See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/259120772

ABI – Art-based Intervention Questionnaire

Article in The Arts in Psychotherapy · July 2013

DOI: 10.1016/j.aip.2013.06.005

CITATIONS READS

8 976

2 authors:

Sharon Snir Dafna Regev

Tel-Hai Academic College University of Haifa
29 PUBLICATIONS 160 CITATIONS 30 PUBLICATIONS 101 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

joint drawings View project

The personal meaning of recovery among adolescents with comorbidity between depression and
anxiety: phenomenological art based research View project

All content following this page was uploaded by Sharon Snir on 09 October 2017.

The user has requested enhancement of the downloaded file.

 The Arts in Psychotherapy 40 (2013) 338–346

Contents lists available at SciVerse ScienceDirect

The Arts in Psychotherapy

ABI – Art-based Intervention Questionnaire

Sharon Snir a,∗ , Dafna Regev b
a
Department of Education, Tel Hai College, and The Graduate School of Creative Art Therapies, University of Haifa, Israel
b
The Graduate School of Creative Art Therapies and the Emili Sagol C.A.T. Research Center, University of Haifa, Israel

a r t i c l e i n f o a b s t r a c t

Keywords: This study describes the development of the Art-based Intervention (ABI) Questionnaire, a self-report
Art-therapy instrument that examines the creative process as experienced in art therapy and in educational inter-
Therapeutic-interventions ventions. A qualitative analysis of reflections documented by 120 students who participated in workshop
Self-report-questionnaire
sessions for creating with art materials rendered 50 questionnaire items, which pertain to four main cate-
gories: 1. Feelings and thoughts preceding the artistic process, 2. Feelings and thoughts occurring during
the artistic process, 3. Attitude toward the artistic product, and 4. Approaches to the material. In the
second stage of the study, 291 volunteers created with art materials and filled the ABI and the Session
Evaluation Questionnaire (SEQ). Factor analysis demonstrated ten factors in all four categories together.
Internal reliabilities for the ABI scales ranged between ˛ = .453 and ˛ = .909. Significant correlations found
between the ABI and the SEQ subscales indicate that overall, the two instruments are similar but not iden-
tical, thus demonstrating concurrent validity of the ABI. The validation process and the meaning of the
resulting subscales are discussed.
© 2013 Elsevier Inc. All rights reserved.

The need to research and create reliable and validated scales for
the measuring, diagnosis, and assessment of therapeutic processes
in art-based therapies has been raised by art therapists who wish
to base their work on research based theories rather than on beliefs
(c.f., Hinz, 2008; Kapitan, 2010; Lusebrink, 2010; Rubin, 1984).
These therapists aim to organize information and knowledge accu-
mulated through practice into common themes, in order to create a
professionally acceptable conventional language (Revess-Shenhav,
1999).
Evidence-based studies evaluating the efficacy of art-based
therapies have indicated the need for instruments that would make
it possible not only to measure the effects of therapy on external
criteria, but also to assess the therapeutic processes that lead to
the desired outcome (Regev & Guttmann, 2005; Slayton, D’Archer,
& Kaplan, 2010). At the same time, other mental health profes-
sionals who work together with art therapists would like to have
evidence regarding the efficacy of the therapeutic modality and
the validity of the diagnoses and assessments attained through this
modality (Feder & Feder, 1998). The context for this demand is the
fact that art-based assessments are often used to complement and
substantiate a diagnosis attained through verbal therapy (Rubin,
1999), as well as the fact that conducting a scientific review and
evaluation of therapeutic methods and change processes is a pro-
fessionally acceptable and existing procedure among practitioners
∗ Corresponding author at: Narkisim 12a Street, Tivon,
Israel. Tel.: +972 77 41100608; fax: +972 77 4110060.
E-mail address: sharonsnir@gmail.com (S. Snir).
of psychotherapy (e.g., Bergin & Garfield, 1994; Luborsky & Crits-
Christoph, 1998; Roth & Fonagy, 1996; Safran & Muran, 2000).
As part of these efforts, a number of important studies suggested
a formal analysis of structured visual expression as a standard-
ized tool for diagnosis (DDS; Cohen, Hammer & Singer, 1988;
Cohen, Mills & Kijak, 1994; PPAT; Gantt & Tabone, 1998). An
important theory that address the problem of measuring and stan-
dardized evaluation is the Expressive Therapies Continuum (ETC)
(Hinz, 2008; Lusebrink, 2004, 2010), which provides a systematic
approach to understand the relationship between client and mate-
rials.
Recently, given the increased recognition in the impact of
research and the desire to further develop and academize the field
and the profession, there has been a rise in the number of art thera-
pist conducting research, which in turn has led to a discussion of the
specific difficulties involved in the scientific research of art thera-
pies (Carolan, 2001; Deaver, 2002; Kapitan, 2010). This discussion
suggests that many of the difficulties described regarding the con-
struction of measurement and assessment scales and the practice of
research in this field in general stem from the complexities involved
in art therapy, as such, the particular richness and sophistication of
the language of art, as well as the non-verbal characteristics of art-
based therapy (Linesch, 1995; McNiff, 1998; Rosal, 1998; Wadeson,
1980, 1987).
Furthermore, it is commonly claimed that it is difficult to inte-
grate art-based therapy, which underscores the uniqueness of each
individual and the many viable ways to view and perceive the world
into a research model that views human beings as a source of objec-
tive and predictable knowledge (Junge & Linesch, 1993).

0197-4556/$ – see front matter © 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.aip.2013.06.005
 S. Snir, D. Regev / The Arts in Psychotherapy 40 (2013) 338–346
One possible solution to some of the difficulties encountered is
to apply quantitative tools used for research in allied fields, such
as psychotherapy and interpersonal relationships. These instru-
ments facilitate the observation of complex and abstract processes,
such as therapeutic relationships, the emotional progress of ther-
apy, and transference, which can be difficult to measure. Using
methods of self-reporting, these quantitative tools make it possi-
ble to focus on the experience of the client or participant, thereby
observing the factors, connections, and implications that play a
part in the therapeutic process, as well as the potential for altering
these experiences through therapeutic interventions (c.f., Stiles,
Reynolds, Hardy, Rees, Barkham & Shapiro, 1994; Wiggins, Elliott, &
Cooper, 2012; Wiseman et al., 2002). The focus on the individual’s
experience as relayed through self-reporting, provides an oppor-
tunity to observe and analyze the therapeutic processes involved
in art-based therapy and in art-based interventions in the field of
education.
There have been several attempts in recent years to implement
quantitative instruments, e.g. the Session Evaluation Ques-
tionnaire, a self-reporting instrument developed to investigate
processes in psychotherapy (Stiles et al., 1994) in the research of
art-based interventions (c.f., Snir & Wiseman, 2010). These efforts
have helped identify the factors that influence one’s reaction to a
work with art materials. However, they have done little to explain
the unique contribution of the creative experience that takes place
in the process of art-based therapies.
The current study aims to address this gap, by proposing
a self-reporting instrument, the Art-based Intervention (ABI)
Questionnaire, which aims to focus on the creative process as expe-
rienced in the making of an artistic object. Integrating a quantitative
instrument that can be subjected to statistical tests in the field of
art-based therapy would make it possible to research and measure
developments and changes that occur in the therapeutic process,
assess reactions to various art materials and art-based interven-
tions (in the educational framework as well as in the clinic), and
examine the possible effects of patient variables, therapist vari-
ables and different stages in the therapeutic process and in the
therapeutic relations, on these reactions.
Study Number 1
The main goal
The development of the Art-based Intervention Questionnaire
was based on the reactions of college students to the experience
of working with various art materials. The creative task was per-
formed in a free but supported manner, in a framework similar to
that used in art therapy creative processes. Afterwards, students
were asked to reflect on the experience in free written form. The
first stage of the study was to identify and define the statements
which were most relevant in terms of their ability to express the
creators’ experiences. Once these statements were gathered, a sec-
ond, more refined selection process was initiated.
Participants
This research phase was based on reflections written by 120
bachelor’s degree students, 20–30 years old, who were enrolled in
the course “Art materials in educational and therapeutic work.” The
majority of students were in their first year of studies. The reflec-
tions were collected over a three-year period from three different
groups (about 40 participants in each group), which participated in
workshop sessions where they created a work with art materials.
At the beginning of the course, the students gave their consent to
339
have their written reflections used for research purposes. Most of
the participants had no special background in art.
Instruments
At the end of each session, participants were asked to write their
reflections about their experience with the material they had used,
including their associations with the material, and to reconstruct
the feelings and thoughts that had arose while they were work-
ing. The written responses, about half a page in length each, were
prepared on the same day as the session and sent to the course
instructor by e-mail with no identifying details.
Procedure
The sessions, each of which lasted two academic hours, were
held once a week throughout one semester (13 weeks). The stu-
dents could choose to work according to their own style and
technique with the material offered or to follow specific instruc-
tions that were designed to expose them to different ways of
working with each material. In most cases, the students sat around
tables, with 4–6 participants per table, as they chose. Each of the art
materials: markers, oil-pastels, gouache paints, finger paints and
clay, was offered at least once or twice during the entire semester,
according to other considerations related to the course.
Participants’ reflections were read and reviewed by two judges,
who selected the statements that described the encounter with
the various materials. After reviewing a total of 10–15 reflections
per art material, the judges determined that reviewing additional
reflections would no longer affect the list of pertinent statements.
The selected statements were analyzed and processed. Statements
that were relevant only to a specific material were eliminated, as
were those that appeared unclear or ambiguous and those which
described an unusual experience.
Results
The process of identifying and selecting statements from the
pool of students’ reflections rendered a list of 50 items, pertaining to
four main categories: 1. Feelings and thoughts preceding the artis-
tic process (8 items); 2. Feelings and thoughts occurring during the
artistic process (28 items); 3. Attitude toward the artistic product
(6 items); 4. Approaches to the material (8 items). These four cate-
gories corresponded with the four observational foci proposed by
the University of Haifa phenomenological approach (Hesse, 1981,
1998).
Study Number 2
The main goal
The purpose of the second stage of the study was to define the
subscales for each of the four parts of the questionnaire, and then
validate them and test their reliability.
Participants
Participants in the second stage of the study were 291 volun-
teers, 109 (37.5%) men and 182 (62.5%) women, of ages ranging
between 18 and 77 (M = 29.7, SD = 11.024). Of the total number of
participants, 189 (64.0%) were single and 90 (30.9%) were mar-
ried, while the majority of participants had no children at the
time (74.2%). Participants’ education levels were as follows: 118
(40.5%) were high school graduates and 114 (39.2%) had obtained an
undergraduate degree. Demographic information regarding Study
2 participants is presented in detail in Table 1.
340 S. Snir, D. Regev / The Arts in Psychotherapy 40 (2013) 338–346
Table 1
Participants’ demographic variables (n = 291).
The variable Values
Family status Married
Single
Other
Number of children 0 74.2
1 3.8
2 4.5
3 11.0
4 or more
Level of education High school dropouts
High school graduates
Undergraduates
Graduates
Other
Work Freelance
Full-time employees
Part-time employees
Unemployed
Other
Participants’ average score on a question about their involve-
ment with art, rated on a scale from 1 (not at all) to 7 (to a great
extent), was 2.88 (SD = 2.14). Answering the question “Have you
ever studied art?”, participants’ average score, using the same scale,
was 2.81 (SD = 2.16). In other words, participants in this study had
very little prior interaction with art, either as a hobby or as a subject
of study.
Instruments
Art-based Intervention (ABI) Questionnaire
Participants received a questionnaire, composed of the 50 items
that were defined in Study 1 and were asked to indicate the degree
to which each statement describes their artistic experience in the
workshop session, using a scale of 1 (not at all) to 7 (to a great
extent).
Session Evaluation Questionnaire (SEQ; Stiles et al., 1994)
The SEQ is a self-report instrument used to assess the reactions
of clients and therapists to specific counseling or psychother-
apy sessions. The SEQ has two parts, each consisting of 12 pairs
of adjectives of polar-opposite meanings presented on a 7-point
semantic differential scale. The first part assesses participants’ per-
ceptions of the session using two evaluative dimensions: depth and
smoothness. The depth dimension assesses the session’s perceived
power and value (e.g., valuable-worthless; shallow-deep), and the
smoothness dimension assesses the degree to which the session’s
atmosphere was perceived as comfortable, relaxed, and pleasant
(e.g., difficult-easy; unpleasant-pleasant). On this part of the SEQ,
respondents were instructed to rate their evaluation of the cre-
ative process. The second part of the SEQ assesses participants’
post-session mood on two dimensions, positivity and arousal. The
positivity dimension refers to feelings of confidence and clarity, as
well as happiness and the absence of fear or anger (e.g., confident-
afraid; happy-sad), and the arousal dimension refers to feeling
active and excited as opposed to quiet and calm (e.g., quiet-aroused;
wakeful-sleepy). We adapted the instructions for this second part
of the SEQ in that the respondents were instructed to rate their
mood during the creative process. Internal reliabilities for the SEQ
scales were depth ˛ = .76, smoothness ˛ = .72, positivity ˛ = .79, and
arousal ˛ = .56.
Percentage Number of participants
30.9 90
64.9 189
3.8 11
216
11
13
32
5.6 19
2.4 7
40.5 118
39.2 114
11.0 32
6.8 20
4.5 13
35.7 104
26.8 78
17.2 50
13.7 40
Procedure
Participants were recruited through ads posted around campus
and in the community, which were worded as follows.
“Required: Participants for a study that examines reactions
to situations. Criteria: Over 18 years of age, native speakers of
Hebrew. No prior knowledge or preparation necessary. Partici-
pation involves attending a short meeting and filling out forms.
Approximate duration: 1 h maximum. Payment: NIS 30.00 per par-
ticipant”.
During a phone conversation, potential participants were
informed that the “situation” involved creating a work of art and
that no prior artistic experience was necessary. This information
was purposely eliminated from the ads posted, in order not to
attract art aficionados only. It was emphasized that the artistic
work would remain anonymous and that confidentiality would be
maintained throughout.
Data collection was carried out by six assistants, students in the
Faculty of Education, who had been briefly trained by the authors
of the study. They were instructed to follow a single structured pro-
tocol, which each received in written form. The protocol included
clear and uniform guidelines about how the art materials were to
be distributed, emphasizing the presentation of new materials in
an organized fashion and the need to provide a wide range of colors.
The task took place in classrooms prepared especially for this
purpose in two different college campuses or in art studios in the
community. Participants were asked to arrive at a predetermined
time, such that at any given time, there were between 2 and 6 par-
ticipants in the room working individually at separate tables, with
no eye contact or interaction among them. Participants received
an explanation regarding the format of the meeting, specifically,
30 min would be devoted to creative artwork, after which they
would be asked to complete a series of questionnaires. They were
told they would be working with a particular art material and were
reminded that no prior knowledge or experience was required, and
that they were not expected to create an aesthetic product; rather,
they were to try and become more familiar with the material, expe-
rience it, and use it to express themselves. The type of material used
was predetermined by the experimenter, so that participants work-
ing simultaneously in one room were all using the same material
(markers, oil-pastels, gouache paints, finger paints or clay). Before
approaching the material, participants signed an informed consent
form, which included a description of the study, a confidentiality
 S. Snir, D. Regev / The Arts in Psychotherapy 40 (2013) 338–346
statement, and the fact that they were free to withdraw from the
study at any time. Then the experimenter distributed the mate-
rial, modeled the basic principles of working with this material,
and told participants that they were free to create two separate
works if they desired. The rationale for allowing for two works per
participant was to give them the opportunity to “warm-up” to the
material.
Participants were given aprons to protect their clothing, two
sheets of different sized paper so that they could choose the canvas
base for their creation. The paper sizes offered were tailored to the
material that would be used.
Participants decided individually when the artistic process was
complete, and after handing the product to the assistant, they were
given the questionnaire to be completed. Once the completed ques-
tionnaire was handed in, each participant received a remuneration
of NIS 30.00 (Israeli currency).
Results
Factor analysis and reliability
Because of the primacy of the analysis, and based on the assump-
tion that the variables are independent, we used factor analysis
with varimax rotation to locate which factors exist in the four parts
of the questionnaire. Factor analysis was performed separately for
each one of the four ABI parts, since each part was viewed as
addressing a separate and unique aspect of the artistic experience.
On the first part of the ABI Questionnaire, which reviews the
feelings and thoughts that preceded the artistic process, the 8 items
fell under 3 main factors reflecting 65.15% of the variance (see
Table 2 for factor loading and reliability).
The first factor, excitement in anticipation of the artistic process,
included three items, with factor loadings greater than .836; the
second factor, sense of confidence, included two items with factor
loadings greater than .807; and the third factor, aversion, included
two items with factor loadings greater than .769. Following the fac-
tor analysis results, one item was removed from the questionnaire,
as it did not correspond to any of the factors. Reliability for the three
factors ranged between .474 and .846 Cronbach’s alpha. In addition,
a good general reliability rate of .746 Cronbach’s alpha was found
for all seven items on this part of the questionnaire.
Factor analysis on the second part of the questionnaire, feelings
and thoughts during the artistic process, found four factors, which
together explained 65.15% of the variance (Table 3). The first fac-
tor, pleasantness and therapeutic value, included ten items, with
factor loadings greater than .561 per factor. The second factor, com-
petence, included six items, with factor loadings greater than .432
per factor. The third factor, difficulty carrying out the artistic task,
included four items with factor loadings greater than .400 per fac-
tor. The fourth factor, playfulness, included two items, with factor
loadings greater than .689 per factor. Two additional factors, of one
item each, were eliminated from the final analysis. In addition, four
other items were eliminated from the questionnaire, as they did not
correspond significantly with any of the factors. Reliability for the
four factors ranged between .507 and .909 Cronbach’s alpha. Cal-
culating the reliability of all 22 items in this part, and allowing for
inverted order of items, rendered an acceptable reliability rate of
.856 Cronbach’s alpha.
The six items of the third part of the questionnaire, regarding
attitudes toward the artistic product, were all found to pertain
to one factor, positive attitude toward the artistic product, which
explained 53.12% of the variance (Table 4) and had factor load-
ing greater than −.625. Calculating the reliability of all 6 items of
this section and allowing for inverted order of items, found a good
reliability rate of .820 Cronbach’s alpha.
The fourth part of the questionnaire, regarding attitudes toward
the materials, had six items pertaining to two factors, which
341
explained 61.74% of the general variance (Table 5). The first fac-
tor, meaningful material, included three items with factor loadings
greater than −.617. The second factor, pleasant material, included
three items with factor loadings greater than .485. Two items were
eliminated from the analysis, after factor loading did not render
a significant correlation with any single factor. The reliability of
the first factor, meaningful material, was .735 Cronbach’s alpha,
and the reliability of the second factor, pleasant material, was .453
Cronbach’s. Calculating the reliability of all six items of this factor,
allowing for inverted items, found a good reliability rate of .635
Cronbach’s alpha.
Finally the reliability of all four parts of the questionnaire
together was calculated and was found to be .914 Cronbach’s alpha.
The correlations found among the various factors from all
four parts of the questionnaire are presented in Table 6. Most of
the correlations were of statistical significance (p < .001), of low-
to-medium values, indicating that most factors relate to similar
content fields, yet they are clearly distinguishable. A higher corre-
lation value of .614 was found between the factor of pleasantness
and therapeutic value and that of excitement, indicating similar
content, a finding which should be explored in future studies.
Testing concurrent-forms reliability
Concurrent validity was assessed by calculating the correspon-
dence between answers to the ABI questionnaire and answers to the
four parts of the SEQ questionnaire. Table 7 presents the findings. As
shown, the correlation between SEQ and ABI was calculated taking
into consideration the three types of ranking: general ranking of the
questionnaire, overall ranking of all four parts of the questionnaire,
and the rankings for each of the subscales. The correlations were
found to be significant (p < .001) for most, ranging from weak corre-
lations (.189) to strong correlations (.655). An acceptable or strong
correlation between the overall ABI score and the SEQ scales indi-
cated concurrent reliability for the entire questionnaire. Assessing
correlations between SEQ scales and the four scores per each part
of the ABI questionnaire found good correlations (mostly higher
than .45) with the dimensions of depth and arousal, while weaker
correlations were found for the dimensions of smoothness and pos-
itivity (between .25 and .40). Similarly, weak to medium strength
significant correlations were found between the SEQ scales and the
ten subscales on the ABI questionnaire. As can be seen in Table 7,
the scales of therapeutic value and pleasure rendered the strongest
correlations, indicating good concurrent reliability for this scale.
Testing discriminative ability
Testing for the discriminative ability was done by examining
the questionnaire’s ability to identify differences in participants’
responses to creating with different materials. For this purpose, ini-
tially an ANOVA test was performed using the total ABI score on the
questionnaire. The discriminative ability test found that the type
of material had a significant effect on the responses to the ques-
tionnaire (F(4, 282) = 7.22, p < .001, 2 = .09). This indicates that the
questionnaire is capable of differentiating participants’ reactions to
the various materials. A subsequent Bonferroni test was conducted
in order to adjust the observed significance level for the multiple
comparisons that were made. The subsequent Bonferroni test found
that the average response to working with the markers was sig-
nificantly lower than the average response to working with the
other materials. In other words, using the markers was perceived
as negative compared to the use of other materials (Table 8).
MANOVA tests were conducted on the first, second and fourth
parts of the ABI. In addition, an ANOVA was conducted on the third
part which contains only one subscale.
342 S. Snir, D. Regev / The Arts in Psychotherapy 40 (2013) 338–346

Table 2
Factor loadings for the 7-item first part of the ABI: feelings and thoughts preceding the artistic process (n = 291).

Questionnaire Items Loadings for factors

Excitement Confidence Aversion

I was eager to begin the creative task .875

I was curious about the creative task .848
I was glad to have an opportunity to engage in a creative task .836
I had several ideas about what I would want to make .819
I felt confident that whatever I chose to do would go well .807
I was upset at the prospect of getting dirty .792
I was averse to participating in this situation .769

Total explained variance: 65.15% 29.22% 18.99% 16.94%

Eigenvalue 2.34 1.52 1.35
Cronbach’s alpha: .746 .846 .671 .474

Table 3
Factor loadings for the 22-item second part of the ABI: Feelings and thoughts during the artistic process (n = 291).

Questionnaire Items Loadings for factor

Pleasantness Competence Difficulty Playfulness

and carrying out
therapeutic the artistic task
value

Working on my art project, I felt a sense of inner peace and warmth .736
Working on my art project released the tension I’d been carrying .728
Working on my art project felt to me like occupational therapy .726
I find that this type of work makes me concentrate .701
I learned about myself in the process .697
Working gave me a sense of confidence .663
I found it pleasant to be creating something .646
While I was creating, I came up with all kinds of ideas for my art project .633
I enjoyed working on my art project .631
I felt I could keep on going for hours .561
I had a difficult time executing my ideas −.794
I felt I wasn’t good at this kind of activity −.777
I felt I wasn’t being creative −.742
I felt I had managed to make a nice and esthetic product without making an effort .732
I felt limited −.570
I was able to let go and let the work take me anywhere .432
I encountered lots of technical difficulties in performing the artistic task .775
It took me a while to understand how to work with the materials .683
I had a hard time sitting still and wanted to get up and move around .479
I knew exactly how to handle the materials −.400
I felt it was OK to make mistakes .779
I felt I was playing with the materials .689

Total explained variance: 65.15% 21.65% 16.75% 7.64% 6.81%

Eigenvalue 6.06 4.69 2.14 1.91
Cronbach’s alpha: .856 .909 .863 .598 .507

The MANOVA test on the first part, Feelings and thoughts
preceding the artistic process, found no effect of material type
on the responses to the questionnaire (F(12, 741) = .946, n.s.,
2 = .018).
The MANOVA test on the second part, Reaction during the
artistic process, found a significant effect of material type on the
response to the questionnaire (F(16, 849) = 4.38, p < .001, 2 = .06).
The ANOVA tests demonstrated a significant effect of material type
on each of the subscales (see Table 8). Subsequent Bonferroni tests
demonstrated for example that the sense of therapeutic value and
pleasantness was rated higher in response to working with gouache
and finger paints than in response to working with the markers,
and that difficulties performing the artistic task were reported at
a higher level in response to working with oil pastels and clay,

Table 4
Factor loadings for the 6-item third part of the ABI: attitude toward the artistic product (n = 291).

Questionnaire items Loadings for factor: positive attitude toward the artistic product

I wanted to keep what I had made .849

I was excited to see what I was created .763
I completed the project with a sense of satisfaction .757
I was not interested in the final product −.713
I was surprised by what I had made .642
I wasn’t satisfied with what I had made −.625

Total explained variance: 53.12%

Eigenvalue 3.19
Cronbach’s alpha: .820
 S. Snir, D. Regev / The Arts in Psychotherapy 40 (2013) 338–346 343

Table 5
Factor loadings for the 6-item fourth part of the ABI: Attitude toward the materials (n = 291).

Loadings for factor

Questionnaire Items Meaningful material Pleasant material

Flat material, lacks depth −.858

Material communicates cold detachment −.811
A powerful material .617
User-friendly material .763
Material is pleasant to work with .749
The material has a soothing effect .485

Total explained variance: 61.74% 33.95% 27.78%

Eigenvalue 2.72 2.22
Cronbach’s alpha: .635 .735 .453

Table 6
Correlations between sub scale of the ABI.

1 2 3 4 5 6 7 8 9 10

1. Excitement
2. Confidence .395***
3. Aversion −.139 −.199***
4. Pleasantness and therapeutic value .614*** .579*** −.201***
5. Ability .311*** .594*** −.257*** .551***
6. Difficulty carrying out the artistic task −.173** −.329*** .299*** −.261*** −.461***
7. Playfulness .249*** .285*** −.104 .369*** .269*** −.099
8. Positive attitude toward the artistic product .434*** .537*** −.150 .595*** .550*** −.215*** .145
9. Meaningful material .240*** .255*** −.118 .455*** .394*** −.222*** .191*** .307***
10. Pleasant material .326*** .321*** −.196*** .463*** .343*** −.167** .180** .252*** .339***
M 4.63 4.58 1.91 4.74 5.01 2.329 5.499 4.29 5.13 5.68
SD 1.49 1.57 1.57 1.24 1.35 1.04 1.40 1.43 1.44 1.80
***
p < .001.

compared to the response to working with markers and gouache
paints (additional differences are shown in Table 8).
The ANOVA test on the third part of the ABI, Attitude toward the
artistic product, found no effect to material type (F(4,280) = 1.564,
n.s., 2 = .22).
Finally, in the fourth part, Attitude toward the materials, a
significant effect was found in the response to creating with var-
ious materials (F(8, 560) = 10.657, p < .000, 2 = .132), while the
ANOVA results showed a significant effect of the type of material,
as reported by participants’ responses on the two subscales. Subse-
quent Bonferroni tests found that the gouache paints, oil pastels and
clay were considered the most meaningful art materials. Addition-
ally, finger-paints were found to be significantly less meaningful
than gouache paints, oil pastels and clay, but more meaningful
than the markers. Markers were also perceived as less pleasant to
work with than the gouache paints, the oil pastels or the finger
paints.
General discussion
The current study focused on developing a tool for self-reporting
on one’s participation in an art based intervention. The purpose
of this tool is to facilitate our understanding of developments and
changes that occur through therapeutic processes and in response
to art-based interventions, by observing the manner in which par-
ticipants experience the creative process.
The process of validating this tool included two studies. The first
study focused on selecting and defining statements to be included
in the self-report questionnaire. These were selected from a pool
of participants’ written verbal reactions to the creative experience,
in which various art materials were used. The purpose of the sec-
ond study was to define, validate, and assess the reliability of the
subscales used in the four parts of the self-report questionnaire.
The first study analyzed participants’ reactions to creating
freely with various art materials and identified fifty statements,

Table 7
Correlations between responses to ABI and responses on the four parts of the SEQ.

Depth (n = 290) Smoothness (n = 290) Positivity (n = 290) Arousal (n = 290)

Overall score .655*** .400*** .415*** .506***

Feelings and thoughts preceding the artistic process .489*** .360*** .347*** .453***
Feelings and thoughts during the artistic process .605*** .383*** .401*** .468***
Positive attitude toward the artistic product .548*** .279*** .208*** .451***
Attitude toward the materials .454*** .248*** .357*** .254***
Excitement .451*** .263*** .241*** .395***
Confidence .383*** .282*** .302*** .398***
Aversion −.147 −.194*** −.182 −.104
Pleasantness and therapeutic value .628*** .367*** .351*** .462***
Ability .455*** .383*** .435*** .377***
Difficulty carrying out the task −.205*** −.384*** −.315*** −.141
Playfulness .123 .294*** .233*** .141
Meaningful material .456*** .212*** .316*** .234***
Pleasant material .307*** .198*** .274*** .189***
***
p < .001.
344 S. Snir, D. Regev / The Arts in Psychotherapy 40 (2013) 338–346

Table 8
Assessing differential validity using ANOVA tests.

Variables Gouache paints Oil pastels Magic markers Clay (n = 52) D Finger paints F 2 Bonferroni
(n = 59) A (n = 60) B (n = 61) C (n = 55) E

Total ABI score 4.83(.71) 4.79(.84) 4.17(.86) 4.82(.82) 4.80(.87) 7.22*** .093 C,A,B,D,E

Excitement 4.73(1.30) 4.733(1.41) 4.30(1.66) 4.51(1.53) 5.0(1.43) 1.79 .025

Confidence 4.77(1.45) 4.80(1.49) 4.00(1.54) 4.51(1.69) 4.90(1.59) 3.22* .044 E>C
Aversion 1.80(1.93) 1.81(1.25) 2.07(1.71) 1.82(1.61) 1.90(1.19) .313 .004

Pleasantness and 4.98(1.08) 4.79(1.35) 4.23(1.28) 4.84(1.22) 4.97(1.10) 3.78** .051 A,E > C
therapeutic value
Ability 5.16(1.30) 5.15(1.19) 4.47(1.38) 5.27(1.19) 5.03(1.58) 3.28* .045 B,D > C
Difficulty carrying out 2.03(.94) 2.23(1.00) 2.07(0.95) 2.68(1.11) 2.58(1.07) 4.64*** .062 B,D > A,C
the task
Playfulness 5.71(1.53) 5.27(1.51) 4.94(1.37) 5.63(1.27) 5.95(1.06) 4.90*** .065 A,E > C

Positive attitude 4.35(1.39) 4.50(1.24) 3.92(1.53) 4.47(1.53) 4.28(1.44) 1.564 0.22

toward the artistic
product
Meaningful material 5.57(1.20) 5.72(1.11) 3.99(1.43) 5.68(1.23) 4.80(1.43) 20.12*** .223 A,B,D > E > C
Pleasant material 6.15(2.31) 5.94(2.05) 4.91(1.34) 5.55(1.54) 5.94(1.22) 4.63*** .062 A,B,E > C
*
p < .05.
**
p < .01.
***
p < .001.

which were found to correspond to four categories: feelings and
thoughts preceding the artistic process; feelings and thoughts dur-
ing the artistic process; attitude toward the artistic product; and
approaches to the material. These four categories are mentioned in
the phenomenological approach (Betensky, 1995; Hazut & Siano,
2008), as areas to be investigated by the therapist as a resource for
assessing the therapeutic processes of art-based therapies. Accord-
ing to this approach, the therapist observes the client prior to the
latter’s engagement in the creative process at the first moments of
contact with the art materials. Afterwards, the therapist observes
the client’s emotional, verbal, behavioral and artistic reactions dur-
ing the creative process, and the client’s attitude toward the art
material. Finally, the therapist observes the client’s manner of relat-
ing to the artistic product he or she made. The ABI questionnaire
asks the participants who performed the artistic task to relate to
these very same categories and assess their own creative expe-
rience, by reporting on the feelings, thoughts, and perceptions
related to the various stages of the creative process.
In the second study, the items from each of the content cat-
egories were submitted to varimax factor analysis, in order to
identify the factors pertinent to each category. For the first part
of the questionnaire, which focuses on the category feelings and
thoughts preceding the creative artistic process, three factors were
identified: excitement in anticipation of the artistic process, sense
of confidence, and aversion. These factors are related to knowl-
edge from previous experience, such as expectations garnered in
anticipation of creating with art materials, which in turn affect par-
ticipants’ confidence, as they face the creative task. The third factor,
initial aversion to the art material when facing the task, can explain
later developments that can be seen as the process unfolds. This fac-
tor may be related also to factors of personality or to past traumatic
events (Moon, 2010).
Factor analysis on the second part of the questionnaire, feel-
ings and thoughts during the artistic process, found four factors:
pleasantness and therapeutic value, competence, difficulty carry-
ing out the artistic task, and playfulness. The first factor connects
between two central dimensions of the artistic experience: an over-
all feeling of pleasantness during the process and the feeling that
the creative process is therapeutic and beneficial to the creator, for
example, by imbuing in the creator a sense of calm and release.
The connection between these two dimensions is supported by
theories of positive psychology, whereby pleasant sensations and
positive moods are related to a positive sense of self and men-
tal health (Fredrickson, 1998, 2009). Also the existing theoretical
assumptions about art-based therapy attribute an important ther-
apeutic role to the pleasure experienced through creativity, since it
serves to increase the individual’s involvement in a creatively valu-
able process (Perl-Dekel, 1998; Regev & Snir, submitted; Wadeson,
1980). Hence, this factor is of vital importance as part of the self-
reporting tool. Two additional factors in this category, namely,
competence and difficulty carrying out the artistic task, reflect the
manner in which participants perceive through their experience,
their own strengths and their ability to cope with the creative
process. According to the Phenomenological approach, these per-
ceptions are very important (Betensky, 1995; Hazut & Siano, 2008)
in the therapeutic process, as the emphasis on the client’s strengths
empowers the client to use and apply these strengths also in life in
general, beyond the therapeutic venue. Finally, the fourth factor
in this category, playfulness, touches on the unique quality of the
therapeutic process, the ability to create and play and examine the
self with the help of art materials (Luria, 2002; Winnicott, 1995).
The third part of the questionnaire was found to include only one
factor, namely, positive attitude toward the artistic product, which
generally describes the degree to which the client views the cre-
ative product positively or negatively. In the therapeutic context,
this attitude is expressed by the client’s reaction to the product:
whether to preserve the work or destroy it. Theorists of art-based
therapy (e.g., Schaverien, 1998; Shapira, 2005) consider this dimen-
sion to be representative of relationships, with self and with other
significant people in one’s life, as well as one’s relationship with
the therapeutic process in general.
Finally, the fourth part of the questionnaire describes
approaches to the materials and includes two factors: pleasant
material and meaningful material. Together, these two factors
describe the manner in which the client perceives the material:
whether it is pleasant to work with, on the one hand, and whether
it provides a viable vehicle for self-expression, on the other hand.
This part of the questionnaire is particularly important for defin-
ing the attributes as well as the therapeutic value of the various
art materials (Moon, 2010). Interestingly, a factor analysis that
reviewed clients’ responses to verbal therapy found similar dimen-
sions (Stiles et al., 1994). These dimensions served as the basis for
the design of the SEQ (discussed herein), which was the question-
naire the current study used to validate the ABI questionnaire. In
 S. Snir, D. Regev / The Arts in Psychotherapy 40 (2013) 338–346
that study, the authors found that clients related to the therapy in
terms of the quality of the therapeutic session on a scale from pos-
itive to negative, using qualities such as good, pleasant, calm, and
in terms of the depth of the therapeutic experience, relating to the
therapeutic value and the significance of the session. It appears that
also in the current study, pleasant and meaningful were two sig-
nificant factors in the participants’ perceptions of the therapeutic
intervention, especially at the level of the materials used.
It should be noted that six of the factors defined were comprised
of only 2 or 3 items. While for some of the factors the internal con-
sistency between items was not negatively affected by the small
number of items, in three factors, playfulness, aversion, and pleas-
ant materials, reliability was relatively low. Nevertheless, due to
the need at this early stage of the research to rely on factors for-
mulated on a theoretical basis, we decided to continue using the
subscales. However, to allow the information from the subscales of
the questionnaire to be scored and used, the reliability of the four
parts of the questionnaire as a single unit was tested, and the results
indicated good internal consistency reliability. A reliability test of
all of the items of the questionnaire together further improved the
reliability values.
Thus, the questionnaire offers three options for analyzing the
data. An overall score can be obtained, based on the average cal-
culated for all of the questionnaire items; four separate scores can
be obtained from the four parts of the questionnaire; or separate
scores can be obtained from each of the ten subscales. An overall
score obtained after inverting the necessary items, provides a gen-
eral assessment of the degree to which the creative process was
perceived as a positive experience. This is of clinical importance, as
it provides a broad and general view of the positive impact of art-
based interventions. In terms of research, this score can provide
information for differentiating between clients’ responses follow-
ing various interventions, and assist in making decisions such as
whether to transition to a different activity, such as joint paint-
ing with the therapist. A more detailed scoring is possible, by
using specific questions that focus on more complex and nuanced
aspects of the therapeutic process, or on more complex theoreti-
cal assumptions. Thus, for example, the relationship between one’s
perceptions of the art material and perceptions related to the ther-
apeutic process can be assessed by utilizing the first and second
parts of the questionnaire. The questionnaire can be used to assess
responses to different interventions, for example, comparing the
reactions to artistic products created with free access to various
art materials compared to the reactions to artistic products cre-
ated in a circumscribed framework. A more complex picture can
emerge if information is collected about the clients, e.g., informa-
tion about personality, relationships, or mental health. A study
incorporating such data is under way, currently in the analysis
stage. Findings from the forthcoming study could help us under-
stand the preferences and responses of individuals to interventions
with various art materials, while considering the emotional, inter-
personal or personality-type profile of the client. This information,
may serve as well to evaluate the connection between dimensions
of the therapeutic relationship, specifically, the client-therapist
relationship to the client’s reaction to the creative product and the
significance attributed to it. The ability of the ABI to measure com-
plicated naturalistic art based interventions is consistent with the
ETC (Lusebrink, 2004, 2010), which provides a theoretical model
for art-based assessments and applications of media in art ther-
apy. This ability distinguishes the ABI and the ETC from standard
projective techniques that can give a wealth of information about
diagnostic categories and personality traits but may too strictly
limit media and content choices and thus restrict the information
needed to proceed in therapy (Hinz, 2008).
In the process of validating this instrument, several interesting
findings emerged. One such finding relates to the SEQ, which was
345
used also in previous studies in this field (Freilich, 2010; Snir &
Wiseman, 2010). The findings of the current study demonstrated
that overall, the subscales of the proposed instrument correlated
significantly in varying strengths with the subscales of the SEQ.
Finding correlations of medium strength indicated that, in general,
the two instruments are similar but not identical, which supported
the process of defining concurrent validity. Consequently, the new
instrument can be used to examine, more accurately and in a man-
ner more suited to the creative experience, the various factors that
influence clients’ responses to the creative process.
Another interesting finding was related to the study’s ability to
make distinctions from among the variety of responses to the range
of art materials. A previous study by the current authors (Snir &
Regev, 2012) demonstrated the unique properties of each of the
materials, by observing the reflections of students after using the
various materials. The current study shows that beyond qualitative
observations, a quantitative instrument can be used for the purpose
of making differential observations. The most prominent difference
was the clients’ response to working with the markers, compared
to working with the other art materials; notwithstanding, more
refined observations are also possible with this instrument.
The main limitation of this study is that it was not performed
within a therapeutic setting, and thus it did not take into account
the art based intervention when undertaken in the presence of the
treating therapist within the framework of a therapeutic relation-
ship, or in connected series of creative experiences. Thus, future
studies are required in order to test the relevance of the question-
naire and that of its parts when administered to clinical populations
and within a therapeutic framework. One follow up study, which
was recently undertaken in art therapy training program, examined
the effect of the therapist’s presence and the pattern of the client-
therapist relationship on the client’s response to working with the
art materials.
Further studies can serve to confirm the validity of the proposed
factors in the context of a therapeutic setting. Confirmatory Factor
Analysis can examine the associations between the various parts
of the questionnaire and determine their relative significance. Fur-
thermore, follow up studies could focus on the factors comprised
of only two or three items, which were not conclusively validated,
in order to assess their efficacy, comparing the advantages and dis-
advantages of their application. In addition, this instrument was
implemented with the use of basic art materials and individual
work within a group. Finally, it should be noted that the ABI was
based on students’ reflections to art materials. Hence, the ABI’s
validity should be tested with other populations along with more
complex materials and in more complex settings, such as group
work, dyads, or other formats in future research.
In summary, the field of research focused on art-based inter-
vention is still in its infancy, and the possibility of expanding it to
include quantitative assessments of responses to the creative expe-
rience as part of the therapeutic process may open new and varied
directions of investigation. Art therapists often feel that the expe-
rience of artistic creation may be entirely subjective and, hence,
unquantifiable. The current study is an attempt to provide thera-
pists in this field with a new method for assessing and observing
the creative process.
References
Bergin, A. E., & Garfield, S. L. (Eds.). (1994). Handbook of psychotherapy and behavior
change. New York: Wiley.
Betensky, M. G. (1995). What do you see? - Phenomenology of therapeutic art expres-
sion. London: Jessica Kingsley Publishers.
Carolan, R. (2001). Models and paradigms of art therapy research. Art Therapy: Journal
of the American Art Therapy Association, 18(4), 190–206.
Cohen, B. M., Hammer, J. S., & Singer, S. (1988). The diagnostic drawing series: A
systematic. approach to art therapy evaluation and research. The Arts in Psy-
chotherapy, 15(1), 11–21.
346 S. Snir, D. Regev / The Arts in Psychotherapy 40 (2013) 338–346
Cohen, B. M., Mills, A., & Kijak, A. K. (1994). An introduction to the Diagnostic Drawing
Series: A standardized tool for diagnostic and clinical use. Art Therapy: Journal
of the. American Art Therapy Association, 11(2), 105–110.
Deaver, S. P. (2002). What constitutes art therapy research? Art Therapy: Journal of
the American Art Therapy Association, 19(1), 23–27.
Feder, B., & Feder, E. (1998). The art and science of evaluation in the arts therapies.
How do you know what’s working? Springfield, Illinois, USA: Charles C Thomas,
Publisher, LTD.
Fredrickson, B. L. (1998). What good are positive emotions? Review of General Psy-
chology, 2, 300–319.
Fredrickson, B. L. (2009). Positivity. New York: Crown.
Freilich, R. (2010). The contribution of art therapy to the social, emotional, and
academic adjustment of children with learning disabilities. The Arts in Psy-
chotherapy, 37, 97–105.
Gantt, L., & Tabone, C. (1998). The Formal Elements Art Therapy Scale: The rating
manual. Morgantown, WV: Gargoyle Press.
Hazut, T., & Siano, Y. (2008). The Haifa Approach: Observation, assessment and anal-
ysis (in Hebrew). Department of Continuing Education, The Program for Visual-Art
Therapy. Internal Publication of the University of Haifa.
Hesse, P. (1981). Formal analysis form and accompanying explanation sheet for the
formal analysis form. Handout prepared by the Unit for Training Art Therapists.
University of Haifa.
Hesse, P. (1998). A few phenomenological-style, non-verbal, Art Therapy principles. The
Division of Extension Studies and Continuing Education. University of Haifa.
Hinz, L. D. (2008). Walking the line between passion and caution in art therapy:
Using the Expressive Therapies Continuum to avoid therapist errors. Art Therapy:
Journal of the American Art Therapy Association, 25(1), 38–40.
Junge, N. B., & Linesch, D. (1993). Our own voices: New paradigms for art therapy
research. Arts in Psychotherapy, 20(1), 61–67.
Kapitan, L. (2010). Introduction to art therapy research. N.Y. and London: Routledge.
Linesch, D. (1995). Art therapy research: Learning from experience. Art Therapy:
Journal of the American Art Therapy Association, 12(4), 261–265.
Luborsky, L., & Crits-Christoph, P. (1998). Understanding transference: The CCRT
method (2nd ed.). Washington, DC: American Psychological Association.
Lusebrink, V. B. (2004). Art therapy and the brain: An attempt to understand the
underlying processes of art expression in therapy. Art Therapy: Journal of the
American Art Therapy Association, 21(3), 125–135.
Lusebrink, V. B. (2010). Assessment and therapeutic application of the Expressive
Therapies Continuum: Implications for brain structures and functions. Art Ther-
apy: Journal of the American Art Therapy Association, 27(4), 168–177.
Luria, L. (2002). Play and the tension it involves: In the footsteps of Winnicott. In A.
Ferroni (Ed.), Play: The Psychoanalytic Perspective and a Different View (in Hebrew).
Tel-Aviv, Yediot Aharonot.
McNiff, S. (1998). Enlarging the vision of art therapy research. Art Therapy: Journal
of the American Art Therapy Association, 15(2), 86–92.
Moon, C. H. (2010). Materials and media in art therapy. New York: Routledge.
View publication stats
Perl-Dekel, A. (1998). Comparing therapeutic factors of two kinds of group therapy
– verbal therapy vs. expressive-creative therapy (in Hebrew), Thesis submitted
towards a Master’s degree in Education. Haifa: University of Haifa.
Regev, D., & Guttmann, J. (2005). The psychological benefits of artwork: The case of
children with learning disorders. The Arts in Psychotherapy, 32, 302–312.
Regev, D., & Snir, S. (2013). Art therapy for treating children with Autism Spectrum
Disorders (ASD): The unique contribution of art materials. Submitted to The Arts
in Psychotherapy.
Revess-Shenhav, N. (1999). Diagnosing and identifying pictorial characteristics of
Borderline Personality Disorder, MA Thesis. University of Haifa, Department of
Education.
Rosal, M. L. (1998). Research thoughts: Learning from the literature and from expe-
rience. Art Therapy: Journal of the American Art Therapy Association, 15(1), 47–50.
Roth, A., & Fonagy, P. (1996). What works for whom?: A critical review of psychotherapy
research. New York: Guilford.
Rubin, J. A. (1984). The Art of Art Therapy. New York: Brunner/Mazel, Inc.
Rubin, J. A. (1999). Art Therapy – An Introduction. Philadelphia: Taylor & Francis.
Safran, J. D., & Muran, J. C. (2000). Negotiating the therapeutic alliance. New York:
Guilford.don: Jessica Kingsley.
Schaverien, J. (1998). The scapegoat: Jewish experience and art psychotherapy
groups. In J. Campbell, J. Campbell, et al. (Eds.), Art Therapy Race and Culture.
London: Jessica Kingsley.
Shapira, G. (2005). Destruction as material for creating. In H. Davyash (Ed.), Adoles-
cence and injury. Jerusalem: The Summit Institute.
Slayton, S. C., D’Archer, J., & Kaplan, F. (2010). Outcome studies on the efficacy of art
therapy: A review of findings. Art Therapy, 27(3), 108–118.
Snir, S., & Regev, D. (2012). A dialogue with materials: Creators share their experi-
ences. The Arts in Psychotherapy, 40, 94–100.
Snir, S., & Wiseman, H. (2010). Attachment in romantic couples and perceptions of
a joint drawing session. Family Journal, 18, 116–126.
Stiles, W. B., Reynolds, S., Hardy, G. E., Rees, A., Barkham, M., & Shapiro, D. A.
(1994). Evaluation and description of psychotherapy sessions by clients using
the Session Evaluation Questionnaire and the Session Impacts Scale. Journal of
Counseling Psychology, 41(2), 175–185.
Wadeson, H. (1980). Art psychotherapy. John Wiley & Sons.
Wadeson, H. (1987). The Dynamics of Art Psychotherapy. New York: John Wiley &
Sons.
Wiggins, S., Elliott, R., & Cooper, M. (2012). The prevalence and characteristics of rela-
tional depth events in psychotherapy. Journal of Psychotherapy Research, 22(2),
139–158.
Winnicott, D. W. (1995). Playing and Reality (in Hebrew). Tel Aviv: Am-Oved.
Wiseman, H., Barber, J. P., Raz, A., Yam, I., Foltz, C., & Livne-Snir, S. (2002). Parental
communication of Holocaust experiences and interpersonal patterns in off-
spring of Holocaust survivors. International Journal of Behavioral Development,
26(4), 371–381.