Prediction of GPA with Educational Psychology Grades and Critical Thinking Scores
Gadzella, Bernadette M., Baloglu, Mustafa, Stephens, Rebecca, Education
Research studies have shown significant relationships between different measures and academic achievement (GPA). For instance, Steward and Al-abdula (1989) reported relationships between critical thinking and academic performances for 237 undergraduates. These researchers indicated that, in general, students who scored high on the Watson-Glaser Critical Thinking Appraisal, WGCTA (Watson & Glaser, 1980a) also had high GPAs. These researchers indicated that not all aspects of the critical thinking abilities (of the WGCTA) contributed equally to the overall academic performance. They reported that Inference, Interpretation, and Evaluation of Arguments (WGCTA subscales) correlated significantly with the GPA, but that only Inference and Evaluation of Arguments scores contributed (9.2%) to the variance within the GPAs.
In a study of three groups (25 subjects in Elementary Education, 25 subjects in Special Education, and 10 subjects in Speech Correction) of preservice professionals, Holmgren and Coven (1984) reported relationships among six subtests of the Survey of Inter-Personal Values (SIV), total WGCTA score, grade-point averages (GPA), English Proficiency test scores (EP), and age of the participants. To determine the best predictors of EP and GPA, they computed stepwise multiple regression equations and found that the total WGCTA scores and age were the best predictors of the students' EP and GPA. However, when only one criterion was studied at a time, the multiple correlation coefficient between GPA and WGCTA was R = .50. Thus (R- = .25), 25% of the total variance in GPA was accounted for by the total WGCTA scores. The correlation coefficient between GPA and EP was R = .46, (R- = .21) accounting for 21% of the variance.
The above studies (Steward & Al-abdulla, 1989; Holmgren & Coven, 1984) provided some information of how critical thinking scores are related and predict GPA. Students' overall GPA is an average of their course grades. One course, Educational Psychology, is required for teacher certification in most teacher training institutions of higher learning. The content covered in Educational Psychology provides the foundation in using sound strategies and constructive thinking skills, which are useful in the teaching career.
The purpose of the present study was to determine the relationships between the Educational Psychology four test scores, course grades, and the critical thinking (using WGCTA subscales and total) scores with students' GPA. In addition, the aim was to determine which of these variables were the best predictors of GPA for students majoring in Education.
There were 114 students (majoring in Education), enrolled in the Educational Psychology course at a southwestern state university, who participated in the study. Generally, the course is taken in the students' junior or senior year. However, because Educational Psychology is also a required course in teacher certification, graduate students who wanted to teach were required to take the course. In this study, there were 24 (21.1%) were men and 90 (78.9%) women, 14 (12.3%) were sophomores, 69 (60.5%) juniors, 11 (9.6%) seniors, 17 (14.9%) graduates, and three students did not report their college status. Their ages ranged from 19 years to 54 years (M = 27.80, SD = 8.12).
Numerous studies have used the Watson-Glaser Critical Thinking Appraisal, WGCTA (Watson & Glaser, 1980a) in determining critical thinking at the post secondary level (McMillan, 1987). It was chosen in the present study to determine students' critical thinking abilities. The WGCTA is a paper and pencil questionnaire that has 80 items from which five subscales (16 items each) are derived (Watson & Glaser, 1980b). The Inference subscale discriminates among the degree of troth and falsity of inference drawn from the provided data. The Recognition of Arguments subscale recognizes unstated assumptions on presuppositions in given statements or assertions. The Deduction subscale determines whether certain conclusions necessarily follow from the provided information. The Interpretation subscale weights evidence and determines if generalizations based on given data are warranted. The Evaluation of Arguments subscale distinguishes between arguments that are strong and relevant and those that are weak or irrelevant to particular questions. The Total Critical Thinking score is the summation of the five subscale scores.
Psychometric properties of the WGCTA have been reported previously. The split-half reliability correlated for 10 norm groups ranged from .69 to .85, test-retest reliability for 96 students' responses was .73, and alternate-form reliability for 228 students' responses to Forms A and B of the WGCTA was .75 (Watson & Glaser, 1980b). In a recent study (Gadzella, Baloglu, & Stephens, 2001), analyses on the WGCTA showed that the internal consistency was .86 for 135 students (majoring in Education), .91 for 30 men, and .83 for 105 women. The split-half reliability for the same group was .65 for the total group, .87 for men, and .55 for women. The concurrent validity for the 135 students' course grades and the total WGCTA scores was r = .42 (p < .001). Using a confirmatory factor analysis, construct validity of the WGCTA (based on the theoretical model consisting of the five subscales) was supported (Normed Fit Index = .97 and Goodness-of-Fit Index = .99, Root Mean Square Error of Approximation = .03). Of the five subscales, Interpretation and Deduction subscale scores had the highest factor loadings (.87 and .68, p < .001, respectively) onto the total WGCTA.
In the present study, four objective-type tests in Educational Psychology were administered. The test items were taken from the Assessment Package for Woolfolk's Educational Psychology (Linden, 1998), which accompanies the text, Educational Psychology (Woolfolk, 1998) that was used in the class. The four tests included areas covering: Research Methods and Social and Moral Development (Test I), Cognitive Development and Behavioral Learning (Test II), Cognitive Learning (Test III), and Measurement and Evaluation (Test IV). To determine the students' Educational Psychology course grades, the four tests were averaged.
Subjects responded to the WGCTA, Form A, and the four Educational Psychology tests during class periods. They signed research release forms indicating that their WGCTA scores, course grades, and grade-point averages (GPA) may be used for research studies. They were given bonus points for their participation. The subjects' GPAs were retrieved from the university records. Means, standard deviations, intercorrelations, and simple and multiple regression equations were used to analyze the data.
The means, standard deviations, and intercorrelations among the subscales and total WGCTA scores and the GPA are presented in Table 1. All subscales and total WGCTA scores correlated significantly (p < .05) with GPA, except Recognition of Assumptions and Evaluation of Arguments subscales. Inference and Recognition of Assumptions scores did not correlate significantly (p > .05) with Evaluation of Arguments scores (Table 1).
To determine which of the WGCTA subscales best predicted GPA, a simultaneous multiple regression analysis was computed. The regression equation computed, using all of the WGCTA subscales, was Y = 2.3 + .03 (Inference) + .003 (Recognition of Assumptions) + .03 (Deduction) + .06 (Interpretation) - .03 (Evaluation of Arguments). The equation significantly predicted GPA [F (5, 113) = 4.3,12 < .001] and accounted for 17% of the total variance. However, out of the five subscales, only Interpretation (t = 2.41, p < .02) and Evaluation of Arguments (t = -1.97, p < .05) subscales significantly predicted GPA.
The means, standard deviations, and intercorrelations among the Educational Psychology test scores and course grades, and the GPAs are presented in Table 2. All test scores, course grades, and the GPA correlated significantly (p < .01) among each other (Table 2). To determine which of the test scores best predicted the GPA, a simultaneous multiple regression analysis was computed. The model, Y= -.01 + .01 (Test I) + .003 (Test II) + .005 (Test III) + .02 (Test IV), significantly predicted the GPA [F (4. 110) = 16.50, p < .0005]. It accounted for 36% of the total variance in the GPA. Out of the four test scores, Test IV (t = 3.44, p < .001) and Test 1 (t = 2.79, p < .01) were the only significant predictors of the GPA.
To determine whether Educational Psychology course grades and the total WGCTA scores predicted the GPA, another simultaneous multiple regression analysis was computed. The model, Y = -.01 + .03 (Course Grades) + .01 (WGCTA), significantly predicted the GPA [F (2, 110) = 31.58, p < .0005], accounting for 37% of the total variance. When only the Educational Psychology course grade was used as a predictor, the simple regression equation, Y = .7 + .4 (Course Grades), significantly predicted 36% of the variance in the GPA [F (1, 110) = 60.8, p < .0005]. When only the WGCTA total score was used as a predictor, the equation, Y = 2.37 + .01 (WGCTA), significantly predicted the GPA [F (1, 113) = 9.49, p < .003]; however, it accounted for less than one percent (.08%) of the variance.
The aims of the present study were to determine (a) the correlations between the Watson-G laser Critical Thinking Appraisal (WGCTA) scores and Educational Psychology course grades with GPAs, and (b) which variables best predicted the GPAs for students enrolled in Education. Two studies (Steward & Al-abdulla, 1989; Holmgren & Coven, 1984) were found that used the WGCTA to predict the subjects' GPAs.
In a study of 237 students' responses, Steward and Al-abdulla (1989) reported that some WGCTA subscales (Inference, Interpretation, and Evaluation of Arguments) correlated significantly with the GPA for students with high GPAs. However only Inference and Evaluation of Arguments (WGCTA subscales) contributed significantly (9.2%) to the variance in the GPA. In the present study, for 114 students majoring in Education, the total WGCTA and three subscale (Inference, Deduction, Interpretation) scores correlated significantly with the GPA; but, only Interpretation and Evaluation of Arguments accounted for 17% of the variance. When only the total WGCTA score was used to predict the GPA, it accounted for only .08% of the variance.
With three special preservice professional groups, Holmgren and Coven (1984) reported that the total WGCTA scores predicted 25% (R = .50, [R.sup.2] = .25) of the variance in the students' GPAs. However, the results in the present study showed that the total WGCTA accounted for less than one percent (.08%) of the variance in the GPA. This is quite different from that reported by Holmgren and Coven (1984). What might be the difference? If one looks at the data provided for the three groups in Holmgren and Coven's study, the students' means on the WGCTA ranged from 70.2 to 74.6 and standard deviations ranged from 6.2 to 11.4. These data are quite different from that reported in the present study (M = 51.42, SD = 9.84).
Although the WGCTA is a frequently used inventory in determining students' critical thinking abilities at the postsecondary level (McMillan, 1987), it is a long test and requires time to respond to it accurately. It is suggested that if the WGCTA is used in future studies, that participants be encouraged to take time and do their very best in every section of the inventory.
No study was found in which Educational Psychology test scores and course grades were used to predict the students' GPAs majoring in Education. In the present study, four objective tests and course grades were used to determine whether they predicted the students' GPAs. The Research Methods and Social and Moral Development (Test I) and Measurement and Evaluation (Test IV) significantly predicted the GPA. However, when the course grade (average of the four tests) was used as a predictor of the GPA, it predicted 36% of the total variance. When the course grade and the total WGCTA score were used together, they accounted for 37% of the variance of the GPA.
In examining the information provided, one notes that the Educational Psychology course grade uniquely accounted for over one-third (36%) of the variance in the GPAs for students majoring in Education. This information should be most valuable to administrators and instructors who teach Educational Psychology. The content of the course serves as a foundation for other Education courses. It offers the types of strategies and thinking abilities, which can be transferred to other courses and the teaching careers in general.
Table 1 Means, Standard Deviations, and Intercorrelations Among the Subscales and Total WGCTA Scores and GPA WGCTA Subscales Variables 1 2 3 4 1. Inference 2.Recognition of Assumptions .20 ** 3. Deduction .21 ** .24 ** 4. Interpretation .32 ** .24 ** .57 ** 5. Evaluation of Arguments .14 ** .18 ** .38 ** .50 ** Total WGCTA .54 ** .65 ** .69 ** .75 ** GPA .23 ** .13 ** .28 ** .34 ** Means 8.44 10.61 10.08 11.24 Std. Deviations 2.72 14.05 12.57 2.52 WGCTA Subscales Variables 5 Total GPA WGCTA 1. Inference 2.Recognition of Assumptions 3. Deduction 4. Interpretation 5. Evaluation of Arguments Total WGCTA .66 ** GPA .02 ** .28 ** Means 11.05 51.42 3.12 Std. Deviations 3.17 9.84 0.51 * p < .05 ** p < .01 Table 2 Means, Standard Deviations, and Intercorrelations Among the Four Objective Tests, Course Grade, and GPA Educational Psychology Test Variables Test I Test II Test III Test II .46 ** Test III .53 ** .48 ** Test IV .47 ** .50 ** .52 ** Course grade .80 ** .78 ** .79 ** GPA .50 ** .39 ** .45 ** Means 82.35 85.03 85.47 Std. Deviations 11.66 11.00 9.80 Educational Psychology Test Variables Test IV Course GPA Grade Test II Test III Test IV Course grade .77 ** GPA .52 ** .60 ** Means 86.89 84.97 3.12 Std. Deviations .9.27 8.22 0.51 ** p < .01
Gadzella, B. M. Baloglu, M., & Stephens, R. (2001). Validity and reliability of the Watson-Glaser Critical Thinking Appraisal for teachers. Paper presented at the annual meeting of the Southwestern Psychological Association, Houston, TX.
Holmgren, B. R., & Coven, T. M. (1984). Selective characteristics of preservice professionals. Education, 104 (3), 321-328.
Linden, K. (1998). Assessment package for Woolfolk Educational Psychology, (7th Ed.), Boston: Allyn and Bacon.
McMillan, J. H. (1987). Enhancing college student critical thinking: a review of studies. Research of Higher Education, 26, 3-29.
Steward, R., & Al-abdulla, Y. (1989). An examination of the relationships between critical thinking and academic success on a university campus. ERIC Document Reproduction Service No ED 318936.
Watson, G., & Glaser, E. (1980a). Watson-Glaser Critical Thinking Appraisal. San Antonio, TX: Psychological Corporation.
Watson, G., & Glaser, E. (1980b). Watson-Glaser Critical Thinking Appraisal, Manual, San Antonio, TX: Psychological Corporation.
Woolfolk, A. (1998). Educational Psychology, 7th Ed., Boston: Allyn and Bacon.
BERNADETTE M. GADZELLA, MUSTAFA BALOGLU, AND REBECCA STEPHENS Psychology and Special Education Texas A&M University Commerce, TX 75429…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Prediction of GPA with Educational Psychology Grades and Critical Thinking Scores. Contributors: Gadzella, Bernadette M. - Author, Baloglu, Mustafa - Author, Stephens, Rebecca - Author. Journal title: Education. Volume: 122. Issue: 3 Publication date: Spring 2002. Page number: 618+. © 1999 Project Innovation. COPYRIGHT 2002 Gale Group.
This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.