imagination facet was most strongly predicted by
Extraversion. Adult OE measures cover more Fantasy/Imagination content than Intellect content
(Gjerde & Cardilla, 2009) and are, subsequently, most strongly related to Extraversion (Markon
et al., 2005), supporting our findings in the current study. Finally, the hypothesis that Sensitivity
would be most strongly predicted by Neuroticism and Agreeableness was also supported. This
supports previous findings that Perceptual Sensitivity and Absorption relate more strongly to
Neuroticism and Agreeableness/Affiliation in middle childhood (Ellis & Rothbart, 2001;
Simonds & Rothbart, 2004; Tackett et al., 2008) and into adulthood, where Absorption has
moderate positive loadings on Negative Emotionality (Tellegen, 1978/1982).
In summary, the present results support the broader hypothesis that child OE shows theoretically
expected patterns of convergence with other personality traits. The differential relationships
between the OE facets and other personality traits further underscore the importance of
17
broadening the construct of child OE to include aspects other than Intellect (which dominates
most current child OE measures) in order to examine these unique facet-level associations.
4.3 Relationship between Openness to Experience and Behavior in Middle Childhood
Finally, our hypotheses about the relationship between OE and behavior were also largely
supported. Specifically, the higher-order OE factor did not significantly predict any problem
scales, in line with overall inconsistent findings in the literature (e.g., John et al., 1994; Ehrler et
al., 1999; Victor, 1994; Prinzie et al., 2004). At the facet level, however, Total Problems was
negatively predicted by Intellect and positively predicted by Imagination. Furthermore, both
Externalizing and Internalizing Behaviors were negatively predicted by Intellect, after
controlling for the other OE facets.
Contrary to our initial hypothesis, the Imagination facet positively predicted only Total
Problems, after controlling for the other OE facets. Although we expected to find a relationship
between Imagination and Internalizing Behaviors, the rate of internalizing problems is still
relatively low in middle childhood and increases through adolescence (Bongers, Koot, van der
Ende & Verhulst, 2003). This may have limited our power to detect the predicted association in
this age group, emphasizing the need for future studies of these relationships in adolescent
samples. The finding that Imagination appears to serve as a potential risk factor for behavioral
problems maps on to research finding a link between creativity and psychopathology in adult
populations (Nettle, 2005; Post, 1994). Overall, these results suggest that, as hypothesized,
broadening the OE construct to include both personality and temperament items is relevant to
better understanding the important role that facets of OE play in predicting problem behaviors in
children. Importantly, these findings suggest that previous results from studies relying on only
broadband measures of child OE may have masked such relationships at the facet level (e.g.,
John et al., 1994; Ehrler et al., 1999; Prinzie et al., 2004).
Our hypotheses about the relationship between OE and behavioral competencies were
supported. Specifically, all CBCL competency domains were positively predicted by the higher-
18
order OE factor. Total and Social Competencies were predicted by Intellect and Sensitivity,
after controlling for the other OE facets. Furthermore, School and Extracurricular Competencies
were predicted by Intellect, after controlling for the other OE facets. Taken together, these
findings suggest that Intellect might be both a protective factor against behavioral problems as
well as a positive predictor of behavioral competencies in childhood.
Somewhat surprisingly, Sensitivity was a positive predictor of Social Competencies and
unrelated to behavioral problems. This was unexpected given that Neuroticism was the best
predictor of Sensitivity in this sample and Neuroticism is the most psychopathology-laden trait
of the big five (e.g., De Pauw & Mervielde, 2010). It may be that heightened sensitivity to social
cues leads to increased social competence, which can protect from behavioral problems (Burt,
Obradovic, Long, & Masten, 2008). This hypothesis is further strengthened by the fact that
Sensitivity was also positively predicted by Agreeableness (see Table 3)—a personality trait that
plays a major role in social competency issues (Shiner, 2000). It may also be that heightened
sensitivity only begins to lead to more negative outcomes such as overreaction to interpersonal
slights in middle childhood. As peers become more important during adolescence (Steinberg et
al., 2006), such overreaction may then result in depression and anxiety for some children.
Longitudinal research is needed to investigate this hypothesis. To sum up, our more nuanced
understanding of OE at the facet-level provided a much more detailed understanding of the
relationship between OE and childhood behaviors.
4.4 Limitations and Future Directions
Some limitations of our study deserve note. First, a general limitation of factor analysis is that
the factors extracted depend on the content of the initial pool of items. A strength of this study
was the inclusion of relevant content from both personality and temperament measures in
examining the OE construct, which has not yet been examined in previous studies. It is still
possible, however, that the items included here do not fully measure the construct. For example,
two other facets of OE in the adult NEO-PI-R (Costa & McCrae, 1992) are Actions and Values.
The Actions facet describes an individual's openness to new activities, such as visiting new
19
places or eating new food, whereas the Values facet describes an individual's openness to re-
examining traditional values with regard to society, religion, or politics. Most measures of child
individual differences, including the measures we used here, do not measure this type of OE
content. It remains to be seen in future studies whether such content is appropriate and
applicable to children and, if so, whether it can be measured reliably at these ages and
incorporated with the structure presented here. Second, the present study is limited by its cross-
sectional design, which can answer questions only about concurrent relationships between OE
and personality and behavior. Longitudinal research is needed to investigate when the
hierarchical structure of child OE established here emerges in development and whether the
middle-childhood OE construct predicts the same important life outcomes over time that it
predicts concurrently.
Finally, such work will greatly benefit from the use of other informants and other methods to
test the robustness of the structural model of child OE that we present here. In Mervielde and De
Fruyt’s (2000) study of the structure of 9-to-12-year-old children’s peer nominations, the OE
factor was less differentiated and emerged only combined with Conscientiousness, suggesting
that the use of younger informants may yield different answers to this research question. A study
using the TMCQ as a self-report questionnaire (Simonds & Rothbart, 2004), did find that
Fantasy/Openness and Perceptual Sensitivity were both reliably measured in 7-to-10-year-old
children with Cronbach’s αs of .78 and .80, respectively. Future research is needed utilizing
informants with varied roles (e.g., peer, teacher, self, unacquainted observer) to better
triangulate on the conceptualization of OE in childhood.
One of the most pressing questions remaining is when OE first emerges in development. The
present study marks an important step toward that end in establishing that child OE is at least
partly defined by the temperament construct Perceptual Sensitivity, which can be reliably
measured as early as infancy (Gartstein & Rothbart, 2003). In addition, the three-factor structure
demonstrated here is reliably measured by middle childhood, dispelling previous suggestions
that OE does not emerge until adolescence. Future research should investigate whether
Perceptual Sensitivity that emerges in infancy eventually develops into the OE construct that we
20
identified in middle childhood or OE as measured in adults. Furthermore, Perceptual Sensitivity
has been shown to have different primary loadings at different ages: on Extraversion/Surgency
in infancy (Gartstein & Rothbart, 2003), on Effortful Control in pre-school (Rothbart et al.,
2001), and on Affiliativeness in adolescence (Ellis & Rothbart, 2001). Therefore, examining
how OE and Perceptual Sensitivity relate to other traits at younger ages might be useful to fully
understand the longitudinal nature of OE.
Another interesting suggestion raised by the present results is that a facet-level approach to
childhood OE may help establish OE’s clinical relevance. A recent meta-analysis with adult
samples found OE to be largely unrelated to such psychological disorders as anxiety,
depression, and substance use disorders (Kotov, Gamez, Schmidt, & Watson, 2010). As
previously noted, however, other studies that examined facet-level associations between OE and
different psychological disorders found that facets of OE are predictive of psychological
disorders in divergent directions (e.g., Carrillo, et al., 2001; DeYoung et al., 2008; Wolfenstein
& Trull, 1997). These findings in adults, in combination with the present investigation, suggest
the need for facet-level investigations to fully elucidate links between OE and psychopathology.
Finally, it would also be interesting to conduct cluster analyses of children’s OE facet scores to
see if distinct clusters of “OE signatures” predict distinct developmental outcomes. Personality
signatures are consistent patterns of within-individual variability across situations in the
expressions of personality (Mischel, 2004) and are exhibited by both adults (Fournier,
Moskowitz, & Zuroff, 2007) and children (Shoda, Mischel, & Wright, 1994). Preliminary data
suggests that children exhibit the most within-individual variance across situations in their
expression of OE compared to the other FFM personality traits (Gao, Jawa, Fournier, & Tackett,
2011), suggesting that analyzing OE signatures might be especially relevant. In particular, OE
signatures might have great utility in furthering our understanding of the relationship between
the Imagination facet and internalizing problems. Previous research has shown that sad mood
induction via mental imagery of a sad memory is more effective in imaginative children than it
is in unimaginative children (Stegge, Meerum Terwogt, & Koops, 1995), suggesting that the
tendency toward a vivid imagination has direct effects on sad mood in sad situations. Future
21
research should examine whether repeated exposure to sad situations might increase the risk for
internalizing problems only for the group of children who express a lot of imagination during
sad situations. Furthermore, future research should also explore additional relationships between
clusters of OE signatures and developmental outcomes such as psychopathology as such
knowledge is needed to fully understand the predictive utility of OE.
4.5 Conclusion
Drawing from both personality and temperament taxonomies, the present study attempted to
establish initial construct validity for OE in middle childhood. Specifically, both temperament
and personality content emerged as important to define the broader OE domain in childhood,
suggesting that research with younger populations may be limited by examining constructs
using only one measure or model. The three-factor model of child OE that emerged
demonstrated facet-level associations between OE and behavioral problems and competencies
that previous studies—relying on domain-level measures of child OE—may have masked. The
results of the present study suggest that future research on child traits would be improved by (1)
measuring child traits with both personality and temperament measures and (2) examining not
only domain-level but also facet-level associations between child traits and important life
outcomes. These findings provide clear support for the validity and reliability of child OE,
hopefully reconciling lingering questions about whether the fifth factor of personality has
emerged as a robust and meaningful trait domain by this early age.
22
References
Abe, J. A. A., (2005). The predictive validity of the Five-Factor Model of personality with
preschool age children: A nine year follow-up study. Journal of Research in Personality,
39, 423–442.
Achenbach, T. M. (1978). The child behavior profile: I Boys Aged 6 – 11. Journal of
Consulting and Clinical Psychology, 46, 478–488.
Achenbach, T. M., & Rescorla, L. A. (2001). Manual for ASEBA School-Age Forms &
Profiles. Burlington, VT: University of Vermont, Research Center for Children,
Youth, & Families.
Bongers, I. L., Koot, H. M., van der Ende, J., & Verhulst, F. C. (2003). The normative
development of child and adolescent problem behavior. Journal of Abnormal
Psychology, 112, 179–192.
Burt, K. B., Obradović, J., Long J. D., & Masten, A. S. (2008). The Interplay of Social
Competence and Psychopathology Over 20 Years: Testing Transactional and Cascade
Models. Child Development, 79, 359–374.
Carrillo, J. M, Rojo, N., Sanchez-Bernardos, M. L., & Avia, M. D. (2001). Openness to
Experience and depression. European Journal of Psychological Assessment, 17, 130–
136.
Cronbach, L. J. & Meehl, P. E. (1955). Construct validity in psychological tests. Psychological
Bulletin, 52, 281–302.
Costa, P. T. & McCrae, R. R. (1992). NEO personality Inventory professional manual. Odessa,
FL: Psychological Assessment Resources.
Costa, P. T. & McCrae, R. R. (1995). Domains and facets: Hierarchical personality assessment
using the Revised NEO Personality Inventory. Journal of Personality Assessment, 64,
21–50.
De Pauw, S. S. W., & Mervielde, I. (2010). Temperament, personality and developmental
psychopathology: A review based on the conceptual dimensions underlying childhood
traits. Child Psychiatry Human Development, 41, 313–329.
De Pauw, S. S. W., Mervielde, I., & Van Leeuwen K. G. (2009). How are traits related to
problem behaviour in preschoolers? Similarities and contrasts between temperament and
personality. Journal of Abnormal Child Psychology, 37, 309–325.
DeYoung, C. G., Peterson, J. B., Séguin, J. R., & Tremblay, R. E. (2008). Externalizing
behavior and the higher-order factors of the big five. Journal of Abnormal Psychology,
117, 947–953.
23
DeYoung, C. G., Peterson, J. B., & Higgins, D. M. (2002). Higher-order factors of the Big Five predict conformity: Are there neuroses of health? Personality and Individual Differences, 33, 533–552.
Digman, J. M. (1989). Five robust trait dimensions: Development, stability, and utility. Journal
of Personality, 57, 195–214.
Ehrler, D. J., Evans, J. G., & McGhee, R. L. (1999). Extending Big-Five theory into childhood:
A preliminary investigations into the relationship between Big-Five personality traits and
behavior problems in children. Psychology in the School, 36, 451–472.
Ellis, L. K., & Rothbart, M. K. (2001). Revision of the Early Adolescent Temperament
Questionnaire. Poster presented at the 2001 Biennial Meeting of the Society for
Research in Child Development, Minneapolis, Minnesota.
Fournier, M. A., Moskowitz, D. S., & Zuroff, D. C. (2008). Integrating dispositions, signatures,
and the interpersonal domain. Journal of Personality and Social Psychology, 94, 531-
545.
Gao, W., Jawa, N. A., Fournier, M. A., & Tackett, J. L. (June, 2011). Situation-based
contingencies in personality expression in middle childhood. Poster presented at the 72nd
Annual Convention of the Canadian Psychological Association, Toronto, ON.
Gartstein, M. A., & Rothbart, M. K. (2003). Studying infant temperament via the Revised Infant
Behaviour Questionnaire. Infant Behavior & Development, 26, 64–86.
Gjerde, P. F., & Cardilla, K. (2009). Developmental Implications of Openness to Experience in
Preschool Children: Gender Differences in Young Adulthood. Developmental
Psychology, 45, 1455–1464.
Goldberg, L. R. (2001). Analyses of Digman’s child-personality data: Derivation of Big-Five
factor scores from each of six samples. Journal of Personality, 69, 709–743.
Graziano, W. G., & Ward, D. (1992). Probing the Big Five in adolescence: Personality and
adjustment during a developmental transition. Journal of Personality, 60, 425–439.
Halverson, C. F., Havill, V. L., Deal, J., Baker, S. R., Victor, J. B., Pavlopoulos, V., … Wen, L.
(2003). Personality structure as derived from parent ratings of free descriptors of
children: The Inventory of Children's Individual Differences. Journal of Personality, 71,
995–1026.
John, O. P., Caspi, A., Robins, R. W., Moffitt, T. E., Stouthamer-Loeber, M. (1994). The “Little
Five”: Exploring the nomological network of the Five-Factor Model of personality in
adolescent boys. Child Development, 65, 160–178.
Kotov, R., Gamez, W., Schmidt, F., & Watson, D. (2010). Linking “big” personality traits to
anxiety, depressive, and substance use disorders: A meta-analysis. Psychological
Bulletin, 136, 768–821.
24
Lamb, M. E., Chuang, S. S., Wessels, H., Broberg, A. G., & Hwang, C. P. (2002). Emergence
and construct validation of the big five factors in early childhood: A longitudinal
analysis of their ontogeny in Sweden. Child Development, 73, 1517–1524.
Markon, K. E., Kruefer, R. F., & Watson, D. (2005). Delineating the structure of normal and
abnormal personality: An integrative hierarchical approach. Journal of Personality and
Social Psychology, 88, 139–157.
Masten, A. S., Curtis, W. J. (2000). Integrating competence and psychopathology: Pathways
toward a comprehensive science of adaptation in development. Development and
Psychopathology, 12, 529–550.
McCrae, R. R. (1994). Openness to Experience – Expanding the boundaries of Factor-V.
European Journal of Personality, 8, 251–272.
Mervielde, I., Buyst, V., & De Fruyt, F. (1995). The validity of the Big-Five as a model for
teachers’ ratings of individual differences among children aged 4-12 years. Personality
and Individual Differences, 18, 525–534.
Mervielde, I., & De Fruyt, F. (1999). Construction of the Hierarchical Personality Inventory for
Children. In I. Mervielde, I. J. Deary, F. De Fruyt, & F. Ostendorf (Eds.), Personality
psychology in Europe (Vol 7, pp. 107–127). Tilburg: Tilburg University Press.
Mervielde, I. & De Fruyt, F. (2000). The Big Five personality factors as a model for the
structure of children's peer nominations. European Journal of Personality, 14, 91–106.
Mervielde, I., De Fruyt, F., Jarmuz, S. (1998). Linking openness and intellect in childhood and
adulthood. In G. A. Kohnstamm, C. F. Halverson, I. Mervielde, & V. Havill (Eds.)
Parental descriptions of child personality: Developmental antecedents of the Big Five?
(pp. 105–126). Mahwah, NJ: Erlbaum.
Mischel, W. (2004). Toward an integrative science of the person. Annual Review of Psychology,
55, 1-22.
Muris, P., Meesters, C., & Diederen, R. (2005). Psychometric properties of the Big Five
Questionnaire for Children (BFQ-C) in a Dutch sample of young adolescents.
Personality and Individual Differences, 38, 1757–1769.
Muthén, L. K., & Muthén, B. O. (1998 –2006). Mplus user’s guide (4th ed.). Los Angeles, CA:
Muthén & Muthén.
Nettle, D. (2006). Schizotypy and mental health amongst poets, visual artists, and
mathematicians. Journal of Research in Personality, 40, 876–890.
Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. (2002). To parcel or not to
parcel: Exploring the questions, weighing the merits. Structural Equation Modeling, 9,
151–173.
Post, F. (1994). Creativity and psychopathology: A study of 291 world-famous men. British
Journal of Psychiatry, 165, 22–34.
25
Prinzie, P., Onghena, P., Hellinckx, W., Grietens, H., Ghesquiere, P., & Colpin, H. (2004).
Parent and child personality characteristics as predictors of negative discipline and
externalizing problem behavior in children. European Journal of Personality, 18, 73–
102.
Ring, A. J., Ryder, A. G., Lavigne, K., Bagby, R. M. (2008, July). Schizotypal personality and
the five-factor model: Two aspects of openness make opposite predictions. Poster
presented at the 29th
meeting of the International Congress of Psychology, Berlin,
Germany.
Rothbart, M. K., Ahadi, S. A., Hershey, K. L., & Fisher, P. (2001). Investigations of
temperament at three to seven years: The Children’s Behavior Questionnaire. Child
Development, 72, 1394–1408.
Shiner, R. (2000). Linking childhood personality with adaptation: Evidence for continuity and
change across time into late adolescence. Journal of Personality and Social Psychology,
78, 310–325.
Shiner, R. & Caspi, A. (2003). Personality differences in childhood and adolescence:
Measurement, development, and consequences. Journal of Child Psychology and
Psychiatry, 44, 2–32.
Shoda, Y., Mischel, W., & Wright, J. C. (1994). Intraindividual stability in the organization and
patterning of behavior: Incorporating psychological situations into the idiographic
analysis of personality. Journal of Personality and Social Psychology, 67, 674-687.
Simonds, J. & Rothbart, M. K. (2004, October). The Temperament in Middle Childhood
Questionnaire (TMCQ): A computerized self-report measure of temperament for ages 7–
10. Poster session presented at the Occasional Temperament Conference, Athens, GA.
Stegge, H., Meerum Terwogt, M., & Koops, W. (1995). Mood congruity in children: Effects of
age, imagery capability, and demand characteristics. International Journal of Behavioral
Development, 18, 177–191.
Steinberg, L., Dahl, R., Keating, D., Kupfer, D., Masten, A., & Pine, D. (2006).
Psychopathology in adolescence: Integrating affective neuroscience with the study of
context. In D. Cicchetti & D. Cohen (Eds.), Developmental psychopathology, Vol. 2:
Developmental neuroscience (2nd
ed., pp. 710–741). New York: Wiley.
Tackett, J. L. (2006). Evaluating models of the personality-psychopathology relationship in
children and adolescents. Clinical Psychology Review, 26, 584–599.
Tackett, J. L., Krueger, B. F., Iacono, W. G., & McGue, M. (2008). Personality in middle
childhood: A hierarchical structure and longitudinal connections with personality in late
adolescence. Journal of Research in Personality, 42, 1456–1462.
Tackett, J. L., Slobodoskaya, H. R., Mar, R. A., Deal, J., Halverson, C. F., Baker, S. R., …
Besevegis, E. (2011). The Hierarchical Structure of Childhood Personality in Five
26
Countries: Continuity from Early Childhood to Early Adolescence. Manuscript
submitted for publication.
Tellegen, A. (1978/1982). Brief manual for the Multidimensional Personality Questionnaire.
Unpublished manuscript, University of Minnesota.
Victor, J. B. (1994). The five-factor model applied to individual differences in school behavior.
In C. F. Halverson, G. A. Kohnstamm, & R. P. Martin (Eds.) The developing structure of
temperament and personality from infancy to adulthood. (pp. 355–366). Hillsdale, NJ:
Lawrence Erlbaum Associates.
Wolfenstein, M., & Trull, T. J. (1997). Depression and Openness to Experience. Journal of
Personality Assessment, 69, 614–632.
27
Tables Table 1
Item Loadings > .40 for Exploratory Factor Analysis of Mother-Reported Openness to Experience Items
Item Stem: My child… Intellect Imagination Sensitivity
… is quick to learn .88
… has good thinking abilities .85
… is intelligent .84
… is quick to understand what is said or going on .76
… has a large vocabulary .74
… is good at problem solving .73
… is eager to learn .71
… is slow to learn -.70
… has difficulty solving problems -.66
… has a good memory .62
… speaks well .58
… likes to ask questions .53
… shows interest in everything .50
… is curious .49
… has a sense of humor .42
… has a lot of imagination .72
… has a big imagination .68
… is creative .63
… likes to pretend .61
… enjoys drawing pictures .57
… likes to make up stories .53
… likes the sound of poems .51
… likes to think of new ideas .48
… is unimaginative -.48
… likes poems .46
… likes to make things .46
… likes reading or listening to make believe stories .43
… notices small changes in the environment, like lights .73
… notices things others don’t notice .67
… notices the color of people’s eyes .58
… notices the sound of birds .56
… notices when parents are wearing new clothing .56
… notices odors like perfume, smoke, and cooking smells .49
… notices even little specks of dirt on objects .46
… likes to run hand over things to see if they are smooth … .44
Note. Items in bold are from the Inventory of Child Individual Differences (Halverson et al., 2003); items in italics
are from the Temperament in Middle Childhood Questionnaire (Simonds & Rothbart, 2004).
All factor loadings significant at p < .01.
28
Table 2
Summary of Correlations between the Openness to Experience, ICID, and CBCL scales Based on Averaged Mother
and Father Report.
Openness Intellect Imagination Sensitivity
Agreeableness .26** .25** .16** .16**
Conscientiousness .57** .64** .24** .27**
Extraversion .60** .54** .46** .32**
Neuroticism -.35** -.41** -.15** -.11
Total Problems -.10 -.19** .05 .01
Externalizing Behaviors -.09 -.14** .03 -.03
Internalizing Behaviors .01 -.06 .08 .07
Total Competencies .38** .39** .20** .22**
School Competencies .35** .44** .12 .11
Social Competencies .29** .29** .16** .19**
Extracurricular Competencies .22** .21** .14 .14**
Note. ICID = Inventory of Child Individual Differences (Halverson et al., 2003); CBCL = Child Behavior Checklist
(Achenbach & Rescorla, 2001)
** p < .01.
29
Table 3
Hierarchical Multiple Regression Analyses Predicting the Lower-Order Openness to Experience Facets from Mother- and Father-Reported Personality Traits.
Lower-order Openness facets
Intellect Imagination Sensitivity
Variable ∆R2
B [95% CI] β ∆R2 B [95% CI] β ∆R
2 B [95% CI] β
Step 1 .21*** .27*** .16***
Controla
Step 2 .36*** .06*** .07***
A -0.13 [-0.20, -0.06] -.28*** 0.06 [0.00, 0.11] .19 0.07 [0.03, 0.10] .37***
C 0.36 [0.29, 0.43] .49*** -0.02 [-0.08, 0.05] -.04 0.08 [0.04, 0.12] .28***
E 0.12 [0.08, 0.15] .29*** 0.07 [0.04, 0.10] .28*** 0.03 [0.01, 0.05] .19**
N -0.08 [-0.16, 0.00] -.18* 0.07 [0.01, 0.13] .27* 0.09 [0.05, 0.13] .52***
Total R2 .57 .33 .24
Note. A = Agreeableness; C = Conscientiousness; E = Extraversion; N = Neuroticism; CI = Confidence Interval.
a Control variables in step 1 included the other two lower-order Openness to Experience facets.
* p < .05. ** p < .01. *** p < .001.
30
Table 4
Multiple Regression Analyses Predicting CBCL Scales from Mother- and Father-Reported Openness to Experience.
CBCL Problem Scales
Total Problems Externalizing Behaviors Internalizing Behaviors
Predictor B [95% CI] β B [95% CI] β B [95% CI] β
Openness -0.08 [-0.17, 0.01] -.10 -0.02 [-0.05, 0.01] -.09 0.00 [-0.02, 0.03] .01
Total R2 .01 .01 .00
Intellect -0.35 [-0.50, -0.20] -.27*** -0.08 [-0.13, -0.03] -.19*** -0.05 [-0.09, 0.00] -.12*
Imagination 0.34 [0.08, 0.60] .16* 0.08 [0.00, 0.17] .12 0.06 [-0.01, 0.14] .11
Sensitivity 0.07 [-0.32, 0.45] .02 -0.03 [-0.16, 0.10] -.03 0.06 [-0.05, 0.18] .07
Total R2 .06*** .03* .02
CBCL Competency Scales
Total Competencies School Competencies Social Competencies Extracurricular Comp.
Predictor B [95% CI] β B [95% CI] β B [95% CI] β B [95% CI] β
Openness 0.08 [0.06, 0.10] .38*** 0.02 [0.01, 0.02] .35*** 0.04 [0.02, 0.05] .29*** 0.02 [0.01, 0.03] .22***
Total R2 .15*** .13*** .09*** .05***
Intellect 0.12 [0.08, 0.15] .36*** 0.04 [0.03, 0.05] .47*** 0.05 [0.03, 0.07] .25*** 0.03 [0.01, 0.05] .17**
Imagination 0.00 [-0.06, 0.06] .00 -0.01 [-0.03, 0.00] -.09 0.00 [-0.04, 0.04] .01 0.01 [-0.03, 0.04] .03
Sensitivity 0.09 [0.01, 0.19] .11* 0.00 [-0.02, 0.02] .01 0.06 [0.00, 0.11] .11* 0.04 [-0.01, 0.08] .08
Total R2 .17*** .20*** .10*** .05*
Note. CBCL = Child Behavior Checklist (Achenbach & Rescorla, 2001).
* p < .05. ** p < .01. *** p < .001
31
Figures
Figure 1. Structure of Openness to Experience in middle childhood based on confirmatory factor
analysis of father report. Values in brackets indicate 95% confidence intervals. All loadings and
correlations are significant at p < .01. RMSEA = .06; CFI = .99.