Journal of Psychoeducational

Assessment
Volume 26 Number 3
September 2008 215-230
The Test Anxiety Inventory for © 2008 Sage Publications
10.1177/0734282907303760

Children and Adolescents (TAICA) http://jpa.sagepub.com

hosted at
http://online.sagepub.com
Examination of the Psychometric
Properties of a New Multidimensional
Measure of Test Anxiety Among
Elementary and Secondary School Students
Patricia A. Lowe
Steven W. Lee
Kristin M. Witteborg
Keri W. Prichard
Megan E. Luhr
Christopher M. Cullinan
Bethany A. Mildren
Jennifer M. Raad
Rebecca A. Cornelius
Melissa Janik
University of Kansas, Lawrence

The Test Anxiety Inventory for Children and Adolescents (TAICA) is a new multidimensional
measure used to assess test anxiety in elementary and secondary school students. The TAICA
is a 45-item self-report measure consisting of a Total Test Anxiety scale, four debilitating test
anxiety subscales (Cognitive Obstruction/Inattention, Physiological Hyperarousal, Social
Humiliation, and Worry), a facilitating test anxiety scale (Performance Enhancement/
Facilitation Anxiety), and a Lie scale. In the present study, the psychometric properties of the
TAICA scores are examined with a volunteer sample of 206 children and adolescents. Results
of the study indicate that the TAICA scores have strong to very strong internal consistency
reliability and temporal stability (1- to 3-week test-retest interval). Evidence supporting the
construct validity of the TAICA scores was found.

Keywords: test anxiety; reliability; validity; Test Anxiety Inventory for Children and Adolescents

T esting is a common practice in contemporary society. Results of tests are used to make
important decisions about an individual’s status in school, college, and work (Zeidner,
1998). Testing is widely used in U.S. schools. When one considers the impact that tests

Authors’ Note: Please address correspondence to Patricia A. Lowe, Department of Psychology and Research
in Education, 634 JRP Hall, Lawrence, KS 66045; e-mail: [email protected].

215
216 Journal of Psychoeducational Assessment

have on students’ lives, it is no wonder that tests and testing situations may produce anxiety
(Zeidner, 1998) in many U.S. elementary and secondary school students.

Prevalence and Negative Effects of Test Anxiety

Students with test anxiety feel tense, fearful, and worried in evaluative situations
(Spielberger & Vagg, 1995). These students do not perform optimally on tests (Hancock,
2001). The number of students who experience test anxiety has been difficult to estimate
(Zeidner, 1998). However, Methia (2004) indicated that more than one third of the students
in U.S. schools experience some test anxiety, and the number of students who experience
test anxiety has increased with the increase in testing and testing requirements (e.g., the No
Child Left Behind Act of 2001) in U.S. schools (Casbarro, 2005).
Test anxiety has negative effects on students’ learning and academic performance (Sub &
Prabha, 2003). Test-anxious students do not perform well on standardized achievement
tests (Everson, Millsap, & Rodriguez, 1991), receive poorer grades (Chapell et al., 2005),
and are more likely to be retained (Beidel & Turner, 1988) and to drop out of school
(Tobias, 1979).

Definition of the Test Anxiety Construct

As a scientific construct, different definitions of test anxiety have evolved as a function
of both the changes in the zeitgeist and theoretical perspectives of key investigators
involved in research in the field (Zeidner, 1998). In the early days, the construct was
defined in motivational terms (Spence & Spence, 1966), and test anxiety was simply a
reflection and expression of general anxiety in evaluative situations (Taylor, 1956) in which
test-anxious students emitted more task-irrelevant responses (i.e., incorrect responses),
resulting in poorer test performance (Spence & Spence, 1966). A shift from a behavioral to
a cognitive orientation occurred in the late 1960s and early 1970s (Hembree, 1988). During
this period of time, test anxiety was viewed as a cognitive-attentional phenomenon (Wine,
1971). According to the cognitive-attentional model, test-anxious students divide their
attention between task-relevant thoughts (e.g., thoughts about how to solve a problem on a
test) and task-irrelevant thoughts (e.g., worries about oneself and one’s performance). The
task-irrelevant thoughts interfere with the students’ ability to focus on the test, thus lower-
ing their test performance (Wine, 1971). The study and test-taking skills deficit paradigm
(e.g., Benjamin, McKeachie, Lin, & Holinger, 1981; Kirkland & Hollandsworth, 1980)
became popular in the early 1980s. According to this model, students with poor study and
test-taking skills experience difficulty encoding and organizing material presented in class,
and because of this difficulty, these students do not perform well on tests. For these
students, test anxiety occurs after repeated failings on tests because they have inadequate
study and test-taking skills (e.g., Benjamin et al., 1981; Kirkland & Hollandsworth, 1980).
More recent conceptualizations of test anxiety were derived from the self-regulation
(Carver & Scheier, 1991), self-worth (Covington, 1992), and transactional process
(Spielberger & Vagg, 1995) models. The self-regulation paradigm views human behavior
 Lowe et al. / Test Anxiety Inventory for Children and Adolescents 217

as a continuous process of matching current behaviors to desired goals (Carver &

Scheier, 1991). Test anxiety occurs when students experience a major discrepancy
between current behaviors and progress toward desired goals (Carver & Scheier, 1991).
In evaluative situations associated with desired goals, test-anxious students experience
self-doubts about their ability to perform, engage in self-depreciating thoughts, and mis-
interpret and do not attend to relevant cues, resulting in poor test performance (Zeidner,
1998). In contrast, the self-worth model suggests that test-anxious students are not able
to maintain their self-worth or protect their academic competency because of repeated
academic failures (Covington, 1992). According to this model, test anxiety is viewed as
a symptom of “perceived intellectual incompetency and a defensive ploy to ward off low
self-evaluation” (Zeidner, 1998, p. 81). In the transactional process model, the dynamic
interaction between students’ personality traits and specific evaluative situations are
highlighted. Test-anxious students have a propensity to experience high levels of trait
anxiety across testing situations, and they perceive specific evaluative situations as
threatening and experience different degrees of state (i.e., situational) anxiety. Worry and
emotionality are symptoms of state anxiety, and these symptoms interfere with students’
test performance (Spielberger & Vagg, 1995; Zeidner, 1998).
A review of the most popular test anxiety models developed over the past 50 years shows
that these paradigms differ in how they conceptualize test anxiety. Yet there appears to be
agreement among the researchers in the field that the test anxiety construct is complex
(Benson, 1998; Friedman & Bendas-Jacob, 1997; Zeidner, 1998). The current view of the
test anxiety construct is that it is a complex construct consisting of multiple dimensions
(Benson, 1998; Zeidner, 1998). Because it is multidimensional in nature, no single theo-
retical perspective has been able to account for its complexity at the present time (Zeidner,
1998). More comprehensive models of test anxiety are needed that (a) define the test anx-
iety construct more broadly, (b) include many key components found in earlier models, and
(c) incorporate current research in the field (Zeidner, 1998). We present such a model (i.e.,
a working model) below.

New Test Anxiety Model

Figure 1 shows our current conception of the influences on and results of test anxiety.
The model is based fundamentally on the biopsychosocial model (Engel, 1977; Schwartz,
1982) that views psychological, social, and biological processes as influential in health sta-
tus. The test anxiety model presented here recognizes the proximal and distal influences of
social systems on the development of test anxiety. The extant research to date enables us to
specify some of the most critical variables that influence the development and maintenance
of test anxiety.
The within-child variables shown on the left side of the test anxiety model, such as intel-
ligence (e.g., Hembree, 1988), social–emotional functioning (e.g., Beidel & Turner, 1988),
trait anxiety (e.g., McDonald, 2001), study skills and habits (e.g., Cassady, 2004), acade-
mic ability (e.g., Sub & Prabha, 2003), and academic self-efficacy (e.g., McIlroy, Bunting,
& Adamson, 2000), are viewed as variables that play an important role in the perceived
218 Journal of Psychoeducational Assessment

Figure 1
Test Anxiety Model

threat of an upcoming test. The degree to which the test is perceived as a threat (i.e., diffi-
culty, possibility of negative evaluation from others, importance of the test) appears to
influence the level of test anxiety (Hancock, 2001).
When test anxiety occurs (see bulb of thermometer in Figure 1), the individual’s behav-
ior, cognitions, and physiology are affected. Behaviorally, test anxiety has the capability of
producing task-relevant behavior (e.g., attention to a task or studying for a test) or task-
irrelevant behavior (e.g., skipping items or “cramming”; Alpert & Haber, 1960). Test anx-
iety has long been known to produce physiological arousal, originally termed emotionality
(Liebert & Morris, 1967; Spielberger & Vagg, 1995), now termed physiological hyper-
arousal (e.g., Joiner et al., 1999). In test anxiety, physiological hyperarousal may produce
sweaty palms, increased heart rate, and shallow or rapid breathing when the individual
encounters a test. Finally, the cognitive dimension of test anxiety seems to include worry
(Morris & Liebert, 1969) and cognitive obstruction (e.g., Swanson & Howell, 1996; Wine,
1971). Although most are familiar with worry, cognitive obstruction refers to the degree to
which test anxiety disrupts the individual’s ability to organize his or her thoughts or to con-
centrate on the task at hand. Social humiliation also refers to cognitions related to fear or
worry that others will deride or disparage the student’s performance on a test. Friedman and
Bendas-Jacob’s (1997) factor analytic and cross-validation study showed that social dero-
gation was a salient feature of test anxiety.
 Lowe et al. / Test Anxiety Inventory for Children and Adolescents 219

On the basis of the classic study by Yerkes and Dodson (1908), Alpert and Haber (1960)
studied test anxiety in undergraduates and found that it has the capacity to enhance or
facilitate test performance at low levels or degrade performance at high levels. This aspect
of test anxiety is depicted in the model as the rise of test anxiety in the thermometer.
Finally, test anxiety is seen as influencing the individual in two ways: the immediate
appraisal of test item performance and the final score or results on the test. When an indi-
vidual’s immediate response to items is perceived as incorrect or inadequate, this percep-
tion may further increase test anxiety (Berger, Munz, Smouse, & Angelino, 1969). Recent
research has shown that test scores are related to academic self-efficacy (Nielsen &
Moore, 2003). On the basis of this working model of test anxiety, a new multidimensional
measure of test anxiety was developed, the Test Anxiety Inventory for Children and
Adolescents (TAICA).

New Test Anxiety Measure

The TAICA is a new multidimensional measure used to assess test anxiety in children
and adolescents in Grades 4 through 12. The TAICA is a 45-item self-report measure (see
Table 1 for examples of the items) and consists of four debilitating test anxiety subscales
(Cognitive Obstruction/Inattention, Physiological Hyperarousal, Social Humiliation, and
Worry), a facilitating test anxiety scale (Performance Enhancement/Facilitation Anxiety),
and a Lie scale. The Cognitive Obstruction/Inattention subscale (7 items) assesses memory
and attention difficulties and obstructed cognitions associated with test anxiety. The
Physiological Hyperarousal subscale (7 items) measures physical symptoms associated
with the test-taking process. The Social Humiliation subscale (10 items) measures fear
associated with failing a test and being belittled or ridiculed by significant others, and the
Worry subscale (7 items) assesses negative thoughts and worries that adversely affect test
performance. The Performance Enhancement/Facilitation Anxiety scale (7 items) measures
facilitating test anxiety. The Lie scale (7 items) is a validity index and measures “ideal” test
behavior. There is a fierce debate in the literature about lie scales and whether they (a) mea-
sure true or error variance and (b) reduce validity coefficients of scale scores (see Mueller-
Hanson, 2002; Piedmont, McCrae, Riemann, & Angleitner, 2000). However, lie scales have
become more sophisticated, and many test developers continue to include lie scales when
they develop new or revise old tests.
In addition to these scales and subscales, the TAICA has a Total Test Anxiety scale. The
Total Test Anxiety scale (31 items) provides an overall measure of test anxiety. The Total
Test Anxiety scale score is obtained by summing the raw scores from the four debilitating
test anxiety subscales (Lowe, Lee, & DeRuyck, 2004).
Limited information is available on the psychometric properties of the TAICA scores.
Thus, the present study examined the sources of validity evidence supporting the use of the
TAICA. Measures used to assess external relations included the Behavior Assessment
System for Children–Self-Report of Personality (BASC-SRP; Reynolds & Kamphaus,
1992a), Revised Children’s Manifest Anxiety Scale (RCMAS; Reynolds & Richmond,
1978), Reynolds Intellectual Screening Test (RIST; Kamphaus & Reynolds, 2002), Test
Anxiety Inventory (TAI; Spielberger, 1980a), Test Anxiety Scale for Children (TASC;
220 Journal of Psychoeducational Assessment

Table 1
Examples of Items Found on Each Scale or Subscale of the
Test Anxiety Inventory for Children and Adolescents
Item No. Item Scale or Subscale

1 I am worried that people will make fun of me if I fail a test. Social Humiliation
37 I fear that my teacher will think I am stupid if I fail a test. Social Humiliation
18 I feel shaky before taking a test. Physiological Hyperarousal
36 I feel sick to my stomach while I am taking a test. Physiological Hyperarousal
3 I worry about how well I did on a test until I receive my grade. Worry
25 I worry about a test before taking it. Worry
5 I do better on a test if I am slightly worried before I take it. Performance Enhancement/
Facilitation Anxiety
43 If I feel slightly nervous when taking a test, Performance Enhancement/
I know I will do well. Facilitation Anxiety
30 I earn a perfect score on each test I take. Lie
35 I do well on all tests. Lie
10 I forget the material I know when taking a test. Cognitive Obstruction/
Inattention
17 My mind wanders when I am taking a test. Cognitive Obstruction/
Inattention

Sarason, Davidson, Lighthall, Waite, & Ruebush, 1960), and Wide Range Achievement
Test–Third Edition (WRAT-3; Wilkinson, 1993a). We hypothesized that (a) the TAICA
debilitating test anxiety scores would correlate most highly with the TAI and TASC scores,
(b) the TAICA subscale scores would correlate more highly with the scores on a similar
than a dissimilar dimension on the RCMAS, (c) the TAICA debilitating test anxiety scores
would correlate more highly with the TAI and TASC scores than would the BASC-SRP
Depression scores, and (d) the TAICA debilitating test anxiety scores would have negligi-
ble to weak correlations with the RIST and WRAT-3 scores.

Method

Participants
Participants were 206 children and adolescents (100 male students and 106 female stu-
dents). The individuals ranged in age from 9 to 19 years (M = 13.6 years, SD = 2.46) and
were in Grades 4 through 12 (M = 8.13, SD = 2.48). The sample included Asian American
(n = 5, 2.4%), African American (n = 2, 1.0%), Hispanic (n = 7, 3.4%), Native American
(n = 3, 1.5%), White (n = 176, 85.4%), and other (n = 13, 6.3%) students. These students
were recruited through word of mouth, flyers, and newspaper ads from rural, suburban, and
urban areas in a midwestern state. A small percentage of the sample (2.9%) received special
education, and 10.7% of the sample participated in free or reduced lunch programs.
 Lowe et al. / Test Anxiety Inventory for Children and Adolescents 221

Instruments

Behavior Assessment System for Children–Self-Report of Personality. The BASC-SRP is

a comprehensive measure used to assess children’s and adolescents’ perceptions and feel-
ings about school, home, peers, and their own behavior (Kamphaus & Frick, 2002). The
BASC-SRP has 10 clinical scales, 4 adaptive scales, and 4 composite scales. For the pur-
pose of this study, only the BASC-SRP Depression scale was used. Respondents rate their
responses on the BASC-SRP using a true–false format (Reynolds & Kamphaus, 1992b).
Reynolds and Kamphaus (1992b) reported coefficient alphas for the majority of BASC-
SRP clinical scale scores of .70 and above, with only two reliability estimates falling below
this value. Temporal stability coefficients (4-week test-retest interval) for the BASC-SRP
clinical scale scores ranged from .57 to .85, with the majority of the test-retest reliability
estimates being more than .70. Evidence supporting the construct validity of the BASC-
SRP scores has been found through factor analytic studies and convergent and discriminant
validity studies (see Reynolds & Kamphaus, 1992b).

Revised Children’s Manifest Anxiety Scale. The RCMAS is a 37-item self-report

measure of chronic, manifest anxiety. The RCMAS consists of a Total Anxiety scale,
three anxiety subscales (Physiological Anxiety, Social Concerns/Concentration, and Worry/
Oversensitivity), and a Lie scale. Respondents rate their responses on a dichotomous scale
using a yes–no format (Reynolds & Richmond, 1985).
Reynolds and Richmond (1985) reported an internal consistency reliability estimate of
.83 for the Total Anxiety scale scores. Coefficient alphas for the three anxiety subscale
scores and Lie scale scores ranged from the .60s to the .80s. Pela and Reynolds (1982)
reported temporal stability coefficients (3-week test-retest interval) of .98 for the Total
Anxiety scale scores and .94 for the Lie scale scores for a group of Nigerian children.
Results of factor analyses and correlational analyses comparing the RCMAS scores to
scores of measures external to the test provide support for the construct validity of the
RCMAS scores (Reynolds & Richmond, 1985).

Reynolds Intellectual Screening Test. The RIST is a brief screening instrument used to
assess general intelligence. The RIST consists of a verbal subtest (Guess What) that pro-
vides a measure of crystallized intelligence and a nonverbal subtest (Odd Item Out) that
provides a measure of fluid intelligence. The scores from these two subtests are summed
and converted to a standard score, the RIST Index score. The RIST Index score provides an
overall estimate of an individual’s general intelligence (Reynolds & Kamphaus, 2002). For
the purpose of this study, only the RIST Index score was used.
Reynolds and Kamphaus (2002) reported a median coefficient alpha of .95 and a temporal
stability coefficient of .84 corrected for restriction of range for the RIST Index scores. Results
from factor analytic studies and criterion-related and discriminant validity studies provide
support for the construct validity of the RIST scores (see Reynolds & Kamphaus, 2002).

Test Anxiety Inventory. The TAI is a 20-item self-report measure of test anxiety for
adolescents and adults. Respondents rate their responses on a 4-point Likert-type scale
ranging from 1 (almost never) to 4 (almost always; Spielberger, 1980b).
222 Journal of Psychoeducational Assessment

Spielberger (1980b) reported coefficient alphas of .92 or higher for the TAI Total scores.
Temporal stability coefficients (2-4 week test-retest interval) of .80 to .81 were found for
the TAI Total scores. Support for the construct validity of the TAI scores has been reported
via factor analyses and correlational analyses comparing the TAI scores to scores external
to the test (see Spielberger, 1980b).

Test Anxiety Inventory for Children and Adolescents. The TAICA is a 45-item self-report
measure. Respondents rate their responses on a 5-point Likert-type scale ranging from 1
(never true about me) to 5 (always true about me; Lowe et al., 2004).
The items developed for the TAICA were based on a review of the relevant literature,
including the different dimensions (i.e., behavioral, physiological hyperarousal, facilitation
anxiety, cognitive obstruction, social derogation, and worry) purported to reflect the test
anxiety construct, the clinical and school experience of the investigators, and the researchers’
experience in the development of measures used to assess anxiety. The work of Rafferty,
Smith, and Ptacek (1997) also provided guidance in the development of the TAICA items.
According to Rafferty and colleagues, different symptoms of test anxiety may be mani-
fested at different times in the assessment period (before, during, or after a test) or the
levels of the different symptoms of test anxiety may vary across the assessment period.
Once the 132 items were developed, the items were reviewed by two measurement experts
for their clarity, readability, and congruity with the test anxiety construct. Some items
reviewed were rewritten at the suggestion of the measurement experts because they
included two statements or ideas in one item or a qualifier. Sixteen items were deleted
because they were redundant and did not add to the quality of the measure. These steps
were taken before piloting the measure. The piloted version of the TAICA consisted of 116
items. The measure was then administered to children and adolescents. Once the data were
collected, an exploratory factor analysis was performed using the method of principal axis
factoring and rotating the factors via the promax procedure. Results of the exploratory fac-
tor analysis yielded a six-factor solution consisting of 45 items. The six factors were named
Cognitive Obstruction/Inattention, Physiological Hyperarousal, Social Humiliation, Worry,
Performance Enhancement/Facilitation Anxiety, and Lie. A principal-components analysis
was also conducted on the test anxiety factors, and the results suggested the presence of a
higher order or general debilitating test anxiety factor. This factor was named the Total Test
Anxiety factor. Coefficient alphas were computed for these factor scores. The reliability
estimates ranged from .77 to .90 (Lowe et al., 2004).

Test Anxiety Scale for Children. The TASC is a 30-item self-report measure used to
assess test anxiety in elementary school-age children. Children use a true–false format to
respond to the TASC statements (Sarason et al., 1960).
Sarason et al. (1960) reported internal consistency reliability estimates of .88 to .90 and
a temporal stability coefficient of .75 for the TASC scores. Evidence supporting the con-
struct validity of the TASC scores has been found via convergent and discriminant validity
studies (Sarason et al., 1960).

Wide Range Achievement Test–Third Edition. The WRAT-3 is a brief measure of acade-
mic achievement. The WRAT-3 consists of three subtests: reading, spelling, and arithmetic.
For the purpose of this study, only the reading and arithmetic subtests were used.
 Lowe et al. / Test Anxiety Inventory for Children and Adolescents 223

Wilkinson (1993b) reported median coefficient alphas of .82 to .95 and temporal stabil-
ity coefficients of .91 or higher for the WRAT-3 subtest scores. Results from discriminant
function analyses and correlational analyses provide support for the construct validity of
the WRAT-3 scores (Wilkinson, 1993b).

Procedures
Participants completed a packet of measures in a small group setting or on an individual
basis during two testing sessions in a clinic. The two testing sessions were approximately
7 to 21 days apart (M = 12.29 days, SD = 5.12). All tests were counterbalanced to prevent
an order effect. Participants completed the TAICA, BASC-SRP, RCMAS, RIST, and TAI
or TASC in the first testing session and the TAICA and WRAT-3 in the second testing ses-
sion. The TAI was administered to adolescents in Grades 9-12, whereas the TASC was
administered to children in Grades 4 to 6. Children ages 9 to 11 were administered the
BASC-SRP Child form, and adolescents ages 12 to 18 were administered the BASC-SRP
Adolescent form. The number of children and adolescents who completed each measure is
shown in Table 4. Test administrators followed standardized procedures for the administra-
tion of these measures according to the instructions provided in the test manuals or printed
on the instruments. The research project was conducted over an 18-month period.

Results

Sources of Validity
Evidence based on internal structure. Coefficient alphas and the 95% confidence
intervals (see Fan & Thompson, 2001; Feldt, 1990) for the TAICA scores for the first and
second testing sessions are presented in Table 2. Coefficient alphas for the TAICA scores
ranged from .8088 to .9410 and from .8600 to .9482 for the first and second testing ses-
sions, respectively. These reliability estimates are in the strong to very strong range.1
Examination of the confidence intervals around each reliability estimate indicates that
the plausible reliability estimates given the current data set are also in the strong to very
strong range.
Temporal stability coefficients for the TAICA scores were computed over a 7- to 21-day
test-retest period. The test-retest reliability estimates are shown in Table 3 along with the
means, standard deviations, and ranges of scores for the first and second testing sessions.
Temporal stability coefficients of .81 to .90 for the TAICA scores were found, suggesting
that the TAICA scores have strong to very strong test-retest reliability over a 1 to 3 week
period. Intraclass correlation coefficients were also computed for the TAICA scores (see
Table 3) and were found to be similar in magnitude to the temporal stability coefficients
computed using the Pearson r statistic. Paired-samples t tests and effect sizes (d) were cal-
culated between the means of the TAICA scale and subscale scores during the first and sec-
ond testing sessions. All means decreased from the first to the second testing session. All t
tests were significant, but the effect sizes were negligible to small. These results suggest that
the TAICA scores are sensitive to change, leading to finer discrimination in measurement.
224 Journal of Psychoeducational Assessment

Table 2
Coefficients Alphas and 95% Confidence Intervals (CIs) for the
First and Second Testing Sessions for the Test Anxiety Inventory
for Children and Adolescents Scores
First Session Second Session

Scale or Subscale α CI CI Width α CI CI Width

Cognitive Obstruction/Inattention .8864 .8631-.9076 .0445 .9071 .8881-.9244 .0363

Physiological Hyperarousal .8674 .8402-.8921 .0519 .8988 .8781-.9177 .0396
Social Humiliation .8909 .8685-.9112 .0427 .9214 .9053-.9360 .0307
Worry .8651 .8374-.8902 .0528 .8733 .8473-.8969 .0496
Performance Enhancement/ .8455 .8138-.8743 .0605 .9004 .8780-.9190 .0410
Facilitation Anxiety
Lie .8088 .7696-.8444 .0748 .8600 .8313-.8861 .0548
Total Test Anxiety .9410 .9289-.9520 .0231 .9482 .9376-.9579 .0203

Note: CIs were computed on the basis of procedures described in Fan and Thompson (2001) and Feldt (1990).

Table 3
Means, Standard Deviations, and Ranges for the First and Second Testing Sessions
and the Effect Sizes, Temporal Stability Coefficients, and Intraclass Correlation
Coefficients (ICCs) for the Test Anxiety Inventory for Children and Adolescents
Scores for the Child and Adolescent Sample
First Session Second Session

Scale or Subscale M SD Range M SD Range d r ICC

Cognitive Obstruction/Inattention 16.57* 6.00 7-35 16.01* 5.96 7-35 .17 .85 .89
Physiological Hyperarousal 11.20* 4.96 7-35 10.82* 4.79 7-35 .16 .88 .87
Social Humiliation 17.55* 7.61 10-43 16.93* 7.75 10-43 .15 .86 .89
Worry 19.31** 6.48 7-35 18.31** 6.21 7-35 .29 .85 .87
Performance Enhancement/ 15.77** 5.55 7-31 15.05** 5.89 7-30 .20 .81 .84
Facilitation Anxiety
Lie 18.67* 5.51 7-30 18.08* 5.82 7-30 .18 .83 .81
Total Test Anxiety 64.61** 20.43 31-131 62.07** 20.05 31-131 .28 .90 .94

*p ≤ .05. **p ≤ .01.

Evidence based on external relations. Validity coefficients between the TAICA scores
and scores external to the test were computed (see Table 4). Examination of the validity
coefficients revealed moderate to strong validity coefficients (rs = .46 to .77) between the
TAICA debilitating test anxiety scale and subscale scores and the TAI and TASC scores.
The TAI and TASC scores accounted for 59.29% and 43.56%, respectively, of the variance
in the TAICA Total Test Anxiety scores. Moderate correlation coefficients of .41 to .61
were reported between the majority of the TAICA debilitating test anxiety scale and sub-
scale scores and the RCMAS anxiety scale and subscale scores, with only three correlation
 Lowe et al. / Test Anxiety Inventory for Children and Adolescents 225

Table 4
Means and Standard Deviations of Scores External to the Test and Validity
Coefficients Comparing the TAICA Scores With the TAI, TASC, BASC-SRP,
RCMAS, RIST, and WRAT-3 Scores
TAICA

Scale or Subscale N M SD 1 2 3 4 5 6 7

TAI 98 45.81f 11.16 .77 .65 .58 .57 .68 .04 –.45
TASC 61 8.23b 7.15 .66 .58 .46 .66 .54 .18 –.21
RCMAS
Total Anxiety 206 48.35e 12.16 .61 .50 .45 .49 .55 .08 –.25
Physiological Anxiety 206 9.51c 3.22 .45 .44 .37 .30 .39 .01 –.21
Social Concerns 206 9.19c 2.98 .57 .48 .41 .50 .47 .04 –.30
Worry/Oversensitivity 206 9.08c 3.28 .58 .50 .37 .43 .59 .10 –.22
Lie 206 10.04c 2.85 –.10 –.15 –.07 –.04 .07 .10 .11
BASC-SRP
Adolescent Depression 151 47.93e 8.36 .44 .35 .30 .44 .30 –.08 –.21
Child Depression 54 48.69e 7.66 .45 .39 .35 .47 .35 –.07 –.11
RIST Index 206 108.96a 11.20 –.31 –.34 –.19 –.19 –.28 .09 .38
WRAT–3
Reading 206 109.04d 12.98 –.11 –.17 –.04 –.04 –.13 .02 .23
Math 206 106.52d 14.01 –.16 –.25 –.12 –.06 –.11 .22 .35

Note: Correlations between the BASC-SRP Depression scores and TAICA scores for children (9-11) and ado-
lescents (12-18) are reported separately. 1 = Total Test Anxiety; 2 = Cognitive Obstruction/Inattention; 3 =
Physiological Hyperarousal; 4 = Social Humiliation; 5 = Worry; 6 = Performance Enhancement/Facilitation
Anxiety; 7 = Lie. TAICA = Test Anxiety Inventory for Children and Adolescents; TAI = Test Anxiety Inventory;
TASC = Test Anxiety Scale for Children; BASC-SRP = Behavior Assessment System for Children–Self-Report
of Personality; RCMAS = Revised Children’s Manifest Anxiety Scale; RIST = Reynolds Intellectual Screening
Test; WRAT-3 = Wide Range Achievement Test–Third Edition.
a. Deviation IQ scores based on age.
b. Raw scores.
c. Scaled scores based on age.
d. Standard scores based on age.
e. T scores based on age.
f. T scores based on grade and gender.

coefficients falling below .40. It is interesting to note that three of the four TAICA debilitating
test anxiety subscale scores (Cognitive Obstruction/Inattention, Physiological Hyperarousal,
and Worry) correlated most highly with the TAI and TASC scores and scores of a similar dimen-
sion on the RCMAS. Weak to moderate correlation coefficients (rs = .30 to .47) were found
between the TAICA debilitating test anxiety scale and subscale scores and the BASC-SRP
Depression scale scores. However, results from Hotelling’s T tests comparing the magnitude of
the correlation coefficients between the scores of the depression and test anxiety scales and sub-
scales indicated that only the correlation coefficients between the scores of the BASC-SRP
Depression scale (adolescent form) and the TAICA Total Test Anxiety scale, Cognitive
Obstruction/Inattention subscale, and Worry subscale scores were statistically significantly
smaller in magnitude than the correlation coefficients between the scores of these same TAICA
scales and subscales and the TAI scores. In contrast, negligible to weak validity coefficients
226 Journal of Psychoeducational Assessment

(rs = –.34 to –.04) were found between the TAICA debilitating test anxiety scale and sub-
scale scores and the RIST, WRAT-3 Reading, and WRAT-3 Arithmetic scores. Negligible
to very weak correlation coefficients (rs = –.08 to .18) were noted between the TAICA
Performance Enhancement/Facilitation Anxiety scores and scores from all measures used
in the study, with the exception of the WRAT-3 Math scores. A weak correlation coefficient
of .22, similar to Hembree’s (1988) findings (M = –.29) between scores of facilitating test
anxiety and academic achievement, was found between the WRAT-3 Math scores and the
TAICA Performance Enhancement/Facilitation Anxiety scores. Correlation coefficients
between the TAICA Lie scale scores and scores of other measures ranged from –.45 to .38,
with a weak correlation of .11 reported with the RCMAS Lie scale scores. These results are
contrary to what was expected and may have been because of the small number of items on
the TAICA Lie scale and a positively skewed distribution. However, overall the majority of
the findings provide support for the construct validity of the TAICA scores.

Discussion
Results from the current study indicate that the TAICA scores have strong to very strong
internal consistency reliability and temporal stability. In addition, evidence supporting the
construct validity of the majority of the TAICA scores was found. Moderate to strong cor-
relation coefficients were expected and found between the TAICA debilitating test anxiety
scale and subscale scores and the TAI and TASC scores. These three instruments were
designed to measure the same construct, test anxiety. The pattern of mostly moderate cor-
relations between the scores of these three measures is consistent with the expectation that
a new measure will share a moderate amount of variance with other instruments purported
to measure the same construct (Reynolds, Richmond, & Lowe, 2003). The amount of
shared variance between the TAICA scores and the scores of the two other instruments in
the current study ranged from 21.16% to 59.29%. Moderate to strong correlation coeffi-
cients have also been reported between scores of other test anxiety measures, including
moderate to strong correlation coefficients between the TAI scores and the Test Anxiety
Scale (Sarason, 1978) scores and between the TAI scores and the Worry and Emotionality
Questionnaire (Liebert & Morris, 1967) scores (see Spielberger, 1980b). The TAICA is
appealing because it is a new measure (most test anxiety measures are more than 20 years
old; Lowe et al., 2004) with new items and norms; it is multidimensional in nature; it
defines the test anxiety construct more broadly; and its scores appear to have good psy-
chometric properties.
Moderate correlation coefficients were also found between the majority of the TAICA
debilitating test anxiety scale and subscale scores and the RCMAS scores. Moderate cor-
relation coefficients were expected between the TAICA scores and RCMAS scores as test
anxiety (Spielberger, 1972) and chronic manifest anxiety (Reynolds, 1985) are viewed as
forms of trait anxiety. Spielberger (1972) defined test anxiety as a situation-specific form
of trait anxiety. Trait anxiety is a stable personality characteristic, whereas state anxiety is
a transitory emotional state determined by the interaction of one’s trait with the present sit-
uation. Thus, a test-anxious student will experience excessive worry and physiological
 Lowe et al. / Test Anxiety Inventory for Children and Adolescents 227

arousal across evaluative situations (trait anxiety), and the level of the student’s anxiety will
vary as a function of the evaluative threat perceived in a specific situation (state anxiety;
Spielberger & Vagg, 1995).
Weak to moderate validity coefficients were noted between the debilitating anxiety scale
and subscale scores and the BASC-SRP Depression scale scores. Weak to moderate valid-
ity coefficients were expected as anxiety and depression are related constructs (Reynolds,
2001) and tend to co-occur (American Psychiatric Association, 2000). Anxiety and depres-
sion are both markers of a higher order factor called Negative Affectivity, and these con-
structs along with other constructs such as self-concept and aggression are related to each
other (Watson & Clark, 1984).
Negligible to weak correlation coefficients were reported between the TAICA debili-
tating test anxiety scores and scores of intelligence and academic achievement tests in the
current study. Negligible to weak relationships between debilitating test anxiety and intel-
ligence and academic achievement tests have generally been reported in the literature
(Hembree, 1988). Hembree conducted a meta-analysis of 61 studies and found a mean cor-
relation coefficient of –.23 between scores of measures of debilitating test anxiety and
scores of intelligence tests. Hembree also conducted a meta-analysis of 44 studies and
reported a mean correlation coefficient of –.29 between scores of debilitating test anxiety
measures and scores of academic achievement tests. These findings are similar to the
results reported in the current study and provide support for the discriminant validity of the
TAICA debilitating test anxiety scores.
According to Hembree (1988), facilitating test anxiety has been much less studied than
debilitating test anxiety. Hembree conducted a meta-analysis of 23 studies that included
2,624 students in Grades 9 through 12 and college and found a mean validity coefficient of
.29 between facilitating test anxiety scores and academic achievement test scores. This
correlation coefficient is similar to the correlation coefficient found in the present study
between the Performance Enhancement/Facilitation Anxiety scale scores and the WRAT-3
Arithmetic scores. In contrast, a negligible validity coefficient was found between the
Performance Enhancement/Facilitation Anxiety scale scores and the WRAT-3 Reading
scores in the present study. It is unclear why a negligible validity coefficient was found. The
WRAT-3 Reading scores measure basic skills in reading such as letter and word recogni-
tion (Wilkinson, 1993b). The studies reported in the literature included samples of older
students in Grades 9 through 12 and college (see Hembree, 1988) and were more likely to
assess students’ reading comprehension skills than letter and word recognition skills. Thus,
the difference in type of reading tests administered may have accounted for the difference
in the validity coefficients reported in the current study and in the literature.
Several limitations are associated with the present study. The proportion of the sample
representing the different racial and ethnic groups from minority backgrounds and students
on free or reduced lunch programs was smaller in comparison to their actual representation
in the U.S. elementary and secondary student population. Of the 48 million students who
attend U.S. elementary and secondary schools, more than 18 million (38.65% of the stu-
dent population) are of minority status and more than 17 million (35.71% of the student
population) participate in free or reduced lunch programs (National Center for Education
Statistics, 2003). Future studies should include a more representative sample of minority
228 Journal of Psychoeducational Assessment

and economically disadvantaged students. Students from different regions of the United
States rather than one state should also be included in future studies.
In addition to studies conducted with samples of students that mirror the U.S. elemen-
tary and secondary student population, future studies are needed to further analyze the con-
struct validity of the TAICA scores. Future studies need to be conducted to evaluate the
internal structure of the TAICA (e.g., using item response theory) and external relations
among constructs (e.g., examining group differences). Clearly, future research with the
TAICA exists.
Although additional research needs to be conducted with the TAICA, the TAICA does
appear to be a promising new measure of test anxiety. With the continued increase in test-
ing required in U.S. schools, the number of students with test anxiety is expected to rise
(Wren & Benson, 2004). Therefore, comprehensive measures of test anxiety of a superior
quality will be needed.

Note
1. We recognize there is no standard rule of thumb for descriptors for r values. We used the following
descriptors: negligible to very weak, .00 to .20; weak, .20 to .40; moderate, .40 to .70; strong, .70 to .90; and
very strong, .90 to 1.00.

References
Alpert, R., & Haber, R. N. (1960). Anxiety in academic achievement situations. Journal of Abnormal and Social
Psychology, 61, 207-215.
American Psychiatric Association. (2000). The diagnostic and statistical manual of mental disorders (4th ed.;
text rev.). Washington, DC: Author.
Beidel, D. C., & Turner, S. M. (1988). Comorbidity of test anxiety and other anxiety disorders in children.
Journal of Abnormal Child Psychology, 16, 275-287.
Benjamin, M., McKeachie, W. J., Lin, Y. G., & Holinger, D. P. (1981). Test anxiety: Deficits in information pro-
cessing. Journal of Educational Psychology, 73, 816-824.
Benson, J. (1998). Developing a strong program of construct validation: A test anxiety example. Educational
Measurement: Issues and Practice, 17(1), 10-22.
Berger, V. F., Munz, D. C., Smouse, A. D., & Angelino, H. (1969). The effects of item difficulty sequencing and
anxiety reaction type on aptitude test performance. Journal of Psychology: Interdisciplinary and Applied,
71, 253-258.
Carver, C. S., & Scheier, M. F. (1991). A control-process perspective on anxiety. In R. Schwarzer & R. A.
Wicklund (Eds.), Anxiety and self focused attention (pp. 3-8). London: Harwood.
Casbarro, J. (2005). Test anxiety and what you can do about it: A practical guide for teachers, parents, and kids.
Port Chester, NY: Dude.
Cassady, J. C. (2004). The influence of cognitive test anxiety across the learning-testing cycle. Learning and
Instruction, 14, 569-592.
Chapell, M. S., Blanding, Z. B., Silverstein, M. E., Takahashi, M., Newman, B., Gubi, A. et al. (2005). Test anx-
iety and academic performance in undergraduate and graduate students. Journal of Educational Psychology,
97, 268-274.
Covington, M. V. (1992). Making the grade. Cambridge, UK: Cambridge University Press.
Engel, G. L. (1977, April 8). The need for a new medical model: A challenge for biomedicine. Science, 196,
129-136.
 Lowe et al. / Test Anxiety Inventory for Children and Adolescents 229

Everson, H. T., Millsap, R. E., & Rodriguez, C. M. (1991). Isolating gender differences in test anxiety: A con-
firmatory factor analysis of the Test Anxiety Inventory. Educational and Psychological Measurement, 51,
243-251.
Fan, X., & Thompson, B. (2001). Confidence intervals about score reliability coefficients, please: An EPM
guidelines editorial. Educational and Psychological Measurement, 61, 517-531.
Feldt, L. S. (1990). The sampling theory for the intraclass reliability coefficient. Applied Measurement in
Education, 3, 361-367.
Friedman, I. A., & Bendas-Jacob, O. (1997). Measuring perceived test anxiety in adolescents: A self-report
scale. Educational and Psychological Measurement, 57, 1035-1045.
Hancock, D. R. (2001). Effects of test anxiety and evaluative threat on students’ achievement and motivation.
Journal of Educational Research, 94, 284-290.
Hembree, R. (1988). Correlates, causes, effects, and treatment of test anxiety. Review of Educational Research,
58, 47-77.
Joiner, T. E., Steer, R. A., Beck, A. T., Schmidt, N. B., Rudd, M. D., & Catanzaro, S. J. (1999). Physiological
hyperarousal: Construct validity of a central aspect of the tripartite model of depression and anxiety. Journal
of Abnormal Psychology, 108, 290-298.
Kamphaus, R. W., & Frick, P. J. (2002). Clinical assessment of child and adolescent personality and behavior
(2nd ed.). Boston: Allyn & Bacon.
Kamphaus, R. W., & Reynolds, C. R. (2002). Reynolds Intellectual Screening Test. Odessa, FL: Psychological
Assessment Resources.
Kirkland, K., & Hollandsworth, J. (1980). Effective test taking: Skills-acquisition versus anxiety-reduction
techniques. Journal of Counseling and Clinical Psychology, 48, 431-439.
Liebert, R. M., & Morris, L. W. (1967). Cognitive and emotional components of test anxiety: A distinction and
some initial data. Psychological Reports, 20, 975-978.
Lowe, P. A., Lee, S. W., & DeRuyck, K. A. (2004). An exploratory factor analysis of a new comprehensive
measure of test anxiety: The Test Anxiety Inventory for Children and Adolescents (TAICA). Unpublished
manuscript.
McDonald, A. S. (2001). The prevalence and effects of test anxiety in school children. Educational Psychology,
21, 89-101.
McIlroy, D., Bunting, B., & Adamson, G. (2000). An evaluation of the factor structure and predictive utility of
a test anxiety scale with reference to students’ past performance and personality indices. British Journal of
Educational Psychology, 70, 17-32.
Methia, R. A. (2004). Help your child overcome test anxiety & achieve higher test scores. College Station, TX:
Virtual Bookwork.com Publishing.
Morris, L. W., & Liebert, R. M. (1969). Effects of anxiety on timed and untimed intelligence tests: Another
look. Journal of Consulting and Clinical Psychology, 33, 240-244.
Mueller-Hanson, R. (2002). Impression management strategy and faking behavior in the self-report measure-
ment of personality. Dissertation Abstracts International, 63(5), 2634B.
National Center for Education Statistics. (2003). Overview of public elementary and secondary schools
and districts: School year 2001-2002. Washington, DC: Institute of Education Sources, U.S. Department of
Education.
Nielsen, I. L., & Moore, K. A. (2003). Psychometric data on the Mathematics Self-Efficacy Scale. Educational
and Psychological Measurement, 63, 128-138.
Pela, O. A., & Reynolds, C. R. (1982). Cross-cultural application of the Revised Children’s Manifest Anxiety Scale:
Normative and reliability data of Nigerian primary school children. Psychological Reports, 51, 1135-1138.
Piedmont, R. L., McCrae, R. R., Riemann, R., & Angleitner, A. (2000). On the validity of validity scales:
Evidence from self-reports and observer ratings in volunteer samples. Journal of Personality and Social
Psychology, 78, 582-593.
Rafferty, B. D., Smith, R. E., & Ptacek, J. T. (1997). Facilitating and debilitating trait anxiety, situational anx-
iety, and coping with an anticipated stressor: A process analysis. Journal of Personality and Social
Psychology, 72, 892-906.
Reynolds, C. R. (1985). Multitrait validation of the Revised Children’s Manifest Anxiety Scale for children of
high intelligence. Psychological Reports, 56, 402.
230 Journal of Psychoeducational Assessment

Reynolds, C. R. (2001). Professional manual for the Clinical Assessment Scales for the Elderly. Odessa, FL:
Psychological Assessment Resources.
Reynolds, C. R., & Kamphaus, R. W. (1992a). Behavior Assessment System for Children–Self-Report of
Personality. Circle Pines, MN: American Guidance Services.
Reynolds, C. R., & Kamphaus, R. W. (1992b). Behavior Assessment System for Children–Self-Report of
Personality manual. Circle Pines, MN: American Guidance Services.
Reynolds, C. R., & Kamphaus, R. W. (2002). Reynolds Intellectual Screening Test manual. Odessa, FL:
Psychological Assessment Resources
Reynolds, C. R., & Richmond, B. O. (1978). What I think and feel: A revised measure of children’s manifest
anxiety. Journal of Abnormal Child Psychology, 6, 271-280.
Reynolds, C. R., & Richmond, B. O. (1985). Revised Children’s Manifest Anxiety Scale manual. Los Angeles:
Western Psychological Services.
Reynolds, C. R., Richmond, B. O., & Lowe, P. A. (2003). Adult Manifest Anxiety Scales manual. Los Angeles:
Western Psychological Services.
Sarason, I. G. (1978). The Test Anxiety Scale: Concept and research. In C. D. Spielberger & I. G. Sarason
(Eds.), Stress and anxiety (Vol. 5, pp. 193-216). Washington, DC: Hemisphere.
Sarason, S. B., Davidson, K. S., Lighthall, F. F., Waite, R. R., & Ruebush, B. K. (1960). Anxiety in elementary
school children. New York: Wiley.
Schwartz, G. E. (1982). Testing the biopsychosocial model: The ultimate challenge facing behavioral medicine?
Journal of Consulting and Clinical Psychology, 50, 1040-1053.
Spence, J. T., & Spence, K. W. (1966). The motivational components of manifest anxiety: Drive and drive stim-
uli. In C. D. Spielberger (Ed.), Anxiety and behavior (pp. 291-326). New York: Academic Press.
Spielberger, C. D. (1972). Conceptual and methodological issues in anxiety research. In C. D. Spielberger (Ed.),
Anxiety (Vol. 2, pp. 481-493). New York: Academic Press.
Spielberger, C. D. (1980a). Test Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press.
Spielberger, C. D. (1980b). Test Anxiety Inventory manual. Palo Alto, CA: Consulting Psychologists Press.
Spielberger, C. D., & Vagg, P. R. (1995). Test anxiety: A transactional process model. In C. D. Spielberger &
P. R. Vagg (Eds.), Test anxiety: Theory, assessment, and treatment (pp. 3-14). Washington, DC: Taylor & Francis.
Sub, A., & Prabha, C. (2003). Academic performance in relation to perfectionism, test procrastination and test
anxiety of high school children. Psychological Studies, 48, 7-81.
Swanson, S., & Howell, C. (1996). Test anxiety in adolescents with learning disabilities and behavior disorders.
Exceptional Children, 62, 389-397.
Taylor, J. A. (1956). Drive theory and manifest anxiety. Psychological Bulletin, 53, 303-328.
Tobias, S. (1979). Anxiety research in educational psychology. Journal of Educational Psychology, 71, 573-582.
Watson, D., & Clark, L. A. (1984). Negative affectivity: The disposition to experience aversive emotional states.
Psychological Bulletin, 96, 465-490.
Wilkinson, G. S. (1993a). Wide Range Achievement Test–Third edition. Wilmington, DE: Wide Range.
Wilkinson, G. S. (1993b). Wide Range Achievement Test–Third edition manual. Wilmington, DE: Wide Range.
Wine, J. (1971). Test anxiety and direction of attention. Psychological Bulletin, 76, 92-104.
Wren, D. G., & Benson, J. (2004). Measuring test anxiety in children: Scale development and internal construct
validation. Anxiety, Stress, and Coping: An International Journal, 17, 227-240.
Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit-formation.
Journal of Comparative and Neurological Psychology, 18, 459-482.
Zeidner, M. (1998). Test anxiety: The state of the art. New York: Plenum.