- Email: [email protected]
Measuring thinking styles in addition to measuring personality traits?
Personality and Individual Differences 33 (2002) 445–458 www.elsevier.com/locate/paid
Measuring thinking styles in addition to measuring personality traits? Li-fang Zhang* Department of Education, The University of Hong Kong, Pokfulam Road, Hong Kong Received 14 February 2001; received in revised form 13 August 2001; accepted 6 September 2001
Abstract This paper intends to join the long-standing debate regarding thinking styles and personality traits— should thinking styles be measured in addition to the measurement of personality traits? The means to achieve this goal was to provide empirical evidence as well as to review other studies in the literature. The Thinking Styles Inventory and the NEO Five-Factor Inventory were administered to 267 (67 male and 200 female) students from a large research university in Beijing, People’s Republic of China. Results showed that thinking styles and personality traits statistically overlap. However, this overlap is limited. Two major arguments are made. First, thinking styles make a unique contribution to the understanding of human individual differences. Second, the necessity for measuring thinking styles apart from measuring personality traits depends on who uses the inventories and for what purposes. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Thinking styles; Personality traits; Measurement
1. Introduction The period between the late 1950s and the early 1970s saw a proliferation of theories of and research on styles that are most often termed variously as cognitive, learning, and thinking styles. Recently, there has been a resurgence of interest in the work on styles. Being dissatisfied with the old theoretical models of styles that mostly address one style dimension with bipolar styles (e.g. field-dependence versus field independence, and reflectivity versus impulsivity), Sternberg (1988, 1997) proposed a theory of thinking styles.
* Tel./Fax: +852-2859-2522. E-mail address: [email protected] 0191-8869/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0191-8869(01)00166-0
446
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
In Sternberg’s theory, the metaphor ‘‘mental self-government’’ is used to portray how the human mind works. Just like there are different ways of governing a society, there are different ways of managing people’s daily activities. These different ways of managing our activities are construed as ‘‘thinking styles’’. The theory proposed 13 thinking styles that fall along five dimensions. These are functions (including the legislative, executive, and judicial thinking styles), forms (including the hierarchical, oligarchic, monarchic, and anarchic styles), levels (including the global and local styles), scopes (including the internal and external styles), and leaning (including the liberal and conservative styles) of the mental self-government. A brief description of the key characteristics of each style is provided in Appendix A. The theory of mental self-government has been operationalized through several inventories, including the Thinking Styles Inventory (TSI, Sternberg & Wagner, 1992). The TSI has been tested among many samples in a few different cultures, including mainland China, Hong Kong, the Philippines, and the United States. With the majority of research participants being university students, these studies have obtained good reliability data, given the heterogeneity of the test items in the inventory. Internal validity data have been obtained through both Pearson’s correlations among the 13 scales and exploratory factor analysis. Furthermore, external validity of the theory has been obtained by testing the scales in the TSI against other constructs such as Biggs’s learning approaches (Zhang, 2000c; Zhang & Sternberg, 2000) and Holland’s personality types (Zhang, 2000a, 2001a) that are in the family of styles work. In addition, the nature of thinking styles also have been examined by investigations of the relationships of thinking styles to such variables as the research participants’ personal characteristics (e.g. age, gender, birth-order, socio-economic status), their situational characteristics (e.g. leadership experience and travel experience), academic achievement, and self-esteem. Detailed findings of this research can be found in Zhang and her colleagues’ previous studies (e.g. Zhang, 2001b; Zhang & Postiglione, 2001; Zhang & Sachs, 1997). However, nothing is known about the relationships of thinking styles to the five-factor personality model, one of the most prominent and heuristic models of personality in psychology. For the TSI, no published work exists that deals specifically with the relationship between the thinking styles as described in the theory of mental self-government and the big five personality traits. For a relatively young science like psychology, there is a somewhat unique consensus about the description of personality based on five universal traits (e.g. Busato, Prins, Elshout, & Hamaker, 1999; Costa & McCrae, 1992; Digman, 1990). Some scholars (e.g. Goldberg, 1993; Taylor & MacDonald, 1999) asserted that the big five personality traits model accounts for most of the variability in personality. The five factor model (FFM) is the product of several decades of factor analytic research centering around trait personality. According to Taylor and MacDonald (1999), the model was originally proposed by Galton (1884) and empirically followed up by Allport and Odbert (1936) and Norman (1963) among many other scholars. The FFM can be understood as a theory of normal personality traits which is composed of five essentially independent dimensions that have been reliably obtained across extensive investigations. The five personality dimensions are Neuroticism (N), Extraversion (E), Openness to Experience (O), Agreeableness (A), and Conscientiousness (C). The following paragraph is a brief introduction of each of the five personality dimensions as illustrated in Costa and McCrae’s (1992) work. Neuroticism is the opposite of emotional stability. People high on the N scale tend to experience such negative feelings as emotional instability, embarrassment, guilt, pessimism, and low
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
447
self-esteem. People scoring high on the E scale tend to be sociable and assertive, and they prefer to work with other people. Openness to Experience is characterized by such attributes as openmindedness, active imagination, preference for variety, and independence of judgment. In addition, people who score high on the O scale tend to be less conservative and traditional. People high on the A scale tend to be tolerant, trusting, accepting, and easily moved. Furthermore, they value and respect other people’s beliefs and conventions. People high on the C scale are characterized as being organized, purposeful, strong-willed, responsible, and trustworthy. Also, they tend to be task-focused and achievement-oriented. In literature, the relationships between the style construct and the personality construct have been explicitly dealt with at both the conceptual level (e.g. Adorno, Frenkel-Brunswick, Levinson, and Sanford, 1950; Eysenck, 1978; Hashway, 1998; Messick, 1996) and the empirical level (e.g. Busato et al., 1999; Furnham, 1992, 1996a, 1996b; Furnham, Jackson, & Miller, 1999; Gadzella, 1999; Jackson & Lawty-Jones, 1996; Pacini & Epstein, 1999; Riding & Wigley, 1997). However, in the realm of empirical research, two different conclusions have been drawn regarding the necessity of the assessment of styles. Some scholars (e.g. Busato et al., 1999; Pacini & Epstein, 1999; Riding & Wigley, 1997) concluded that although there was some systematic overlap between cognitive/learning styles and personality, it certainly makes sense to mention cognitive/ learning styles and personality separately in educational settings. On the contrary, other scholars (e.g. Furnham, 1992, 1996a, 1996b; Furnham,et al, 1999; Jackson & Lawty-Jones, 1996) argued that since cognitive/learning style is a sub-set of personality, there is no need to measure cognitive/ learning styles independently, unless when cognitive/learning style is of interest in its own right. The present study aims at investigating the relationships between thinking styles as measured by Sternberg and Wagner’s (1992) Thinking Styles Inventory and the five personality dimensions as measured by the NEO Five-Factor Inventory (Costa & McCrae, 1992). Based on the empirical findings from this research, the author will present her own view regarding the necessity of assessing thinking styles in addition to assessing personality traits. Based on the definition of each of the 13 thinking styles and of the personality dimensions, the following predictions were made. First, Neuroticism should be positively related to the executive and conservative thinking styles, but negatively related to the legislative, hierarchical, and liberal thinking styles. Second, Extraversion should be positively related to the external style, but negatively to the internal style. Third, the Openness dimension should be positively correlated with the legislative, judicial, and the liberal thinking styles, but negatively correlated with the executive and conservative thinking styles. Fourth, the Agreeableness dimension should be positively correlated with the external style, but negatively with the legislative, judicial, liberal, and internal styles. Fifth, the Conscientiousness dimension should be positively related to the majority of the thinking styles, especially to the hierarchical style.
2. Method 2.1. Sample The sample comprised 267 students (67 males and 200 females) from a large research university in Beijing, People’s Republic of China. The participants’ ages ranged from 17 to 29 years, with 93.2%
448
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
of the participants falling between 19 and 22 years—the ages for traditional university students in mainland China. The participants were studying in the areas of biology, education, and geography. 2.2. Materials All research participants responded to the TSI (Sternberg & Wagner, 1992) and the NEO FiveFactor Inventory (NEO-FFI, Costa & McCrae, 1992). The TSI is a self-report test consisting of 65 statements. Each five statements assess one of the 13 thinking styles in the theory of mental self-government. For each statement, the participants rated themselves on a seven-point Likert scale, with one indicating that the statement does not at all represent the way they normally carry out their tasks, and seven indicating that the statement describes extremely well the way they normally carry out their tasks. Examples of items on the inventory are ’’I like tasks that allow me to do things my own way’’ (legislative), ‘‘I like situations in which it is clear what role I must play or in what way I should participate’’ (executive), and ‘‘I like to evaluate and compare different points of view on issues that interest me’’ (judicial). The present study employed a Chinese version of the inventory that was translated and backtranslated between Chinese and English (Zhang, 1996). Since no significant relationship was expected of the oligarchic and anarchic thinking styles to any of five personality dimensions, these two thinking styles were not included in the present study. As mentioned earlier, the TSI has been tested on numerous occasions in a number of cultures. Good reliability data and validity data have been obtained. In the present study, the Cronbach’s alphas are 0.74, 0.68, 0.71, 0.71, 0.58, 0.85, 0.82, 0.75, 0.80, 0.81, and 0.53, respectively for the legislative, executive, judicial, global, local, liberal, conservative, internal, external, hierarchical, and monarchic thinking styles. These reliability estimates are comparable with the ones obtained in previous studies. The NEO-FFI also is a self-report test. Composed of 60 short items, the inventory is considered as a brief and comprehensive measure of the five personality dimensions. Each dimension is assessed by 12 items. For each item, the participants indicated one of five choices: strongly disagree, disagree, neutral, agree, and strongly agree. The scoring points are 0, 1, 2, 3, and 4. The items are phrased both positively and negatively. Therefore, whereas indicating ‘‘strongly agree’’ to one item may result in a score of zero, indicating ‘‘strongly agree’’ to another item may result in a score of four. The total score for each personality dimension is the sum of the scores on the 12 items in each scale. Appendix B provides 10 sample items. Each two items belong to one of the five scales, with one item scoring zero and the other scoring four. The NEO-FFI has good reliability and validity data that are reported in the NEO Personality Inventory-R Professional Manual (Costa & McCrae, 1992). The present study, again, employed a Chinese version of the NEO-FFI that was translated and back-translated between Chinese and English (Zhang, 2000b). The Cronbach’s alphas for the present study are 0.86, 0.77, 0.52, 0.65, and 0.81, respectively for the neuroticism, extraversion, openness, agreeableness, and conscientiousness personality dimensions. 2.3. Data analysis Previous research has suggested that both thinking styles (e.g. Sternberg & Grigorenko, 1995; Zhang & Sachs, 1997) and the five personality traits (e.g. Costa & McCrae, 1992; Taylor &
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
449
MacDonald, 1999) could vary as a function of a variety of variables, including age, gender, birthorder, and students’ major fields of studies. Therefore, in the present study, preliminary statistical tests were conducted to identify possible significant differences based on the aforementioned variables. Statistically significant differences were found in both thinking styles and personality traits based on gender, but not on the remaining variables. Two major findings were obtained regarding gender differences. First, male students scored significantly higher than did their female counterparts on a number of thinking style scales, including the legislative, executive, judicial, global, liberal, and monarchic styles. Second, female students scored significantly higher than did their male counterparts on the agreeableness personality scale. Because of the gender differences found, the remaining statistical analyses were conducted for males and females separately as well as for the entire sample. To examine the relationships between thinking styles and personality traits, the following statistical analyses were performed. First, a zero-order correlation matrix was computed with the 11 thinking styles scales as one set of the variables and the five personality scales as the other. The aim of this analysis was to provide a clear picture of how the scales from the two inventories are correlated with one another. Second, an exploratory factor analysis (principal components, oblimin rotated) was performed on the 11 thinking styles scales and the five personality scales. This analysis was performed based on the assumption that if the thinking style construct and the personality construct overlap, high factor loadings for a particular resulting factor should be from scales of both inventories. In other words, the scales from the two inventories should share common variance in the data. Third, in order to further examine the extent to which the five personality traits can statistically predict thinking styles, stepwise multiple-regression analyses were performed with the five personality scales being the independent variables and the thinking styles scales being the dependent variables.
3. Results By visual inspection, the patterns of the relationships resulted from the gender-separated and gender-combined analyses are similar. Therefore, only results from the gender-combined analyses are reported later. 3.1. Zero-order TSI and NEO-FFI scale correlations Pearson’s correlation coefficients among the scales from the two inventories are provided in Table 1. These correlations lend strong support to the hypothesized relationships among the scales across the two inventories. Specifically, the predicted relationships regarding neuroticism, openness, and conscientiousness and their respective related thinking styles were fully supported. Partial support was obtained for the predicted relationships regarding the extraversion and the agreeableness scales and their respective related thinking styles. First, as predicted, the Extraversion scale was positively related to the external style; however, unlike predicted, the Extraversion scale was not negatively related to the internal style, although the two scales were negatively correlated, as predicted. Second, as expected, the agreeableness scale was positively correlated with the external style; however, unlike expected, the agreeableness scale was not negatively
450
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
related to the legislative, judicial, liberal, and internal styles, although the correlations were in the predicted direction, that is, negative. 3.2. Factor analysis Exploratory principal-component analysis (with an oblique rotation) yielded five factors with eigenvalues greater than one. These five factors accounted for 70% of the variance in the data. The variance for each of the factors was shared by scales from both inventories, suggesting significant overlap between the two measures. Factor one was dominated by high loadings of the openness dimension as well as of the legislative, judicial, liberal, and internal thinking styles. Factor two was dominated by the neuroticism dimension as well as by the executive, local, and conservative thinking styles. Factor three was dominated by a positive factor loading from the neuroticism dimension as well as by the negative loadings of the hierarchical thinking style and the extraversion, agreeableness, and conscientiousness dimensions. Factor four was dominated by positive loadings of the external thinking style and the extraversion and agreeableness dimensions as well as by the negative loadings of the legislative and the internal thinking styles. Finally, factor five was dominated by a positive loading on the agreeableness dimension and negative loadings of the global and monarchic thinking styles. Table 2 provides the detailed results of this factor analysis. 3.3. Multiple-regression analysis To determine the extent to which personality traits and thinking styles overlap, stepwise multiple-regression analyses were conducted, using the five personality scales as predictors of each of the thinking style scales. As shown in Table 3, all five personality dimensions were significant predictors of all of the thinking styles, although it varied from 1 to 5 personality scales predicting each of the 11 thinking styles. For example, the conscientiousness dimension alone predicted the hierarchical thinking style (R2=0.27, = 0.52). Three personality dimensions (extraversion, openness, and agreeableness) contributed to the prediction of the liberal thinking style (R2=0.19, extraversion=0.33, openness=0.26, agreeableness= 0.18). Yet, another example, all
Table 1 Overlaps between Thinking Styles and Big Five Personality Traits (N=267) Scale Neuro Extra Openness Agree Conscien
Leg 0.31** 0.21** 0.21** 0.06 0.19**
Exe 0.18** 0.04 0.25** 0.06 0.26**
Jud 0.09 0.24** 0.34** 0.01 0.13*
Global Local 0.06 0.07 0.14* 0.14* 0.08
0.09 0.11 0.17** 0.04 0.25**
Lib 0.24** 0.33** 0.27** 0.02 0.14*
Con 0.35** 0.22** 0.26** 0.00 0.06
Internal External Hier 0.15* 0.02 0.23** 0.11 0.20**
0.02 0.37** 0.03 0.24** 0.13*
0.20** 0.21** 0.09 0.04 0.51**
Mona 0.02 0.12* 0.09 0.10 0.19**
Leg, Legislative; Exe, Executive; Jud, Judicial; Lib, Liberal; Con, Conservative; Hier, Hierarchical; Mona, Monarchic; Neuro, Neuroticism; Extra, Extraversion; Agree, Agreeableness; Conscien, Conscientiousness. * P< 0.05. ** P< 0.01.
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
451
five personality dimensions contributed to the significant prediction of the internal thinking style (R2=0.18, openess=0.25, conscientiousness=0.21, agreeableness= 0.26, neuroticism= 0.25, extraversion= 0.21).
4. Discussion The present study aimed at answering the question about the necessity for measuring thinking styles in addition to measuring personality traits by exploring the relationships between the thinking styles proposed in Sternberg’s theory of mental self-government and the big five personality traits. The hypothesized relationships were tested by three statistical procedures. As presented above, results from both the zero-order correlation matrix and the exploratory factor analysis suggested fairly strong, although not complete, support for the predictions. Regarding scale correlations across the two inventories, complete support was obtained for the predictions on three of the five personality dimensions (neuroticism, openness, and conscientiousness) and their counterpart scales in the TSI. These significant relationships found between the scales of the two measures make substantive sense. Take the significant relationship of the openness dimension
Table 2 Oblimin-rotated Principal Component Factor Analysis of the Thinking Styles and Personality Traits Scales (N=267) Scale TSI Legislative Executive Judicial Global Local Liberal Conservative Internal External Hierarchical Monarchic NEO-FFI Neuroticism Extraversion Openness Agreeableness Conscientiousness % Variance Cumulative Variance Eigenvalues
Factor 1
Factor 2
Factor 3
Factor 4
Factor 5
0.33 0.63 0.80 0.75 0.86 0.83 0.69 0.74 0.49
0.65 0.78 0.50 0.72
0.35
0.84 0.68
0.38
0.66
24.92 24.92 3.99
16.88 41.83 2.70
0.38 0.77 12.79 54.62 2.05
0.35
0.38
8.49 63.12 1.36
6.57 69.69 1.05
Scales with factor loadings of less than 0.30 are omitted. TSI, Thinking Styles Inventory; NEO-FFI, NEO Five-Factor Inventory.
452
Thinking Styles Legislative Executive Judicial Neuroticism Extraversion Openness Agreeableness Conscientiousness R2 F d.f. * P< 0.05. ** P< 0.01. *** P< 0.001.
0.35*** 0.20** 0.13* 0.15 14.18*** (3, 240)
0.41***
Global
Local
Liberal
0.21** 0.17** 0.34***
Conservative Hierarchical Monarchic Internal 0.49***
0.33*** 0.26*** 0.18**
0.23*** 0.15* 0.15* 0.23*** 0.19** 0.19** 0.20** 0.39*** 0.32*** 0.22** 0.24 0.16 0.05 0.12 0.19 0.23 19.02*** 22.33*** 6.93** 10.32*** 18.27*** 17.97*** (4, 240) (2, 239) (2, 241) (3, 239) (3, 237) (4, 240)
0.52*** 0.27 89.76*** (1, 243)
0.15* 0.24*** 0.07 8.82*** (2, 242)
External
0.25** 0.37*** 0.21** 0.52*** 0.25*** 0.26*** 0.27*** 0.21** 0.18 0.26 10.17*** 27.78*** (5, 239) (3, 241)
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
Table 3 Predicting Thinking Styles scores using the Big Five Personality Traits (N=267)
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
453
to the legislative, judicial, and liberal thinking styles, for example. Openness is characterized by such attributes as open-mindedness, active imagination, and preference for variety. In a similar vein, people who are high on the legislative, judicial, and liberal thinking styles also tend to be open-minded, imaginative, and perceptive. Another example is the significant relationship between the Neuroticism dimension and the executive and conservative thinking styles. People who are high on the neuroticism dimension tend to be emotionally unstable, easily embarrassed, and pessimistic, and they often suffer from low self-esteem. Understandably, individuals who experience such negative feelings would be more comfortable with working under highly structured situations, with being told what to do, and with carrying out tasks by going along existing rules. All these ways of dealing with tasks best represent the ways by which individuals with the executive and conservative thinking styles would deal with their tasks. However, only partial support was obtained for the predictions on the remaining two personality dimensions (extraversion and agreeableness). Both extraversion and agreeableness were positively correlated with the external thinking style, as expected. However, the to-be-expected relationship between the extraversion scale and the internal style and the relationships of agreeableness to the legislative, judicial, liberal, and the internal styles were not found. This indicates that one may still prefer to work independently (using the internal thinking style) even though one scores high on extraversion. Moreover, this finding suggests that people who are agreeable may still be legislative, judicial, liberal, and internal in their thinking styles. Therefore, the fact that complete support for the to-be-expected relationships was not obtained suggests that one cannot infer, with confidence, people’s thinking styles from their scores on the five personality dimensions. Instead, both the personality measure and the thinking styles measure play a unique role in the data. Results from the factor analysis lent less support to the hypothesized relationships between the thinking styles and the personality dimensions than did the zero-order correlation coefficients. Whereas three of the hypotheses were fully supported (with the remaining two being partially supported) by the computation of correlation matrix, all five predictions were partially supported by the factor analysis results. Furthermore, variance in each of the five factors was shared by scales from both inventories; and none of the five factors was solely loaded by scales from the NEO-FFI. These results from the factor analysis, like those from the zero-order correlations, suggest that one cannot predict, with confidence, people’s thinking styles from their scores on the five personality scales. Finally, results from the multiple-regression analyses revealed the extent to which the big five personality dimensions can statistically predict thinking styles. According to these results, the five personality dimensions accounted for only 5–27% of the variance in the thinking styles scales. The remaining large amount of variance was left unexplained. Therefore, once again, it is evident that the Five-Factor Personality Inventory cannot be used to predict thinking styles with great confidence. It should be noted, however, a study such as this has a major limitation that renders its findings as suggestive rather than conclusive. That is, the findings are completely based on self-report data. It is understood that self-report data are not always highly related to data obtained through behavioral measures. To verify the findings from this study, not only replications of the present study should be conducted among a wide range of populations, but also research involving the use of behavioral measures should be undertaken.
454
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
5. Conclusion Through an empirical investigation, the present study attempted to join the debate on whether or not it would be necessary to assess styles (cognitive/learning/thinking) in addition to assessing personality. Substantial overlap was identified between a thinking styles measure and a measure of the big five personality dimensions. This substantial overlap supported both the conceptual argument for (e.g. Adorno et al., 1950; Eysenck, 1978; Hashway, 1998; Messick, 1996) and the empirical support to (e.g., Busato et al., 1999; Furnham, 1992, 1996a, 1996b; Furnham et al., 1999; Gadzella, 1999; Jackson & Lawty-Jones, 1996; Pacini & Epstein, 1999; Riding & Wigley, 1997) the relationships between personality and styles. It seems undeniable that personality and styles have a strong relationship. The question that one needs to answer is whether or not it makes sense to measure styles separately from personality. The answer that derives from the present study is affirmative. There are three reasons for this affirmative answer. First, in the empirical literature on the relationship between personality and styles, conclusions about the necessity of measuring styles separately have been subjective. Some writers (e.g. Busato et al., 1999) may conclude that personality and styles are two different constructs (and thus need to be measured separately) even when a strong relationship has been found, whereas other scholars (e.g. Furnham et al., 1999) may assert that styles is a subset of personality (and thus do not need to be measured separately) when only a moderately significant relationship has been identified. To this end, findings about the extent to which styles and personality correlate have been inconsistent. For example, Furnham, one of the strong advocates for parsimony of measurement, identified different degrees of relationships in different investigations. Whereas he found substantial overlap between the two constructs in his 1992 study, he only identified moderate correlations in one of his 1996 studies (Furnham, 1996b). In their 1999 study, Furnham and his colleagues found no significant relationship between learning styles and neuroticism, one of the major personality dimensions. Second, results from the present study indicate that the big five personality traits and the thinking styles are not the same since thinking styles made a unique contribution to the variance in the data. Personality dimensions can account for only a little over a quarter of the variance in thinking styles at the most. Indeed, a careful examination of the semantics of the two inventories shows that the two inventories have different orientations. Whereas the items in the TSI assess how the participants deal with different tasks, the items in the Five-Factor Personality Inventory focus on finding out how the participants feel about their environments. Therefore, it is not surprising that thinking styles make a unique contribution to the variance in the data. Third, there is no evidence that personality determines thinking styles. Instead, it could be argued that personality and thinking styles are inextricably entwined. Thus, the fact that the two measures overlap should not warrant the futility of the assessment of either one of the two constructs. On the contrary, knowledge about thinking styles as well as about personality contributes to our understanding of individual differences among human beings.
6. Future directions The present study, like many other studies in the literature, has demonstrated clearly that styles and personality traits are strongly related. It also seems to be fair to say that styles have their
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
455
unique value in explaining human behavior. This conclusion is supportive of Busato, Prins, Elshout, and Hamaker (2000) assertion that a styles inventory has a diagnostic value for students at an individual level for identifying both strengths and weaknesses in, say, their individual study behavior. Then, the ultimate goal of this paper is to answer the question of whether or not we should measure thinking styles in addition to measuring personality traits. The answer is that it depends on who the ‘‘we’’ refers to and for what purposes the inventories would be used. If the ‘‘we’’ refers to such groups as university teachers and counselors, the answer would be that: ‘‘It depends on the purpose and time available’’. If one’s purpose is to have a rough understanding about a student’s thinking style or personality profile and he/she cannot afford the time using both inventories, or he/she does not have access to both inventories, then, by all means, using either one of the inventories should help the teacher/counselor to have a basic understanding of his/her student’s thinking style or personality profile. However, if the assessment has implications for serious decisions such as placing students in particular programs or classes, I would argue for an assessment of both thinking styles and personality traits, among the use of other assessment procedures. Yet, if the ‘‘we’’ refers to researchers who aim at understanding both thinking styles and personality traits, then the question seems not to be whether or not styles should be measured in addition to the measurement of personality, but how to measure each of the two constructs. To achieve this goal, our future research needs to focus on how styles and personality traits work together to affect human behaviors–students’ behaviors, in the context of educational settings.
Acknowledgements I want to thank the Committee on Research and Conference Grants at The university of Hong Kong for having made this research possible. My sincere thanks go to L. L. Zhang for her kind assistance in the data collection and to all those anonymous students whose participation is indispensable to this research.
Appendix A Thinking Styles in the Theory of Mental Self-government Thinking Style
Description
Legislative
One prefers to work on tasks that require creative strategies; One prefers to choose one’s own activities. One prefers to work on tasks with clear instructions and structures; One prefers to implement tasks with set guidelines. One prefers to work on tasks that allow for one’s evaluation; One prefers to evaluate and judge the performance of other people. One prefers to distribute attention to several tasks that are prioritized according to one’s valuing of the tasks.
Executive Judicial Hierarchical
(continued on next page)
456
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
Appendix continued
Thinking Style
Description
Monarchical
One prefers to work on tasks that allow complete focus on one thing at a time. One prefers to work on multiple tasks in the service of multiple objectives, without setting priorities. One prefers to work on tasks that would allow flexibility as to what, where, when, and how one works. One prefers to pay more attention to the overall picture of an issue and to abstract ideas. One prefers to work on tasks that require working with concrete details. One prefers to work on tasks that allow one to work as an independent unit. One prefers to work on tasks that allow for collaborative ventures with other people. One prefers to work on tasks that involve novelty and ambiguity. One prefers to work on tasks that allow one to adhere to the existing rules and procedures in performing tasks.
Oligarchic Anarchic Global Local Internal External Liberal Conservative
Appendix B Sample items from the NEO Five-Factor Inventory Scale Neuroticism
Extraversion
Openness
Agreeableness
Conscientiousness
a
Scorea
Sample item I am not a worrier I often feel inferior to others
0
I do not consider myself especially ‘‘light-hearted’’ I like to have a lot of people around me
0
I don’t like to waste my time daydreaming I am intrigued by the patterns I find in art and nature
0
I often get into arguments with my family and co-workers I try to be courteous to everyone I meet
0
I waste a lot of time before settling down to work I keep my belongings clean and neat
0
Scores are based on verbal anchor ‘‘Strongly Agree’’.
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
457
References Adorno, T. W., Brunswik, E. F., Levinson, D. J., & Sanford, R. N. (1950). The authoritarian personality. New York, NY: Harper & Row. Allport, G. W., & Odbert, H. S. (1936). Trait names: a psycho-lexical study. Psychological Monographs, 47(1). Busato, V. V., Prins, F. J., Elshout, J. J., & Hamaker, C. (1999). The relation between learning styles, the Big Five personality traits and achievement motivation in higher education. Personality and Individual Differences, 26(1), 129– 140. Busato, V. V., Prins, F. J., Elshout, J. J., & Hamaker, C. (2000). Intellectual ability, learning style, personality, achievement motivation and academic success of psychology students in higher education. Personality and Individual Differences, 29, 1057–1068. Costa, P. T. Jr., & McCrae, R. R. (1992). The NEO-PI-R: professional manual. Odessa, FL: Psychological Assessment Resources. Digman, J. M. (1990). Personality structure: emergence of the five-factor model. Annual Review of Psychology, 41, 417– 440. Eysenck, H. J. (1978). The development of personality and its relation to learning. In S. Murray-Smith (Ed.), Melbourne studies in education (pp. 134–181). Australia: Melbourne University Press. Furnham, A. (1992). Personality and learning style: a study of three instruments. Personality and Individual Differences, 13(4), 429–438. Furnham, A. F. (1996a). The big five versus the big four: the relationship between the Myers-Briggs type indicator (MBTI) and NEO-PI five factor model of personality. Personality and Individual Differences, 21(2), 303–307. Furnham, A. F. (1996b). The FIRO-B, the learning style questionnaire and the five-factor model. Journal of Social Behavior and Personality, 11(2), 285–299. Furnham, A., Jackson, C. J., & Miller, T. (1999). Personality, learning style and work performance. Personality and Individual Differences, 27, 1113–1122. Gadzella, B. M. (1999). Differences among cognitive-processing styles groups on personality traits. Journal of Instructional Psychology, 26(3), 161–166. Galton, F. (1884). Measurement of character. Fortnightly Review, 36, 179–185. Goldberg, L. R. (1993). The structure of phenotypic personality traits. American Psychologist, 48(1), 26–34. Hashway, R. M. (1998). Developmental cognitive styles: a primer to the literature including an introduction to the theory of developmentalism. Austin & Winfield: Bethesda, MD. Jackson, C., & Lawty-Jones, M. (1996). Explaining the overlap between personality and learning style. Personality and Individual Differences, 20(3), 293–300. Messick, S. (1996). Bridging cognition and personality in education: the role of style in performance and development. European Journal of Personality, 10, 353–376. Norman, W. T. (1963). Toward an adequate taxonomy of personality attributes: replicated factor structure in peer nomination personality ratings. Journal of Abnormal and Social Psychology, 66, 574–583. Pacini, R., & Epstein, S. (1999). The relation of rational and experiential information processing styles to personality, basic beliefs, and the ratio-bias phenomenon. Journal of Personality and Social Psychology, 76(6), 972–987. Riding, R. J., & Wigley, S. (1997). The relationship between cognitive style and personality in further education students. Personality and Individual Differences, 23(3), 379–389. Sternberg, R. J. (1988). Mental self-government: a theory of intellectual styles and their development. Human Development, 31, 197–224. Sternberg, R. J. (1997). Thinking styles. New York: Cambridge University Press. Sternberg, R. J., & Grigorenko, E. L. (1995). Styles of thinking in the school. European Journal for High Ability, 6, 201–219. Sternberg, R. J., & Wagner, R. K. (1992). Thinking styles inventory. (Unpublished test, Yale University). Taylor, A., & MacDonald, D. A. (1999). Religion and the five factor model of personality: an exploratory investigation using a Canadian university sample. Personality and Individual Differences, 27, 1243–1259. Zhang, L. F. (1996). Chinese translation of the Thinking Styles Inventory. (Unpublished test, The University of Hong Kong).
458
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
Zhang, L. F. (2000a). Are thinking styles and personality types related?. Educational Psychology, 20(3), 271–283. Zhang, L. F. (2000b). Chinese translation of the NEO Five-Factor Inventory. (Unpublished test, The University of Hong Kong). Zhang, L. F. (2000c). Relationship between Thinking Styles Inventory and Study Process Questionnaire. Personality and Individual Differences, 29, 841–856. Zhang, L. F. (2001a). Thinking styles and personality types revisited. Personality and Individual Differences, 31, 883– 894. Zhang, L. F. (2001b). Thinking styles, self-esteem, and extracurricular experiences. International Journal of Psychology, 36(2), 100–107. Zhang, L. F., & Postiglione, G. A. (2001). Thinking styles, self-esteem, and socio-economic status. Personality and Individual Differences (in press). 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. (2000). Are learning approaches and thinking styles related? A study in two Chinese populations.. The Journal of Psychology, 134(5), 469–489.
Measuring thinking styles in addition to measuring personality traits? Li-fang Zhang* Department of Education, The University of Hong Kong, Pokfulam Road, Hong Kong Received 14 February 2001; received in revised form 13 August 2001; accepted 6 September 2001
Abstract This paper intends to join the long-standing debate regarding thinking styles and personality traits— should thinking styles be measured in addition to the measurement of personality traits? The means to achieve this goal was to provide empirical evidence as well as to review other studies in the literature. The Thinking Styles Inventory and the NEO Five-Factor Inventory were administered to 267 (67 male and 200 female) students from a large research university in Beijing, People’s Republic of China. Results showed that thinking styles and personality traits statistically overlap. However, this overlap is limited. Two major arguments are made. First, thinking styles make a unique contribution to the understanding of human individual differences. Second, the necessity for measuring thinking styles apart from measuring personality traits depends on who uses the inventories and for what purposes. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Thinking styles; Personality traits; Measurement
1. Introduction The period between the late 1950s and the early 1970s saw a proliferation of theories of and research on styles that are most often termed variously as cognitive, learning, and thinking styles. Recently, there has been a resurgence of interest in the work on styles. Being dissatisfied with the old theoretical models of styles that mostly address one style dimension with bipolar styles (e.g. field-dependence versus field independence, and reflectivity versus impulsivity), Sternberg (1988, 1997) proposed a theory of thinking styles.
* Tel./Fax: +852-2859-2522. E-mail address: [email protected] 0191-8869/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0191-8869(01)00166-0
446
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
In Sternberg’s theory, the metaphor ‘‘mental self-government’’ is used to portray how the human mind works. Just like there are different ways of governing a society, there are different ways of managing people’s daily activities. These different ways of managing our activities are construed as ‘‘thinking styles’’. The theory proposed 13 thinking styles that fall along five dimensions. These are functions (including the legislative, executive, and judicial thinking styles), forms (including the hierarchical, oligarchic, monarchic, and anarchic styles), levels (including the global and local styles), scopes (including the internal and external styles), and leaning (including the liberal and conservative styles) of the mental self-government. A brief description of the key characteristics of each style is provided in Appendix A. The theory of mental self-government has been operationalized through several inventories, including the Thinking Styles Inventory (TSI, Sternberg & Wagner, 1992). The TSI has been tested among many samples in a few different cultures, including mainland China, Hong Kong, the Philippines, and the United States. With the majority of research participants being university students, these studies have obtained good reliability data, given the heterogeneity of the test items in the inventory. Internal validity data have been obtained through both Pearson’s correlations among the 13 scales and exploratory factor analysis. Furthermore, external validity of the theory has been obtained by testing the scales in the TSI against other constructs such as Biggs’s learning approaches (Zhang, 2000c; Zhang & Sternberg, 2000) and Holland’s personality types (Zhang, 2000a, 2001a) that are in the family of styles work. In addition, the nature of thinking styles also have been examined by investigations of the relationships of thinking styles to such variables as the research participants’ personal characteristics (e.g. age, gender, birth-order, socio-economic status), their situational characteristics (e.g. leadership experience and travel experience), academic achievement, and self-esteem. Detailed findings of this research can be found in Zhang and her colleagues’ previous studies (e.g. Zhang, 2001b; Zhang & Postiglione, 2001; Zhang & Sachs, 1997). However, nothing is known about the relationships of thinking styles to the five-factor personality model, one of the most prominent and heuristic models of personality in psychology. For the TSI, no published work exists that deals specifically with the relationship between the thinking styles as described in the theory of mental self-government and the big five personality traits. For a relatively young science like psychology, there is a somewhat unique consensus about the description of personality based on five universal traits (e.g. Busato, Prins, Elshout, & Hamaker, 1999; Costa & McCrae, 1992; Digman, 1990). Some scholars (e.g. Goldberg, 1993; Taylor & MacDonald, 1999) asserted that the big five personality traits model accounts for most of the variability in personality. The five factor model (FFM) is the product of several decades of factor analytic research centering around trait personality. According to Taylor and MacDonald (1999), the model was originally proposed by Galton (1884) and empirically followed up by Allport and Odbert (1936) and Norman (1963) among many other scholars. The FFM can be understood as a theory of normal personality traits which is composed of five essentially independent dimensions that have been reliably obtained across extensive investigations. The five personality dimensions are Neuroticism (N), Extraversion (E), Openness to Experience (O), Agreeableness (A), and Conscientiousness (C). The following paragraph is a brief introduction of each of the five personality dimensions as illustrated in Costa and McCrae’s (1992) work. Neuroticism is the opposite of emotional stability. People high on the N scale tend to experience such negative feelings as emotional instability, embarrassment, guilt, pessimism, and low
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
447
self-esteem. People scoring high on the E scale tend to be sociable and assertive, and they prefer to work with other people. Openness to Experience is characterized by such attributes as openmindedness, active imagination, preference for variety, and independence of judgment. In addition, people who score high on the O scale tend to be less conservative and traditional. People high on the A scale tend to be tolerant, trusting, accepting, and easily moved. Furthermore, they value and respect other people’s beliefs and conventions. People high on the C scale are characterized as being organized, purposeful, strong-willed, responsible, and trustworthy. Also, they tend to be task-focused and achievement-oriented. In literature, the relationships between the style construct and the personality construct have been explicitly dealt with at both the conceptual level (e.g. Adorno, Frenkel-Brunswick, Levinson, and Sanford, 1950; Eysenck, 1978; Hashway, 1998; Messick, 1996) and the empirical level (e.g. Busato et al., 1999; Furnham, 1992, 1996a, 1996b; Furnham, Jackson, & Miller, 1999; Gadzella, 1999; Jackson & Lawty-Jones, 1996; Pacini & Epstein, 1999; Riding & Wigley, 1997). However, in the realm of empirical research, two different conclusions have been drawn regarding the necessity of the assessment of styles. Some scholars (e.g. Busato et al., 1999; Pacini & Epstein, 1999; Riding & Wigley, 1997) concluded that although there was some systematic overlap between cognitive/learning styles and personality, it certainly makes sense to mention cognitive/ learning styles and personality separately in educational settings. On the contrary, other scholars (e.g. Furnham, 1992, 1996a, 1996b; Furnham,et al, 1999; Jackson & Lawty-Jones, 1996) argued that since cognitive/learning style is a sub-set of personality, there is no need to measure cognitive/ learning styles independently, unless when cognitive/learning style is of interest in its own right. The present study aims at investigating the relationships between thinking styles as measured by Sternberg and Wagner’s (1992) Thinking Styles Inventory and the five personality dimensions as measured by the NEO Five-Factor Inventory (Costa & McCrae, 1992). Based on the empirical findings from this research, the author will present her own view regarding the necessity of assessing thinking styles in addition to assessing personality traits. Based on the definition of each of the 13 thinking styles and of the personality dimensions, the following predictions were made. First, Neuroticism should be positively related to the executive and conservative thinking styles, but negatively related to the legislative, hierarchical, and liberal thinking styles. Second, Extraversion should be positively related to the external style, but negatively to the internal style. Third, the Openness dimension should be positively correlated with the legislative, judicial, and the liberal thinking styles, but negatively correlated with the executive and conservative thinking styles. Fourth, the Agreeableness dimension should be positively correlated with the external style, but negatively with the legislative, judicial, liberal, and internal styles. Fifth, the Conscientiousness dimension should be positively related to the majority of the thinking styles, especially to the hierarchical style.
2. Method 2.1. Sample The sample comprised 267 students (67 males and 200 females) from a large research university in Beijing, People’s Republic of China. The participants’ ages ranged from 17 to 29 years, with 93.2%
448
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
of the participants falling between 19 and 22 years—the ages for traditional university students in mainland China. The participants were studying in the areas of biology, education, and geography. 2.2. Materials All research participants responded to the TSI (Sternberg & Wagner, 1992) and the NEO FiveFactor Inventory (NEO-FFI, Costa & McCrae, 1992). The TSI is a self-report test consisting of 65 statements. Each five statements assess one of the 13 thinking styles in the theory of mental self-government. For each statement, the participants rated themselves on a seven-point Likert scale, with one indicating that the statement does not at all represent the way they normally carry out their tasks, and seven indicating that the statement describes extremely well the way they normally carry out their tasks. Examples of items on the inventory are ’’I like tasks that allow me to do things my own way’’ (legislative), ‘‘I like situations in which it is clear what role I must play or in what way I should participate’’ (executive), and ‘‘I like to evaluate and compare different points of view on issues that interest me’’ (judicial). The present study employed a Chinese version of the inventory that was translated and backtranslated between Chinese and English (Zhang, 1996). Since no significant relationship was expected of the oligarchic and anarchic thinking styles to any of five personality dimensions, these two thinking styles were not included in the present study. As mentioned earlier, the TSI has been tested on numerous occasions in a number of cultures. Good reliability data and validity data have been obtained. In the present study, the Cronbach’s alphas are 0.74, 0.68, 0.71, 0.71, 0.58, 0.85, 0.82, 0.75, 0.80, 0.81, and 0.53, respectively for the legislative, executive, judicial, global, local, liberal, conservative, internal, external, hierarchical, and monarchic thinking styles. These reliability estimates are comparable with the ones obtained in previous studies. The NEO-FFI also is a self-report test. Composed of 60 short items, the inventory is considered as a brief and comprehensive measure of the five personality dimensions. Each dimension is assessed by 12 items. For each item, the participants indicated one of five choices: strongly disagree, disagree, neutral, agree, and strongly agree. The scoring points are 0, 1, 2, 3, and 4. The items are phrased both positively and negatively. Therefore, whereas indicating ‘‘strongly agree’’ to one item may result in a score of zero, indicating ‘‘strongly agree’’ to another item may result in a score of four. The total score for each personality dimension is the sum of the scores on the 12 items in each scale. Appendix B provides 10 sample items. Each two items belong to one of the five scales, with one item scoring zero and the other scoring four. The NEO-FFI has good reliability and validity data that are reported in the NEO Personality Inventory-R Professional Manual (Costa & McCrae, 1992). The present study, again, employed a Chinese version of the NEO-FFI that was translated and back-translated between Chinese and English (Zhang, 2000b). The Cronbach’s alphas for the present study are 0.86, 0.77, 0.52, 0.65, and 0.81, respectively for the neuroticism, extraversion, openness, agreeableness, and conscientiousness personality dimensions. 2.3. Data analysis Previous research has suggested that both thinking styles (e.g. Sternberg & Grigorenko, 1995; Zhang & Sachs, 1997) and the five personality traits (e.g. Costa & McCrae, 1992; Taylor &
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
449
MacDonald, 1999) could vary as a function of a variety of variables, including age, gender, birthorder, and students’ major fields of studies. Therefore, in the present study, preliminary statistical tests were conducted to identify possible significant differences based on the aforementioned variables. Statistically significant differences were found in both thinking styles and personality traits based on gender, but not on the remaining variables. Two major findings were obtained regarding gender differences. First, male students scored significantly higher than did their female counterparts on a number of thinking style scales, including the legislative, executive, judicial, global, liberal, and monarchic styles. Second, female students scored significantly higher than did their male counterparts on the agreeableness personality scale. Because of the gender differences found, the remaining statistical analyses were conducted for males and females separately as well as for the entire sample. To examine the relationships between thinking styles and personality traits, the following statistical analyses were performed. First, a zero-order correlation matrix was computed with the 11 thinking styles scales as one set of the variables and the five personality scales as the other. The aim of this analysis was to provide a clear picture of how the scales from the two inventories are correlated with one another. Second, an exploratory factor analysis (principal components, oblimin rotated) was performed on the 11 thinking styles scales and the five personality scales. This analysis was performed based on the assumption that if the thinking style construct and the personality construct overlap, high factor loadings for a particular resulting factor should be from scales of both inventories. In other words, the scales from the two inventories should share common variance in the data. Third, in order to further examine the extent to which the five personality traits can statistically predict thinking styles, stepwise multiple-regression analyses were performed with the five personality scales being the independent variables and the thinking styles scales being the dependent variables.
3. Results By visual inspection, the patterns of the relationships resulted from the gender-separated and gender-combined analyses are similar. Therefore, only results from the gender-combined analyses are reported later. 3.1. Zero-order TSI and NEO-FFI scale correlations Pearson’s correlation coefficients among the scales from the two inventories are provided in Table 1. These correlations lend strong support to the hypothesized relationships among the scales across the two inventories. Specifically, the predicted relationships regarding neuroticism, openness, and conscientiousness and their respective related thinking styles were fully supported. Partial support was obtained for the predicted relationships regarding the extraversion and the agreeableness scales and their respective related thinking styles. First, as predicted, the Extraversion scale was positively related to the external style; however, unlike predicted, the Extraversion scale was not negatively related to the internal style, although the two scales were negatively correlated, as predicted. Second, as expected, the agreeableness scale was positively correlated with the external style; however, unlike expected, the agreeableness scale was not negatively
450
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
related to the legislative, judicial, liberal, and internal styles, although the correlations were in the predicted direction, that is, negative. 3.2. Factor analysis Exploratory principal-component analysis (with an oblique rotation) yielded five factors with eigenvalues greater than one. These five factors accounted for 70% of the variance in the data. The variance for each of the factors was shared by scales from both inventories, suggesting significant overlap between the two measures. Factor one was dominated by high loadings of the openness dimension as well as of the legislative, judicial, liberal, and internal thinking styles. Factor two was dominated by the neuroticism dimension as well as by the executive, local, and conservative thinking styles. Factor three was dominated by a positive factor loading from the neuroticism dimension as well as by the negative loadings of the hierarchical thinking style and the extraversion, agreeableness, and conscientiousness dimensions. Factor four was dominated by positive loadings of the external thinking style and the extraversion and agreeableness dimensions as well as by the negative loadings of the legislative and the internal thinking styles. Finally, factor five was dominated by a positive loading on the agreeableness dimension and negative loadings of the global and monarchic thinking styles. Table 2 provides the detailed results of this factor analysis. 3.3. Multiple-regression analysis To determine the extent to which personality traits and thinking styles overlap, stepwise multiple-regression analyses were conducted, using the five personality scales as predictors of each of the thinking style scales. As shown in Table 3, all five personality dimensions were significant predictors of all of the thinking styles, although it varied from 1 to 5 personality scales predicting each of the 11 thinking styles. For example, the conscientiousness dimension alone predicted the hierarchical thinking style (R2=0.27, = 0.52). Three personality dimensions (extraversion, openness, and agreeableness) contributed to the prediction of the liberal thinking style (R2=0.19, extraversion=0.33, openness=0.26, agreeableness= 0.18). Yet, another example, all
Table 1 Overlaps between Thinking Styles and Big Five Personality Traits (N=267) Scale Neuro Extra Openness Agree Conscien
Leg 0.31** 0.21** 0.21** 0.06 0.19**
Exe 0.18** 0.04 0.25** 0.06 0.26**
Jud 0.09 0.24** 0.34** 0.01 0.13*
Global Local 0.06 0.07 0.14* 0.14* 0.08
0.09 0.11 0.17** 0.04 0.25**
Lib 0.24** 0.33** 0.27** 0.02 0.14*
Con 0.35** 0.22** 0.26** 0.00 0.06
Internal External Hier 0.15* 0.02 0.23** 0.11 0.20**
0.02 0.37** 0.03 0.24** 0.13*
0.20** 0.21** 0.09 0.04 0.51**
Mona 0.02 0.12* 0.09 0.10 0.19**
Leg, Legislative; Exe, Executive; Jud, Judicial; Lib, Liberal; Con, Conservative; Hier, Hierarchical; Mona, Monarchic; Neuro, Neuroticism; Extra, Extraversion; Agree, Agreeableness; Conscien, Conscientiousness. * P< 0.05. ** P< 0.01.
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
451
five personality dimensions contributed to the significant prediction of the internal thinking style (R2=0.18, openess=0.25, conscientiousness=0.21, agreeableness= 0.26, neuroticism= 0.25, extraversion= 0.21).
4. Discussion The present study aimed at answering the question about the necessity for measuring thinking styles in addition to measuring personality traits by exploring the relationships between the thinking styles proposed in Sternberg’s theory of mental self-government and the big five personality traits. The hypothesized relationships were tested by three statistical procedures. As presented above, results from both the zero-order correlation matrix and the exploratory factor analysis suggested fairly strong, although not complete, support for the predictions. Regarding scale correlations across the two inventories, complete support was obtained for the predictions on three of the five personality dimensions (neuroticism, openness, and conscientiousness) and their counterpart scales in the TSI. These significant relationships found between the scales of the two measures make substantive sense. Take the significant relationship of the openness dimension
Table 2 Oblimin-rotated Principal Component Factor Analysis of the Thinking Styles and Personality Traits Scales (N=267) Scale TSI Legislative Executive Judicial Global Local Liberal Conservative Internal External Hierarchical Monarchic NEO-FFI Neuroticism Extraversion Openness Agreeableness Conscientiousness % Variance Cumulative Variance Eigenvalues
Factor 1
Factor 2
Factor 3
Factor 4
Factor 5
0.33 0.63 0.80 0.75 0.86 0.83 0.69 0.74 0.49
0.65 0.78 0.50 0.72
0.35
0.84 0.68
0.38
0.66
24.92 24.92 3.99
16.88 41.83 2.70
0.38 0.77 12.79 54.62 2.05
0.35
0.38
8.49 63.12 1.36
6.57 69.69 1.05
Scales with factor loadings of less than 0.30 are omitted. TSI, Thinking Styles Inventory; NEO-FFI, NEO Five-Factor Inventory.
452
Thinking Styles Legislative Executive Judicial Neuroticism Extraversion Openness Agreeableness Conscientiousness R2 F d.f. * P< 0.05. ** P< 0.01. *** P< 0.001.
0.35*** 0.20** 0.13* 0.15 14.18*** (3, 240)
0.41***
Global
Local
Liberal
0.21** 0.17** 0.34***
Conservative Hierarchical Monarchic Internal 0.49***
0.33*** 0.26*** 0.18**
0.23*** 0.15* 0.15* 0.23*** 0.19** 0.19** 0.20** 0.39*** 0.32*** 0.22** 0.24 0.16 0.05 0.12 0.19 0.23 19.02*** 22.33*** 6.93** 10.32*** 18.27*** 17.97*** (4, 240) (2, 239) (2, 241) (3, 239) (3, 237) (4, 240)
0.52*** 0.27 89.76*** (1, 243)
0.15* 0.24*** 0.07 8.82*** (2, 242)
External
0.25** 0.37*** 0.21** 0.52*** 0.25*** 0.26*** 0.27*** 0.21** 0.18 0.26 10.17*** 27.78*** (5, 239) (3, 241)
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
Table 3 Predicting Thinking Styles scores using the Big Five Personality Traits (N=267)
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
453
to the legislative, judicial, and liberal thinking styles, for example. Openness is characterized by such attributes as open-mindedness, active imagination, and preference for variety. In a similar vein, people who are high on the legislative, judicial, and liberal thinking styles also tend to be open-minded, imaginative, and perceptive. Another example is the significant relationship between the Neuroticism dimension and the executive and conservative thinking styles. People who are high on the neuroticism dimension tend to be emotionally unstable, easily embarrassed, and pessimistic, and they often suffer from low self-esteem. Understandably, individuals who experience such negative feelings would be more comfortable with working under highly structured situations, with being told what to do, and with carrying out tasks by going along existing rules. All these ways of dealing with tasks best represent the ways by which individuals with the executive and conservative thinking styles would deal with their tasks. However, only partial support was obtained for the predictions on the remaining two personality dimensions (extraversion and agreeableness). Both extraversion and agreeableness were positively correlated with the external thinking style, as expected. However, the to-be-expected relationship between the extraversion scale and the internal style and the relationships of agreeableness to the legislative, judicial, liberal, and the internal styles were not found. This indicates that one may still prefer to work independently (using the internal thinking style) even though one scores high on extraversion. Moreover, this finding suggests that people who are agreeable may still be legislative, judicial, liberal, and internal in their thinking styles. Therefore, the fact that complete support for the to-be-expected relationships was not obtained suggests that one cannot infer, with confidence, people’s thinking styles from their scores on the five personality dimensions. Instead, both the personality measure and the thinking styles measure play a unique role in the data. Results from the factor analysis lent less support to the hypothesized relationships between the thinking styles and the personality dimensions than did the zero-order correlation coefficients. Whereas three of the hypotheses were fully supported (with the remaining two being partially supported) by the computation of correlation matrix, all five predictions were partially supported by the factor analysis results. Furthermore, variance in each of the five factors was shared by scales from both inventories; and none of the five factors was solely loaded by scales from the NEO-FFI. These results from the factor analysis, like those from the zero-order correlations, suggest that one cannot predict, with confidence, people’s thinking styles from their scores on the five personality scales. Finally, results from the multiple-regression analyses revealed the extent to which the big five personality dimensions can statistically predict thinking styles. According to these results, the five personality dimensions accounted for only 5–27% of the variance in the thinking styles scales. The remaining large amount of variance was left unexplained. Therefore, once again, it is evident that the Five-Factor Personality Inventory cannot be used to predict thinking styles with great confidence. It should be noted, however, a study such as this has a major limitation that renders its findings as suggestive rather than conclusive. That is, the findings are completely based on self-report data. It is understood that self-report data are not always highly related to data obtained through behavioral measures. To verify the findings from this study, not only replications of the present study should be conducted among a wide range of populations, but also research involving the use of behavioral measures should be undertaken.
454
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
5. Conclusion Through an empirical investigation, the present study attempted to join the debate on whether or not it would be necessary to assess styles (cognitive/learning/thinking) in addition to assessing personality. Substantial overlap was identified between a thinking styles measure and a measure of the big five personality dimensions. This substantial overlap supported both the conceptual argument for (e.g. Adorno et al., 1950; Eysenck, 1978; Hashway, 1998; Messick, 1996) and the empirical support to (e.g., Busato et al., 1999; Furnham, 1992, 1996a, 1996b; Furnham et al., 1999; Gadzella, 1999; Jackson & Lawty-Jones, 1996; Pacini & Epstein, 1999; Riding & Wigley, 1997) the relationships between personality and styles. It seems undeniable that personality and styles have a strong relationship. The question that one needs to answer is whether or not it makes sense to measure styles separately from personality. The answer that derives from the present study is affirmative. There are three reasons for this affirmative answer. First, in the empirical literature on the relationship between personality and styles, conclusions about the necessity of measuring styles separately have been subjective. Some writers (e.g. Busato et al., 1999) may conclude that personality and styles are two different constructs (and thus need to be measured separately) even when a strong relationship has been found, whereas other scholars (e.g. Furnham et al., 1999) may assert that styles is a subset of personality (and thus do not need to be measured separately) when only a moderately significant relationship has been identified. To this end, findings about the extent to which styles and personality correlate have been inconsistent. For example, Furnham, one of the strong advocates for parsimony of measurement, identified different degrees of relationships in different investigations. Whereas he found substantial overlap between the two constructs in his 1992 study, he only identified moderate correlations in one of his 1996 studies (Furnham, 1996b). In their 1999 study, Furnham and his colleagues found no significant relationship between learning styles and neuroticism, one of the major personality dimensions. Second, results from the present study indicate that the big five personality traits and the thinking styles are not the same since thinking styles made a unique contribution to the variance in the data. Personality dimensions can account for only a little over a quarter of the variance in thinking styles at the most. Indeed, a careful examination of the semantics of the two inventories shows that the two inventories have different orientations. Whereas the items in the TSI assess how the participants deal with different tasks, the items in the Five-Factor Personality Inventory focus on finding out how the participants feel about their environments. Therefore, it is not surprising that thinking styles make a unique contribution to the variance in the data. Third, there is no evidence that personality determines thinking styles. Instead, it could be argued that personality and thinking styles are inextricably entwined. Thus, the fact that the two measures overlap should not warrant the futility of the assessment of either one of the two constructs. On the contrary, knowledge about thinking styles as well as about personality contributes to our understanding of individual differences among human beings.
6. Future directions The present study, like many other studies in the literature, has demonstrated clearly that styles and personality traits are strongly related. It also seems to be fair to say that styles have their
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
455
unique value in explaining human behavior. This conclusion is supportive of Busato, Prins, Elshout, and Hamaker (2000) assertion that a styles inventory has a diagnostic value for students at an individual level for identifying both strengths and weaknesses in, say, their individual study behavior. Then, the ultimate goal of this paper is to answer the question of whether or not we should measure thinking styles in addition to measuring personality traits. The answer is that it depends on who the ‘‘we’’ refers to and for what purposes the inventories would be used. If the ‘‘we’’ refers to such groups as university teachers and counselors, the answer would be that: ‘‘It depends on the purpose and time available’’. If one’s purpose is to have a rough understanding about a student’s thinking style or personality profile and he/she cannot afford the time using both inventories, or he/she does not have access to both inventories, then, by all means, using either one of the inventories should help the teacher/counselor to have a basic understanding of his/her student’s thinking style or personality profile. However, if the assessment has implications for serious decisions such as placing students in particular programs or classes, I would argue for an assessment of both thinking styles and personality traits, among the use of other assessment procedures. Yet, if the ‘‘we’’ refers to researchers who aim at understanding both thinking styles and personality traits, then the question seems not to be whether or not styles should be measured in addition to the measurement of personality, but how to measure each of the two constructs. To achieve this goal, our future research needs to focus on how styles and personality traits work together to affect human behaviors–students’ behaviors, in the context of educational settings.
Acknowledgements I want to thank the Committee on Research and Conference Grants at The university of Hong Kong for having made this research possible. My sincere thanks go to L. L. Zhang for her kind assistance in the data collection and to all those anonymous students whose participation is indispensable to this research.
Appendix A Thinking Styles in the Theory of Mental Self-government Thinking Style
Description
Legislative
One prefers to work on tasks that require creative strategies; One prefers to choose one’s own activities. One prefers to work on tasks with clear instructions and structures; One prefers to implement tasks with set guidelines. One prefers to work on tasks that allow for one’s evaluation; One prefers to evaluate and judge the performance of other people. One prefers to distribute attention to several tasks that are prioritized according to one’s valuing of the tasks.
Executive Judicial Hierarchical
(continued on next page)
456
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
Appendix continued
Thinking Style
Description
Monarchical
One prefers to work on tasks that allow complete focus on one thing at a time. One prefers to work on multiple tasks in the service of multiple objectives, without setting priorities. One prefers to work on tasks that would allow flexibility as to what, where, when, and how one works. One prefers to pay more attention to the overall picture of an issue and to abstract ideas. One prefers to work on tasks that require working with concrete details. One prefers to work on tasks that allow one to work as an independent unit. One prefers to work on tasks that allow for collaborative ventures with other people. One prefers to work on tasks that involve novelty and ambiguity. One prefers to work on tasks that allow one to adhere to the existing rules and procedures in performing tasks.
Oligarchic Anarchic Global Local Internal External Liberal Conservative
Appendix B Sample items from the NEO Five-Factor Inventory Scale Neuroticism
Extraversion
Openness
Agreeableness
Conscientiousness
a
Scorea
Sample item I am not a worrier I often feel inferior to others
0
I do not consider myself especially ‘‘light-hearted’’ I like to have a lot of people around me
0
I don’t like to waste my time daydreaming I am intrigued by the patterns I find in art and nature
0
I often get into arguments with my family and co-workers I try to be courteous to everyone I meet
0
I waste a lot of time before settling down to work I keep my belongings clean and neat
0
Scores are based on verbal anchor ‘‘Strongly Agree’’.
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
457
References Adorno, T. W., Brunswik, E. F., Levinson, D. J., & Sanford, R. N. (1950). The authoritarian personality. New York, NY: Harper & Row. Allport, G. W., & Odbert, H. S. (1936). Trait names: a psycho-lexical study. Psychological Monographs, 47(1). Busato, V. V., Prins, F. J., Elshout, J. J., & Hamaker, C. (1999). The relation between learning styles, the Big Five personality traits and achievement motivation in higher education. Personality and Individual Differences, 26(1), 129– 140. Busato, V. V., Prins, F. J., Elshout, J. J., & Hamaker, C. (2000). Intellectual ability, learning style, personality, achievement motivation and academic success of psychology students in higher education. Personality and Individual Differences, 29, 1057–1068. Costa, P. T. Jr., & McCrae, R. R. (1992). The NEO-PI-R: professional manual. Odessa, FL: Psychological Assessment Resources. Digman, J. M. (1990). Personality structure: emergence of the five-factor model. Annual Review of Psychology, 41, 417– 440. Eysenck, H. J. (1978). The development of personality and its relation to learning. In S. Murray-Smith (Ed.), Melbourne studies in education (pp. 134–181). Australia: Melbourne University Press. Furnham, A. (1992). Personality and learning style: a study of three instruments. Personality and Individual Differences, 13(4), 429–438. Furnham, A. F. (1996a). The big five versus the big four: the relationship between the Myers-Briggs type indicator (MBTI) and NEO-PI five factor model of personality. Personality and Individual Differences, 21(2), 303–307. Furnham, A. F. (1996b). The FIRO-B, the learning style questionnaire and the five-factor model. Journal of Social Behavior and Personality, 11(2), 285–299. Furnham, A., Jackson, C. J., & Miller, T. (1999). Personality, learning style and work performance. Personality and Individual Differences, 27, 1113–1122. Gadzella, B. M. (1999). Differences among cognitive-processing styles groups on personality traits. Journal of Instructional Psychology, 26(3), 161–166. Galton, F. (1884). Measurement of character. Fortnightly Review, 36, 179–185. Goldberg, L. R. (1993). The structure of phenotypic personality traits. American Psychologist, 48(1), 26–34. Hashway, R. M. (1998). Developmental cognitive styles: a primer to the literature including an introduction to the theory of developmentalism. Austin & Winfield: Bethesda, MD. Jackson, C., & Lawty-Jones, M. (1996). Explaining the overlap between personality and learning style. Personality and Individual Differences, 20(3), 293–300. Messick, S. (1996). Bridging cognition and personality in education: the role of style in performance and development. European Journal of Personality, 10, 353–376. Norman, W. T. (1963). Toward an adequate taxonomy of personality attributes: replicated factor structure in peer nomination personality ratings. Journal of Abnormal and Social Psychology, 66, 574–583. Pacini, R., & Epstein, S. (1999). The relation of rational and experiential information processing styles to personality, basic beliefs, and the ratio-bias phenomenon. Journal of Personality and Social Psychology, 76(6), 972–987. Riding, R. J., & Wigley, S. (1997). The relationship between cognitive style and personality in further education students. Personality and Individual Differences, 23(3), 379–389. Sternberg, R. J. (1988). Mental self-government: a theory of intellectual styles and their development. Human Development, 31, 197–224. Sternberg, R. J. (1997). Thinking styles. New York: Cambridge University Press. Sternberg, R. J., & Grigorenko, E. L. (1995). Styles of thinking in the school. European Journal for High Ability, 6, 201–219. Sternberg, R. J., & Wagner, R. K. (1992). Thinking styles inventory. (Unpublished test, Yale University). Taylor, A., & MacDonald, D. A. (1999). Religion and the five factor model of personality: an exploratory investigation using a Canadian university sample. Personality and Individual Differences, 27, 1243–1259. Zhang, L. F. (1996). Chinese translation of the Thinking Styles Inventory. (Unpublished test, The University of Hong Kong).
458
L.F. Zhang / Personality and Individual Differences 33 (2002) 445–458
Zhang, L. F. (2000a). Are thinking styles and personality types related?. Educational Psychology, 20(3), 271–283. Zhang, L. F. (2000b). Chinese translation of the NEO Five-Factor Inventory. (Unpublished test, The University of Hong Kong). Zhang, L. F. (2000c). Relationship between Thinking Styles Inventory and Study Process Questionnaire. Personality and Individual Differences, 29, 841–856. Zhang, L. F. (2001a). Thinking styles and personality types revisited. Personality and Individual Differences, 31, 883– 894. Zhang, L. F. (2001b). Thinking styles, self-esteem, and extracurricular experiences. International Journal of Psychology, 36(2), 100–107. Zhang, L. F., & Postiglione, G. A. (2001). Thinking styles, self-esteem, and socio-economic status. Personality and Individual Differences (in press). 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. (2000). Are learning approaches and thinking styles related? A study in two Chinese populations.. The Journal of Psychology, 134(5), 469–489.