Measurement in Physical Education and Exercise Science

ISSN: 1091-367X (Print) 1532-7841 (Online) Journal homepage: https://shapeamerica.tandfonline.com/loi/

hmpe20

The Sport Imagery Questionnaire for Children

(SIQ-C)

C. R. Hall , K. J. Munroe-Chandler , G. J. Fishburne & N. D. Hall

To cite this article: C. R. Hall , K. J. Munroe-Chandler , G. J. Fishburne & N. D. Hall (2009)

The Sport Imagery Questionnaire for Children (SIQ-C), Measurement in Physical Education and
Exercise Science, 13:2, 93-107, DOI: 10.1080/10913670902812713

To link to this article: https://doi.org/10.1080/10913670902812713

Published online: 10 Apr 2009.

Submit your article to this journal

Article views: 1554

View related articles

Citing articles: 5 View citing articles

Full Terms & Conditions of access and use can be found at

https://shapeamerica.tandfonline.com/action/journalInformation?journalCode=hmpe20
Measurement in Physical Education and Exercise Science, 13: 93–107, 2009
Copyright © Taylor & Francis Group, LLC
ISSN: 1091-367X print / 1532-7841 online
DOI: 10.1080/10913670902812713

The Sport Imagery Questionnaire for Children (SIQ-C)

1532-7841 in Physical Education and Exercise Science
1091-367X
HMPE
Measurement Science, Vol. 13, No. 2, March 2009: pp. 1–39

C. R. Hall
HALL, MUNROE-CHANDLER, FISHBURNE, HALL
SIQ-C

School of Kinesiology, University of Western Ontario, London, Canada

K. J. Munroe-Chandler
Department of Kinesiology, University of Windsor, Windsor, Canada

G. J. Fishburne and N. D. Hall

Faculty of Education, University of Alberta, Edmonton, Canada

Athletes of all ages report using imagery extensively to enhance their sport performance. The Sport
Imagery Questionnaire (Hall, Mack, Paivio, & Hausenblas, 1998) was developed to assess cognitive and
motivational imagery used by adult athletes. No such instrument currently exists to measure the use of
imagery by young athletes. The aim of the present research was to modify the Sport Imagery Question-
naire for use with children aged 7–14 years. This was accomplished through three phases. Across these
phases, evidence was generated showing adequate factorial, convergent, and discriminant validity and
reliability of the instrument, which is termed the Sport Imagery Questionnaire—Children’s Version. In
addition, the relationships of scores on the Sport Imagery Questionnaire—Children’s Version to gender
and age were examined. While it was found that male and female athletes employed imagery to about
the same extent, there were some age group differences in the use of imagery.
Key words: athletes, children, imagery, questionnaire

INTRODUCTION

Imagery is often viewed as “the cornerstone of sport psychology interventions” (Cornelius,

2002, p. 206). It is a popular mental training tool used by athletes at all levels and advocated by
coaches (Munroe-Chandler & Hall, 2007). Imagery is a volitional experience that involves the
use of one or more of the senses to create, or recreate, a particular sporting skill or situation
(White & Hardy, 1998). Imagery has been a well-researched topic with adult athletes, especially
elite ones (cf. Hall, 2001), and previous sport literature has established that athletes can benefit
from using imagery in sports to enhance performance (Morris, Spittle, & Watt, 2005).
Much of the recent imagery research has stemmed from Paivio’s (1985) analytic framework
suggesting that imagery has cognitive and motivational functions that operate on either a
specific or a general level. Thus, the cognitive general (CG) function entails imaging strategies,

Correspondence should be sent to Dr. C. R. Hall, School of Kinesiology, University of Western Ontario, London,
N6A 3K7, Canada. E-mail: [email protected]
94 HALL ET AL.

game plans, or routines (for example, a fast break in basketball), while the cognitive specific (CS)
function involves imaging specific sport skills (for example, a tennis serve). The motivational
general (MG) function of imagery includes imaging physiological arousal levels and emotions
(for example, feeling calm and relaxed in front of a crowd), and the motivational specific (MS)
function of imagery includes imaging individual goals (for example, winning a medal). More
recently, Hall et al. (1998) divided the MG function into an MG-arousal (MG-A) function,
encompassing imagery associated with arousal and stress, and an MG-mastery (MG-M) function,
representing imagery associated with being mentally tough, in control, and self-confident.
Hall and colleagues (1998) used Paivio’s framework as the basis for developing the Sport
Imagery Questionnaire (SIQ), which assesses the five functions of imagery. The factorial
validity of the SIQ has been supported (Hall et al., 1998; Hall, Stevens, & Paivio, 2005).
Various indices have been used to assess the goodness-of-fit of the model, and a satisfactory fit
has been found (Non-Normed Fit Index [NNFI] = .91, Comparative Fit Index [CFI] = .90, and
Root Mean Square Error of Approximation [RMSEA] = .06). The reliability for the SIQ has
also been demonstrated as acceptable with internal consistency estimates for the five functions
ranging from .70 to .88 (Hall et al., 2005).
The SIQ has been extensively employed to examine adult athletes’ use of cognitive and
motivational imagery. Researchers have consistently found that athletes of higher skill levels employ
all five functions of imagery more frequently than athletes participating at lower skill levels. For
example, Gregg and Hall (2006) examined the relationship of an objective measure of skill level, a
golf handicap, and the five functions of imagery. It was found that as the handicap decreased, there
was a corresponding increase in imagery use. In addition, Cumming and Hall (2002) found provin-
cial- and national-level athletes engaged in significantly more imagery than regional-level athletes
during the off-season. In order to determine whether imagery use changes over the course of a com-
petitive season, Munroe, Hall, Simms, and Weinberg (1998) found that varsity athletes employ imag-
ery more in the later stages of the season than in the early stages of the season. They also found that
cognitive and motivational functions of imagery are used to varying degrees in different varsity
sports, but athletes report utilizing more motivational than cognitive imagery.
Although most of the sport imagery research has been conducted with adults, there have been
a number of studies that have examined youth athletes. The most comprehensive of these studies
was conducted by Munroe-Chandler, Hall, Fishburne, and Strachan (2007), in which they
qualitatively examined young athletes’ imagery use from a developmental perspective. The par-
ticipants in the study were 111 male and female athletes competing in both team and individual
sports. They represented four different age cohorts (i.e., 7–8, 9–10, 11–12, and 13–14). Sixteen
focus groups, two for each age category and gender, were used as the method of data collection,
with each focus group consisting of 6–8 participants. All athlete groups, regardless of age or
gender, reported using imagery for both training and competition. Results also revealed that
athletes from all age cohorts used imagery for all five imagery use functions (i.e., cognitive and
motivational purposes). Given that these young athletes were between the ages of 7 and 14 years
and undoubtedly learning new skills and strategies, it is understandable that the cognitive
functions of imagery emerged from the focus groups. Due to the fact that these athletes were all
competitive, it is also reasonable that they would use imagery for motivational purposes such as
setting goals, regulating arousal, and increasing confidence.
Munroe-Chandler et al. (2007) also reported a number of gender differences in young
athletes’ imagery use. For example, male athletes did not report using imagery related to arousal
 SIQ-C 95

and anxiety (MG-A). In contrast, all female athletes reported using this function as a means of
controlling arousal and anxiety levels. In addition, only the older female cohorts reported using
MG-A imagery for relaxation and excitement. Overall, the results revealed that gender differ-
ences in young athletes’ use of imagery do exist, and further examination of these differences is
warranted.
In an attempt to investigate the effects of CG imagery on learning and performance of
game plans and strategies, Munroe-Chandler, Hall, Fishburne, and Shannon (2005) analyzed
three distinct soccer strategies (defending a direct free kick, taking a direct free kick, and
defending a corner kick) in a sample of under 13 female soccer players. Results revealed that
CG and CS imagery, as well as MG-A, significantly increased from baseline to post-imagery
intervention. However, the execution of soccer strategies was not enhanced with the introduc-
tion of a CG imagery intervention for the soccer strategies. Moreover, the findings of this
particular study indicated that (a) young athletes use imagery quite extensively and (b) their
use of imagery increased over the duration of the study, most likely as a result of the imagery
intervention.
Furthermore, Munroe-Chandler and Hall (2004) investigated the effects of MG-M imagery
on the collective efficacy of a youth female soccer team of ages 10–12 years. The soccer team
was divided into three groups—forward, mid-fielders, and defensive/goalkeeper—and modified
imagery scripts were developed for each group. Results of the 13-week imagery intervention
indicated that collective efficacy increased with the implementation of the MG-M imagery
intervention for both training and competition for forwards and midfielders, as well as an
increase of the imagery use on a daily basis.
Given that the above studies have clearly demonstrated that young athletes aged 7–14 years
use cognitive and motivational imagery, to further explore their imagery use, it would be worth-
while to have a valid and reliable assessment tool. While such a tool currently exists for adult
athletes, namely the SIQ, there is no similar instrument available to assess young athletes’ use of
imagery. One possible solution is to adapt the SIQ for children. However, the main concern in
adapting an adult measure is to ensure the modified items are appropriate, both in language and
in content, for children. Stadulis, MacCracken, Eidson, and Severance (2002) commented that
the “reliability and validity of administering the adult version (of a questionnaire) to children
would be suspect due to the child’s inability to comprehend terminology and concepts” (p. 148).
Developmental research (Brustad, 1998; Whaley, 2007) also notes that measurements should
reflect the cognitive stage of a sample, be written in a language and format appropriate for a
sample, and address concerns that are relevant to a sample. Furthermore, the instrument should
be assessed to see if a stable factor structure exists across developmental groups (Whaley, 2007).
Therefore, the purpose of the present research was to adapt the adult version of the SIQ for children.
Three experimental phases were undertaken to achieve this end.

PHASE 1

Similar to the approach used by Stadulis et al. (2002) in their adaptation of a children’s form of
the Competitive State Anxiety Inventory (CSAI; Martens, Vealey, & Burton, 1990), we first
focused on the potential rewording of items so that they were language appropriate to the
sample. We then examined the validity of the instrument using factor analytical techniques.
96 HALL ET AL.

Method

Participants. A convenient sample of participants included 428 young athletes (68% female
and 32% male) ranging in age from 7–14 years old (M = 10.88, SD = 1.87). Munroe-Chandler
et al. (2007) have found children in this age range report using both cognitive and motivational
imagery for sports. The athletes participated in 21 different sports. Participation in the sport of
hockey was reported most frequently (33%), followed by soccer (26%) and basketball (10%).
The athletes competed at a variety of competitive levels.
Measure. The SIQ (Hall et al., 1998) is employed to assess the cognitive and motivational
imagery use of adult athletes (i.e., CG, CS, MG-A, MG-M, and MS imagery). It is composed of
30 questions (six questions per imagery function). Each item is rated on a 7-point Likert-type
scale with anchors of 1 (rarely) and 7 (often) to indicate use of that function of imagery. The
SIQ has been shown to have acceptable internal consistencies with alpha coefficients from .70 to
.89 and to possess adequate factorial validity (Hall et al., 2005).
Given the sound psychometric properties of the SIQ and its extensive use in sport imagery
research, it was decided to adapt the SIQ for measuring young athletes’ imagery use rather than
develop a completely new instrument. Morton-Williams and Sykes (1984) suggested that
problems in comprehension might come from difficulties in understanding concepts rather than
from unknown vocabulary. Therefore, it is important to make concepts as clear as possible, and
the inclusion of sport-specific terms may aid in the young athletes’ comprehension. This was
accomplished with the aid of a language specialist, and the revised items were pilot tested with a
small sample of 7-year-old athletes. Examples of the revisions made were: “I image the
audience applauding my performance” was changed to “I see the audience cheering for me,” and
“I imagine the excitement associated with performing” was changed to “I imagine how exciting
it is to be in a competition.” The rating scale was also shortened from 7 to 5 points, and
descriptives were added at each point along the scale rather than just at the anchors; 1 = Not At All,
2= A Little Bit, 3=Sometimes, 4= Often, and 5 = Very Often. Justification for this reduction is
provided by research (Maurer & Price, 1998; Pajares, Hartley, & Valiante, 2001) in which it is
argued that a 5-point Likert scale is more practical and provides greater prediction of theoreti-
cally relevant external variables than the response structures that are typically advocated.
Procedure and data analysis. Upon receiving ethics clearance from the universities’
research ethics boards, athletes from various sports organizations and summer sports camps
were contacted to participate in the research. Permission and consent were obtained from the
athletes’ parents. After completing an individual assent form, athletes volunteering to be part of
the research were administered the questionnaire at a variety of locations, including their
practice and sport camp facilities (e.g., fields, arenas, pools, courts, gyms). After completing the
questionnaire, the athletes handed it back to the investigator who was on site to answer any
questions the athletes may have had. The amount of time to complete the questionnaire varied
with the age of the athletes, the older ones (i.e., 13 and 14 years) taking about 10 min while the
younger ones (i.e., 7 and 8 years) took up to 20 min.
Given that the present questionnaire was adapted from the SIQ, a confirmatory factor
analysis (CFA) was conducted in an attempt to validate the questionnaire’s factor structure and
composition. A five-factor model of imagery use—CG, CS, MG-A, MG-M, and MS imagery—
was hypothesized. The guidelines for judging the fit of models in the CFA have been the subject
 SIQ-C 97

of much discussion within the literature (e.g., Barrett, 2007; Hayduk & Glaser, 2000; Hu &
Bentler, 1999; Markland, 2007). Consequently, a combination of criteria, recognizing that these
are not “golden rules,” was employed to provide a balanced approach to testing model fit. The
c2/df (Q) ratio was used as an index of absolute model fit (Kelloway, 1998). The c2 statistic,
however, has been criticized for being overly sensitive to sample size and being ambiguous in
terms of how close the implied and observed covariance matrices must be to indicate that the
model fits the data (Kelloway, 1998). Due to these limitations, the CFI and the Goodness-of-Fit
Index (GFI) were employed, given their suitability as indicators of a global model in research
with small sample sizes (West, Finch, & Curran, 1995). The RMSEA was also considered to
assess the discrepancy between the implied and observed correlation matrices (Kelloway, 1998).
Fit indices greater than .90 for CFI and a GFI and less than .10 for RMSEA were considered
indicative of acceptable model fit.
For purposes of testing the model, the five factors were allowed to covary with one another,
items were loaded exclusively on relevant factors, and the variance for each latent factor was
fixed at one. Maximum likelihood estimation was employed to estimate the model.

Results and Discussion

It has been suggested by Short, Monsma, and Short (2004) and Short and Short (2005) that some
items on the SIQ, namely those dealing with anxiety, may be problematic because they can be
viewed as debilitative rather than facilitative. Accordingly, we examined the items on the Sport
Imagery Questionnaire—Children’s Version (SIQ-C) entailing anxiety. Two items about
imagining nervousness (“I see myself being nervous at competitions” and “When I think about
an upcoming competition, I imagine myself being nervous”) proved problematic and were
eliminated since most of the younger athletes (<9 years) gave these items a rating of 1 (i.e., do
not use imagery at all). While these items were problematic for only the younger athletes, the
goal was to have a single questionnaire that was applicable to all athletes aged 7–14 years and
not to have separate questionnaires for different age groups. In addition, before undertaking the
CFA, an examination of the distributional properties of the remaining items was conducted.
Concerns were only evident for two of the items. Both MS items concerning winning (i.e.,
“I imagine how it would feel to win a medal or trophy” and “I see myself winning a medal or
trophy”) were skewed to the high end of the rating scale and, therefore, were eliminated.
As a result of item deletions, the questionnaire now consisted of 25 items: 6 CG, 6 CS, 4 MG-A,
5 MG-M, and 4 MS. While the questionnaire was shorter than the original 30-item version, there
was still a concern about overall length, especially for the younger athletes. This concern with
length had to be balanced with having an adequate number of items to represent each factor. The
desirable minimum number of items depends to some extent on the quality of the items, but at
least three are recommended by Marsh (2007). It was felt that 6 items were probably not neces-
sary to adequately measure CS and CG imagery use. Based on the inter-item correlation matrix
for the CS items, two items were eliminated (i.e., “It’s easy for me to change how a skill looks”
and “I can correct a skill in my head”). They were highly correlated (>.70) with each other, and
both were highly correlated with the item “I can usually control how a skill looks in my head.”
Based on the inter-item correlation matrix for the CG items, two items were eliminated (i.e.,
“I see every part of a game, for example offence and defense” and “I see myself making plays
exactly how I want them to happen in competition”), since they were not that highly correlated
98 HALL ET AL.

with the other four items. It should be noted that subsequent analyses were conducted with both
the 25-item and 21-item versions of the questionnaire, and virtually the same results were found.
Only the results for the shorter version are reported, since having the shortest possible questionnaire
while maintaining an adequate number of items for each factor was believed to be optimum
given the age range of the athletes being targeted (i.e., 7–14 years).
The results of the CFA approached a reasonable fit for the hypothesized five-factor model;
Q = 3.08, CFI = .89, GFI = .89, and RMSEA = .07. Q represents an adequate fit to the data
(Marsh & Hocevar, 1985). Both CFI and GFI had values very close to .90, also indicating an
adequate fit (Bentler, 1990; Bollen, 1989). The RMSEA value of .07 suggests a reasonable error
of approximation (Browne & Cudeck, 1993). The overall fit of the structural model was corrob-
orated by the moderate to strong standardized parameter loadings (ls ranged from .55 to .78).
These values indicate each item had a meaningful contribution to its respective subscale (Hair,
Anderson, Tatham, & Black, 1998).
The alpha coefficients for the 21-item SIQ-C are given in Table 1. Three of these coefficients
(CS, MS, and MG-M) were considered acceptable as they exceeded the suggested criteria of .70
(Nunnally, 1978), while CG and MG-A were just below this criteria.
The descriptive statistics for the 21-item version of the questionnaire, now labeled the SIQ-C,
are presented in Table 1. The correlations between the five subscales of the SIQ-C ranged from
.48 to .72; similar to the subscales on the SIQ, the subscales on this instrument were moderately
correlated. A repeated-measures ANOVA indicated that the various functions of imagery were
employed to differing extents; F(4, 424) = 162.28, p < .01, h2 = .61. More specifically, using a
Bonferroni adjustment to control for Type 1 errors (p < .01), participants used MG-M, MS, and
MG-A imagery more than CS and CG imagery.
The results of the CFA supported the multidimensional factor structure of the SIQ-C from the
perspective of the framework outlined by Paivio (1985) and interpreted more broadly within the
sport imagery literature (Hall, 2001; Hall et al., 1998). Similar to the SIQ employed to measure
imagery use by adult athletes, the SIQ-C assesses young athletes’ (7–14 years of age) use of CG,
CS, MG-A, MG-M, and MS imagery in sport. While the present research provided some
evidence that the SIQ-C has acceptable psychometric properties and may be a useful instrument
for the assessment of both cognitive and motivational imagery use, further examination of the
instrument was desirable. Since the present results could be sample specific, we decided to
cross-validate them in an independent sample.

TABLE 1
Phase 1 Descriptive Statistics for the SIQ-C Subscales

Subscale Mean SD Alpha

CS 3.45 .92 .80

CG 3.22 .82 .69
MS 3.87 .86 .75
MG-A 3.55 .88 .69
MG-M 3.75 .82 .82

Note: SIQ-C scored on a 5-point scale. Higher scores

indicate more frequent use of imagery.
 SIQ-C 99

PHASE 2

The primary purpose of Phase 2 was to further assess the factor structure of the 21-item
version of the SIQ-C, again using confirmatory techniques. A secondary purpose was to
examine the influence of gender and age on the use of imagery. Age was separated into four
cohorts (7–8, 9–10, 11–12, and 13–14) based on Piaget’s stages of cognitive development
(Piaget, 1971). Given the differences in cognitive development (Piaget, 1971) as well as
previous imagery research in which age differences have been reported in young athletes
aged 7–14 years (Munroe-Chandler et al., 2007), it was hypothesized that older athletes (i.e.,
13–14 years) would indicate using more cognitive and motivational imagery than younger
athletes (i.e., 7–8 years). Results from Munroe-Chandler et al.’s (2007) study have also
indicated gender differences. More specifically, girls used imagery related to arousal regula-
tion and confidence more than their male counterparts. As such, the current study hypothe-
sized that female participants would report using more MG-A and MG-M imagery than male
participants.

Method

Participants. The participants were 628 young athletes (55% female and 45% male) repre-
senting four age cohorts: 7–8 (n = 86), 9–10 (n = 170), 11–12 (n = 168), and 13–14 (n = 204). A
wide variety of sports were represented by the sample, with 34 different sports being reported.
Participation in the sport of soccer was reported most frequently (35%), followed by hockey
(12%), basketball (10%), and swimming (9%) with the athletes competing at a variety of com-
petitive levels.
Measure. The 21-item SIQ-C from Phase 1 was employed in the present phase. The
questionnaire consists of five subscales (CG, CS, MG-A, MG-M, and MS), and participants
responded on a 5-point scale with 1 = Not At All and 5 = Very Often. The individual items
comprising each subscale are shown in Table 2.
Procedure and data analysis. Once ethics clearance was obtained from the universities’
research ethics boards, athletes from various sports organizations and summer sports camps
were contacted to participate in the research. Permission and consent were obtained from the
athletes’ parents. After completing an individual assent form, athletes volunteering to be part of
the research were administered the questionnaire. They completed the SIQ-C at a variety of
locations, including their practice and sport camp facilities, and following completion of the
questionnaire, handed it back to the investigator. The SIQ-C on average took about 15 min to
complete with the younger athletes taking slightly longer than the older athletes.
A CFA was conducted to examine the factor structure of the SIQ-C using the same proce-
dures as in Phase 1. Since the research of Munroe-Chandler et al.(2007) indicated there are some
gender differences in the use of imagery by young athletes, separate CFAs were also conducted
for males and females. A MANOVA was undertaken to determine whether gender and age
influence the use of imagery. To have sufficient participants in each age group, the same four
age categories employed by Munroe-Chandler et al. (2007) were used: 7–8, 9–10, 11–12, and
13–14 years. The five subscale scores of the SIQ-C (i.e., CG, CS, MG-A, MG-M, and MS)
served as the dependent variables.
100 HALL ET AL.

TABLE 2
Phase 2 Descriptive Statistics for the SIQ-C Subscales

Subscale
Individual Items Mean (SD) Alpha

CS 3.53 (.88) .77

I can usually control how a skill looks in my head.
When I think of doing a skill, I always see myself doing it perfectly.
Before trying a skill, I see myself doing it perfectly.
When learning something new, I see myself doing it perfectly.
CG 3.32 (.76) .62
I make up new game plans or routines in my head.
I see what I would do if my game plans or routines do not work out.
I imagine continuing with my game plan or routine even if it is not going well.
I see myself following the game plan or routine at competitions.
MS 3.84 (.83) .70
I see myself doing my very best.
I imagine other people telling me that I did a good job.
I see the audience cheering for me.
I see myself as a champion.
MG-A 3.55 (.86) .77
In my head, I imagine how calm I feel before I compete.
I imagine myself staying calm in competitions.
When I think of a competition, I imagine myself getting excited.
I imagine how exciting it is to be in a competition.
MG-M 3.79 (.75) .70
I imagine myself being confident in competition.
I see myself being mentally strong.
I see myself being focused in a tough situation.
I see myself being in control in tricky situations.
I see myself getting through tough situations with good results.

TABLE 3
Phase 2 Goodness of Fit Indices

Sample Q CFI GFI RMSEA

Entire sample 3.33 .89 .91 .06

Males only 2.13 .88 .89 .06
Females only 2.42 .90 .89 .06

Results and Discussion

Means and standard deviations for the five subscales of the SIQ-C are reported in Table 3. The
subscales were moderately correlated with values ranging from .41 to .57. All subscales had
acceptable internal consistencies with the exception of CG, each exceeding a Cronbach alpha
coefficient of .70 (Nunnally, 1978). A repeated-measures ANOVA indicated significant differences
 SIQ-C 101

in the athletes’ use of the five functions of imagery; F(4, 624) = 95.68, p < .01, h2 = .38. More
specifically, using a Bonferroni adjustment to control for Type 1 errors (p < .01), pairwise
comparisons indicated that participants used MG-M and MS imagery the most, followed by CS
and MG-A imagery, and finally, CG imagery the least.
Previous research with adult athletes has shown that the five functions of imagery are used to
different degrees, with MG-M imagery being used the most and MS imagery the least (Munroe
et al., 1998). The current study with young athletes supports the finding that the various
functions of imagery are used differentially, and more specifically, that MG-M and MS are used
the most and CG is used the least. The finding that MG-M imagery was used extensively is not
surprising given the previous adult data (Hall et al., 1998; Moritz, Hall, Martin, & Vadocz,
1996) and that MG-M imagery has the potential to increase athletes’ sport confidence (Callow,
Hardy, & Hall, 2001). This is important given that sport confidence is a major determinant of
success (Gould, Weiss, & Weinberg, 1981).
The current finding for the MS function of imagery may be largely due to the younger
athletes’ use of this function. The relationship between age and MS imagery is examined and
discussed below. Age may also help to explain the current findings for CG imagery. Given the
age range of the sample being 7–14 years, it is possible that they are more concerned with
learning the basic skills rather than putting those skills together to form a detailed game plan
or routine. This may be further explained by the participants sampled in the current study.
Many of the athletes were recruited from sport camps in which the dominant goal is skill
learning. Therefore, it is possible that sport camp athletes may use more CS imagery rather
than CG imagery.
CFA. CFA was used to assess the factor structure of the SIQ-C, with a five-factor model of
imagery use being hypothesized. Reasonable fit indices (Q = 3.33, CFI = .89, GFI = .91,
RMSEA = .06) and the moderate to strong standardized parameter loadings (ls = .42 to .77)
again indicated that the measurement model was tenable. Separate CFAs were then conducted
for males and females, and the fit indices were almost identical to those found for the entire sample
(see Table 3).
Multivariate analysis of variance. A MANOVA with gender and age group (i.e., 7–8, 9–10,
11–12, and 13–14 years) as the independent variables and the five subscales of the SIQ-C as
the dependent variables revealed a significant main effect for age group; Pillai’s Trace = .087,
F(15, 1854) = 3.69, p < .01, h2 = .03. The main effect for gender and the gender X age group
interaction failed to be significant (p > .05). The means and standard deviations for each of the
five functions of imagery for gender and age group are shown in Table 4.
Follow-up univariate tests using a Bonferroni adjustment to control for Type 1 errors (p < .0125)
indicated significant effects of age group for both MS imagery, F(3, 620) = 11.42, p < .01, h2 = .05,
and MG-A imagery, F(3, 620) = 4.25, p < .01, h2 = .02. Tukey post hoc tests (p < .05) indicated
that the two youngest age groups (i.e., 7–8 and 9–10 years) employed significantly more MS
imagery than the oldest age group (i.e., 13–14 years). Also, 9–10-year-olds used more MG-A
imagery than 13–14-year-olds.
It was hypothesized that female athletes would use more MG-A and MG-M imagery than
males and that the older age cohorts would use more cognitive and motivational imagery than
the younger athletes, but neither of these hypotheses were supported. No gender differences
were evident. Furthermore, younger athletes used more MS imagery than the older athletes. This
102 HALL ET AL.

TABLE 4
Phase 2 SIQ-C Subscale Means and Standard Deviations
for Gender and Age

Subscale Gender Age Mean SD

CS Male 7–8 3.44 .82

9–10 3.72 .75
11–12 3.60 .98
13–14 3.60 .90
Female 7–8 3.72 .96
9–10 3.55 .76
11–12 3.33 .88
13–14 3.41 .95
CG Male 7–8 3.29 .78
9–10 3.35 .79
11–12 3.29 .73
13–14 3.23 .72
Female 7–8 3.30 .85
9–10 3.43 .78
11–12 3.30 .72
13–14 3.33 .77
MS Male 7–8 3.86 .90
9–10 4.12 .64
11–12 3.86 .83
13–14 3.63 .78
Female 7–8 4.18 .82
9–10 4.03 .78
11–12 3.75 .86
13–14 3.60 .88
MG-A Male 7–8 3.26 .79
9–10 3.72 .89
11–12 3.59 .86
13–14 3.34 .98
Female 7–8 3.73 .71
9–10 3.77 .82
11–12 3.44 .84
13–14 3.53 .80
MG-M Male 7–8 3.66 .70
9–10 3.84 .75
11–12 3.93 .72
13–14 3.76 .76
Female 7–8 3.85 .78
9–10 3.84 .71
11–12 3.72 .80
13–14 3.72 .75

finding may be due to the more egocentric stage of development of the younger children
(Fishburne, 2005); that is, younger athletes focus mainly on themselves. This finding is also in
line with recent results reported by Munroe-Chandler et al. (2007). Younger athletes were found
to extensively employ MS imagery as it relates to individual goals, which is precisely what the
MS subscale of SIQ-C evaluates. With respect to the finding that MG-A imagery was used more
 SIQ-C 103

by 9–10-year-olds than 13–14-year-olds, we are currently unable to offer any explanation.

Given this finding only accounted for 2% of the variance, the meaningfulness of this effect is
questionable and could be the subject of future research.

PHASE 3

The aim of Phase 3 was to assess the convergent and discriminant validity of the SIQ-C. When
measures of constructs that theoretically should be related to each other are, in fact, observed to
be related to each other, this is referred to as convergent validity. Conversely, measures of
constructs that theoretically should not be related to each other are referred to as discriminant
validity (Trochim, 2006). In order to examine these two types of validity, correlations were run
between the subscales of the SIQ-C and measures of self-confidence and self-efficacy. In accord
with the applied model of imagery use (Martin, Moritz, & Hall, 1999) and previous research that
has shown a relationship between MG-M imagery use and self-confidence (Moritz et al., 1996)
and self-efficacy (Mills, Munroe, & Hall, 2000), it was expected MG-M would be the most
strongly correlated function of imagery with both self-confidence and self-efficacy. In contrast,
it was expected that the cognitive imagery functions (i.e., CS and CG) would show a weak
relationship with both self-confidence and self-efficacy. Again, this weak relationship is in
accord with the applied model of imagery use and findings in previous imagery research (Mills
et al., 2000; Moritz et al., 1996).

Method

Participants. The participants were 82 young soccer athletes (61% female and 21% male)
ages 8–14 years (M = 11.53, SD = 1.95). Athletes competed at a variety of levels including
house league (n = 21), intermediate (n = 25), and competitive (n = 36).
Measure. The 21-item SIQ-C from Phase 1 was employed in the present phase. The ques-
tionnaire consists of five subscales (CG, CS, MG-A, MG-M, and MS), and participants
responded on a 5-point scale with 1 = Not At All and 5 = Very Often.
Children’s overall confidence was measured with the confidence subscale of the Competitive
State Anxiety Inventory 2 for Children (CSAI-2C). The CSAI-2C (Stadulis et al., 2002)
comprises 15 items assessing cognitive anxiety, somatic anxiety, and self-confidence. Five
items are included in each of the subscales, and participants rate the intensity of their anxiety
symptoms on a 4-point Likert scale (1 = Not At All to 4 = Very Much So). Internal consistency
for the subscales ranges from .73 to .78. Construct validation of the CSAI-2C was supported
using the Sport Competition Anxiety Test for Children (SCAT-C). For the present study, only
the confidence subscale was employed.
Children’s self-efficacy was assessed with the Self-Efficacy Questionnaire—Soccer (SEQ-S).
The questionnaire consists of five items designed to measure the perceived self-efficacy of an
athlete using the procedures recommended by Bandura (1997). This questionnaire is a modification
of the instrument used by Munroe-Chandler and Hall (2004) in their study with youth soccer
athletes. Participants record the strength of their belief on a 100-point scale, ranging in 10-unit
intervals from 0 (no confidence) to 100 (complete confidence).
104 HALL ET AL.

Procedure and data analysis. Once ethics clearance was obtained from the universities’
research ethics boards, athletes from various soccer organizations were contacted to participate
in the research. Permission and consent were obtained from the athletes’ parents. After completing
an individual assent form, athletes volunteering to be part of the research were administered the
questionnaires. They completed the SIQ-C, the confidence subscale of the CSAI-2C, and the
SEQ-S at their place of practice. The questionnaires took approximately 10–20 min to complete.
Pearson’s product–moment correlations were calculated to assess the strength of relationships
between the scores on each of the subscales of the SIQ-C and the measure of self-confidence
(CSAI-2C) and self-efficacy (SEQ-S).

Results and Discussion

For convergent validity, it was expected that MG-M imagery use would be strongly correlated
with both self-confidence and self-efficacy. This proved to be the case; r = .73 (p < .01) for MG-M
and self-confidence and r = .61 (p < .01) for MG-M and self-efficacy. For discriminant validity,
it was expected that the cognitive functions of imagery (i.e., CS and CG) would show a weaker
relationship with self-confidence and self-efficacy than the MG-M imagery function. This also
was evident. The correlations were as follows; CS imagery and self-confidence (r = .39, p < .01)
and self-efficacy (r = .41, p < .01), CG imagery and self-confidence (r = .38, p < .01) and self-
efficacy (r = .38, p < .01). See Table 5 for all correlations among the five imagery functions and
the measures of self-confidence and self-efficacy.
Taken together the results of Phase 3 provide support for both the convergent and discriminant
validity of the SIQ-C. As predicted in the applied model of imagery use (Martin et al., 1999), the
results revealed strong correlations between the MG-M function of imagery and the measures of
self-confidence and self-efficacy. In contrast, both cognitive functions of imagery (CS and CG)
were not as strongly associated with the scores on the CSAI-2C and our measure of self-
efficacy. Although convergent and discriminant validity are important, they are but two types of
validity. Future research may consider examining other types of validity including construct and
predictive validity (Thomas, Nelson, & Silverman, 2005).

TABLE 5
Bivariate Correlations Among the SIQ-C Subscales and SEQ-S and CSAI-2C Confidence Subscale

Subscales 1 2 3 4 5 6

1. CS 1.00
2. CG 0.49* (.31–.65) 1.00
3. MS 0.53* (.36–.67) 0.36* (.16–.54) 1.00
4. MG-A 0.47* (.29–.63) 0.52* (.35–.66) 0.54* (.37–.68) 1.00
5. MG-M 0.64* (.50–.76) 0.58* (.42–.71) 0.58* (.42–.71) 0.70* (.58–.80) 1.00
6. CSAI-2C 0.39* (.19–.56) 0.38* (.18–.56) 0.52* (.35–.66) 0.60* (.44–.73) 0.61* (.46–.74) 1.00
(confidence)
7. SEQ-S 0.41* (.22–.58) 0.38* (.18–.56) 0.47* (.29–.63) 0.49* (.31–.65) 0.73* (.62–.82) 0.58* (.42–.71)

Note: 95% confidence intervals are shown in brackets.

*p < .01.
 SIQ-C 105

GENERAL DISCUSSION

Athletes as young as 7 years of age report using imagery in sport for both cognitive and
motivational purposes (Munroe-Chandler et al., 2007). The SIQ has been successfully employed to
investigate cognitive and motivational imagery use by adults (Hall et al., 2005). The purpose of the
present research was to modify the SIQ for use with children. Results of the CFA indicated that the
five-factor model approached a reasonable fit for the 21-item SIQ-C, the instrument consisting of
CS, CG, MS, MG-M, and MG-A imagery, with each factor assessed on a 5-point rating scale. In
addition, favorable internal consistencies for most of the items designed to assess each factor were
found. In the current study, the CG and MG-A subscales demonstrated reliabilities just slightly
lower than the recommended .70 cutoff (Nunnally, 1978). However, Devellis (1991) has suggested
that although alpha values below .60 are generally considered unacceptable, it is not uncommon to
see published scales with alphas in the .60–.69 range. Thus, the present research provides some
evidence that the SIQ-C has reasonable structural validity and may be a useful instrument for the
assessment of cognitive and motivational imagery use by athletes aged 7–14 years.
In general, the present research supports that of Munroe-Chandler et al. (2007), demonstrating
that athletes between the ages of 7 and 14 years, similar to adult athletes, report using all five of
Paivio’s (1985) cognitive and motivational functions of imagery use. Also similar to the findings
for adult athletes (Munroe et al., 1998) but contrary to those of Munroe-Chandler et al. (2007),
males and females reported using imagery to about the same extent. Some age group differences
were evident in the present study, and these should be investigated in more detail in future work.
The overall means for the five subscales of the SIQ-C were between 3 (sometimes use
imagery) and 4 (often use imagery) on a 5-point Likert-type scale. These scores would suggest
that young athletes are making use of imagery to enhance their sport performance. These scores
also indicate that these young athletes were using both the cognitive and motivational functions
of imagery with a frequency that would be comparable to adults (Hall et al., 2005). Although
their use of imagery is encouraging, increases in their reported use of this valuable psychologi-
cal skill are possible and could be realized given the right coaching and guidance.
Despite the appeal of the present findings, several limitations of the present study need to be
acknowledged. First, this research was correlational and cross-sectional in design, and therefore,
longitudinal research examining the use of imagery over time is warranted. Second, no attempt
was made to link scores on the SIQ-C with other measures of imagery such as instruments that
measure imagery ability. It must be noted, however, that no such instruments specifically
designed for use with children currently exist.
The SIQ-C has the potential to be a valuable tool for measuring imagery use by young
athletes that could be employed by researchers and, possibly, applied sport psychologists.
Before this can be done, however, further psychometric analysis of the SIQ-C should to be
undertaken, including replication of the present findings, an assessment of test–retest reliability,
and an assessment of other types of validity (e.g., concurrent).

REFERENCES

Bandura, A. (1997). Self-efficacy. The exercise of control. New York: W. H. Freeman.

Barrett, P. (2007). Structural equation modeling: Adjudging model fit. Personality and Individual Differences, 42, 815–824.
106 HALL ET AL.

Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107, 238–246.
Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley.
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). Newbury Park, CA: Sage.
Brustad, R. (1998). Developmental considerations in sport and exercise psychology measurement. In J. L. Duda (Ed.),
Advances in sport and exercise psychology measurement, (pp. 461–470). Morgantown, WV: Fitness Information
Technology Inc.
Callow, N., Hardy, L., & Hall, C. (2001). The effect of a motivational general-mastery imagery intervention on the sport
confidence of high-level badminton players. Research Quarterly for Exercise and Sport, 72, 389–400.
Cornelius, A. (2002). Interventions techniques in sport psychology. In. J. M. Silva & D. E. Stevens (Eds.), Psychological
foundations of sport (pp. 177–196). Boston, MA: Allyn and Bacon.
Cumming, J., & Hall, C. (2002). Athletes’ use of imagery in the off-season. The Sport Psychologist, 16, 160–172.
Devellis, R. F. (1991). Scale development: Theory and applications. Newbury Park: Sage Publications.
Fishburne, G. J. (2005). Developmentally appropriate physical education for children and youth. Edmonton, AB: Ripon
Publishing
Gould, D., Weiss, M., & Weinberg, R. (1981). Psychological characteristics of successful and non-successful big ten
wrestlers. Journal of Sport Psychology, 3, 69–81.
Gregg, M., & Hall, C. (2006). The relationship of skill level and age to the use of imagery by golfers. Journal of Applied
Sport Psychology, 18, 363–375.
Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis. Upper Saddle River, NJ:
Prentice Hall.
Hall, C., Mack, D., Paivio, A., & Hausenblas, H. (1998). Imagery use by athletes: Development of the Sport Imagery
Questionnaire. International Journal of Sport Psychology, 29, 73–89.
Hall, C. R. (2001). Imagery in sport and exercise. In R. N. Singer, H. A. Hausenblas, & C. M. Janelle (Eds.), Handbook
of sport psychology (2nd ed., pp. 529–549). New York: John Wiley & Sons.
Hall, C. R., Stevens, D. E., & Paivio, A. (2005). The Sport Imagery Questionnaire: Test manual. Morgantown, WV:
Fitness Information Technology.
Hayduk, L., & Glaser, D. N. (2000). Jiving the four-step, waltzing around factor analysis, and other serious fun.
Structural Equation Modeling, 7, 1–35.
Hu, L., & Bentler, P. M. (1999). Cut off criteria for fit indices in covariance structure analysis: Conventional criteria
versus new alternatives. Structural Equation Modeling, 6, 1–55.
Kelloway, E. K. (1998). Using LISREL for structural equation modeling—A researcher’s guide. Thousand Oaks, CA: Sage.
Markland, D. (2007). The golden rule is that there are no golden rules: A commentary on Paul Barrett’s recommenda-
tions for reporting model fit in structural equation modeling. Personality and Individual Differences, 42, 851–858.
Marsh, H. W. (2007). Application of confirmatory factor analysis and structural equation modeling in sport and exercise
psychology. In G. Tenenbaum & R. C. Eklund (Eds.), Handbook of sport psychology (3rd ed., pp. 774–798).
Hoboken, NJ: John Wylie and Sons.
Marsh, H. W., & Hocevar, D. (1985). Application of confirmatory factor analysis to the study of self-concept: First- and
higher-order factor models and their invariance across groups. Psychological Bulletin, 97, 562–582.
Martens, R., Vealey, R. S., & Burton. D. (1990). Competitive anxiety in sport. Champaign, IL: Human Kinetics.
Martin, K. A., Moritz, S. E., & Hall, C. R. (1999). Imagery use in sport: A literature review and applied model. The
Sport Psychologist, 13, 245–268.
Maurer, T. J., & Price, H. R. (1998). A comparison of Likert scale and traditional measures of self-efficacy. Journal of
Applied Psychology, 83, 324–329.
Mills, K. D., Munroe, K. J., & Hall, C. R. (2000). The relationship between imagery and self-efficacy in competitive
athletes. Imagination, Cognition, and Personality, 20(1), 33–39.
Moritz, S. E., Hall, C. R., Martin, K. A., & Vadocz, E. (1996). What are confident athletes imaging?: An examination of
image content. The Sport Psychologist, 10, 171–179.
Morris, T., Spittle, M., & Watt, A. P. (2005). Imagery in sport. Champaign, IL: Human Kinetics.
Morton-Williams, J., & Sykes, W. (1984). The use of interaction coding and follow-up interviews to investigate compre-
hension of survey questions. Journal of Market Research Society, 26(2), 109–127.
Munroe, K., Hall, C., Simms, S., & Weinberg, R. (1998). The influence of type of sport and time of season on athletes’
use of imagery. The Sport Psychologist, 12, 440–449.
 SIQ-C 107

Munroe-Chandler, K. J., & Hall, C. R. (2004). Enhancing the collective efficacy of a soccer team through motivational
general-mastery imagery. Imagination, Cognition and Personality, 24, 51–67.
Munroe-Chandler, K. J. & Hall, C. R. (2007). Psychological interventions in sport. In P. Crocker (Ed.), Introduction to
sport psychology: A Canadian perspective (pp. 184–213). Toronto, ON: Pearson.
Munroe-Chandler, K. J., Hall, C. R., Fishburne, G. J., & Shannon, V. (2005). Using cognitive general imagery to
improve soccer strategies. European Journal of Sport Science, 5, 41–49.
Munroe-Chandler, K. J., Hall, C. R., Fishburne, G. J., & Strachan, L. (2007). Where, when and why young athletes use
imagery: An examination of developmental differences. Research Quarterly for Exercise and Sport, 78, 103–116.
Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill.
Paivio, A. (1985). Cognitive and motivational functions of imagery in human performance. Canadian Journal of
Applied Sport Sciences, 10, 22S–28S.
Pajares, F., Hartley, J., & Valiante, G. (2001). Response format in writing self-efficacy assessments. Greater discrimination
increases prediction. Measurement and Evaluation in Counseling and Development, 33, 214–221.
Piaget, J. (1971). Science of education and the psychology of the child. New York: Viking.
Short, S. E., Monsma, E. A., & Short, M. W. (2004). Is what you see really what you get? Athletes’ perceptions of imagery’s
functions. The Sport Psychologist, 18, 341–349.
Short, S. E., & Short, M. W. (2005). Differences between high and low confident football players on imagery functions:
A consideration of athletes?. Journal of Applied Sport Psychology, 17, 197–208.
Stadulis, R. E., MacCracken, M. J., Eidson, T. A., & Severance, C. (2002). A children’s form of the competitive state
anxiety inventory: The CSAI-2C. Measurement in Physical Education and Exercise Science, 6(3), 147–165.
Thomas, J. R., Nelson, J. K., & Silverman, S. J. (2005). Research methods in physical activity (5th ed.). Champaign, IL:
Human Kinetics.
Trochim, W. (2006). Convergent and discriminant validity. Retrieved October 11, 2007, from www.socialresearchmen-
tods.net/kb/convdisc.php.
West, S. G., Finch, J. F., & Curran, P. J. (1995). Structural equation models with nonnormal variables: Problems and
remedies. In R. H. Hoyle (Ed.), Structural equation modeling: Concepts, issues, and applications (pp. 56–75).
Thousand Oaks, CA: Sage.
Whaley, D. E. (2007). A life span developmental approach to studying sport and exercise behavior. In G. Tenenbaum &
R. C. Eklund (Eds.), Handbook of sport psychology (3rd ed., pp. 645–661). Hoboken, NJ: John Wylie and Sons.
White, A., & Hardy, L. (1998). An in-depth analysis of the uses of imagery by high-level slalom canoeists and artistic
gymnasts. The Sport Psychologist, 12, 387–403.