Study I.

The ESEA Title IV-C Project:

Developing Critical and Creative Thinking Through Chess

The ESEA Title IV-C federally funded research project was approved for three years
(1979-82). It was extended for one school year (82-83) at local expense for a combined total of
four years. The primary goal of the study was to provide challenging experiences that would
stimulate the development of critical and creative thinking.
The Title IV-C project was an investigation of students identified as mentally gifted with an
IQ of 130 or above. Students in the nonchess groups exceeded those in the chess group in Mean
IQ by 2.3 points, which is not significantly different. All participants were students in the Bradford
Area School District in grades 7 through 9. The individuals sampled in this study could not be
randomly assigned to groups because the students' individualized education plans prescribed
activities based on interests.
The primary independent variables reviewed in this summary are the chess treatment, the
computer treatment, and all nonchess treatments combined. Each group met once a week for 32
weeks in the gifted resource room at Bradford Area High School to pursue its interest area under
the leadership of the Coordinator of Secondary Gifted Education (Robert Ferguson). Most groups
spent a total of 60-64 hours pursuing their preferred activity.
The dependent variables were the differences in the means of the posttests from the
pretests. Data were collected from the Watson-Glaser Critical Thinking Appraisal and the
Torrance Tests of Creative Thinking. The chi square test and the t test were applied to determine
the level of statistical significance.

PRE & POSTTEST SCORES FOR CHESS GROUP

RAW SCORES ON CRITICAL THINKING TEST
90
80
R
70
A
W 60

S 50
C 40
O
R 30
E
20
S
10
0
A B C D E F G H I J K L M N O
STUDENT ID

PRETEST SCORE POSTTEST SCORE

FIGURE 1. A comparison of the pre and posttest scores for the chess
group on the Critical Thinking Appraisal

page 1
 Results and Data Analysis
It is important to note that all scores reported for the Watson-Glaser Critical Thinking
Appraisal (WCTA or CTA) are equivalent raw scores. Watson and Glaser (1964, p. 8) used a
procedure called equi-percentile equating to determine equivalent raw scores. These scores were
all based on norms for high school students and beyond. Since this study was testing junior high
level students and no norms exist for seventh and eighth graders, the project director was forced to
use the high school norms and equivalent raw scores. In some cases pupils in the study actually
scored more correct answers on the posttest than on the pretest and still showed a loss due to the
equivalent raw score procedure.
Inspection of the pre and posttest results in the figure on page one shows that all but one
chessplayer demonstrated gains in raw scores. The average annual increase in equivalent raw
scores for the chess group was 10.53.
The average annual increase in percentile score for the chess group was 17.3%. Nationally,
students who take this test at yearly intervals do not show a gain in percentile ranking. This
comparison shows that the Bradford chess group significantly outperformed the average student in
the country four years in a row!
A 50% score means the student is average in the country for that grade level on the
Watson-Glaser Critical Thinking Appraisal. A score of 99% means the student is one of the best
critical thinkers in that grade for the skills assessed by the Watson-Glaser Critical Thinking
Appraisal. A Student who scores in the 50th percentile in 1979 and who continues to perform in
average fashion, will score in the 50th percentile in 1980. An increased percentile score indicates
an above average performance.
Percentile scores are inappropriate for statistical analysis. In order to have an appropriate
metric, the percentile scores were converted to equivalent raw scores.
The t test was used to test statistical significance of the gains on the Watson-Glaser Critical
Thinking Appraisal. The t test measures the quantity of the gain to assess whether it is significant.

TABLE 1. Dependent t test evaluating significance of gains on the

Critical Thinking Appraisal (CTA) by chessplayers

VARIABLE NUMBER MEAN

Pretest Scores 15 62.80
Posttest Scores 15 73.33
Difference Standard t value
Error
10.53 2.2 4.786
Significant beyond the .001 level

page 2
 Table 1 on the preceding page demonstrates that the chessplayers achieved a very
significant gain (p < .001) from the pretest to the posttest in critical thinking skills as measured by
the Watson-Glaser Critical Thinking Appraisal. The level of significance tells us that there is less
than one possibility in a thousand that this result could have occurred by chance.
Just as the dependent t test illustrated above is extremely significant, so too is the
independent t test illustrated in Table 2, which indicates that the chess group's performance is
notably superior to that of the nonchess group's. The results, which are statistically significant at
the .001 level, are shown in Table 2.

TABLE 2. Independent t test evaluating significance of difference on

the Watson-Glaser Critical Thinking Appraisal between the
chessplayers and nonchessplayers

VARIABLE NUMBER MEAN

Nonchess Group Gains 79 1.86
Chess Group Gains 15 10.53
Difference Standard t value
Error
8.67 2.4 3.61
Significant at the .001 level

The data were also evaluated using a nonparametric, or distribution-free, test of

significance. For Study I, the chi square test of statistical significance was used to evaluate the
gains/losses on the Watson-Glaser Critical Thinking Appraisal. The chi square test evaluates the
significance of the number of chessplayers demonstrating gains on the CTA compared to the
number of non-chessplayers showing gains. Because the chi square test is nonparametric, it is
insensitive to the size of gains; it considers a gain of one point in the same manner as a gain of 30
points or 100 points.
The chess group was compared to the nonchess group, the computer group, and the
nonparticipants. The chi square test results ranged from marginally significant at .072 to very
significant at .002. A complete listing of the chi square test results may be found in Table 3 on the
next page.
Particular attention should be given to the results comparing the gains of the eighth graders
on the CTA. These are perhaps the most significant of all the critical thinking results because eighth
graders comprised 46% of the total number of students participating in the project. Out of a total
of ninety-four pupils who completed both the pre and posttests, forty-three were eighth graders.
Because this was the largest grade sample, it becomes more statistically important and increases our
level of confidence in the results.

page 3
TABLE 3. Statistical summary for CTA

TABLES t Test Chi Square X2

p< p<
MALES & FEMALES COMBINED:
Chess Group 0.001
Chess vs. Nonchess 0.001 0.008
Chess vs. Computer 0.003 0.008
Chess vs. Nonparticipants 0.025 0.002

MALES:
Chess Group 0.003
Chess vs. Nonchess 0.072 0.056
Chess vs. Computer 0.017 0.023

FEMALES:
Chess Group 0.043
Chess vs. Nonchess 0.085 0.071
Chess vs. Computer 0.195 0.104

ALL 8TH GRADERS:

Chess Group 0.003
Chess vs. Nonchess 0.006 0.009
Chess vs. Computer 0.142 0.05

In a Fidelity Electronics' article entitled "The Minds of Tomorrow" (1993), the company
states: "In light of chess playing's ability to shape future minds, schools all across the United States
view chess as a powerful educational tool. Thousands of pre-teens and teens understand that chess
coheres the mind to anticipate, make decisions, and react in a way no other game can."
Dr. R.J. Topping (1988), the Coordinator of the Gifted/Talented Programs for the White
Plains Public Schools, agrees with Fidelity and states:
Chess is an integral part of the logic and creative problem-solving segment of our
More Able Student curriculum. It cultivates critical thinking skills in our students,
enhancing their personal growth and academic learning. We encourage other
school systems to consider offering their students experiences in this dynamic
content area (Chess in the Schools, 1988, p. 3).
Many teachers use chess as a vehicle to teach critical thinking skills. They stress to students
that learning how to think is more important than learning the solution to a specific problem.
Through chess, pupils learn how to invent creative solutions to problems. They learn how to
analyze a situation by focusing on the important factors. Chess is effective because it is
self-motivating. The game is intrinsically fascinating, and the goals of attack and defense, climaxing

page 4
in checkmate, motivate young people to delve deep into their mental resources (Chess in the
Schools, 1988, p. 2).
The next portion of the results and data analysis summary reviews the different aspects of
creativity tested in this research: fluency, flexibility, and originality.
Verbal fluency is an individual's ability to generate a large number of ideas with words. Chessplayers
often have a running dialogue within their minds reviewing the checklist for important strategic and
tactical factors or mentally calculating: "If I go there, then he'll move . . ."
Flexibility represents a person's ability to produce a variety of types of ideas, to shift from one approach to
another, or to use a variety of strategies. Originality is skill at producing ideas that are different from the obvious.

Torrance (1974) defined creative thinking as: "a process of becoming sensitive to
problems, deficiencies, gaps in knowledge, missing elements, disharmonies, and so on; identifying
the difficulty; searching for solutions, making guesses, or formulating hypotheses about the
deficiencies: testing and retesting these hypotheses and possibly modifying and retesting them; and
finally communicating the results."
It is important to note that all scores reported for the Torrance Tests of Creative Thinking
are standard T-scores. All raw scores were converted in accordance with the recommendations in
the Torrance Tests of Creative Thinking Norms-Technical Manual (1974, pp. 48, 56). These
scores were all based on creative thinking norms established for junior high school students.

INCREASE IN CREATIVITY
COMPARISON BETWEEN CHESS AND NONCHESS
59.78
60

50

40
G
A
I 30
N 24.206
S 18.5619.48
20

10
5.568
3.089
0
Chess Group Other Groups
THREE ASPECTS OF CREATIVITY
Fluency Flexibility Originality

FIGURE 2. A comparison of the chess group gains to the nonchess

group gains

page 5
 Creativity is a major aspect of chess at the master level, but can chess influence creativity at
the amateur level? Summary Table 4 sheds some light on this question. It would appear from the
data collected and the statistical test results listed in the table below that there can be little doubt
that chess does enhance creativity in gifted adolescents. Dr. Stephen Schiff's claim that creativity
can be taught through the art of chess has been confirmed.
While the entire chess group made superior gains over the other groups in all three areas,
the aspect that demonstrated the most significant growth was originality. It should be noted that
several researchers have found that gains in originality are usual for those receiving creativity
training, whereas gains in fluency are often slight or nonexistent. The fact that the chess group's
gains in fluency were significant beyond the .05 level when compared to the national norms is an
important discovery.
It appears that chess is superior to many currently used programs for developing creative
thinking and, therefore, could logically be included in a differentiated program for mentally gifted
students.

TABLE 4. Statistical summary of t tests on Creativity

TABLES FLUENCY FLEXIBILITY ORIGINALITY

p< p< p<
MALES & FEMALES COMBINED:
Dependent Chess 0.077 0.024 0.01
Population Mean Chess vs. Norms 0.039 0.002 0.001
Independent Chess vs. Nonchess 0.049 0.05 0.018
Independent Chess vs. Computer 0.038 0.08 0.022
MALES:
Dependent Chess 0.142 0.03 0.016
Population Mean Chess vs. Norms 0.07 0.008 0.003
Independent Chess vs. Nonchess 0.039 0.007 0.002
Independent Chess vs. Computer 0.076 0.018 0.007
ALL 8TH GRADERS:
Dependent Chess 0.32 0.088 0.018
Population Mean Chess vs. Norms 0.171 0.037 0.019
Independent Chess vs. Nonchess 0.305 0.061 0.009
Independent Chess vs. Computer 0.606 0.12 0.027
ALL 8TH GRADE MALES:
Dependent Chess 0.32 0.088 0.018
Population Mean Chess vs. Norms 0.171 0.037 0.019
Independent Chess vs. Nonchess 0.383 0.014 0.006
Independent Chess vs. Computer 0.561 0.107 0.02

page 6
 Conclusions
It is evident from the above tables and data that chess had a definite impact on developing
both critical and creative thinking skills. Because the sample size of the treatment group was only
15 students, the author would encourage replication of this study using a larger N.
It was also evident that there were significant gains in the participants' chess skills. Six of
the pupils involved in this study participated in the annual Pennsylvania State Scholastic
Championship beginning in 1980. Three of those six excelled. Two of the boys became candidate
masters and one of the girls made the top 50 list for all women chessplayers in the United States.
The project director concurs wholeheartedly with Dr. Stephen M. Schiff (1991), who
wrote: ". . . the study of chess is one of the most critically important additions to the curriculum
that schools can offer to our pre-adolescent gifted and talented student population." Based on the
results of Study I and others, this researcher urges the inclusion of chess in the curriculum to
augment the skills of the mentally gifted.
The USA Junior Chess Olympics Training Program used in each of Ferguson's studies
undeniably demonstrated effectiveness in bringing about the desired changes in the participating
students. This author would strongly recommend the adoption or adaptation of the USA Junior
Chess Olympics Training Program within the school curriculum throughout the country.

For Those Who Haven't Studied Statistics

"Tradition holds that the level of significance must be expressed as the probability that a
true null hypothesis is being rejected. That means that the lower the significance level, the higher
is our confidence that the effect we have observed is real." (Phillips, Statistical Thinking: A
Structural Approach, p. 85, 1973)
Some researchers hold that a probability of .1 (10%) is significant; however, in this study
and Ferguson's other research, a significant difference is equal to or less than .05 (often written
p<.05). A very significant difference is one for which the probability of having occurred by
sampling error is less than 1% (.01) and is frequently written p<.01. In the statistical summary
(Table 4), the significant and very significant levels have been bolded.

For Additional Information

The preceding material is a brief synopsis of the information found in a paper (200+ pages)
by Robert Ferguson entitled Teaching the Fourth "R" (Reflective Reasoning) Through Chess. If
you would like a more comprehensive review of this research and his other studies, send a check
for $39.95 payable to the American Chess School at the address below. All profits from the sale
of this publication are used to support chess in the schools.
American Chess School
140 School Street
Bradford, PA 16701
814-368-8009
www.amchess.org

page 7