ArticlePDF Available

Web-Based Assessment of Children's Social-Emotional Comprehension

Authors:

Abstract and Figures

Social-emotional comprehension includes the ability to encode, interpret, and reason about social-emotional information. The better developed children's social-emotional comprehension, the more positive their social interactions and the better their peer relationships. Many clinical tools exist to assess children's social behavior. In contrast, fewer clinically interpretable tools are available to assess children's social-emotional comprehension. This study evaluated the psychometric properties of a group of direct assessments of social-emotional comprehension. Scores on these assessments reflected children's performance on challenging tasks that required them to demonstrate their social-emotional comprehension. In 2 independent samples, including a general education school sample (n = 174) and a clinic sample (n = 119), this study provided evidence that (a) individual assessments yield variably reliable scores, (b) composite scores are highly reliable, (c) direct assessments demonstrate a theoretically coherent factor structure and convergent and discriminant validity, and (d) composite scores yield expected age- and diagnostic-group differences. Implications for clinical practice, theory, and assessment development are discussed. (PsycINFO Database Record (c) 2013 APA, all rights reserved).
Content may be subject to copyright.
Direct Assessment of Children’s Social-Emotional Comprehension
Clark McKown, Adelaide M. Allen, Nicole M. Russo-Ponsaran, and Jason K. Johnson
Rush University Medical Center
Social-emotional comprehension includes the ability to encode, interpret, and reason about social-
emotional information. The better developed children’s social-emotional comprehension, the more
positive their social interactions and the better their peer relationships. Many clinical tools exist to assess
children’s social behavior. In contrast, fewer clinically interpretable tools are available to assess
children’s social-emotional comprehension. This study evaluated the psychometric properties of a group
of direct assessments of social-emotional comprehension. Scores on these assessments reflected chil-
dren’s performance on challenging tasks that required them to demonstrate their social-emotional
comprehension. In 2 independent samples, including a general education school sample (n 174) and
a clinic sample (n 119), this study provided evidence that (a) individual assessments yield variably
reliable scores, (b) composite scores are highly reliable, (c) direct assessments demonstrate a theoretically
coherent factor structure and convergent and discriminant validity, and (d) composite scores yield
expected age- and diagnostic-group differences. Implications for clinical practice, theory, and assessment
development are discussed.
Keywords: social-emotional learning, child assessment, emotion recognition, theory of mind, social
problem solving
Social-emotional comprehension includes the ability to encode,
interpret, and reason about social-emotional information (Lipton &
Nowicki, 2009). Prior research has demonstrated that the better
developed children’s social-emotional comprehension, the more
positive their social interactions and the better their peer relation-
ships (Baron-Cohen, 1989; Bauminger, Edelsztein, & Morash,
2005; Collins & Nowicki, 2001; Crick & Dodge, 1994; Denham,
2006; Duke, Nowicki, & Walker, 1996; McKown, 2007; McK-
own, Gumbiner, Russo, & Lipton, 2009; Nowicki, Duke, & van
Buren, 2008). Better peer relationships are in turn associated with
better mental health and other important life outcomes (Parker &
Asher, 1987).
Practitioners working with socially marginalized children
would thus be well-served to include assessments of children’s
social-emotional comprehension as part of a social assessment
battery and use assessment findings to guide treatment plans.
Yet few clinical tools are suited to assessing social-emotional
comprehension. With noteworthy exceptions (e.g., the Meyer-
Salovey-Caruso Emotional Intelligence Test, Youth Version
[MSCEIT-YV]; Mayer, Caruso, & Salovey, 2005), existing
assessment strategies are better suited to assessing social be-
havior than social-emotional comprehension. Examples include
behavior rating scales (e.g., Social Skills Intervention System;
Gresham & Elliott, 2008), behavioral observations (e.g., Peer
Social Behavior Code; Walker & Severson, 1992), peer nomi-
nations (Chan & Mpofu, 2001; Moreno, 1933), and self-report
(e.g., Trait Emotional Intelligence Questionnaire-Child Form;
Mavroveli, Petrides, Shove, & Whitehead, 2008). In contrast,
social-emotional comprehension involves the activation and
deployment of internal mental processes. Accordingly, it would
be better measured with direct assessments requiring children to
solve social– emotional tasks.
Most of the few existing direct assessments narrowly assess a
specific dimension of social-emotional comprehension, which lim-
its their clinical utility. For example, the Diagnostic Analysis of
Nonverbal Accuracy (DANVA; Nowicki & Duke, 1994) measures
emotion recognition from faces, posture, and affective prosody.
Two of the DANVA’s strengths are that it includes child faces for
the facial emotion recognition subtest and that it can be used from
preschool through adulthood. The Mind in the Eyes Test Child
Version (Baron-Cohen, Wheelwright, Scahill, Lawson, & Spong,
2001) is a mental-state inference task in which children view
photos of the eye region of faces and indicate what the person is
thinking or feeling. It is currently used for research purposes only,
and there is no age range provided. The NEPSY-II (Korkman,
Kirk, & Kemp, 2007) includes Affect Recognition and Theory of
Mind (ToM) subtests for children ages 5 to 16 years old.
Two assessments measure social-emotional comprehension
more broadly, but focus on targeted populations. The Social In-
formation Processing Application (SIP-AP; Kupersmidt, Stelter, &
Dodge, 2011) measures several dimensions of children’s ability to
solve hypothetical social problems and was designed to assess
This article was published Online First July 1, 2013.
Clark McKown, Adelaide M. Allen, Nicole M. Russo-Ponsaran, and
Jason K. Johnson, Department of Behavioral Sciences, Rush University
Medical Center.
This research reported here was supported by grants from the Dean and
Rosemarie Buntrock Family Foundation and the Institute of Education
Sciences through Grant R305A110143 to Rush University Medical Center.
The opinions expressed are those of the authors and do not represent views
of the Institute or the U.S. Department of Education.
Correspondence concerning this article should be addressed to Clark
McKown, Department of Behavioral Sciences, Rush University Medical
Center, RNBC 4711 Golf Road, Suite 1100, Skokie, IL 60076. E-mail:
Clark_A_McKown@rush.edu
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Psychological Assessment © 2013 American Psychological Association
2013, Vol. 25, No. 4, 1154–1166 1040-3590/13/$12.00 DOI: 10.1037/a0033435
1154
social-cognitive correlates of aggression in boys ages 8 to 12 years
old (Kupersmidt et al., 2011). The MSCEIT-YV (Mayer et al.,
2005) assesses the ability to perceive emotion from faces, match
emotions to physical sensations, identify the cause of emotions,
and identify effective ways to manage emotions and was designed
for youth ages 10 to 17 years old (Rivers et al., 2012). Both of
these assessments sample social-emotional comprehension
broadly. Each is appropriate for a specific sex, age range, or
clinical population.
Thus, existing direct assessments vary in the domain coverage,
age range, and population for which they are appropriate. As a
result, depending on a child’s age, sex, and presenting issue,
appropriate social-emotional comprehension assessments may not
be available. With a limited range of tools suitable for the direct
clinical assessment of social-emotional comprehension, clinicians
may not be able to come to a fully informed understanding of
contributors to social impairment in many clinical populations.
Building on prior work, this study evaluated the psychometric
properties of a group of social-emotional comprehension assess-
ments in general education students and children with neurobe-
havioral disorders, including attention-deficit/hyperactivity disor-
der (ADHD), autism spectrum disorder (ASD), and reading
disorder (RD), across a wide age range (5 to 14 years). The
assessments are designed to measure children’s ability to encode,
interpret, and reason about social-emotional information. Evaluat-
ing the psychometric properties of assessments in two distinct
samples with a broad age range permits us to test the potential
clinical usefulness of the assessments across diverse groups.
A Model of Social-Emotional Comprehension
Lipton and Nowicki (2009) proposed the Social Emotional
Learning Framework (SELF) to conceptualize social-emotional
comprehension. In this context “social-emotional” refers to a mix-
ture of social and emotional phenomena. The SELF incorporates
features of prominent theories of social-emotional comprehension,
including social neuroscience (Adolphs, 2003), social information
processing (Crick & Dodge, 1994), affective social competence
(ASC; Halberstadt, Denham, & Dunsmore, 2001), and emotional
intelligence (EI; Salovey & Mayer, 1989 –1990). The SELF em-
phasizes three broad constructs that each resemble features of
other models.
First, encoding of social-emotional information, or Social
Awareness, is defined as the ability to label others’ emotions from
nonverbal cues. Social Awareness draws upon research on non-
verbal communication (Nowicki & Duke, 1994) and is similar to
“emotion perception” from EI (Salovey & Mayer, 1989 –1990) and
“receiving affective messages” from ASC (Halberstadt et al.,
2001). Second, interpreting others’ perspectives from their behav-
ior and language, or Social Meaning, draws from research that
finds that children’s theory of mind and pragmatic judgment both
involve interpreting intentions underlying others’ words and ac-
tions (Capps, Kehres, & Sigman, 1998; Landa, 2000; McKown,
2007; Peterson, Wellman, & Liu, 2005; Wellman & Liu, 2004).
Third, Social Reasoning is defined as the ability to reason about
social problems and draws from research on social information-
processing (Bauminger et al., 2005; Crick & Dodge, 1994; Den-
ham, 2006; Dirks, Treat, & Weersing, 2007).
Empirical evidence supports the conclusion that these three
dimensions of social-emotional comprehension are important.
When examined separately, Social Awareness, Social Meaning,
and Social Reasoning are each associated with social outcomes
including peer status, interpersonal negotiating skill, aggression,
and social withdrawal (Banerjee & Watling, 2005; Chung &
Asher, 1996; Dodge & Price, 1994; Erdley & Asher, 1999; Hughes
& Ensor, 2007; Nowicki & Duke, 1994; Slaughter, Dennis, &
Pritchard, 2002; Yeates, Schultz, & Selman, 1991; Yeates,
Schultz, & Selman, 1990). Work examining Social Awareness,
Social Meaning, and Social Reasoning together suggests that the
better children perform on assessments of several dimensions of
social-emotional comprehension, the more others report they en-
gage in socially competent behavior (McKown et al., 2009).
Hypotheses
A main goal of this study was to demonstrate the psychometric
properties of a group of direct assessments of social-emotional
comprehension. To this end, we tested several hypotheses. First,
we hypothesized that scores on a group of social-emotional com-
prehension assessments would exhibit high internal consistency,
test–retest, and interrater reliabilities. Second, we hypothesized
that assessments selected to reflect Social Awareness, Social
Meaning, and Social Reasoning would reflect three separate but
correlated latent variables. We tested this hypothesis using confir-
matory factor analysis (CFA). Finally, we hypothesized that Social
Awareness, Social Meaning, and Social Reasoning latent variables
would each be more strongly associated with corresponding latent
variables created using alternate measures (convergent validity)
than with latent variables reflecting different dimensions of social-
emotional comprehension (divergent validity). We tested this hy-
pothesis using structural equation modeling (SEM; Campbell &
Fiske, 1959; Messick, 1995).
Another goal was to determine whether the assessments yielded
expected age and diagnostic differences, which would support
their validity and clinical relevance. We hypothesized that (a)
greater age would be associated with better performance on as-
sessments of Social Awareness, Social Meaning, and Social Rea-
soning skills (Izard & Harris, 1995; Nowicki & Duke, 1994;
Pillow, 2011; Wellman, Cross, & Watson, 2001; Zuckerman,
Blanck, DePaulo, & Rosenthal, 1980); (b) compared to typical
peers, children with ASD, ADHD, and RD would demonstrate
significantly poorer Social Awareness, Social Meaning, and Social
Reasoning skills; and (c) children with ASD would demonstrate
greater social-emotional comprehension deficits than children with
ADHD or RD (Channon, Charman, Heap, Crawford, & Rios,
2001; Embregts & van Nieuwenhuijzen, 2009; Hall & Richmond,
1985; Mikami, Lee, Hinshaw, & Mullin, 2008; H. L. Swanson &
Malone, 1992; Tur-Kaspa & Bryan, 1994).
Method
Recruitment and Sample
The Rush University Medical Center Institutional Review Board
(IRB) approved all study procedures and consent processes. All
but two of the assessments were administered by research staff
with bachelor’s degrees or higher who were experienced in work-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
1155
ASSESSING SOCIAL-EMOTIONAL COMPREHENSION
ing with children and who were supervised by senior members of
the research team. The measures used for quantifying autism
characteristics were administered by either a clinical psychologist
or research psychologist who was reliable in the administration
and scoring of those measures.
Eligibility. All general education students in kindergarten
through eighth grade at an urban parochial school and a suburban
public school were invited to participate.
In addition, children with a diagnosis of ASD, ADHD, or RD
who had average or above average cognitive abilities (full scale
IQ 85) as measured by the Wechsler Abbreviated Scale of
Intelligence (WASI; Wechsler, 1999) were eligible to participate.
When no prior diagnosis had been documented, ASD was opera-
tionalized as a score at or above 12 on the Social Communication
Questionnaire (SCQ; Rutter, Bailey, Lord, & Berument, 2003) and
scores above the diagnostic cutoff for autism spectrum disorder on
the Autism Diagnostic Interview-Revised (ADI-R; Lord, Rutter, &
Le Couteur, 1994) and the Autism Diagnostic Observation Sched-
ule (ADOS; Lord, Rutter, DiLavore, & Risi, 1999). Children who
had a prior diagnosis of ASD were required to score above the
diagnostic cutoff for ASD on either the ADI-R or ADOS.
ADHD was operationalized as scoring in the clinical range on
one teacher and one parent report of inattention or hyperactivity on
either the Behavior Assessment System for Children (2nd ed.;
BASC-2; Reynolds & Kamphaus, 2004) or the Swanson, Nolan and
Pelham Scale (4th ed.; SNAP-IV; J. M. Swanson, 1995) and a history
of difficulties before age 7 reported by a parent on the Kiddie-
Schedule for Affective Disorders and Schizophrenia (K-SADS; Kauf-
man et al., 1997).
RD was operationalized as at least one core reading skill score
on the Wechsler Individual Achievement Test (2nd ed.; WIAT-II;
Wechsler, 2005) at least one standard deviation below full scale IQ
on the WASI and a score at least one standard deviation below
average on the phonological awareness subtest from the Compre-
hensive Test of Phonological Processing (CTOPP; Wagner,
Torgesen, & Rashotte, 1999).
Recruitment and informed consent. For the school sample,
informational parent meetings were held as part of back-to-school
nights at the beginning of the school year. A consent form along
with a letter from the principal was mailed to all parents. The clinic
sample was recruited from a suburban outpatient mental health
center associated with the research study, and from schools in the
surrounding area and through therapists, physicians, and other
professionals in the area working with the study population. Flyers
were posted in the waiting rooms of the recruitment sites, e-mailed
to parents in a clinical research data base, e-mailed to professionals
in the community, and advertised in a newsletter. Parent informed
consent and child assent were obtained for all participants.
Sample characteristics. Parents of 186 general education stu-
dents ages four to 14 years old (M 8.8 years, SD 2.6 years)
consented to have their children participate. One child was under
5 years old (4.9 years). The school sample included 84 boys
(45.2%). Eighty children were White (43%), 53 were Black
(28.5%), 23 were Latino (12.4%), 12 were Asian (6.5%), and 18
identified as multiracial (9.7%). The 119 participants in the clinic
sample ranged from ages 5 to 14 (M 10.3 years old, SD 2.2
years). In total, 44 met criteria for ASD (36.7%), 46 met criteria
for ADHD (38.3%), and 24 met criteria for RD (24.2%). The clinic
sample included 96 boys (80.7%). Ninety children in this sample
were White (75.6%), seven were Black (5.9%), two were Latino
(1.7%), three were Asian (2.5%), and three were multiracial
(2.5%). Fourteen children identified their ethnicity as “other” or
did not indicate their ethnicity (11.8%).
The school sample included more girls than the clinic sample,
2
(1) 37.6, p .05, was more ethnically diverse,
2
(5) 65.7,
p .05, and was older, F(1, 291) 243, p .05. On the Social
Skills Rating System (SSRS; Gresham & Elliott, 1990), teachers
reported that children in the school sample displayed average
levels of socially skilled and problem behavior (Social Skills
Standard Score 100.0; Problem Behavior Standard Score
100.2). In contrast, teachers reported that children in the clinic
sample were below average in social skills and above average in
problem behaviors (Social Skills Standard Score 89.3; Problem
Behavior Standard Score 110.5). Sample differences in social
skills and problem behaviors were significant (F(1, 256) 28.7,
p .05, d 0.69 for Social Skills, and F(1, 256) 35.4, p .05,
d ⫽⫺0.76 for Problem Behaviors). In addition, on a three-item
questionnaire, teachers reported that children in the school sample
were significantly more well accepted by peers than children in the
clinic sample, F(1, 260) 17.3, p .05, d 0.53.
Procedures
To minimize fatigue, more and less demanding assessments
were alternated and were administered in the same order for all
participants. School sample participants completed social-
emotional comprehension assessments individually at their school.
Each child participated in an average of two and one half hours of
testing, broken into two or three sessions on different days. Breaks
were offered on an as-needed basis to prevent testing fatigue.
Participants from the clinic sample were tested individually at an
outpatient clinic. Testing was broken into two sessions. In the first
session, which lasted approximately 2 hr for children suspected of
ADHD or RD diagnoses and up to 3 hr for children suspected of
an ASD diagnosis, assessments were administered to confirm
diagnostic eligibility. In the second session, which lasted approx-
imately 2.5 hr, social-emotional comprehension assessments were
administered.
Core Social-Emotional Comprehension Measures
Core measures were selected for which prior research had
provided evidence of good reliability and validity.
Social Awareness. Four measures were used to assess So-
cial Awareness. First, for the 24-item child faces subtest of the
Diagnostic Analysis of Nonverbal Accuracy (DANVA; Now-
icki & Duke, 1994), children viewed photographs of child faces
and indicated whether each child was happy, sad, angry, or
scared. Second, children completed the 22-item Match Emo-
tional Prosody to Emotional Face (MEPEF) subtest of the
Comprehensive Affect Testing System (CATS; Weiner, Greg-
ory, Froming, Levy, & Ekman, 2006). Per CATS standard
administration procedures, for each item, children listened to an
audio recording of an adult making a statement. Children then
selected one of five faces that displayed the same emotion
conveyed by the speaker’s tone of voice. Third, children com-
pleted a 22-item posture recognition task (Heberlein, Gläescher,
& Adolphs, 2007). For this task, children viewed photographs
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
1156
MCKOWN, ALLEN, RUSSO-PONSARAN, AND JOHNSON
of adults with blurred faces and indicated whether the person
was happy, sad, angry, or scared. Finally, children completed a
20-item point-light display task (Heberlein, Adolphs, Tranel, &
Damasio, 2004). For this task, children viewed brief video clips
of abstracted human forms walking. For each clip, children
indicated whether the person was happy, sad, angry, or scared.
Children received one point for each correct response.
Social Meaning. One theory of mind and one pragmatic
judgment measure were used to assess Social Meaning. Theory
of mind was assessed in children in kindergarten through fifth
grade using 12 vignettes from Strange Stories (Happé, 1994;
White, Hill, Happé, & Frith, 2009). In each vignette, a character
states one thing but means something else. Respondents are
asked whether what the character said is true and why the
character said what he or she said. Children received 1 point for
correctly inferring the speaker’s intention or mental state. To
assess pragmatic judgment, the 60-item pragmatic judgment
subtest of the Comprehensive Assessment of Spoken Language
(CASL; Carrow-Woolfolk, 1999) was administered to all par-
ticipants. For this test, children were asked what they would do
or say in a range of situations with social language demands,
including greetings, requesting information, expressing sympa-
thy, joining a conversation, and polite interruption. Scoring on
the CASL followed standard procedures.
Social Reasoning. Five social problem-solving vignettes in-
volving peer entry, peer pressure, peer provocation, and differ-
ences of opinion were administered to assess Social Reasoning
(McKown et al., 2009). Vignettes were read aloud and chil-
dren’s answers were written down verbatim. Independent raters
coded children’s responses. For problem identification, children
were asked to define the problem. Responses were awarded two
points when the child accurately described the problem and
referred to a consequence (e.g., “They are not talking to me and
I feel sad.”), one point when the child accurately described the
problem but did not mention a consequence (e.g., “They are not
talking to me.”), and zero points were given when the child
gave an inaccurate or implausible response. For goal genera-
tion, children were asked “How would you like things to turn
out?” Children were awarded one point for a prosocial goal
(e.g., “I want us to become friends.”) and zero points for all
other goals (e.g., “I want him to get in trouble”). Children were
then asked to generate possible solutions. Each assertive, com-
petent solution was awarded a point (e.g., “I would walk up and
say hi.”), while all other responses (aggressive, passive-
avoidant or 3rd party intervention) were scored a “0.” A final
raw score for each child for each code was the average score
across raters and vignettes.
Alternate Measures of Social-Emotional
Comprehension
To aid in establishing construct validity, alternate child mea-
sures that reflected Social Awareness, Social Meaning, and Social
Reasoning were selected and administered.
Social Awareness. The Davis Set of Emotion Expressions
(Tracy & Robins, 2004) served as an alternate measure of Social
Awareness. Children viewed photographs and indicated whether
the emotion expressed by the person in each photograph matched
a target emotion word (happy, sad, angry, or afraid). Eighteen
photographs, reflecting varied emotion displays, were presented in
random order for each target emotion. Children received one point
each time they correctly indicated whether a photograph reflected
the target emotion.
Social Meaning. The first 15 items of the NEPSY-II ToM
(Korkman et al., 2007) served as an alternate measure of Social
Meaning and was administered to children in kindergarten through
fifth grade. Items measured false belief understanding, nonliteral
language comprehension, and appearance-reality distinction.
Social Reasoning. An independent assessment of Social
Reasoning was not administered. However, as part of the Social
Reasoning task, children were asked to provide as many pos-
sible solutions to address the problem presented in each of the
five vignettes. The number of solutions generated was not part
of our model of social-emotional comprehension but is ac-
knowledged by other theorists as an important indicator of
social problem-solving skill (Spivack & Shure, 1974). Accord-
ingly, the total number of solutions generated for each vignette
was averaged across the five vignettes and used as an alternate
measure of Social Reasoning.
Data for Temporal Stability Analyses
Children who participated in the first year of the school study
and who remained at their respective school during the course
of the study were invited to participate in each subsequent year.
As a result, two waves of data were available for a subset of the
school sample (n 72). Children repeating the assessment
completed a shortened group of assessments. The average in-
terval between testing was 12.0 months (SD 2.0 months). The
DANVA Child Faces and Davis Set of Emotion Expressions
were not administered during the subsequent years to reduce the
burden on participants. Participants in the clinic sample were
not retested.
Results
Descriptive Statistics
Table 1 includes descriptive statistics. Table 2 shows correla-
tions between variables.
Missing Data
Cross-sectional data. In 12 cases from the school sample,
because of absences or participant dropout, five or more assess-
ment scores were missing, including all assessment scores from at
least one, and often two, dimensions of social-emotional compre-
hension. Those cases were omitted from subsequent analyses. The
remaining samples included 174 students in the school sample and
119 in the clinic sample.
Next, the percentage of participants completing each measure in
each sample was calculated (Table 1). For 27 of 30 measures, data
were collected from more than 95% of participants. However, in
the school sample, 74% of children completed the Strange Stories
and 70% completed the NEPSY-II ToM. These assessments were
not administered to students above fifth grade because of devel-
opmentally expected ceiling effects.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
1157
ASSESSING SOCIAL-EMOTIONAL COMPREHENSION
Controlling for age, children who were missing Strange Stories
were slightly better at defining social problems, 1.2 versus 1.0,
2
.03, F(1, 164) 4.16, p .05, and produced slightly more
competent potential solutions to hypothetical social problems, 1.5
versus 1.2,
2
.03, F(1, 164) 4.19, p .05. As with Strange
Stories, controlling for age, children who were missing NEPSY-II
ToM data were slightly better at defining social problems, 1.2
versus 1.0,
2
.02, F(1, 164) 3.96, p .05, and produced
Table 1
Description of Assessments
Social-emotional dimension
and assessment Items
School sample Clinic sample % Available
n Stability MSDn MSDSchool Clinic
Social Awareness
Awareness composite .95 .96
DANVA Accuracy 24 174 .71 82.2 12.4 118 .70 78.4 13.4 100 99
CATS MEPEF 22 174 .67 .70 49.6 15.8 118 .72 51.8 17.5 100 99
Postures Accuracy 24 173 .80 .42 80.7 16.5 118 .82 79.2 17.9 99 99
PLW 20 173 .60 .33 65.4 15.4 118 .64 64.8 16.4 99 99
Social Meaning
Meaning composite .97 .97
Strange Stories 12 128 .74 .64 5.2 2.8 119 .75 6.2 2.9 74 100
CASL Pragmatics 60 174 .96 .76 41.3 16.1 109 .94 45.5 14.3 100 92
Social Reasoning
Reasoning composite .95 .90
Problem Identification 5 169 .61 .24 1.1 0.3 115 .50 1.1 0.3 98 100
Goal Quality 5 174 .69 .17 0.8 0.2 114 .49 0.8 0.2 97 100
No. Competent Solutions 5 169 .72 .31 1.4 0.6 108 .59 1.2 0.5 98 100
Criterion Measures
Davis composite .97 .96
Davis Happiness 18 174 .64 84.9 10.6 119 .64 82.4 11.5 100 100
Davis Sadness 18 174 .76 85.3 15.3 119 .72 83.0 14.9 100 100
Davis Anger 18 174 .83 90.3 14.8 119 .75 88.4 13.8 100 100
Davis Fear 18 174 .82 86.4 16.8 119 .83 84.4 18.3 100 100
NEPSY-II ToM 15 122 .74 .67 15.4 4.0 113 .74 17.3 3.8 70 95
No. Solutions 5 169 .84 .24 2.3 0.8 108 .75 2.1 0.8 97 100
Note. DANVA Diagnostic Analysis of Nonverbal Accuracy (Nowicki & Duke, 1994); CATS MEPEF Match Emotional Prosody to Emotional Face
subtest of the Comprehensive Affect Testing System (Weiner, Gregory, Froming, Levy, & Ekman, 2006); PLW point-light walker; CASL
Comprehensive Assessment of Spoken Language (Carrow-Woolfolk, 1999); NEPSY-II ToM NEPSY-II theory of mind (Korkman et al., 2007).
Table 2
Zero-Order Correlations Between Variables
Variable 1 2 345678910111213141516
1. Age .37
.43
.50
.42
.53
.63
.39
.46
.33
.29
.34
.29
.29
.55
.05
2. DANVA .46
.32
.59
.39
.33
.46
.31
.20
.23
.34
.36
.36
.37
.37
.06
3. CATS MEPEF .53
.49
.47
.26
.40
.47
.28
.24
.24
.31
.36
.31
.36
.29
.14
4. Posture .42
.60
.40
.46
.49
.55
.31
.31
.12 .39
.43
.50
.54
.53
.06
5. PLW .46
.45
.38
.48
.35
.43
.33
.31
.17
.21
.32
.41
.33
.31
.08
6. Strange Stories .48
.35
.37
.42
.44
.70
.50
.42
.37
.34
.34
.37
.38
.63
.26
7. CASL .77
.51
.54
.51
.53
.79
.59
.56
.51
.35
.35
.37
.43
.78
.34
8. Prob ID .27
.28
.18
.28
.27
.39
.39
.44
.47
.26
.17
.22
.26
.39
.28
9. Goal Quality .26
.21
.21
.13
.19
.27
.30
.10 .34
.24
.27
.24
.32
.39
.03
10. Competent Sol .52
.33
.41
.32
.34
.49
.60
.49
.21
.13 .20
.27
.26
.37
.63
11. Davis Happy .45
.37
.39
.41
.29
.39
.50
.21
.08 .24
.35
.36
.35
.33
.06
12. Davis Sad .32
.39
.33
.42
.34
.34
.40
.13
.20
.22
.25
.58
.51
.30
.06
13. Davis Angry .32
.36
.32
.35
.32
.34
.42
.15
.22
.26
.35
.54
.61
.33
.07
14. Davis Afraid .40
.36
.36
.44
.52
.29
.46
.29
.19
.29
.41
.37
.42
.39
.04
15. NEPSY-II ToM .56
.30
.30
.32
.27
.68
.72
.43
.29
.58
.39
.20
.34
.36
.20
16. # Solutions .26
.14
.19
.02 .17
.23
.34
.38
.04 .72
.16
.01 .10
.14
.37
Note. DANVA Diagnostic Analysis of Nonverbal Accuracy (Nowicki & Duke, 1994); CATS MEPEF Match Emotional Prosody to Emotional Face subtest
of the Comprehensive Affect Testing System (Weiner, Gregory, Froming, Levy, & Ekman, 2006); PLW point-light walker; CASL Comprehensive
Assessment of Spoken Language (Carrow-Woolfolk, 1999); Prob ID problem identification; Sol solutions; NEPSY-II ToM NEPSY-II theory of mind
(Korkman et al., 2007). Correlations below the diagonal are from the school sample; correlations above the diagonal are from the clinical sample.
p .10.
p .05.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
1158
MCKOWN, ALLEN, RUSSO-PONSARAN, AND JOHNSON
slightly lower quality goals in response to hypothetical social
problems, 0.9 versus 1.0,
2
.05, F(1, 164) 8.53, p .05.
Children who were missing Strange Stories and NEPSY-II ToM
did not differ from children who were not missing those measures
on any other available measures. Having missing data from other
variables was not associated with performance on other measures.
Because data from most variables were missing at random
(MAR), confirmatory factor analyses (CFAs) and structural equa-
tion models (SEMs) were estimated with full information maxi-
mum likelihood (FIML). When data are MAR and variables asso-
ciated with missingness are included in the model, FIML results
are equivalent to stochastic multiple imputation (McCartney, Bub,
& Burchinal, 2006).
Temporal Stability. We evaluated the extent to which chil-
dren in the school sample with two waves of data differed from
those with one. In the context of independent samples t tests, on 12
of 16 observed variables, there were no significant differences
between children who did and did not have two waves of data.
Compared to children who participated in the study for only 1 year,
children with two waves of data were younger, 8.7 versus 9.7 years
old, t(291) 2.75, p .05, and provided more competent potential
solutions to hypothetical social problems, 1.5 versus 1.2, t(275)
3.10, p .05. There were no other significant differences between
children who did and who did not have two waves of data avail-
able.
Reliability
Interrater reliability. Interviewers transcribed children’s ver-
batim responses to Strange Stories during the assessment. A team
of four raters independently coded children’s verbatim responses
to each item. Across the 12 Strange Stories items, average pairwise
kappa was .72, and intraclass correlation was .90. A team of five
raters coded all social problem-solving responses for two consec-
utive years of data collection. Average pairwise kappa for problem
identification, goal quality, competence of solutions generated, and
competence of best responses was .67, .54, .66, .70, and .97,
respectively. The intraclass correlation for these codes was .92,
.84, .95, .92, and .99, respectively.
Reliability over 1 year. One-year measurement stability is
presented in Table 1 for measures and participants for which
successive waves of data were available. The average 1-year
stability was r .45, with a range of .17 to .76. All stability
coefficients were statistically significant (p .05) except for Goal
Quality and the number of solutions generated.
Internal consistency. Cronbach’s alpha was computed for
each assessment and sample (see Table 1). In the school sample,
internal consistency reliability ranged from .60 to .96. Ten out of
15 assessments achieved an alpha of .70 or greater and average
internal consistency was .75. In the clinical sample, internal con-
sistency reliability ranged from .49 to .94. Ten out of 15 scales
achieved an alpha of .70 or greater and average internal consis-
tency was .70. For four of the 14 measures, differences in alphas
between samples (Feldt, 1969) were statistically significant (W
.67, p .05 for CASL; W .61, p .05 for Goal Quality;
W .68, p .05 for the number of competent solutions gener-
ated; W .68, p .05 for Davis Anger recognition).
Composite reliability. When the elements of a composite
score are correlated, the reliability of the composite score is higher
than the average reliability of those elements (Nunnally & Bern-
stein, 1994). Factor scores for Nonverbal Awareness, Social Mean-
ing, and Social Reasoning were created and standardized and used
in all validity analyses. Nunnally and Bernstein (1994, p. 271)
defined the reliability of a composite that is a weighted sum as
follows:
r
yy
1
b
i
2
i
2
b
i
2
i
2
r
ii
Y
2
,
where r
yy
is the reliability of the latent variable, b
i
is factor weights
of each measure i associated with the latent variable,
i
2
is the
variance of each measure i, r
ii
is the reliability of measure i, and
Y
2
is the summed variance of the obtained scores.
The reliability of the Social Awareness composite was .95 for
the school sample, and .96 for the clinic sample. The reliability of
the Social Meaning composite was .97 for the school and clinic
samples. The reliability of the Social Reasoning composite was .95
for the school sample, and .90 for the clinic sample. Sample
differences in the reliabilities of Social Awareness and Social
Meaning scores were not statistically significant (W .80 for
Social Awareness, ns,W 1.00 for Social Meaning). The sample
difference in the reliability of the Social Reasoning composite was
statistically significant (W .50, p .05).
Validity
Factor structure. We used Amos (17.0.2; Arbuckle, 2008)to
construct and test confirmatory factor analysis (CFA) and struc-
tural equation models (SEM). The fit of all models was evaluated
with overall
2
goodness-of-fit,
2
/df adjusted goodness-of-fit, the
comparative fit index (CFI), the incremental fit index (IFI), and
root-mean-square error of approximation (RMSEA). Models were
interpreted as a good fit with the data if the
2
/df adjusted
goodness-of-fit statistic was 2, CFI and IFI were both .90,
RMSEA .08 (Browne & Cudeck, 1993), and hypothesized
coefficients were significant and in the predicted direction.
We compared three models in each sample. Summary findings
can be found in Table 3. First, we tested a one-factor model in
which all indicator variables loaded onto a single factor. The fit of
this model to the data was marginal. Next, we tested a two-factor
model in which the DANVA, CATS MEPEF, posture recognition
and point-light display scores loaded on one factor, and Strange
Stories, pragmatic judgment, problem identification, goal quality,
and the number of competent solutions loaded on a second factor.
The fit of the data to this model in both samples was acceptable
and was significantly better than the one-factor model. Next, we
tested a three-factor model corresponding to our theory of social-
emotional comprehension, which is depicted in Figure 1. The data
fit the three-factor model was significantly better than the two-
factor model in both samples, and, as seen in Table 2, the fit of the
data to the model was excellent. Across samples, the three-factor
model coefficients were the same valence and similar magnitude.
Because the three-factor model fit our conceptualization of social-
emotional comprehension and was superior to simpler models in
both samples, the remaining analyses used this model. Factor
scores reflecting Social Awareness, Social Meaning, and Social
Reasoning were saved, standardized, and used in validity analyses
focused on age- and diagnostic differences.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
1159
ASSESSING SOCIAL-EMOTIONAL COMPREHENSION
Post hoc analyses also revealed that, in both samples, the data fit
a model in which Social Awareness, Social Meaning, and Social
Reasoning loaded on a second-order Social-Emotional Compre-
hension latent variable, IFI/CFI .98, RMSEA .055 (.000
.089),
2
(24) 36.4, ns,
2
/df 1.52 for the school sample; and
IFI 1.00, CFI 1.00, RMSEA .058 (.029 –.082),
2
(25)
24.8, ns,
2
/df 0.99 for the clinic sample. This second-order
factor may be useful to achieve some research goals (see McKown
et al., 2009). We were interested in the convergent and discrimi-
nant validity of Social Awareness, Social Meaning, and Social
Reasoning. Accordingly, we opted to examine the three-factor
model without consideration of the higher order factor.
Convergent and discriminant validity. We next evaluated
the convergent and discriminant validity of the social-emotional
comprehension assessments. For each sample, SEM was used to
evaluate the extent to which latent variables reflecting Social
Awareness, Social Meaning, and Social Reasoning created with
core assessments were more related to parallel latent variables
created with alternate assessments than they were related to other
latent variables created with alternate assessments. Convergent
validity was demonstrated by strong associations between Social
Awareness and the alternate Social Awareness latent variable,
between Social Meaning and the alternate Social Meaning latent
variable, and between Social Reasoning and the alternate Social
Reasoning latent variable. Discriminant validity was demonstrated
by weaker associations between Social Awareness and the alter-
nate Social Meaning and Social Reasoning latent variables, be-
tween Social Meaning and the alternate Social Awareness and
Social Reasoning latent variables, and between Social Reasoning
and the alternate measures of Social Awareness and Social Mean-
ing latent variables.
SEMs corresponding to this model were fit separately to the
school sample data and the clinic sample data. In both cases, the
overall fit of the model to the data was very good, IFI/CFI .96,
RMSEA .055 (.035–.073),
2
(81) 126.0, p .05,
2
/df
1.56 for the school sample; and IFI/CFI .95, RMSEA .058
(.029 –.082),
2
(81) 113.4, p .05,
2
/df 1.40 for the clinic
sample. These models are depicted in Figure 2. Inspection of
model paths revealed that in both samples, the structural coeffi-
cients linking parallel latent variables with alternate assessments
were large and statistically significant. Furthermore, most of the
structural coefficients linking latent variables in different domains
were nonsignificant. Those that were statistically significant were
nevertheless much smaller in magnitude than coefficients linking
parallel latent variables.
Age differences in social-emotional comprehension. As
shown in Table 2, performance on all social-emotional compre-
hension assessments were significantly and positively associ-
ated with age in both samples. Not surprisingly, therefore, as
illustrated in Figure 3, factor scores reflecting Social Aware-
ness, Social Meaning, and Social Reasoning were positively
associated with age in both samples. What is clear from this
Table 3
Goodness-of-Fit Indices of Alternative Models, by Sample
Model df
2
2
/df
⌬␹
2
(df)
IFI CFI RMSEA (90% CI)
School sample
One factor 27 83.5 3.09 0.90 .90 .110 (.084–.137)
Two factor 26 46.1 1.77 29.3 (1)
0.97 .96 .067 (.033–.098)
Three factor 24 36.4 1.52 9.7 (2)
0.98 .98 .055 (.000–.089)
Clinic sample
One factor 27 64.7 2.40 0.90 .89 .068 (.047–.089)
Two factor 26 30.1 1.16 34.6 (1)
0.99 .99 .023 (.000–.053)
Three factor 24 22.3 0.93 7.8 (2)
1.01 1.00 .000 (.000–.043)
Note. IFI incremental fit index; CFI comparative fit index; RMSEA root-mean-square error of approximation; CI confidence interval.
p .05.
.78/
.74
.67/
.53
.80/
.90
.72/.84
.65
Faces
Voices
Posture
Gait
Social
Awareness
.64/
.56
/.56
.74/
.67
Social
Reasoning
ID
Goal
Solution
.55/
.71
.41/.63
.78
/.62
Social
Meaning
.96/
.87/
.76
ToM
Prag
.91
Figure 1. Three-factor confirmatory model of social-emotional compre-
hension. Note that coefficients are standardized. All values to the left of the
“/” are from the school sample; values to the right of the “/” are from the
clinic sample. Bold signifies paths and path coefficients reflecting conver-
gent validity. All coefficients are significant at the p .05 level.
2
(24)
36.4/22.3, ns;
2
/df 1.52/0.93; incremental fit index 0.98/1.01; com
-
parative fit index .98/1.00; root-mean-square error of approximation
(90% confidence interval) .055 (.000–.089)/.000 (.000 –.043). ToM
theory of mind; Prag pragmatic judgment subtest of the Comprehensive
Assessment of Spoken Language; ID problem identification.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
1160
MCKOWN, ALLEN, RUSSO-PONSARAN, AND JOHNSON
figure is that the slope of age-related changes in social-
emotional comprehension are very similar in school and clinic
samples, with higher mean scores in the school sample than in
the clinic sample.
The plot of social-emotional comprehension across age sug-
gests a steeper curve at younger ages, gradually flattening with
age. Post hoc analyses evaluated the extent to which age-related
changes in domain scores was curvilinear for each factor score
reflecting Social Awareness, Social Meaning, and Social Rea-
soning. Specifically, each factor score was regressed on age and
an age-squared quadratic term. Analyses were run separately by
dimension of social-emotional comprehension and sample. If
the quadratic term was significant, this was interpreted as
evidence for the presence of a curvilinear relationship between
age and the outcome measure in question. Additional analyses
were run adding participant characteristics that differed across
groups (sex, ethnicity, SSRS Social Skills and Problem Behav-
ior) as covariates. In all cases, findings were unchanged.
We therefore maintained the simpler models reported in
Table 4.
Diagnostic differences in social-emotional comprehension.
To further evaluate validity, we compared children from dif-
ferent diagnostic groups on standardized factor scores reflecting
Social Awareness, Social Meaning, and Social Reasoning. Age-
adjusted standardized means and standard errors are reflected in
Figure 4. Controlling for age, children in the school sample scored
significantly higher than children in the clinic sample on the compos-
ite measures of Social Awareness, Social Meaning, and Social Rea-
soning, F(1, 289) 16.9, p .05 for Social Awareness; F(1, 289)
25.0, p .05 for Social Meaning; and F(1, 289) 19.5, p .05
for Social Reasoning. Pairwise comparisons also revealed that
children with ASD performed worse than typically developing
children and worse than children with either ADHD or RD on
composite measures of Social Awareness, Social Meaning, and
Social Reasoning. Additional analyses were run adding participant
characteristics that differed across groups (sex, ethnicity, SSRS
Social Skills and Problem behavior scores) as covariates. In all
cases, findings were unchanged. We therefore used the simpler
models reported above and depicted in Figure 4.
Discussion
Direct assessments of social-emotional comprehension are crit-
ical elements in the clinical toolkit, enabling assessors to evaluate,
identify, and conceptualize contributors to social impairment. This
study contributes to a growing body of literature finding that direct
.88/
.25
/-.37
-.19
.05/
.66
.80
.45
.65/
.79
.54/.49
.61/
.70
.80/.61
1.0/1.0
Social
Awareness
Scared
Angry Sad Happy
Social
Reasoning
# Solutions
.68/.87
.67/.76
.83/.95
1.0/1.0
-.08/
.18
-.33/
-.32
-.11
/
.36
.15/
.01
Faces
Voices
Posture
Gait
Social
Awareness
.81/
.64/
.52/
.75/.85
.68
.66/
.54
/.56
.74/
.67
.67
.95/
Social
Meaning
Social
Reasoning
ID
Goal
Solution
.49 /
.64
.28/.52
1.0
/.78
ToM
Prag
1.0
Social
Meaning
NEPSY ToM
Figure 2. Convergent and discriminant validity social-emotional comprehension assessments. Note that
coefficients are standardized. All values to the left of the “/” are from the school sample; values to the right of
the “/” are from the clinic sample. Bold signifies paths and path coefficients reflecting convergent validity. All
coefficients are significant at the p .05 level except underlined values, which are nonsignificant.
2
(81)
126.0/113.4, p .05;
2
/df 1.56/1.40; incremental fit index .96/.96; comparative fit index .96/.95;
root-mean-square error of approximation (90% confidence interval) .055 (.035–.073)/.058 (.029 –.082). ToM
theory of mind; Prag pragmatic judgment subtest of the Comprehensive Assessment of Spoken Language;
ID problem identification.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
1161
ASSESSING SOCIAL-EMOTIONAL COMPREHENSION
assessment of social-emotional comprehension is feasible, that
direct assessments yield reliable scores, and that the scores ob-
tained from direct assessments measure what they are designed to
measure. We found that (a) direct assessments of social-emotional
comprehension reflected three latent variables that we called So-
cial Awareness, Social Meaning, and Social Reasoning; (b) the
reliability of the factor scores was excellent; (c) the latent variables
demonstrated convergent and discriminant validity; and (d) the
assessments differentiated children from different age and diag-
nostic groups. Further development is needed to increase the
usability and feasibility of assessments that cover a wide age
range, offer broad construct coverage, and yield subtest scores
with greater reliability.
Relationship to Existing Theory
Findings from this study are consistent with and extend theo-
retical models of social-emotional comprehension. Consistent with
Lipton and Nowicki (2009), this study suggests that three broad
dimensions—Social Awareness, Social Meaning, and Social Rea-
soning—reflect salient aspects of social-emotional comprehen-
sion. Undoubtedly, those constructs may be further subdivided and
assessed. This study suggests that for investigators and clinicians
wishing to obtain a broad assessment of social-emotional compre-
hension, assessing across these three dimensions is feasible, the-
oretically and empirically justified, can be done efficiently, and
includes important factors that affect children’s social behavior.
Thus, the present study, and the model of social-emotional com-
prehension that it supports, offers a straightforward and empiri-
cally supported heuristic with which investigators may organize
their research hypotheses and clinicians may gather data to de-
velop case formulations.
Psychometrics of Social-Emotional
Comprehension Assessment
This study demonstrated that direct assessments of Social
Awareness, Social Meaning, and Social Reasoning yield reliable
composite scores. This is consistent with prior research reporting
the reliability of direct social-emotional comprehension assess-
ments that measure Social Awareness, Social Meaning, or Social
Reasoning (Banerjee & Watling, 2005; Hughes & Ensor, 2007;
Kupersmidt et al., 2011; McKown et al., 2009; Nowicki & Duke,
1994; Slaughter et al., 2002; Yeates et al., 1991). Study findings
also support the conclusion that direct assessment of social-
emotional comprehension is valid for its intended purpose. CFA
and SEM analyses both found that Social Awareness, Social
Meaning, and Social Reasoning are distinct, but correlated con-
Table 4
The Relationship Between Age and Social-Emotional Comprehension Factor Scores, by Sample
Parameter
Criterion
School sample Clinic sample
Social Awareness Social Meaning Social Reasoning Social Awareness Social Meaning Social Reasoning
Intercept 4.05
5.73
5.84
6.13
6.80
6.42
Age B 0.65
0.97
1.00
0.94
1.11
1.07
Age
2
B
0.02
0.03
0.04
0.03
0.04
0.04
R
2
.49
.62
.57
.41
.49
.42
Note. Parameter estimates are unstandardized regression coefficients.
p .05.
Figure 3. Relationship between age (in years) and social-emotional com-
prehension.
Figure 4. Social-emotional comprehension, by sample and diagnostic
group. Error bars indicate one standard error. ADHD attention-deficit/
hyperactivity disorder.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
1162
MCKOWN, ALLEN, RUSSO-PONSARAN, AND JOHNSON
structs that demonstrate convergent and discriminant validity.
Those constructs in turn reflect a higher order social-emotional
comprehension skill. Moreover, the present study provides evi-
dence that measuring broadly across distinct dimensions of social-
emotional comprehension can yield findings that may more com-
prehensively reflect an individual child’s strengths and needs than
assessments that are more narrowly tailored.
The psychometric properties of the social-emotional compre-
hension assessments were consistent in two very different samples
of children—one from a general education setting and one from a
clinic, lending confidence to the conclusion that the social-
emotional comprehension scores are reliable, and that the assess-
ments demonstrate robust construct validity. In addition, direct
assessments discriminated children from different diagnostic and
age groups. As expected, the school sample performed better than
the clinic sample. Furthermore, children with autism-spectrum
disorders performed worse than children with ADHD or RD. This
is consistent with previous research documenting severe social-
emotional comprehension deficits in children with ASD (e.g.,
Bauminger, 2002; Capps, Yirmiya, & Sigman, 1992; Channon et
al., 2001; Clark, Winkielman, & McIntosh, 2008; Dodd, Ocampo,
& Kennedy, 2011; Embregts & van Nieuwenhuijzen, 2009; Flood,
Julian Hare, & Wallis, 2011; Mazefsky & Oswald, 2007; Meyer,
Mundy, Van Hecke, & Durocher, 2006; Volden, Mulcahy, &
Holdgrafer, 1997). These findings also mirror work suggesting less
severe social deficits among children with ADHD and RD (Hall &
Richmond, 1985; Jackson, Enright, & Murdock, 1987; Nowicki &
Duke, 1994; Sciberras, Ohan, & Anderson, 2012; H. L. Swanson
& Malone, 1992; Tur-Kaspa & Bryan, 1994).
Limitations and Future Directions
Construct representation. Social behavior and social func-
tioning are influenced by an array of individual and situational
factors, many of which were beyond the scope of the present study
and were thus not examined. This study presented strong evidence
of the validity of social-emotional comprehension assessments
covering a conceptually coherent set of dimensions. Nevertheless,
alternative models are clearly possible. Further research should
explore other factors that influence social behavior and function-
ing, such as self regulation.
Sample size. This studies included two relatively small sam-
ples (n 174 and n 119 in the school and clinic samples,
respectively) that substantially differed in characteristics such as
sex and racial-ethnic composition. Despite the small sample sizes,
in both studies, the fit of the data to the models was excellent and
the coefficients were robust, in the predicted direction, and of
comparable magnitude. Similar findings in two very different
samples suggest that these findings are robust and generalizable to
children with a broad array of social-emotional challenges. Future
research with larger samples will permit multiple-group analyses
to evaluate whether the model coefficients differ between groups.
Measurement. Certain measurement limitations from this
study leave open questions for future research. For example,
longitudinal data were only available in the school sample and
only for a subset of measures. This limits our understanding of the
temporal stability of social-emotional comprehension in children
with neurobehavioral disorders. In addition, the alternative mea-
sure for Social Reasoning was taken from the same measure as our
core assessments. Thus, the strong relationship between the focal
Social Reasoning latent variable and the alternate Social Reason-
ing latent variable may be inflated by common method variance.
Therefore, the convergent validity findings regarding Social Rea-
soning should be interpreted with caution. Nevertheless, Social
Reasoning did demonstrate good evidence of discriminant validity.
The internal consistency of individual assessments was variable.
The average internal consistency was in the mid .70s, and some
measures were much lower than this. All scores were retained
because of their importance to our model of social-emotional
comprehension. In contrast, the reliabilities of the composite factor
scores was generally well above .90. On a practical level, this
means that scores from most of the individual assessments are
insufficiently reliable for interpreting the strengths and needs of
individual children from those scores and are certainly too low for
high-stakes decision making. At the composite level, reliabilities
are adequate for interpreting individual student scores. An impor-
tant limitation of these assessments, therefore, is that to obtain
reliable estimates of social-emotional comprehension requires the
administration and complex aggregation of many individual as-
sessments, which reduces the usability and utility of such assess-
ments in real-world practice. Further assessment development is
needed to increase the consistency of measurement and reduce the
number of measures required to achieve this goal. Until that
development occurs, interpretation of individual student social-
emotional comprehension test scores should proceed with caution.
The internal consistency of some measures and composite
scores differed between samples. Future work on the development
of direct assessments should be particularly mindful of the poten-
tial for measurement reliability to be lower in clinical populations
and should seek to identify and minimize the sources of unreli-
ability. In the present study, age and diagnostic differences were
tested on composite scores, not on scores on individual assess-
ments. At the composite level, sample differences between esti-
mated reliabilities were low, with a maximum difference of .07.
The likely impact of this difference is that it may have made it
more difficult to detect differences between samples. We con-
clude, therefore, that group differences were robust.
Temporal stability analyses suggest that, over the course of 12
months, general education children generally score consistently on
direct assessments of social-emotional comprehension. However,
reliability coefficients over this span varied greatly and were
generally lower than desirable for clinical measures. Future work
should examine the temporal stability over a shorter time span to
more clearly evaluate the test–retest reliability of these assessment
strategies.
A final limitation is that this study did not address the incre-
mental validity of these assessments, over and above other vari-
ables that may be more readily assessed in clinical practice. If, for
example, the relationship between social-emotional comprehen-
sion and relevant outcomes were mediated entirely by child IQ,
then these assessment approaches would not demonstrate incre-
mental criterion-related validity and, as a result, would be of
limited value. It is reassuring that prior research has suggested that
social-emotional comprehension assessments account for variance
in outcomes, over and above IQ (McKown, 2007; Rivers, Brackett,
& Salovey, 2008). Furthermore, prior research has found that a
higher order factor reflecting Social Awareness, Social Meaning,
and Social Reasoning is more strongly associated with teacher
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
1163
ASSESSING SOCIAL-EMOTIONAL COMPREHENSION
report of social behavior than any one of those factors alone
(McKown et al., 2009). Further research is necessary to fully
evaluate the incremental validity of these assessment approaches.
Complexity. There are practical limitations of the assessments
examined in this study. They require varying procedures and level
of training. Even a shortened group of assessments takes more than
1 hr to administer. The existence and quality of norms varies, so
interpreting scores is challenging. As a result, these assessment
approaches may be useful in the context of clinical practice when
the clinician has obtained sufficient training to administer, score,
and interpret findings. As assessment development moves forward,
clinicians are advised to use the best contemporary standardized
assessments designed to measure important dimensions of social-
emotional comprehension in children and youth. Assessments such
as the NEPSY-II ToM and Affect Recognition subtests, the SIP-
AP, MSCEIT-YV, and the Social Language Development Test are
examples of currently available instruments.
Conclusion
Clinicians and researchers can use direct assessments of social-
emotional comprehension to understand the strengths and limita-
tions that affect each child’s social relationships. The evidence
presented in this article suggests that direct assessment, used in
conjunction with other well-validated strategies, such as behavior
rating scales, may provide important and clinically useful infor-
mation about factors that affect social relationships. In addition to
being able to characterize strengths and limitations, it is important
that assessments be able to guide treatment planning. Accordingly,
an important next step will be to investigate strategies for linking
the careful assessment of social-emotional comprehension to in-
tervention strategies that help children develop the skills they need
to be successful.
References
Adolphs, R. (2003). Cognitive neuroscience of human social behaviour.
Nature Reviews Neuroscience, 4, 165–178. doi:10.1038/nrn1056
Arbuckle, J. L. (2008). Amos (Version 17.0). Chicago, IL: SPSS.
Banerjee, R., & Watling, D. (2005). Children’s understanding of faux pas:
Associations with peer relations. Hellenic Journal of Psychology, 2,
27– 45.
Baron-Cohen, S. (1989). The autistic child’s theory of mind: A case of
specific developmental delay. Child Psychology & Psychiatry & Allied
Disciplines, 30, 285–297. doi:10.1111/j.1469-7610.1989.tb00241.x
Baron-Cohen, S., Wheelwright, S., Scahill, V., Lawson, J., & Spong, A.
(2001). Are intuitive physics and intuitive psychology independent?
Journal of Developmental & Learning Disorders, 5, 47–78.
Bauminger, N. (2002). The facilitation of social-emotional understanding
and social interaction in high-functioning children with autism: Inter-
vention outcomes. Journal of Autism and Developmental Disorders, 32,
283–298. doi:10.1023/A:1016378718278
Bauminger, N., Edelsztein, H. S., & Morash, J. (2005). Social information
processing and emotional understanding in children with LD. Journal of
Learning Disabilities, 38, 45– 61. doi:10.1177/00222194050380010401
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model
fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural equation
models (pp. 136 –162). Beverly Hills, CA: Sage.
Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant
validation by the multitrait–multimethod matrix. Psychological Bulletin,
56, 81–105. doi:10.1037/h0046016
Capps, L., Kehres, J., & Sigman, M. (1998). Conversational abilities
among children with autism and children with developmental delays.
Autism, 2, 325–344. doi:10.1177/1362361398024002
Capps, L., Yirmiya, N., & Sigman, M. (1992). Understanding of simple
and complex emotions in non-retarded children with autism. Child
Psychology & Psychiatry & Allied Disciplines, 33, 1169 –1182. doi:
10.1111/j.1469-7610.1992.tb00936.x
Carrow-Woolfolk, E. (1999). Comprehensive assessment of spoken lan-
guage manual. Circle Pines, MN: American Guidance Service.
Chan, S., & Mpofu, E. (2001). Children’s peer status in school settings:
Current and prospective assessment procedures. School Psychology In-
ternational, 22, 43–52. doi:10.1177/01430343010221004
Channon, S., Charman, T., Heap, J., Crawford, S., & Rios, P. (2001).
Real-life-type problem-solving in Asperger’s syndrome. Journal of Au-
tism and Developmental Disorders, 31, 461–469. doi:10.1023/A:
1012212824307
Chung, T., & Asher, S. R. (1996). Children’s goals and strategies in peer
conflict situations. Merrill-Palmer Quarterly, 42, 125–147.
Clark, T. F., Winkielman, P., & McIntosh, D. N. (2008). Autism and the
extraction of emotion from briefly presented facial expressions: Stum-
bling at the first step of empathy. Emotion, 8, 803– 809. doi:10.1037/
a0014124
Collins, M., & Nowicki, S. J. (2001). African American children’s ability to
identify emotion in facial expressions and tones of voice of European
Americans. The Journal of Genetic Psychology: Research and Theory on
Human Development, 162, 334 –346. doi:10.1080/00221320109597487
Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social
information-processing mechanisms in children’s social adjustment.
Psychological Bulletin, 115, 74–101. doi:10.1037/0033-2909.115.1.74
Denham, S. A. (2006). Social-emotional competence as support for school
readiness: What is it and how do we assess it? Early Education and
Development, 17, 57– 89. doi:10.1207/s15566935eed1701_4
Dirks, M. A., Treat, T. A., & Weersing, V. R. (2007). The situation
specificity of youth responses to peer provocation. Journal of Clinical
Child and Adolescent Psychology, 36, 621–628. doi:10.1080/
15374410701662758
Dodd, J. L., Ocampo, A., & Kennedy, K. S. (2011). Perspective taking
through narratives: An intervention for students with ASD. Communi-
cation Disorders Quarterly, 33, 23–33. doi:10.1177/1525740110395014
Dodge, K. A., & Price, J. M. (1994). On the relation between social
information processing and socially competent behavior in early school-
aged children. Child Development, 65, 1385–1397. doi:10.2307/
1131505
Duke, M. P., Nowicki, S., & Walker, E. (1996). Teaching your child the
language of social success. Atlanta, GA: Peachtree.
Embregts, P., & van Nieuwenhuijzen, M. (2009). Social information pro-
cessing in boys with autistic spectrum disorder and mild to borderline
intellectual disabilities. Journal of Intellectual Disability Research, 53,
922–931. doi:10.1111/j.1365-2788.2009.01204.x
Erdley, C. A., & Asher, S. R. (1999). A social goals perspective on
children’s social competence. Journal of Emotional and Behavioral
Disorders, 7, 156 –167. doi:10.1177/106342669900700304
Feldt, L. S. (1969). A test of the hypothesis that Cronbach’s alpha or
Kuder-Richardson coefficient twenty is the same for two tests. Psy-
chometrika, 34, 363–373. doi:10.1007/BF02289364
Flood, A. M., Julian Hare, D., & Wallis, P. (2011). An investigation into
social information processing in young people with Asperger syndrome.
Autism, 15, 601– 624. doi:10.1177/1362361310387803
Gresham, F. M., & Elliott, S. N. (1990). Social skills rating system manual.
Circle Pines, MN: American Guidance Service.
Gresham, F. M., & Elliott, S. N. (2008). Social skills improvement system:
Rating scales. Bloomington, MN: Pearson Assessments.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
1164
MCKOWN, ALLEN, RUSSO-PONSARAN, AND JOHNSON
Halberstadt, A. G., Denham, S. A., & Dunsmore, J. C. (2001). Affective
social competence. Social Development, 10, 79 –119. doi:10.1111/1467-
9507.00150
Hall, C. W., & Richmond, B. O. (1985). Non-verbal communication,
self-esteem and interpersonal relations of LD and non-LD students.
Exceptional Child, 32, 87–91. doi:10.1080/0156655850320204
Happé, F. G. E. (1994). An advanced test of theory of mind: Understanding
of story characters’ thoughts and feelings by able autistic, mentally
handicapped, and normal children and adults. Journal of Autism and
Developmental Disorders, 24, 129–154. doi:10.1007/BF02172093
Heberlein, A. S., Adolphs, R., Tranel, D., & Damasio, H. (2004). Cortical
regions for judgments of emotions and personality traits from point-light
walkers. Journal of Cognitive Neuroscience, 16, 1143–1158. doi:
10.1162/0898929041920423
Heberlein, A. S., Gläescher, J., & Adolphs, R. (2007). Emotion recognition
from human postures with masked facial expressions. Unpublished man-
uscript.
Hughes, C., & Ensor, R. (2007). Positive and protective: Effects of early
theory of mind on problem behaviors in at-risk preschoolers. Journal of
Child Psychology & Psychiatry & Allied Disciplines, 48, 1025–1032.
doi:10.1111/j.1469-7610.2007.01806.x
Izard, C. E., & Harris, P. (1995). Emotional development and develop-
mental psychopathology. In D. Cicchetti & D. J. Cohen (Eds.), Devel-
opmental psychopathology, Vol. 1: Theory and methods (pp. 467–503).
Oxford, England: Wiley.
Jackson, S. C., Enright, R. D., & Murdock, J. Y. (1987). Social perception
problems in learning disabled youth: Developmental lag versus percep-
tual deficit. Journal of Learning Disabilities, 20, 361–364. doi:10.1177/
002221948702000612
Kaufman, J., Birmaher, B., Brent, D., Rao, U., Flynn, C., Moreci, P.,...
Ryan, N. (1997). Schedule for Affective Disorders and Schizophrenia
for School-Age Children–Present and Lifetime Version (K-SADS-PL):
Initial reliability and validity data. Journal of the American Academy of
Child & Adolescent Psychiatry, 36, 980 –988. doi:10.1097/00004583-
199707000-00021
Korkman, M., Kirk, U., & Kemp, S. (2007). NEPSY–Second Ed. (NEPSY-
II). San Antonio, TX: Pearson Assessment.
Kupersmidt, J. B., Stelter, R., & Dodge, K. A. (2011). Development and
validation of the social information processing application: A web-based
measure of social information processing patterns in elementary school-
age boys. Psychological Assessment, 23, 834 847. doi:10.1037/
a0023621
Landa, R. (2000). Social language use in asperger syndrome and high-
functioning autism. In A. Klin, F. R. Volkmar, & S. S. Sparrow (Eds.),
Asperger syndrome (pp. 125–155). New York, NY: Guilford Press.
Lipton, M., & Nowicki, S. (2009). The social emotional learning frame-
work (SELF): A guide for understanding brain-based social emotional
learning impairments. Journal of Developmental Processes, 4, 99 –115.
Lord, C., Rutter, M., DiLavore, P., & Risi, S. (1999). Autism Diagnostic
Observation Schedule-WPS edition. Los Angeles, CA: Western Psycho-
logical Services.
Lord, C., Rutter, M., & Le Couteur, A. (1994). Autism Diagnostic Inter-
view—Revised: A revised version of a diagnostic interview for caregiv-
ers of individuals with possible pervasive developmental disorders.
Journal of Autism and Developmental Disorders, 24, 659 685. doi:
10.1007/BF02172145
Mavroveli, S., Petrides, K. V., Shove, C., & Whitehead, A. (2008). Inves-
tigation of the construct of trait emotional intelligence in children.
European Child & Adolescent Psychiatry, 17, 516 –526. doi:10.1007/
s00787-008-0696-6;10.1007/s00787-008-0696-6
Mayer, J. D., Caruso, D. R., & Salovey, P. (2005). The Mayer-Salovey-
Caruso Emotional Intelligence Test–Youth Version (MSCEIT-YV) Re-
search Version. Toronto, Ontario, Canada: Multi-Health Systems.
Mazefsky, C. A., & Oswald, D. P. (2007). Emotion perception in Asperg-
er’s syndrome and high-functioning autism: The importance of diagnos-
tic criteria and cue intensity. Journal of Autism and Developmental
Disorders, 37, 1086 –1095. doi:10.1007/s10803-006-0251-6
McCartney, K., Bub, K. L., & Burchinal, M. (2006). Best practices in
quantitative methods for developmentalists: VI. Selection, detection, and
reflection. Monographs of the Society for Research in Child Develop-
ment, 71, 105–126.
McKown, C. (2007). Concurrent validity and clinical usefulness of several
individually administered tests of children’s social-emotional cognition.
Journal of Clinical Child and Adolescent Psychology, 36, 29 41.
McKown, C., Gumbiner, L. M., Russo, N. M., & Lipton, M. (2009).
Social-emotional learning skill, self-regulation, and social competence
in typically developing and clinic-referred children. Journal of Clinical
Child and Adolescent Psychology, 38, 858 871. doi:10.1080/
15374410903258934
Messick, S. (1995). Validity of psychological assessment: Validation of
inferences from persons’ responses and performances as scientific in-
quiry into score meaning. American Psychologist, 50, 741–749. doi:
10.1037/0003-066X.50.9.741
Meyer, J. A., Mundy, P. C., Van Hecke, A. V., & Durocher, J. S. (2006).
Social attribution processes and comorbid psychiatric symptoms in
children with Asperger syndrome. Autism, 10, 383– 402. doi:10.1177/
1362361306064435
Mikami, A. Y., Lee, S. S., Hinshaw, S. P., & Mullin, B. C. (2008).
Relationships between social information processing and aggression
among adolescent girls with and without ADHD. Journal of Youth and
Adolescence, 37, 761–771. doi:10.1007/s10964-007-9237-8
Moreno, J. L. (1933). Psychological and social organization of groups in
the community. Proceedings & Addresses. American Association on
Mental Deficiency, 38, 224–242.
Nowicki, S., & Duke, M. P. (1994). Individual differences in the nonverbal
communication of affect: The diagnostic analysis of nonverbal accuracy
scale. Journal of Nonverbal Behavior, 18, 9 –35. doi:10.1007/
BF02169077
Nowicki, S., Duke, M. P., & van Buren, A. (2008). Starting kids off right.
Atlanta, GA: Peachtree Press.
Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.).
New York, NY: McGraw-Hill.
Parker, J. G., & Asher, S. R. (1987). Peer relations and later personal
adjustment: Are low-accepted children at risk? Psychological Bulletin,
102, 357–389. doi:10.1037/0033-2909.102.3.357
Peterson, C. C., Wellman, H. M., & Liu, D. (2005). Steps in theory-of-
mind development for children with deafness or autism. Child Devel-
opment, 76, 502–517. doi:10.1111/j.1467-8624.2005.00859.x
Pillow, B. H. (2011). Children’s discovery of the active mind: Phenome-
nological awareness, social experience, and knowledge about cognition.
New York, NY: Springer Science and Business Media.
Reynolds, C. R., & Kamphaus, R. W. (2004). Behavior assessment system
for children (2nd ed.). San Antonio, TX: Pearson Assessment.
Rivers, S. E., Brackett, M. A., Reyes, M. R., Mayer, J. D., Caruso, D. R.,
& Salovey, P. (2012). Measuring emotional intelligence in early ado-
lescence with the MSCEIT-YV: Psychometric properties and relation-
ship with academic performance and psychosocial functioning. Journal
of Psychoeducational Assessment, 30, 344 –366. doi:10.1177/
0734282912449443
Rivers, S. E., Brackett, M. A., & Salovey, P. (2008). Measuring emotional
intelligence as a mental ability in adults and children. In G. J. Boyle, G.
Matthews, & D. H. Saklofske (Eds.), The SAGE handbook of personality
theory and assessment (Vol. 2, pp. 440 460). Los Angeles, CA: Sage.
doi:10.4135/9781849200479.n21
Rutter, M., Bailey, A., Lord, C., & Berument, S. K. (2003). Social com-
munication questionnaire. Los Angeles, CA: Western Psychological
Services.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
1165
ASSESSING SOCIAL-EMOTIONAL COMPREHENSION
Salovey, P., & Mayer, J. D. (1989 –1990). Emotional intelligence. Imagi-
nation, Cognition and Personality, 9, 185–211. doi:10.2190/DUGG-
P24E-52WK-6CDG
Sciberras, E. F., Ohan, J. F., & Anderson, V. (2012). Bullying and peer
victimisation in adolescent girls with attention-deficit/hyperactivity dis-
order. Child Psychiatry and Human Development, 43, 254–270.
doi:10.1007/s10578-011-0264-z
Slaughter, V., Dennis, M. J., & Pritchard, M. (2002). Theory of mind and
peer acceptance in preschool children. British Journal of Developmental
Psychology, 20, 545–564. doi:10.1348/026151002760390945
Spivack, G., & Shure, M. B. (1974). Social adjustment of young children:
A cognitive approach to solving real-life problems. Oxford, England:
Jossey-Bass.
Swanson, H. L., & Malone, S. (1992). Social skills and learning disabili-
ties: A meta-analysis of the literature. School Psychology Review, 21,
427– 443.
Swanson, J. M. (1995). The Swanson, Nolan, and Pelham Teacher and
Parent Rating Scale (SNAP-IV). Irvine, CA: Child Development Center,
University of California.
Tracy, J. L., & Robins, R. W. (2004). Show your pride: Evidence for a
discrete emotion expression. Psychological Science, 15, 194 –197. doi:
10.1111/j.0956-7976.2004.01503008.x
Tur-Kaspa, H., & Bryan, T. (1994). Social information-processing skills of
students with learning disabilities. Learning Disabilities Research &
Practice, 9, 12–23.
Volden, J., Mulcahy, R. F., & Holdgrafer, G. (1997). Pragmatic language
disorder and perspective taking in autistic speakers. Applied Psycholin-
guistics, 18, 181–198. doi:10.1017/S0142716400009966
Wagner, R. K., Torgesen, J. K., & Rashotte, C. A. (1999). Comprehensive
Test of Phonological Processing. Austin, TX: PRO-ED.
Walker, H. M., & Severson, H. (1992). Systematic screening for behavior
disorders. Longmont, CO: Sopris West.
Wechsler, D. (1999). Wechsler Abbreviated Scale of Intelligence (WASI).
San Antonio, TX: Psychological Corporation.
Wechsler, D. (2005). Wechsler Individual Achievement Test 2nd ed. (WIAT
II). London, England: Psychological Corporation.
Weiner, S. G., Gregory, A., Froming, K. B., Levy, C. M., & Ekman, P.
(2006). Emotion processing: The comprehensive affect testing system
user’s manual. Sanford, FL: Psychology Software.
Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysis of theory-
of-mind development: The truth about false belief. Child Development,
72, 655– 684. doi:10.1111/1467-8624.00304
Wellman, H. M., & Liu, D. (2004). Scaling of theory-of-mind tasks. Child
Development, 75, 523–541. doi:10.1111/j.1467-8624.2004.00691.x
White, S., Hill, E., Happé, F., & Frith, U. (2009). Revisiting the strange
stories: Revealing mentalizing impairments in autism. Child Develop-
ment, 80, 1097–1117. doi:10.1111/j.1467-8624.2009.01319.x
Yeates, K. O., Schultz, L. H., & Selman, R. L. (1990). Bridging the gaps
in child-clinical assessment: Toward the application of social-cognitive
developmental theory. Clinical Psychology Review, 10, 567–588. doi:
10.1016/0272-7358(90)90097-T
Yeates, K. O., Schultz, L. H., & Selman, R. L. (1991). The development of
interpersonal negotiation strategies in thought and action: A social-
cognitive link to behavioral adjustment and social status. Merrill-Palmer
Quarterly, 37, 369 405.
Zuckerman, M., Blanck, P. D., DePaulo, B. M., & Rosenthal, R. (1980).
Developmental changes in decoding discrepant and nondiscrepant non-
verbal cues. Developmental Psychology, 16, 220 –228. doi:10.1037/
0012-1649.16.3.220
Received August 28, 2012
Revision received May 13, 2013
Accepted May 15, 2013
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
1166
MCKOWN, ALLEN, RUSSO-PONSARAN, AND JOHNSON
... Students' social emotional competence was assessed using the SELweb Early Education assessment (Aksu Dunya et al., 2019;McKown, 2019;McKown et al., 2013), which is designed for students in kindergarten through third grade. This web-based task consists of four subtests assessing the following social-emotional skills: (1) emotion recognition, (2) self-control, (3) social perspective-taking, and (4) social problem-solving. ...
Article
Full-text available
Objectives This study examined the effect of the Settle Your Glitter mindfulness-based curriculum on kindergarten through second-grade students’ emotion recognition in others, self-control, social problem-solving, and perspective-taking. Methods Students from one school (n = 186; 66% African American and 20% Latinx) experienced the mindfulness intervention plus business-as-usual academic curriculum. Students from another school (n = 214; 84% African American and 11% Latinx) experienced the academic curriculum only with inclusion of a literacy coach (active control). Results Intervention students demonstrated better posttest emotion recognition (Cohen’s d = 0.38) and problem-solving (Cohen’s d = 0.26) than control students. The effect on emotion recognition was moderated by pretest levels of self-control, such that intervention students with lower initial self-control had greater gains in emotion recognition than those with higher initial self-control, whose gain scores did not differ from control students. Control students had higher posttest perspective-taking scores than intervention students, which is likely a result of literacy coaching that control teachers received. Pretest self-control scores positively predicted perspective-taking gain scores, regardless of treatment group. Conclusions The curriculum improved skills that are critical for social awareness and managing relationships with others.
... Finally, there are a few additional studies where investigators have used different measurement methods to evaluate the CASEL model. For example, McKown et al., (2013) evaluated the factor structure of a direct measure of SEL competencies that was partially aligned with the CASEL framework with two relatively small but diverse samples of students. A school sample consisted of 174 children ranging in age from 5 to 14 years. ...
Article
Full-text available
Although interest in social-emotional learning (SEL) is growing, there are few efficient measures of SEL constructs, especially those of the popular CASEL Framework. To address this need, the SSIS SEL Brief Scales (SSIS SELb) were developed. Initial evidence for the SSIS SELb is promising, but evidence is needed for its structural validity. The purpose of this study was to provide such evidence. We used confirmatory factor analysis to evaluate six plausible structures of the SSIS SELb-Teacher (n = 643) and SSIS SELb-Student (n = 530) forms with a large sample collected from two Midwestern private/parochial schools. Results indicated each of the models were defensible, but all models pointed to the conclusion that applied use should focus on the overall SEL Composite score. Findings provided support for the SSIS SELb Scales and indicated the need for future research better explicating the theoretical specifics of the CASEL Framework. Highlights • The study provided evidence of the structural validity of the SSIS SEL Brief Scales – Teacher and Student Forms, two efficient, multi-rater assessments of Social Emotional Learning competencies. • Using a large sample of private school students, results suggested that multiple structures were plausible for the SSIS SEL Brief Scales, including hierarchical, correlated factors, and bifactor structures. • Applied use should focus on the SSIS SEL Composite as recommended by the test publishers and future research is needed both on the structure of the SSIS SEL Brief scales and the theoretical implications of the CASEL framework.
... Therefore, assessment measures for preschoolers and school-aged children with sufficient psychometric properties are required. Several instruments have been developed to measure emotional competence in early childhood (e.g., the Assessment of Children's Emotions Skills [30] and the Diagnostic Analysis of Nonverbal Accuracy [DANVA; 31]), for reviews see McKown et al. [32] or McKown [33]; there is no claim for a complete overview on emotional competence measures. Overall, there are only a few validated instruments for the preschool age (e.g., the Children and Adolescents' Recognition of Emotions [30], the Emotion Matching Task [34], and the Affect Knowledge Test shortened [AKT-S, 35]), and these instruments are available only in English, indicating the necessity of a German instrument. ...
Article
Full-text available
A computer-based emotional competence inventory for preschoolers and school-aged children (MeKKi) was developed to assess five components of emotional competence: emotion vocabulary, emotion identification (situational, visual, auditory), emotion understanding, emotion expression, and emotion regulation. Validity, reliability, and factor structure were examined in a community sample of 313 preschoolers and school-aged children (164 boys, 145 girls, 4 n.a.) age 4–11 years ( M = 6.35 years, SD = 1.85). Item statistics and Cronbach’s α were calculated for the subscales. The unidimensionality of the subscales was additionally tested via item response theory or confirmatory factor analysis. Internal consistency (α) was overall satisfactory at 0.82, though the consistencies of the Visual and Auditory Emotion Identification subscales were lower. Unidimensionality was demonstrated for all subscales except Emotion Understanding. Results provide support for the use of the MeKKi in research and clinical settings to assess emotional competence.
... While certainly working memory (e.g., van Nieuwenhuijzen & Vriens, 2012) and language abilities (e.g., Bauminger-Zviely et al., 2019;McKown et al., 2013) support SIP, to reduce cognitive demands, users do not need to rely solely on imagination, working memory, or language skills to access the interactions. Animated and narrated scenarios are played out twice and response options can be repeated through a hovering mechanism. ...
Article
Virtual Environment for Social Information Processing (VESIPTM) is a web‐based social information processing assessment developed for youth in Grades 3‐7. VESIP was developed to address: (a) the broader assessment of social information processing in a variety of socially challenging situations and (b) the need for technically strong and scalable assessments that can be administered universally in schools as a growing number of states adopt social and emotional learning standards. Consistent with the Crick and Dodge theoretical model of social information processing, VESIP assesses six different dimensions: solution preference, problem identification, emotion response, intent attribution, goal preference, and social self‐efficacy. This study summarizes technical properties of VESIP based on the evaluation of two general education samples: a multi‐state group of students whose data were part of a large‐scale norming study (N = 2,156), and a subset of local students from that group who participated in a validation study (n = 334). Confirmatory factor analyses supported a model that has three distinct facets: (a) one that includes five social information processing factors, (b) one that includes five situational factors, and (c) one overall social information processing factor. This model closely parallels the Crick and Dodge model and suggests that social information processing is somewhat situation specific. Internal consistency and test‐retest reliabilities for social information processing factors were good. VESIP scores were consistently associated with an alternate measure of social information processing and other criterion measures. Implications for theory and practice are considered.
... Thus, as an organizing model, it provides a way to connect specific competencies with abstract theory in a predictive, functional way, for a population that exhibits unambiguous social challenges (those with ASD). The SELF posits that social competence is influenced by three elements of social cognition, which cut across both implicit and explicit processes: social reasoning, meaning, and awareness (McKown et al., 2013). Lipton and Nowicki (2009) identified laboratory-based measures to operationalize each of these three elements of social cognition; these measures, which are included in this study, can be categorized into assessments of aspects of explicit or implicit social cognition. ...
Article
Lay abstract: Difficulties with social communication and interaction are a hallmark feature of autism spectrum disorder. These difficulties may be the result of problems with explicit social cognition (effortful and largely conscious processes) such as learning and recalling social norms or rules. Alternatively, social deficits may stem from problems with implicit social cognition (rapid and largely unconscious processes) such as the efficient integration of social information. The goal of this study was to determine how problems in explicit and implicit social cognition relate to social behavior in 34 youth with autism spectrum disorder. We measured aspects of implicit and explicit social cognition abilities in the laboratory using behavioral, cognitive, and brain (electrophysiological) measures. We then used those measures to predict "real-world" social behavior as reported by parents, clinicians, and independent observers. Results showed that overall better aspects of implicit and explicit social cognition predicted more competent social behavior. In addition, the ability to fluidly integrate social information (implicit social cognition) was more frequently related to competent social behavior that merely knowing what to do in social situations (explicit social cognition). These findings may help with the development of interventions focusing on improving social deficits.
... Participants receive standard scores for each of the subtests (M = 100; SD = 15). The CASL has evidenced good internal consistency, contentvalidity, and test-rest reliability (McKown et al. 2013;Hoffman et al. 2011;Reichow et al. 2008). ...
Article
Full-text available
The language and social skill deficits associated with autism spectrum disorder (ASD) warrant further study. Existing research has focused on the contributions of pragmatic language to social skills, with little attention to other aspects of language. We examined the associations across three language domains (semantics, syntax, and pragmatics) and their relations to parent- and teacher-rated social skills among children with ASD. When parent-reported language skills were considered simultaneously, only semantics significantly predicted children’s social skills. For teacher-reported language skills, all three language domains predicted children’s social skills, but none made unique contributions above and beyond one another. Further research should consider the impact of social context on language expectations and interventions targeting semantic language on children’s development of social skills.
... Inattention is becoming increasingly common in elementary school-aged children, and thus, there is a growing demand for SEL approaches (McKown, Allen, Russo-Ponsaran, & Johnson, 2013). Those with ADHD have a four times lower college graduation rate and poorer average socioeconomic status than those without ADHD ( Erskine et al., 2016). ...
Article
Full-text available
To design a social and emotional learning (SEL) skills training program for children with attention deficit hyperactivity disorder (ADHD). Quantitative electroencephalography was analyzed, focusing on the strengths and weaknesses in the brain mapping performance (BMP) of children. A total of 305 participants with and without ADHD were recruited. Theta relative power was analyzed using stepwise multiple regression. Twelve brain locations were further predicted using a design thinking model to develop a reasonable prototype via crossover methods. Both strengths and weaknesses were found at the T5 and F7 brain regions in the ADHD group. The Fz region was a well‐defined predictor in both groups. The co‐predictors of both groups were F3 and F4. No comparison was suggested to enhance positive performance in either group of children. The SEL program could be designed based on a good‐fit model of the BMP of these subcomponents of primary brain functions.
Article
The passage of the No Child Left Behind Act in 2002 and the Every Student Succeeds Act in 2015 spurred changes in the way educators use data. On the one hand, the policies inspired educators’ awareness of large gaps in achievement between subgroups based on gender, race, and socioeconomic status. On the other hand, the policies inspired the use of data-based indicators in day-to-day routines in schools. In some cases, practitioners started working with researchers to analyze data using advanced measurement techniques to develop “early warning indicators.” These indicators identify when students are at risk of not achieving key milestones, such as graduating from high school, in hopes of providing practitioners with time to intervene early enough to change the students’ trajectories. Educational leaders have created “early warning systems” that produce reports summarizing these indicators, provide a process for the leaders to organize interventions based on the indicator reports, and a further process to examine whether students are back on track toward milestones such as high school graduation as a result of the intervention. Given the pace at which the science and use of these indicators is advancing, the topic is in need of a new resource explaining the most up-to-date elements involved with early warning indicators.
Article
Impairments in theory of mind (ToM)—long considered common among individuals with autism spectrum disorder (ASD)—are in fact highly heterogeneous across this population. Although such heterogeneity should be reflected in differential recruitment of neural mechanisms during ToM reasoning, no research has yet uncovered a mechanism that explains these individual differences. In this study, 78 (48 with ASD) adolescents viewed ToM vignettes and made mental-state inferences about characters’ behavior while participant electrophysiology was concurrently recorded. Two candidate event-related-potentials (ERPs)—the late positive complex (LPC) and the late slow wave (LSW)—were successfully elicited. LPC scores correlated positively with ToM accuracy and negatively with ASD symptom severity. Note that the LPC partially mediated the relationship between ASD symptoms and ToM accuracy, which suggests that this ERP component, thought to represent cognitive metarepresentation, may help explain differences in ToM performance in some individuals with ASD.
Article
Full-text available
Se considera que educar en emociones es una tarea fundamental, que deben asumir los padres y profesionales, ya que el desarrollo de su competencia, permite mejorar otros aspectos de su vida. En los últimos años, la investigación en el campo de las emociones ha aumentado, sin embargo, en el área de intervención temprana y educación especial hay pocos estudios específicos. Esta investigación, realizada a través de grupos focales, busca identificar la importancia que tanto los padres como los profesionales le dan a las competencias emocionales y sus implicaciones para el desarrollo del niño. Los resultados sugieren que los profesionales y padres creen manejar los aspectos emocionales, sin embargo, su manejo es sólo en algunas categorías y de forma empírica, basándose en su experiencia personal. Queda clara la necesidad de formación en el tema y la realización de un seguimiento del estado emocional de niños, padres y profesionales.***A Competência Emocional em Intervenção Precoce: Perspectivas dos pais e dos profissionais***Considera-se que educar as emoções é uma tarefa fundamental que deve ser assumida por pais e profissionais, pois o desenvolvimento de sua competência permite aprimorar outros aspectos de sua vida. Nos últimos anos, as investigações neste campo aumentaram, contudo, existem poucos estudos específicos na área de intervenção precoce e educação especial. Esta pesquisa, realizada por meio de grupos focais, procura identificar a importância que os pais e profissionais atribuem às competências emocionais e suas implicações no desenvolvimento da criança. Os resultados sugerem que os profissionais e os pais acreditam que gerem os aspectos emocionais; no entanto, a sua gestão é apenas em algumas categorias e de forma empírica, com base na sua experiência pessoal. Ficou clara a necessidade de formação neste assunto e a monitorização do estado emocional de crianças, pais e profissionais em futuros estudos.Palavras chave: Atenção precoce, práticas centradas na família, habilidades emocionais, educação emocional.
Article
Full-text available
Emotional intelligence (EI) theory provides a framework to study the role of emotion skills in social, personal, and academic functioning. Reporting data validating the importance of EI among youth have been limited due to a dearth of measurement instruments. In two studies, the authors examined the reliability and validity of the Mayer-Salovey-Caruso Emotional Intelligence Test-Youth Version (MSCEIT-YV), a performance test of EI. Study 1 examined psychometric attributes of the MSCEIT-YV in a large sample of fifth- to eighth-grade students (N = 756). Study 2 examined the relationship of the MSCEIT to student and teacher reports of academic, social, and personal functioning among fifth- and sixth-grade students (N = 273). The authors report that EI can be measured reliably with the MSCEIT-YV and that higher scores on the test are related to healthier psychological functioning and greater social competence based on both teacher and student ratings, as well as to academic performance in English language arts.
Article
Typescript (photocopy). Thesis (Ph. D. in education)--University of Illinois at Chicago, 1993. Vita. Includes bibliographical references (leaves 95-105).
Article
This article presents a framework for emotional intelligence, a set of skills hypothesized to contribute to the accurate appraisal and expression of emotion in oneself and in others, the effective regulation of emotion in self and others, and the use of feelings to motivate, plan, and achieve in one's life. We start by reviewing the debate about the adaptive versus maladaptive qualities of emotion. We then explore the literature on intelligence, and especially social intelligence, to examine the place of emotion in traditional intelligence conceptions. A framework for integrating the research on emotion-related skills is then described. Next, we review the components of emotional intelligence. To conclude the review, the role of emotional intelligence in mental health is discussed and avenues for further investigation are suggested.