Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13
Topics
Discussion
Announcements
 

What Are the Thinking Styles of Turkish Student Teachers?


by Seval Fer - 2007

Background: The Thinking Styles Inventory—developed by Sternberg and Wagner based on Sternberg's (1988, 1997) earlier theory of mental self-government—was selected for the research in order to assess thinking styles of student teachers. Another reason is that the theoretical constructs, as well as the inventory generated from the theory, have proved to be valuable to assess thinking styles of people in several studies. For example, previous research reported that students differed in their thinking styles depending on their personal characteristics (e.g., Sternberg & Grigorenko, 1995; Zhang, 1999, 2002b, 2001e). However, research found no distinct patterns between thinking styles and personal characteristics of students (e.g., Grigorenko & Sternberg, 1997; Zhang, 2002a). Little research has focused on the study of non-Western students' thinking styles. Thus, it is reasonable to expect that the thinking styles could also be identified among student teachers in Turkey since the styles are differentiated within a different country and culture.

Objectives of Study: (1) What is the validity and reliability of the TSI in assessing TS among Turkish student teachers? (2) Is the structure of a factor analysis of the TSI consistent with the five dimensions postulated by the theory of mental self-government in a Turkish sample? (3) Are TS of student teachers differentiated based on such socialization variables as gender, age, educational level, type of university attended, and field of study followed?

Research Design: The study reported here used quantitative methods with a survey sampling design.

Subjects: The research subjects were comprised of 402 student teachers enrolled in English, mathematics, and science teaching programs at Yildiz Technical University in Istanbul, Turkey.

Results: The results of factor analysis for construct validity of the inventory addressed 13 subscales under the five dimensional constructs with 104 items. Moreover, the total internal reliability of scale was a .92 reliability coefficient. Findings demonstrated that the 13 subscales had internal consistency reliabilities; Cronbach’s alphas of the 13 subscales for these subjects ranged from .61 to .91, as well as item-scale correlation that ranged from .37 to .88. Test re-test reliability for external reliability of subscales was between .63 to .78. When the ANOVA findings of the research is evaluated as a whole, it might be said that the student teachers’ particular thinking styles were differentiated by the socialized variables of gender, age, type of university attended, as well as field of study followed.

Conclusions/Recommendations: In order to measure thinking styles of students based on a model of broad intellectual styles, the inventory might be used as an efficient instrument. Moreover, the results of this study demonstrated a diversity of thinking styles among the participants. However, further research is needed to clarify the nature of thinking styles as assessed by the inventory at different educational levels and culture to facilitate a better understanding of the thinking styles of students.



INTRODUCTION


The concept of styles has been employed widely for many years in psychological and educational literature to identify individual differences in human performance. In this regard, an important current topic in research is the identification of different style constructs. There have also been various efforts to assess several dimensions of styles based on which constructs have been popular. For example, Sternberg and Grigorenko (1997) propose three general approaches to stylistic aspects of constructs in the style literature. According to them, the first approach is cognition-centered, which deals with individual differences in cognition and perception, resulting in the identification and description of several abilities and cognitive processing. Cognitive styles might be used to illustrate how one organizes certain information. The sample of cognitive styles is the field dependence-independence theory of Witkin, the wholist-analytic and the verbaliser-imager of Riding and Cheema. The second is personality-centered, which is associated with environmental and process-based issues related to addressing individual differences between people. Myers-Briggs type indicator is a sample of the personality style. The third approach is activity-centered, dealing with learning styles, which focuses on various activities, settings, and environments. This approach focuses on the activities that might arise from aspects of cognition and personality. Learning styles might be used to describe how one prefers to learn about certain information. The sample of activity styles is Kolb’s experiential learning style and Grasha-Riechmann’s student learning style. According to Dai and Feldhusen (1999), the theory of mental self-government concerns personality-functioning rather than merely cognitive preferences. On the contrary, Zhang (2002b, 2002f) states that the Thinking Styles Inventory is more cognition-oriented. Sternberg (1994) contended that thinking style is at the interface between intelligence and personality. In this study, the Thinking Styles Inventory is accepted and selected to be used in the current research in a cognition-oriented sense since it refers to a particular and preferred way of thinking, or cognition via its intellectual functioning.


The Thinking Styles Inventory was developed by Sternberg and Wagner (Sternberg, 1988, 1997),  which is based on the earlier theory of mental self-government formulated by Sternberg (1988, 1997), representing stylistic aspects of intellectual functioning. The basic assumption of the theory is that people, like government and societies, have preferences in how they use their skills and how they govern their thought processes to manage their everyday activities, organize or govern themselves and their mental processes, and establish systems and organizations for this governance. According to the theory, mental self-government is used to portray how the human mind works. Thus, there are different ways of managing people's mental activities, within the school and without. These different ways of managing people's activities are defined as thinking styles (Sternberg, 1988, 1994, 1997; Sternberg & Grigorenko, 1997). Styles are thought to be distinct from abilities, involving preferences—not necessarily conscious ones—in the use of whatever abilities one has. These are not connected solely with ability, but rather, are a preferred way of expressing or using one or more abilities. (Armstrong, 2000; Cano-Garcia & Hughes, 2000; Grigorenko & Sternberg, 1997; Sternberg, 1997; Sternberg & Grigorenko, 1993, 1995, 1997; Sternberg & Lubart, 1991a, 1991b; Zhang, 2000; Zhang &  Sternberg, 1998, 2000). The theory proposes 13 thinking styles under five dimensions of mental self-governance: functions, forms, levels, scope, and leanings. In Appendix A, a brief description of the key characteristics of each thinking styles is described.


FUNCTIONS OF MENTAL SELF-GOVERNMENT


As with the different functions of government, people also have three different functions for focusing on different tasks: legislative, executive, and judicial. A person with a legislative style likes being engaged in tasks with creating, imagining, and planning problem solutions and generating new approaches and solutions. An individual with an executive style is concerned with the proper implementation of tasks within a set of guidelines, following rules, figuring out known ways. A person with a judicial style is concerned with judging, analyzing, evaluating, and comparing the process and products of other people's activities.


FORMS OF MENTAL SELF-GOVERNMENT


Just as there are different forms of government, individuals also have four different forms for focusing on different tasks: monarchic, hierarchic, oligarchic, and anarchic. People with a monarchic style deal with a single goal or a way of doing one thing at a time, whereas people with a hierarchic style focus on many prioritized tasks, have a good sense of priorities, and tend to be systematic. People with an oligarchic style prefer to work on several tasks and multiple goals in the same time without setting any priority. Finally, people with an anarchic style enjoy dealing with tasks and situations, which are unstructured, often seem intolerant or unaware of the rules and regulations, and tend to resist authority.


LEVELS OF MENTAL SELF-GOVERNMENT


As in different levels of government such as, federal, state, county, and city, people also have two different levels for focusing on different tasks: global and local. An individual with a global style prefers to deal with problems general in nature, which require abstract thinking about the overall picture of an issue. On the contrary, a person with a local style prefers concrete problems and deals with specific aspects of an issue and its specific details.


SCOPE OF MENTAL SELF-GOVERNMENT


Just as there are different scopes of government such as dealing both with internal, or domestic affairs, similarly, individuals also have two different scopes for focusing on different tasks: internal and external. An individual who is internal in style tends to be introverted, task-oriented, aloof, and works alone and independently. In contrast, a person with an external style prefers to deal with tasks that allow interaction with other people.


LEANINGS OF MENTAL SELF-GOVERNMENT


As for the ways in which government encompasses different political leanings ranging from the most conservative to the most liberal, people might lean in two different ways in order to focus on different tasks: conservative and liberal—referring to styles, not politics. A person with a conservative style likes to adhere to existing rules and procedures, minimizes change, and avoids ambiguous situations when possible. In contrast, an individual with a liberal style enjoys being engaged in tasks that involve novelty, ambiguity, and ambiguous or uncertain situations. (Bernardo, Zhang, & Callueng, 2002; Sternberg & Grigorenko, 1993; Sternberg, 1994, 1997; Zhang, 1999, 2001d).


Curry (1990) states that cognitive style is unchangeable. However, Allison and Hayes (1997; as cited in Hill, Puurula, Sitko-Lutek, & Rakowska, 2000) indicate cognitive style might be acquired through socialization. Moreover, Hill et al. (2000) state that the effects of cultural conditioning are agreed to be internalized through both family influences and the educational system. Similarly, thinking styles are modified, at least in part, by the environment in which people live. Styles are also thought, at least by some, to vary across context and specific tasks, as well as developmental periods, and to be socialized by the predominant culture. People have a style profile, meaning they show varying amounts of each style, but are not locked into any one profile. People choose styles of managing themselves with which they are comfortable (Grigorenko & Sternberg, 1997; Sternberg, 1988, 1994, 1997; Sternberg & Grigorenko, 1993, 1995, 1997). As Sternberg (1997) emphasizes, no single style is bad, and the healthy classroom, or society, creates an environment in which all styles flourish. Nevertheless, he does not mean to say that everyone should adapt by developing the same and most socially accepted mode.


The Thinking Styles Inventory was selected for the current research in order to assess the thinking styles of student teachers. Another reason is that the theoretical constructs, as well as the inventory generated from the theory, have proved to be valuable in order to assess thinking styles of people in several studies. For example, Grigorenko and Sternberg (1997) indicated that students' performance was associated with not only their levels and types of abilities, but also with their judicial, executive, and legislative thinking styles. Researchers also found in the U.S. and Chinese samples that thinking styles statistically predicted academic achievement of students beyond abilities (e.g., Grigorenko & Sternberg, 1997; Zhang & Sternberg, 1998). According to Cano-Garcia and Hughes (2000), students who obtained the best results preferred to think and learn in a certain way, as well as the academic achievement of students being influenced by their thinking and learning styles. O’Hara and Sternberg (2000) revealed that people with a legislative thinking style showed higher creative performance, whereas people with a judicial thinking style indicated lower creative performance when not given any special instructions. Another study (see Sternberg, 1994) indicated that most teachers were best at teaching children who matched their own styles of thinking and learning. Sternberg and Grigorenko (1993, 1995) found that teachers tended to overestimate the extent to which their students share their own styles, and that students in fact received higher grades and evaluations that were more favorable when their styles more closely matched those of their teachers. Sternberg (1994) found that teachers showed some differences in styles across subject-matter areas. For example, humanities teachers tended to be more liberal than science teachers were, and science teachers tended to be more local than humanities teachers were. Sternberg (1994) clearly suggests that different types of assessment methods favor different types of thinkers.


However, previous findings are mixed when attempting to relate the effects of student characteristics such as age, gender, study field, and university class level on thinking styles. For example, Sternberg and Grigorenko (1995) found significant relationships between the teaching styles and the grade taught, the length of the teaching experience, and the subject area taught. Moreover, students differed in their thinking styles depending on their personal characteristics, such as age, birth-order, gender, and their learning environments. Zhang (1999) also revealed significant relationships between creativity-relevant styles and student characteristics of age, work experience, and travel experience. However, participants' thinking styles were not significantly different in terms of the following variables: birth order, college class level, field of study, marital status, gender, type of school attended, and parents' educational level. Furthermore, Zhang (2002b) reported that thinking styles of university students varied as a function of both their personal characteristics such as age, gender, and socio-economic status, as well as their situational characteristics, such as work, travel, and leadership experiences. In another study, Zhang (2001e) indicated that particular thinking styles were related to age, leadership experience, travel experience, and the number of hobbies undertaken. However, Grigorenko and Sternberg (1997) found no distinct patterns of particular thinking styles among gifted students of high school in terms of abilities, gender, or grade. Similarly, Zhang (2002a) reported no statistically significant difference between thinking styles and age, gender, and the previous work experience of university students.


In the theory of mental self-government, little research is focused on the study of non-Western students' thinking styles. The non-Western studies using this theory were carried out in Hong Kong, mainland China, and the Philippines (e.g., Bernardo et al. 2002; Zhang, 1999, 2000; Zhang & Sachs, 1997; Zhang & Sternberg, 2000). Results of these studies are summarized briefly. First, students differ in their thinking styles. However, students’ thinking styles vary depending on their personal characteristics. Second, the reliability and validity of the Thinking Styles Inventory has been demonstrated satisfactorily, and has shown a number of similarities across different cultures. Moreover, Sternberg (1997) suggests that thinking styles are common characteristics across culture, environments, and situations. Furthermore, he has contended that thinking styles are at least partially socialized. However, the results of these studies reported previously have been mixed concerning the effects of student characteristics on thinking styles. Thus, it is reasonable to expect that the thinking styles defined by Sternberg's theory could also be identified among student teachers in Turkey since thinking styles are differentiated within a different country and culture. That is why in the current study, the following research questions were addressed:


1. What is the validity and reliability of the Thinking Styles Inventory in assessing thinking styles among Turkish student teachers?


2. Is the structure of a factor analysis of the Thinking Styles Inventory consistent with the five dimensions postulated by the theory of mental self-government in a Turkish sample?


3. Are thinking styles of student teachers differentiated based on such socialization variables as gender, age, educational level, type of university attended, as well as the field of study followed?


METHODS


PARTICIPANTS


The research subjects comprised 402 student teachers who were enrolled in English, mathematics, and science teaching programs at Yildiz Technical University, Istanbul, Turkey. All students entering these programs were selected as participants for this research. These students were chosen as being convenient for the research objectives since these programs are open to individuals from different universities, fields, and backgrounds, which would allow the thinking styles of these diverse students to be identified.


The participants in this program came from two different kinds of academic backgrounds. First were those who had pursued at least the last year of their undergraduate degree in a wide range of disciplines in which the language of instruction was English (English Teaching Program). Secondly were those who were already enrolled in the Master of Education in Mathematics and Science Teaching Program, in which a prerequisite of entry was possessing an undergraduate degree in mathematics, physics or chemistry. Students earn dual degrees in their subject areas as teachers for secondary schools while they simultaneously obtain master’s degree status on completion of the teacher education program. This program, which lasts three semesters, each of which lasting 16 weeks, aims to prepare student teachers to teach their respective subjects at secondary school level. The program integrates a strong pedagogical knowledge and pedagogical content knowledge with a well-planned allocation of set hours in each semester for professional development.


Of the total number of participants, 43 percent (N=169) had completed a bachelor’s degree, 32 percent (N=127) were pursuing the last year of their bachelor’s degree (and intended upon graduation to proceed to the above-mentioned program), and 26 percent (N=102) were pursuing or already had a postgraduate degree. Among the participants, 64 percent (N=259) and 36 percent (N=143) were male and female, respectively. The age of the participants ranged from the age of 19 to 43, with 89 percent (N= 356) of them falling between the age of 19 and 28.


The student teachers were from various universities including Istanbul (N=93, 24%), Bogazici (N=77, 19%), Yildiz Technical (N=75, 19%), Fatih (N= 74, 19%), and other (N=75, 19%) universities. The student teachers were also from various study fields such as mathematics and the sciences (N=143, 36%), languages (N=88, 22%), business, economy, and administration (N= 71, 18%), engineering and architecture (N=42, 11%), social sciences (N=28, 7%) and other fields (N= 24, 6%). The universities and the study fields of students that had less than 15 students were compounded as “other” for the purpose of analyzing data for the present study.


MEASURES


The Thinking Styles Inventory, developed by Sternberg and Wagner, based on Sternberg's (1988, 1997) earlier theory of mental self-government, was used to examine the nature of thinking styles of student teachers. In addition to the inventory, participants were asked to complete a subject information sheet to examine the relationships between the thinking styles and the student teachers’ demographic characteristics such as gender, age, educational level, type of university attended, and field of study followed.


The inventory is a self-report test with 104 items, which contains 13 subscales, each of which has eight items. It is designed to assess five dimensions of mental self-government: functions, forms, levels, scope, and leanings. Participants were asked how well each item describes themselves and were asked to rate themselves on a seven-point Likert scale ranging from 1 to 7 (1 = not at all well, 7 = extremely well) in order to assess one of the 13 thinking styles defined in the theory of mental self-government.


In the current study, the participants responded to a Turkish version of the inventory translated from English for this research. Reverse translation process between Turkish and English was carried out by three senior lecturers of English who are specialists in the English and Turkish languages. In the reverse translation process, each statement was considered for consistency with respect to the original inventory and its cultural accuracy of translation. Both The English and Turkish versions of the inventory were also applied to the 28 subjects similar to those of the participants of the present research to test the appropriateness between the Turkish and English versions of the inventory via test-retest. In terms of test-retest reliability, Pearson's correlations between two versions of the inventory, ranging from .70 to .86 among the 13 subscales, indicated acceptable reliability. The correlations were significant at the .01 and .05 level.


The reliability and validity of the inventory has been demonstrated satisfactorily, and has shown a number of similarities across different cultures, in the samples from the U.S., Spain, Hong Kong, mainland China, and the Philippines (e.g., Bernardo et al. 2002; Cano-Garcia & Hughes, 2000; Dai & Feldhusen, 1999; Grigorenko & Sternberg, 1997; Sternberg, 1997; Sternberg & Grigorenko, 1993, 1997; Zhang, 1999, 2000; Zhang & Sachs, 1997; Zhang & Sternberg, 2000). In these studies, the inventory had internal consistency reliabilities; Cronbach’s alphas of the 13 subscales ranged from .35 to .90. The results of factor analysis for construct validity of the inventory addressed 13 subscales under the three to five dimensional constructs. The factors yielding variance of the sample ranged from 35 percent to 78 percent in different cultures.


PROCEDURE


Data was collected from the student teachers by the researcher from five classes in the class session during the first week of the semester of the 2003-2004 academic year. Participation was on a voluntary basis in this research. The student teachers were given the opportunity to clarify any question that they had about the administration of the inventory. The participants finished responding to the inventory in a total of 30 to 45 minutes. Four hundred two inventories returned from students were found to be usable for data analysis from of a total of 441.


RESULTS


INTERNAL AND EXTERNAL RELIABILITIES


Testing the first question of whether each subscale is measuring a single idea and whether the items that make up the subscale are internally consistent, internal reliability data were obtained through both Cronbach’s alpha coefficient and Pearson's correlations among the 13 subscales. Internal and external analysis is reported in Table 1.


Table .1. Internal and External Reliabilities for 13 Scales of the Thinking Styles Inventory


Subscales

N

X

SS

a

r1

r2

Legislative

390

5.65

.76

.81

.58-.74

.74

Executive

373

5.09

.96

.81

.55-.75

.72

Judicial

378

5.15

.88

.79

.48-.76

.74

Monarchic

378

4.57

.90

.61

.42-.64

.68

Hierarchic

381

5.46

.98

.88

.61-.80

.74

Oligarchic

370

3.97

.91

.68

.47-.68

.63

Anarchic

387

4.26

.96

.72

.41-.70

.75

Global

380

4.50

.97

.77

.48-.78

.75

Local

385

4.44

.90

.71

.40-.76

.71

Internal

385

4.49

1.07

.82

.54-.73

.73

External

378

4.85

1.02

.84

.37-.82

.78

Liberal

392

5.24

1.01

.91

.73-.84

.71

Conservative

385

3.68

1.18

.90

.49-.88

.72

a1: Cronbach alpha of the subscales.

r1: Item-scale correlation of the subscales. Pearson's correlations were significant at the .01 level.

r2: Test re-tests reliability of the subscales. Pearson's correlations were significant at the .01 level.


Findings demonstrated that the total internal consistency reliability of the inventory for 13 subscales had an average Cronbach alpha coefficient of .92. The Cronbach’s alphas of the 13 subscales for these subjects ranged from .61 to .91. Except for the measure of monarchic style, the internal consistency indicated reasonably high reliability. Internal reliability on each scale was also obtained via item-scale correlation. Findings showed that the13 subscales had internal consistency reliabilities ranging from .37 to .88. Pearson's correlations that were significant at the .01 level indicated acceptable internal reliability.


External reliability was examined to test the degree of consistency of the measure over time via test re-test reliability in each subscale. For this aim, the inventory was applied by administering the same inventory on two occasions, over a period of four weeks, to the same group of 32 subjects who were similar to those of the participants for the present research. As can be seen in Table 1, Pearson’s correlations between two applications of the inventory ranging from .63 to .78 among the 13 subscales were significant at the .01 level, indicating acceptable external reliability.


INTERNAL VALIDITY


Testing the first and second question, internal validity data was obtained through both factor analysis and Pearson's correlations among the 13 subscales.


SCALE INTERCORRELATIONS


Validity of the inventory was assessed to test the question of whether conceptually opposite styles tend to be negatively correlated; relevant pairs of correlations were compared by inter-scale correlations. The detailed result of analysis is reported in Table 2.


Table 2. Interscale Pearson’s Correlation Matrix for 13 Scales of the Thinking Styles Inventory


Subscales

1

2

3

4

5

6

7

8

9

10

11

12

Legislative

            

Executive

.25**

           

Judicial

.42**

.15*

          

Monarchic

.15**

.43**

.08

         

Hierarchic

.45**

.64**

.31**

.39**

        

Oligarchic

.13*

-.02

.20**

.12*

-.12*

       

Anarchic

.29**

.03

.30**

.12*

.04

.58**

      

Global

.09

.37**

.13*

.58**

.32**

.11

.08

     

Local

.33**

.07

.16**

-.03

.14*

.38**

.43**

-.26**

    

Internal

.45**

.07

.18**

.18**

.14*

.33**

.36**

.17**

.25**

   

External

.19**

.18**

.34**

.10

.30**

.24**

.27**

.11

.31**

-.16**

  

Liberal

.45**

.03

.57**

.07

.24**

.24**

.34**

.04

.33**

.30**

.42**

 

Conservative

-.06

.35**

-.18**

.41**

.12*

.17**

.11

.32**

.08

.06

.14*

-.28**

* Correlations were significant at the .05 level.

** Correlations were significant at the .01 level.


As illustrated in Table 2, intercorrelations for the 13 subscales ranged from .02 to .64. Most of the scale intercorrelations were low. Essentially, some correlations pointed in the direction predicted by the theory of mental self-government—that is, global and local styles (r = -.26, p<.01), as well as liberal and conservative styles were negatively correlated (r = -.28, p<.01), as were internal and external styles (r = -.16, p<.01). However, there was a substantial overlap between legislative-executive (r = .25, p<.01) and legislative-liberal style (r =.45, p<.01), just as there was a correlation between executive and conservative styles (r =.35, p< .01). However, two of the significant correlations pointed in a direction that was not predicted by the theory of mental self-government—that is, the correlation between monarchic and hierarchic (.45, p<.01), and between judicial and hierarchic (r =.31, p<.01) styles.


FACTOR ANALYSIS


An exploratory factor analysis was conducted to examine construct validity of the inventory. Visual inspection of eigenvalues with the scree test supported the possible extraction of five factors in line with the theory of mental self-government. Therefore, a varimax rotation method was performed to determine the pattern of relationships for the subscales of the inventory. A five-factor solution also fitted the data of sample, when considering the Kaiser-Guttman criterion of eigenvalues exceeding 1. Five factors yielded and explained variance in the Turkish sample of 76 percent in the data. The detailed result of analysis is reported in Table 3.


Table 3. Results of Factor Analysis for 13 Scales of the Thinking Styles Inventory


Subscales

1

2

3

4

5

Legislative

.19

.50

.51

-.07

.39

Executive

-.01

-.05

.79

.32

-.04

Judicial

.15

.79

.14

.12

-.06

Monarchic

.13

-.04

.37

.72

.06

Hierarchic

-.10

.27

.83

.19

-.01

Oligarchic

.83

.09

-.18

.16

.03

Anarchic

.78

.23

-.00

.07

.09

Global

.01

.08

.15

.88

.03

Local

.66

.08

.36

-.47

.01

Internal

.39

.18

.13

.12

.78

External

.38

.37

.29

.03

-.68

Liberal

.28

.81

.11

-.02

.00

Conservative

.37

-.57

.34

.39

-.14

Eigenvalues

3.63

2.36

1.63

1.25

1.05

% of variance

27.92

18.19

12.55

9.67

8.12

% of  cumulative variance

27.92

46.11

58.67

68.34

76.47


As shown in Table 3, the results of factor analysis for the inventory addressed 13 subscales under the five dimensional constructs with 104 items. Factor one was dominated by the positive factor loadings of the oligarchic (.83), anarchic (.78), and local (.66) styles. Factor two showed the positive factor loadings for the legislative (.50), judicial (.79), and liberal (.81) styles as well as a negative loading for the conservative (-.57) style. Factor three was dominated by the positive loadings of the legislative (.51), executive (.79), and hierarchic (.83) styles. Factor four showed the positive factor loadings for the monarchic (.72) and global (.88) styles. Finally, factor five contrasted internal (.78) with external (-.68) style. Although there were cross-loadings by the legislative style, the five factors generally seemed relatively independent of one another.


THINKING STYLES BY STUDENT TEACHER DEMOGRAPHIC CHARACTERISTICS


Testing the third question, a one-way analysis of variance (ANOVA) was employed in order to identify the mean differences between thinking styles as the dependent variable and the student teacher, socialized characteristics of gender, age, educational level, type of university attended, and field of study as the independent variables. The Scheffe procedure was performed for post hoc comparisons. The mean results are reported in Table 4.


Answering how strong the effect of the different factors is and what their interaction is with the dependent variable requires calculating a measure of the effect size. Effect size is an indicator of how strong or how important research results are, which gives an understanding of the proportion of the variability in the dependent variables. While there are several different measures that exist for the computation of effect sizes within the d family (e.g., Cohen's d, Hedges's g), Cohen’s d is the most common indicator employed in the current research. Cohen’s d is an effect size measure representing the standardized difference between two means on some measured outcome. It should also be noted that Cohen (1988; cited in Thompson, 1994, 1996) gives very rough guidelines about estimating effect sizes that are categorized as small (d = .20), medium (d = .50), or large (d= .80). This guideline was adopted in this research to report the effect size of Cohen’s d in terms of thinking styles and the student teacher characteristics of gender that has two group means.


As Thompson (1996) mentioned, measures of effect size in ANOVA are measures of the degree of association between effect (e.g., a main effect, an interaction, a linear contrast) and the dependent variable. Four of the commonly used measures of effect size in AVOVA are: Eta squared (h2), partial Eta squared (hp2), omega squared, and the Intraclass correlation. Omega squared and the intraclass correlations are estimates of the degree of association in the population. Both Eta squared and partial Eta squared are estimates of the degree of association for the sample, which are expressed as the proportion of the total variance in the dependent variable that is associated with a particular effect. Eta squared d, the most common indicator, was adopted for this analysis. Eta squared is computed as the sum-of-squares explained, divided by the sum-of-squares total (Furr, 2004). SPSS for Windows, as part of an ANOVA process, does not give Eta squared the best measure of effect size, but is intuitively easy to understand. Eta squared varies from zero to one–the bigger the number the greater the effect (just like a correlation coefficient). According to Cohen (1988; cited in Kotrlik & Williams, 2003), Eta squared effect sizes can be categorized as small (h2 = .01) or medium (h2 = .06), or large (h2= .14), which is the metric adopted in this research to report Eta squared effect sizes. In the current research, the effect size of Eta squared reported in terms of thinking styles and the student teacher characteristics of age, educational level, type of university attended, and field of study followed, includes more than only two groups. Cohen’s d and Eta squared scores for each effect, and for all the dependent variables in this study, are shown in Table 4.


Table 4. Means and Effect Sizes in Thinking Styles According to the Demographic Characteristics of Student Teachers


 

leg

exec

judic

mon

hier

olig

anar

glob

loc

inter

exter

liber

cons

Gender

             

 Female

5.72

5.15

5.17

4.50

5.56

3.91

4.24

4.47

4.47

4.50

4.83

5.21

3.57

 Male

5.52

5.01

5.12

4.71

5.29

4.08

4.31

4.57

4.39

4.48

4.91

5.32

3.88

Cohen’s d

.26

.14

.06

-.23

.27

-.19

-.07

-.10

.09

.02

-.08

-.11

.26

Age

             

 19-23

5.58

5.14

5.09

4.53

5.40

3.92

4.20

4.45

4.39

4.39

4.82

5.06

3.74

 24-28

5.83

4.98

5.26

4.62

5.57

4.06

4.32

4.50

4.59

4.69

4.94

5.54

3.60

 29-33

5.59

5.43

5.47

4.79

5.73

4.03

4.39

4.88

4.33

4.48

4.94

5.47

3.58

 34-48

5.67

5.19

4.86

4.35

5.33

3.74

4.68

5.01

4.41

4.33

5.05

5.36

3.71

Eta

.15

.11

.14

.08

.10

.08

.10

.14

.10

.13

.06

.22

.06

Eta2  (h2)

.02

.01

.02

.00

.01

.00

.01

.02

.01

.01

.00

.05

.00

Educational Level

             

Fourth Grade Underg.

5.57

4.86

5.23

4.57

5.30

3.90

4.16

4.33

4.36

4.41

4.77

5.01

3.55

Undergraduate Degree

5.71

5.17

5.19

4.61

5.52

4.03

4.40

4.61

4.56

4.61

4.95

5.37

3.82

Graduate Student

5.74

5.32

5.04

4.51

5.63

3.99

4.17

4.48

4.41

4.41

4.89

5.37

3.56

Graduate Degree

5.64

5.16

5.19

4.66

5.34

3.93

4.50

5.13

4.18

4.74

4.59

5.55

3.75

Eta

.09

.18

.08

.04

.13

.06

.12

.16

.11

.10

.08

.17

.10

Eta2 (h2)

.00

.03

.00

.00

.01

.00

.01

.02

.01

.01

.00

.03

.01

University

             

Bogazici

5.65

5.16

5.23

4.51

5.36

3.95

4.16

4.34

4.50

4.54

4.72

5.31

3.73

Istanbul

5.55

4.79

5.17

4.55

5.17

4.03

4.40

4.38

4.48

4.49

4.85

5.28

3.50

Yildiz

5.81

5.38

5.03

4.59

5.67

3.99

4.21

4.49

4.48

4.41

4.91

5.38

3.57

Fatih

5.73

5.24

5.16

4.76

5.72

3.98

4.12

4.72

4.35

4.59

5.02

5.14

3.88

Other

5.61

4.99

5.22

4.44

5.51

3.93

4.42

4.59

4.43

4.45

4.90

5.22

3.65

Eta

.12

.21

.08

.11

.21

.03

.12

.14

.05

.05

.09

.08

.11

Eta2 (h2)

.01

.04

.00

.01

.04

.00

.01

.02

.00

.00

.00

.00

.01

Study Field

             

Language Studies

5.60

4.80

5.15

4.51

5.32

3.98

4.24

4.28

4.39

4.49

4.68

5.07

3.48

Busin-Econ- Adm

5.69

4.98

5.37

4.43

5.55

3.91

4.18

4.64

4.29

4.63

4.88

5.38

3.49

Engin-Archit.

5.61

5.06

5.11

4.72

5.59

3.93

4.43

4.66

4.44

4.26

5.24

5.16

3.85

Social Studies

5.65

5.30

5.22

4.80

5.68

4.03

4.26

4.68

4.50

4.49

4.79

5.01

3.85

Science-Math

5.70

5.37

5.08

4.64

5.50

4.00

4.26

4.54

4.54

4.51

4.90

5.39

3.78

Other

5.77

4.85

5.19

4.30

5.18

4.02

4.40

4.30

4.59

4.57

4.82

5.23

3.81

Eta

.06

.24

.11

.14

.12

.04

.07

.15

.11

.08

.14

.14

.13

Eta2 (h2)

.00

.06

.01

.02

.01

.00

.00

.02

.01

.00

.02

.02

.01


A one-way analysis procedure showed a statistically significant difference of gender on the legislative and hierarchic styles. Female student teachers scored higher on legislative (F (1, 388) = 5.98, p< .01) and hierarchic styles (F (1, 379) = 6.36, p<.01) than did their male counterparts, the means being, respectively, 5.72 (SD 0.70), and 5.56 (SD 0.94). Furthermore, Cohen's d statistic effect size for this data yielded the result that gender accounted for .26 for legislative style and .27 for hierarchic style, with a moderate effect size, which means that approximately one quarter of the variance in each individual’s score was explained by group differences. This analysis also indicated that there was a statistically significant difference of gender for the monarchic and conservative styles. Male student teachers scored higher on monarchic (F (1, 376) = 4.86, p<.02) and conservative styles (F (1, 383) = 5.91, p<.01) than did their female counterparts, the means being, respectively, 4.71 (SD 0.96), 3.88 (SD 1.28). Furthermore, Cohen's d effect size revealed that student teachers’ gender had a small effect size for monarchic (d=-.23) and for conservative style (d=.26), which means that 23 percent of the variability of the dependent variable of monarchic style and 26 percent of the variability of the dependent variable of conservative style can be explained in terms of the independent variable of gender.  Thus, the impact of the gender variable was not as large as it would be.


An ANOVA showed that there was a statistically significance age difference on the legislative (F (3, 381) = 3.27, p<.02), judicial (F (3, 369) = 2.54, p<.05), global (F (3, 371) = 2.56, p<.05), and liberal (F (3, 384) = 6.36, p<.00) styles. This analysis showed that the younger students who were 19-23 and 24-28 years old scored significantly higher on the legislative and liberal styles than those who were 29 years old and older. The Eta squared statistic indicated that the ages of student teachers explained the small effect size estimate of legislative (h2 =.02), judicial (h2 =.02), global (h2 =.02) and liberal (h2 =.05) styles of student teachers. Note that the effect of thinking style is significant for all age groups, but is larger in magnitude for the youngsters than for the older ones.


Statistically significant differences of educational level were also found among the participants who were pursuing fourth grade undergraduate studies and who possessed bachelor’s degrees superior to those who were in the graduate studies on the executive (F (3, 365) = 4.45, p<.00), global (F (3, 373) = 3.28, p<.02), and liberal (F (3, 385) = 4.10, p<00) styles. This analysis indicated that the participants who were pursuing fourth grade undergraduate studies, as well as those who had undergraduate degrees, scored significantly higher on the liberal styles than those who were in the graduate studies. Moreover, the participants who were pursuing graduate studies scored significantly higher on the executive style than the rest of the participants. Furthermore, Eta squared for this data yielded an effect size estimate of executive (h2 =.03), global (h2 =.02), and liberal (h2 =.03) styles of student teachers. What it has here is a small effect, even though it is statistically significant. Educational level does not seem to affect thinking style in an important way.


Similarly, there was a statistically significant difference among which university a student attended on the executive (F (4, 362) = 4.47, p<.00) and hierarchic (F (4, 371) = 4.38, p<.00) styles. This analysis showed that participants who were from the universities of Istanbul and Yildiz scored significantly higher on the executive style than those who were from other universities. Moreover, participants who were from the University of Fatih scored significantly higher on the hierarchic style in comparison to those from other universities. Note that the effect of executive (h2 =.04) and hierarchic (h2 =.04) style is practically significant with a small effect for student teachers.


There was also a statistically significant difference of type of study field attended on the executive (F (5, 362) = 4.59, p<.00) style. This analysis indicated that student teachers who were from the language, science, and mathematics studies scored significantly higher on the executive style than those who were from other fields. However, the sample estimate of Eta squared (h2 =.06) was just .06, which means that the given field of study is estimated to account for .06 percent of variance in the dependent variable of executive style measure, with a medium effect size.


DISCUSSION AND IMPLICATIONS


In terms of the first objective of the current research, the total internal consistency reliability of the inventory for 13 subscales had an average Cronbach alpha coefficient of .92, higher than those reported in the previous research (e.g., Bernardo et al. 2002; Cano-Garcia & Hughes, 2000; Dai & Feldhusen, 1999; Grigorenko & Sternberg, 1997; Sternberg, 1997; Sternberg & Grigorenko, 1993, 1997; Zhang, 1999, 2000, 2001b, 2002a, 2002d; Zhang &  Sternberg, 2000). The reason for the high reliability of the inventory might be explained by the relative heterogeneity of the Turkish sample in terms of the educational level, type of university attended, as well as the field of study followed by the subjects. These results are congruent with the explanation of Tompson (1994): “[M]ore heterogeneous samples often lead to more variable scores and thus to higher reliability. Therefore, the same measure, when administered to more heterogeneous or more homogeneous sets of subjects, will yield scores with differing reliability” (p. 839). An alternative explanation for this result might also be found in the explanation for the Turkish sample. Other studies used a sample of college or secondary school students, while the present study investigated a sample of university students ranging from undergraduates to those with graduate degrees; the high internal consistency of measures might offer up a function of sample differences. Except for the measure of monarchic style that was similar to those reported by Sternberg (1994; Alpha=.42), the Cronbach’s alphas of the scales for these subjects were reasonably high for the 13 subscales. Previously mentioned research employing the inventory obtained fairly encouraging reliability data, in general, ranging from .35 to .88 for the subscales. The internal and external reliabilities in each scale obtained from the current study were considered sufficient for this research.


In keeping with the first and second objectives of the study, some correlations were found in the direction predicted by the theory of mental self-government. Consistent with previous research (Sternberg, 1997), global and local styles, as well as liberal and conservative styles were negatively correlated, as were internal and external styles. The results confirmed the previous finding (Sternberg, 1997): that there was a substantial overlap between legislative-executive and legislative-liberal styles, as was the correlation established between executive and conservative styles. These results are consistent with those obtained in the previous study conducted in Hong Kong (Zhang & Sachs, 1997). Inter-scale correlation analysis for the current research generally suggests that the 13 subscales are correlated in the directions that would be in keeping with those stipulated by the theory of mental self-government. However, some of the absolute values of the correlations were not as high as those reported in Sternberg's (1997) study. Although the differences in correlations might arise as a result of problems connected to heterogeneity of sample, as previously noted, they might also be attributed to the fact that the subjects in the current research were tested in their second language. The low reliability of the measurement might be due to some problems in the translation of the items in the scale or due to the lack of relevancy of some items used with Turkish students. Sample differences among cultures might also be the result of random fluctuations in the data. One possible explanation for these results might be found in the research of thinking styles, outlined in the introduction of the present study. Since the studies reported by Sternberg (1997) used a sample of college students, while the present study investigated a sample of university students ranging from undergraduates to those with graduate degrees, the low correlations in the Turkish sample might be a function of age, educational level, type of university attended, and field of study. On the other hand, the inventory might be less discriminative of different dimensions of stylistic differences among Turkish student teachers. Still, there are considerable similarities between the results reported in the different studies, as previously noted, covering samples from the U.S., Europe, and Asia.


In terms of the first and second objectives of the study, an exploratory factor analysis with varimax rotation addressed 13 subscales under the five dimensional constructs, consistent with the theory of mental self-government. Although there were cross-loadings by the legislative style, the five factors generally seemed relatively independent of one another. Whereas the five-factor structure shares some features with previous findings, it differs in some ways. For example, Sternberg (1997) explained a five-factor model for 13 thinking styles conducted with the college student sample, partly supportive of the dimensions postulated by the theory. Research employing the inventory has obtained fairly encouraging construct validity of the inventory (e.g., Bernardo at al. 2002; Cano-Garcia & Hughes, 2000; Dai & Feldhusen, 1999; Grigorenko & Sternberg, 1997; Sternberg, 1997; Sternberg & Grigorenko, 1993, 1995, 1997; Zhang 1999, 2000, 2001a, 2001b, 2001c, 2001d, 2002a, 2002b, 2002c, 2002d, 2002e; Zhang &  Sternberg, 2000). Including the current study, the factor analysis of these studies have generally supported dimensions postulated by the theory since 13 subscales yielded under the three-to-five dimensional constructs, and there seem to be cross-loadings by some of the dimensions. Nevertheless, generally speaking, although the findings of the present study are somewhat inconclusive as to the nature and relationships of thinking styles among Turkish student teachers, the reliability and validity of the inventory obtained from the current study were considered sufficient for the purposes of this research. This might be interpreted as an indication that the Thinking Styles Inventory is applicable in samples that differ culturally.


In accordance with the third objective of this study, based on Sternberg's (1988, 1997) notion that styles are in part socialized, one-way analysis of variance procedures indicated that male students scored higher on the monarchic and conservative styles than did their female counterparts who scored higher on the legislative and hierarchic styles. Moreover, the younger students who were 19-23 and 24-28 years old scored significantly higher on the legislative and liberal styles than those who were 29 years old and older. And thus it appears that female, as well as younger Turkish students have over time developed thinking processes that focus on trying out new and creative ways of thinking. This was expected, and in line with previous research that has shown that thinking styles could vary as a function of age and gender (e.g., Sternberg & Grigorenko, 1993, 1995; Zhang, 1999). It should also be noted that, as Zhang (2002f) states, gender and age differences might be attributed to the nature of the psychological development of students.


An alternative explanation for gender differences might also be explained by the effects of culture. Sternberg (1997) indicates how different dimensions of culture might encourage or inhibit the development of specific thinking styles. Thus, individuals in one culture, compared with those in other cultures, might be expected to have a higher or lower preference for certain styles. In this connection, the findings of legislative and hierarchic style among female student teachers, and monarchic and conservative style among male student teachers, might reflect cultural preferences. Therefore, from the findings of the present study, it might be interpreted that different roles are attributed to the members of each gender in Turkish society. As Tezcan (1995) indicated, while females are under social pressure, males are expected to preserve the traditional customs and values in Turkish society. In accordance with the findings from Tezcan (1993), females, contrary to Turkish Law, in general, did not have equal rights in family life since, until recently, Turkish culture ascribed traditional roles to women. However, the changes in the educational system and improvement in the educational level, as well as the opening of new vocational opportunities, have had an influence on the perceived role of women. Moreover, women are influenced in a positive way as a result of the changing nature of urbanization and secularism in the society in Turkey. In keeping with the opinions of Halonen and Santrock (1999), in recent years, the idea that parents are the critical agents in gender role development has come under fire. Culture, education, schools, press, the media, and family members also influence gender behavior. On the other hand, the world is extremely complex, and the use of stereotypes is one way we simplify this complexity. However, this is the first study that identified such a difference in thinking styles between female and male student teachers in Turkey. Therefore, this result can only be viewed as tentative.


The study also revealed that the participants who were pursuing graduate studies scored significantly higher on the executive style than the rest of the participants. Moreover, the less educated participants who were pursuing the fourth grade undergraduate studies and those who had undergraduate degrees scored significantly higher on the liberal style than those who were in graduate studies programs. Similarly, the analysis showed that participants who were from the universities of Istanbul and Yildiz scored significantly higher on the executive style than those who were from other universities. Moreover, participants from Fatih University scored significantly higher on the hierarchic style than those who were from other universities. There was also a significant difference in a student’s field of study for the executive style. The analysis indicated that student teachers who were from the language, science, and mathematics fields of study scored significantly higher on the executive style than those who were from other fields of study. Although the differences might have resulted from Turkish sampling, as previously noted, it might also be attributed to the different educational system and culture. It is reasonable to assume that the cultural and educational concepts and practices affect students. In accordance with the findings of the current study, it is possible to infer that the Turkish educational system places a higher priority on the executive and hierarchical styles in terms of field of study attended and type of university attended. One possible explanation for these results might also be that the executive thinking style might correspond more to the nature of the courses studied in the fields of language, science, and mathematics. Moreover, the form of thinking required might be associated with tasks particular to the different types of courses in the curriculum such as language, science, and mathematics. An alternative explanation for these results might also be found in the way directives for students and teachers are issued through the Turkish education system. For example, some aspects of the curricula applied in schools are organized in a very specific manner (Ministry of National Education, 1992, 1995, 1998, 1999). That is, the rate of compulsory courses in curricula is high, but that of the elective courses is low. This situation restricts flexibility in the offering of the same opportunities in different areas of education. Subject mastery is more valued than pedagogical skills are in education. Furthermore, teachers heavily rely on a didactic method of presentation in class, and appear to be traditionalist (OECD, 1989). Therefore, students of the executive style might get more encouragement to continue operating in the same tried and tested way than those of other styles. On the other hand, as Zhang (2002d) indicated, the executive style is related to analytic thinking for processing information. Most of the tasks in the executive style involve working on tasks with clear guidelines, instructions, and structures. However, Zhang and Sachs (1997) have indicated that students in the fields of the natural sciences and technology tended to be more globally oriented than those in the social sciences and humanities. Another such study (Sternberg & Lubart, 1991a, 1991b) has revealed that students who were most creative generate new ideas in their research, regardless of their discipline.


Two points are worth discussing regarding the findings of thinking styles and the type of the university. First, one possible explanation for these results might be found in the explanations of thinking styles, outlined in the introduction of the present study. As Sternberg (1997) noted, different cultures might value some thinking styles over others. Because each culture has its own values and each educational system has a different reward system, the particular thinking styles that might contribute to students are different for each culture. Even though there is not clear evidence in Turkish studies what the overriding reason might be, one of the possible explanations of the finding is that universities themselves might influence students’ thinking styles since students from different universities are exposed to different learning and teaching environments. Each kind of culture in these universities has unique characteristics that would help shape the thinking styles of university students. For example, Fatih University culture is known for its values of ancestor worship and holding up collective effort, as well as nontraditional assessments. Therefore, students from Fatih University tended to be more oriented to develop hierarchic thinking in an educational context. On the contrary, students from Yildiz and Istanbul universities are known for holding up individualism, the unity of ruling and teaching, as well as traditional assessments. Therefore, students from Yildiz and Istanbul universities tended to be more oriented to develop executive thinking in an educational context. An alternative explanation, based on the results of the study, suggests that teachers at Fatih University put emphasis on the divergent abilities, whereas teachers at Yildiz and Istanbul universities put great emphasis on convergent abilities. Moreover, as Zhang (1999) states, being in a tradition of quantitative assessment, students might be trying to meet the demands of the course requirements. For the same reason, high academic achievement tends to depend on convergent abilities to meet highly structured objectives rather than divergent thinking abilities.


Previous findings have been mixed concerning the effects of student characteristics on thinking styles. For example, the significant differences in thinking styles on the basis of the aforementioned variables are consistent with previous studies (see Sternberg & Grigorenko, 1995; Zhang, 1999, 2001 b, 2002b; Zhang & Sachs, 1997), however, are inconsistent with previous research (Grigorenko & Sternberg, 1997; Zhang, 2002a). For example, Grigorenko and Sternberg (1997) found no distinct patterns of particular thinking styles among the gifted high school students based on abilities, gender, or grade. Since the study reported used a sample of gifted high school students, while the present study investigated a sample of university students, the different findings in the Turkish sample might be seen as a function of age, educational level, type of university attended, as well as field of study followed. Similarly, Zhang (2002a) reported no statistically significant difference between thinking styles and age, gender, and the previous work experience of university students in Hong Kong. Although he used a sample of university students for the current research, the differences might have resulted from sampling. There is a growing interest and tendency that cultural ethnic values, social environment, and cognition are important factors for understanding differences in individual behavior (e.g., Schneider, 1989; Shaw, 1990). The development of thinking styles is also influenced by many variables and factors. For example, Zhang and Watkins (2001) found that cognitive-developmental patterns of the American and Chinese participants differed. However, no clear conclusion can be drawn based on the results of studies already conducted. This discussion attempted to link the specific pattern of the results to Turkish cultural and educational concepts, in particular, since according to the theory of mental self-government (Sternberg, 1988, 1997), cultural factors might influence how thinking styles develop in a socio-cultural and educational context. Researchers have found within the Turkish educational system (e.g., Akdeniz, Yigit, & Kurt, 2002; Demirtas, 1999; Hasanoglu, 2003, Kucuk, Yesim, Saka, & Genel, 2002, Yanpar-Sahin, 2001; Yildirim & Demir, 2003) that there seems to be an emphasis in classroom instruction and learning, which tends to favor the teacher-oriented approach. From a thinking style perspective, this approach helps develop the executive and conservative styles; on the contrary, the student-oriented approach nurtures the legislative, global, and liberal styles (Sternberg, 1994, 1997; Sternberg & Grigorenko, 1995, 1997).


Because the education system stresses the executive and hierarchic thinking style, it is not surprising to find in the current study that thinking styles were differentiated as a function of the educational levels, university types, and study fields of participants. Moreover, in accordance with the opinion of Cano-Garcia and Hughes (2000), the fact of the matter is that because educational systems rewards through the awarding of “good” grades, students adapt their styles to the context as far as teaching methods, evaluation, etc. are concerned. As a result, as is outlined by Sternberg (1997), a bias exists in education, which favors the development of the specific styles of students, such as executive and hierarchic styles, as the current research indicates. Yet even graduate students tend to be executive as shown in this study. However, results in the current study indicated that the female and younger student teachers used more effective thinking styles such as legislative and liberal styles that are associated with using creative strategies and complex information processing. In this sense, styles might not be innate, but instead developed, and thereby reflect task or situational demands as well as individual dispositions (Sternberg, 1997; Sternberg & Grigorenko, 1997).


Generally speaking, in terms of the third objective, the results of current research lent only partial support to Sternberg’s (1988, 1997) notion that styles are in part socialized. Participants had significantly different particular thinking styles based on such socialization variables as gender, age, educational level, university type, and field of study pursued. Nevertheless, the role of biological, psychological, and cognitive development, as well as individual differences might also encourage or inhibit the development of specific styles of students. Therefore, these results need to be verified by future studies, especially those that pay attention to the effects of gender, university type, and field of study. In addition, although the one-way analysis procedures showed the statistically significant differences, effect size statistics, with a small effect, indicated that each independent variable explained a small effect size estimate on the thinking styles of student teachers. Thus, the impact of the educational level, type of university attended, and field of study attended was not as large as it would be. Still, the findings of the present research strengthen and confirm the importance of a claim as well as a tendency of the thinking styles within the university concept. According to the theory of mental self-government (Sternberg, 1988, 1994), thinking styles are partially developed and socialized; moreover, culture is one of the foremost factors that shapes thinking styles. Thus, different cultures and education systems might encourage or inhibit the development of specific styles. Styles are also thought, at least by some, to vary across context and specific tasks, as well as in the developmental period of students, and are considered to be partially socialized by the predominant culture (Sternberg, 1997). As Zhang (1999) says, if styles are partially socialized, then they are teachable.


CONCLUSIONS AND IMPLICATIONS


It should be noted that there was a major limitation to the present study. That is, this is the first study that investigated thinking styles among Turkish students in Turkey. Therefore, the conclusion drawn regarding thinking styles should be considered preliminary. Further investigation is required to specify adequately the thinking styles of students and their socialized variables that might influence their thinking styles.


Nevertheless, the conclusions from the results of the present research might be of interest to educational psychologists, researchers, educators, and the educational process itself. The first conclusion is that in order to measure thinking styles of students based on a model of broad intellectual styles, the inventory might be used as an efficient instrument. This method might open new perspectives in the field of learning and assessment, as much as in the consultation processes. The second conclusion concerns the relationship between thinking styles and socialization variables of student teachers. The results of this study demonstrated a diversity concerning the particular thinking styles of student teachers.


What are the implications of these results? First, diversity should be allowed in teaching and learning in and outside the classroom by allowing for a  wide variation in students' thinking styles. As is outlined by Cano-Garcia and Hughes (2000), schooling should bring about the creation of learners who know how to learn. However, because educational systems reward through the awarding of “good” grades, so might students adapt their styles to the context as far as teaching methods, evaluation, etc. are concerned. As a result, as Sternberg (1997) has argued, a bias exists in education, which favors the development of the specific styles of students. As Zhang (1999) has indicated, too many students tend to use executive and local thinking styles that will render them ineffective for future career survival. These executive and local student thinkers are not there by accident, but rather are a result of socialization. Being in an environment of a quantitative assessment tradition, these students are merely trying to meet the demands of the course requirements. For the same reason, in higher education, high academic achievement tends to depend on convergent abilities to meet highly structured objectives rather than on divergent thinking abilities. As Cano-Garcia and Hughes (2000) have reported, executive and local styles can strengthen performance but do not much favor student learning or an intellectual process from a cognitive viewpoint. According to cognitive theorists (see, Armstrong, 2000; Feldman, 2000; Halonen & Santrock, 1999; Morgan, 1997, Rayner & Riding, 1997; Sternberg & Williams, 2002), however, effective learning occurs via interaction with, and support from, people and objects in the world. Thinking is the manipulation of information perceived, learned, and remembered. It is the mental creation of life. It is therefore necessary to access the styles used by students in learning and thinking, as well as to motivate educators to analyze their own teaching styles. As Sternberg (1997) says, teachers should design a way of teaching which takes into account the diversity of individual styles. This must be done to enrich and favor all the students. As stated by Zhang (2002c), educators should consider the fact that the development of thinking styles of students, as an effort to produce students who are going to be capable of adapting themselves to the ever-changing world, should necessitate educators to cultivate the creative thinking styles of students, such as liberal and legislative styles. What can we do to change the present situation? It has been argued by Sternberg (1997) that thinking styles are partially the result of socialization. If thinking styles are socialized, then they might be modified. As Zhang (2002c) says, in order to modify the thinking styles of students, teachers might re-design their curricular activities by using multiple instructional and evaluation methods to allow for developing multiple thinking styles of students.


The results of this study also point to the direction for future research. Further research is needed to clarify the nature of thinking styles as assessed by the Thinking Styles Inventory at different educational levels and by different cultures to facilitate a better understanding of the thinking styles of students. Further research is also needed to perform cross-cultural comparisons regarding the nature of thinking styles and to clarify the socio-cultural developmental processes that shape how different thinking styles become interrelated.


The author would like to thank and express her gratitude to Li-Fang Zhang for his help on this article. The author also thanks and expresses her gratitude to Paul Hearn and Yetis Ozkan for invaluable proofreading of the English version of this article.


APPENDIX A: THINKING STYLES IN THE THEORY OF SELF-GOVERNMENT


Thinking styles

Descriptions of thinking styles

Sample Item

I. Functions

Legislative


Executive


Judicial



II.Forms

Monarchic


Hierarchic


Oligarchic


Anarchic



III.Levels

Global


Local


IV. Scope

10. Internal


11.External


V.Leanings

12.Liberal



13.Conservative


Working on tasks with creative strategies, own ideas and ways.

Working on tasks with guidelines and clear instructions.

Judging-evaluating process, products and performance of other people.


Pursuing single task at a time.


Setting and deciding priorities to several tasks.

Having multiple tasks without setting priorities.

Taking random and flexible approach.



Dealing with the overall and general picture and abstract ideas.

Dealing with details and concrete issues.


Working on their own and alone.

Working on tasks with other people.


Working on tasks with unfamiliarity, novelty and ambiguity.

Doing tasks in traditional ways with the existing rules.


When working on a task, I like to start with my own ideas.

I like projects that have a clear structure and a set plan and goal.

I enjoy work that involves analyzing, grading, or com­paring things.



When talking or writing about ideas, I stick to one main idea.

I like to set priorities for the things I need to do before I start doing them.

I sometimes have trouble setting priorities for multiple things that I need to get done.

When discussing or writing down ideas, I use what­ever comes to mind.


I like situations or tasks in which I am not concerned with details.

I like problems where I need to pay attention to detail.


I like projects that I can complete independently.

When starting a task, I like to brainstorm ideas with friends or peers.


I like to do things in new ways not used by others in the past.


I like to do things in ways that have been used in the past.

Resource: (Grigorenko & Sternberg, 1997; Sternberg, 1994, 1997; Zhang, 2001d, 2002a).


References


Akdeniz, A.R., Yigit, N., & Kurt, S. (2002). Yeni fen bilgisi ogretim programi ile ilgili Ogretmen dusuncelerini arastırma [Opinions of the teachers concerning science curriculum]. Trabzon: KTU, Fatih Egitim Fakultesi.


Armstrong, S.J. (2000). The influence of individual cognitive style on performance in management education. Educational Psychology, 20(3), 323-339.


Bernardo, A.B.I., Zhang, L.F., & Callueng, C.M. (2002). Thinking styles and academic achievement among Filipino students. Journal of Genetic Psychology, 163(2), 149-163. Retrieved July 15, 2003, from ProQuest database.


Cano-Garcia, F. & Hughes, E.H. (2000). Learning and thinking styles: An analysis of their interrelationship and influence on academic achievement. Educational Psychology, 20(4), 413-430.


Curry, L. (1990, October). A critique of the research on learning styles. Educational Leadership, 50-57.


Dai, D.Y., & Feldhusen, J.F. (1999). A validation study of the thinking styles inventory: Implications for gifted education. Roeper Review, 21(4), 302-401. Retrieved June 19, 2003, from ProQuest database.


Demirtas, A. (1999). Egitimde nitelik: Bugunu ve gelecegi. Egitimde nitelik gelistirme [Quality in education: Development quality in education]. Istanbul: Kultur Koleji Yayinlari.


Feldman, R.S. (2000). Essentials of understanding psychology. Boston: Mc Graw Hill.


Furr, R.M. (2004). Interpreting effect sizes in contrast analysis. Understanding Statistics, 3(1),1-25.


Grigorenko, E.L., & Sternberg, R.J. (1997). Styles of thinking, abilities, and academic performance. Exceptional Children, 63(3), 295-312. Retrieved June 19, 2003, from ProQuest database.


Halonen, J.S., & Santrock, J.W. (1999). Psychology: Context and applications. Boston: McGraw Hill.


Hasanoglu, Y. (2003). Ortaogretimde kimya ogretiminin degerlendirilmesi [Evaluation of chemistry teaching]. Yuksek Lisans Tezi, Ataturk Universitesi.


Hill, J., Puurula, A., Sitko-Lutek, A., & Rakowska, A. (2000). Cognitive style and socialization: An exploration of learned sources of style in Finland, Poland and the UK. Educational Psychology, 20(3), 285-305.


Kotrlik, J.W. & Williams, H.A. (2003). The incorporation of effect size in information technology, learning, and performance research. Information Technology, Learning, and Performance Journal, 21(1), 1-7.


Kucuk, M.,Yesim, M., Saka, C., & Genel, Y. (2002). Kimya egitim ve ogretiminde uygulanan ogretim yontemleri ile etkin kavram ogretimi arasındaki iliskinin incelenmesi [Investigation of relationship between the instructional methods in chemistry teaching and concept teaching]. Van: Yuzuncu Yil Universitesi.


Ministry of National Education. (1992). National Education report to Parliament. Ankara: MONE.


Ministry of National Education. (1995). Education in Turkey. Ankara: MONE.


Ministry of National Education.  (1998). 16. Milli Egitim Surasi Raporlari (Report of the 16th National Education Assemblies). Ankara: MONE.


Ministry of National Education. (1999). The Turkish Educational System. Ankara: MONE.


Morgan, H. (1997). Cognitive styles and classroom learning. London: Praeger.


OECD. (1989). Review of national policies for education: Turkey. Paris: OECD.


O’Hara, L.A., & Sternberg, R.J. (2000). It does not hurt to ask: Effects of instructions to be creative, practical, or analytical on essay-writing performance and their interaction with students’ thinking styles. Creativity Research Journal, 13(2), 197-210. Retrieved September 25, 2003, from ProQuest database.


Rayner, S.G., & Riding, R.J. (1997). Towards a categorization of cognitive styles and learning styles. Educational Psychology, 17(1-2), 5-27.


Schneider, N. J. (1989). Strategy formulation: The impact of national culture. Organization Studies, 10(2), 149-168.


Shaw, J. B. (1990). A cognitive categorization model for the study of intercultural management. Academy of Management Review, I5(4), 626-645.


Sternberg, R.J. (1988). Mental self government: A theory of intellectual styles and their development. Human Development, 31, 197-224.


Sternberg, R. J. (1994). Allowing for thinking styles. Educational Leadership, 52(3), 36-39.


Sternberg, R.J. (1997). Thinking styles. Cambridge: Cambridge University Press.


Sternberg, R.J., & Grigorenko, E.L. (1993). Thinking styles and the gifted. Roeper Review, 16(2), 122 131. Retrieved June 19, 2003, from ScienceDirect database.


Sternberg, R.J., & Grigorenko, E.L. (1995). Styles of thinking in the school. European Journal for High Ability, 6(2), 1¯19.


Sternberg, R.J., & Grigorenko, E.L. (1997). Are cognitive styles still in style? American Psychologist, 52, 700-712. Retrieved June 19, 2003, from ProQuest database.


Sternberg, R.J., & Lubart, T.I. (1991a). An investment theory of creativity and its development. Human Development, 34, 1-31.


Sternberg, R.J., & Lubart, T.I. (1991b). Creating creative minds. Phi Delta Kappa, 608-614.


Sternberg, R.J., & Williams, W.M. (2002). Educational psychology. Boston: Allyn and Bacon.


Tezcan, M. (1993). Egitim sosyolojisinde çağdaş kuramlar ve Tukiye [Contemporary Theories in Educational Sociology in Turkey]. Ankara: Ankara Universitesi.


Tezcan, M. (1995). Sosyolojiye giris: Temel kavramlar [Introduction to Sociology: Foundations]. Ankara: Ankara Universitesi.


Thompson, B. (1994). Guidelenes for authors. Educational and Psychological Measurement, 54(4), 837-847.


Thompson, B. (1996). AERA editorial policies regarding statistical significance testing: Three suggested reforms. Educational Researcher, 25(2), 26-30.


Yanpar-Sahin, T. (2001). Olusturmacı yaklasımın sosyal bilgiler dersinde bilissel ve duyussal ogrenmeye etkisi [The effect of constructivist approach on cognitive and affective learning in social studies]. Kuram ve Uygulamada Egitim Bilimleri, 1(2), 463-482.


Yildirim, Z., & Demir, K. (2003). Burdur il merkezindeki ilkögretim okullarinda gorev yapan fen bilgisi ogretmenlerinin alanlari ve yeterliklerine iliskin gorusleriyle fen bilgisi egitimi ogrencilerinin ogretmenlerle ilgili gozlemleri [Opinions of the teachers and their students concerning fields and competencies in science teaching]. Retrieved July 19, 2004, from http//www. fenokulu.com/fenyeterlik.htm


Zhang, L.F. (1999). Further cross-cultural validation of the theory of mental self-government. The Journal of Psychology, 133(2), 165-181. Retrieved August 03, 2003, from ProQuest database.


Zhang, L.F. (2000). Are thinking styles and personality types related? Educational Psychology, 20 (3), 271-283.


Zhang, L.F. (2001a). Approaches and thinking styles in teaching. The Journal of Psychology, 135(5), 547-561. Retrieved August 03, 2003, from ProQuest database.


Zhang, L.F. (2001b). Do thinking styles contribute to academic achievement beyond self-rated abilities? The Journal of Psychology, 135(6), 621-637. Retrieved August 03, 2003, from ProQuest database.


Zhang, L.F. (2001c). Thinking styles, self-esteem, and socio-economic status. Personality and Individual Differences, 31(8), 1333-1346. Retrieved July 15, 2003, from ScienceDirect database.


Zhang, L.F. (2001d). Thinking styles and personality types revisited. Personality and Individual Differences, 31(6), 883-894. Retrieved August 03, 2003, from ScienceDirect database.


Zhang, L.F. (2001e). Do styles of thinking matter among Hong Kong secondary school students? Personality and Individual Differences, 31(3), 289-301. Retrieved July 15, 2003, from ScienceDirect database.


Zhang, L.F. (2002a). Thinking styles and cognitive development. Journal of Genetic Psychology, 163(2), 179-195. Retrieved August 03, 2003, from Proquest database.


Zhang, L.F. (2002b). Thinking styles and big five personality traits. Educational Psychology, 22(1), 17-31.


Zhang, L.F. (2002c). Thinking styles: Their relationships with modes of thinking and academic performance. Educational Psychology, 22(3), 331-348.


Zhang, L.F. (2002d). Thinking styles and modes of thinking: Implications for education and research. The Journal of Psychology, 136(3), 245-261. Retrieved June 19, 2003, from Proquest database.


Zhang, L.F. (2002e). Measuring thinking styles in addition to measuring personality traits? Personality and Individual Differences, 33(3), 445-458. Retrieved August 03, 2002, from ScienceDirect database.


Zhang, L.F. (2002f). The role of thinking styles in psychosocial development. Journal of College Student Development, 43(5), 696-797.


Zhang, L.F., & Sachs, J. (1997). Assessing thinking styles in the theory of mental self-government: A Hong Kong validity study. Psychological Reports, 81, 915¯928.


Zhang, L.F., & Sternberg, R.J. (1998). Thinking styles, abilities, and academic achievement among Hong Kong university students. Educational Research Journal, 13(1), 41¯62.


Zhang, L.F., & Sternberg, R.J. (2000). Are learning approaches and thinking styles related? A study in two Chinese populations. The Journal of Psychology, 134(5), 469-489.


Zhang, L.F., & Watkins, D. (2001). Cognitive development and student approaches to learning: An investigation of Perry's theory with Chinese and US university students. Higher Education, 41(3), 239-261.




Cite This Article as: Teachers College Record Volume 109 Number 6, 2007, p. 1488-1516
https://www.tcrecord.org ID Number: 13523, Date Accessed: 12/6/2021 10:45:57 AM

Purchase Reprint Rights for this article or review
 
Article Tools
Related Articles

Related Discussion
 
Post a Comment | Read All

About the Author
  • Seval Fer
    Yildiz Technical University, Istanbul
    E-mail Author
    SEVAL FER, Ph.D., is an Associate Professor of the Educational Sciences Department, in the Curriculum and Instruction program, at the Faculty of Education, Yildiz Technical University, Istanbul, Turkey. The author has published five books and several articles related to curriculum and teaching and has taken part in national and international projects of the Ministry of Education in Turkey concerning curriculum and instruction. A current research interest of the author is instructional design.
 
Member Center
In Print
This Month's Issue

Submit
EMAIL

Twitter

RSS