Personality styles and language learning strategies of Chinese and Dutch learners

0

Personality styles and language learning strategies

of Chinese and Dutch learners

Selma Bakker

S1528963

MA in Applied Linguistics Faculty of Liberal Arts University of Groningen

Supervisors: Prof. Dr. C.L.J. de Bot

Dr. F.M. Edens

1

T

ABLE OF CONTENTS

A

1.

I

The current study: objective, relevance, and structure 1

2.

B

2.1 English as a foreign language in the People‟s Republic of China 3

2.1.1 Teaching Approaches 4

2.2 English as a foreign language in the Netherlands 4

2.2.1 Teaching Approaches 5

3. INDIVIDUAL DIFFERENCES 6

3.1 Individual differences 6

3.2 Language learning strategies 6

3.2.1 Language learning strategy use 8

3.3 Personality 9

3.3.1 Jung 9

3.3.2 Myers Briggs 10

3.4 Personality in language learning 11

3.5 Personality and language learning strategies 12

2

4.3 Materials 19

4.3.1 Strategy Inventory Language learning 19

4.3.2 Myers Briggs Type Indicator 20

4.4 Procedure 20

4.5 Design & Analysis 21

5.

R

5.1 Respondents 22

5.2 Results of the SILL 22

5.2.1 Introduction 22

5.2.2 Within the groups 23

5.2.3 Between the groups 24

5.3 Results of the MBTI 25

5.3.1 Introduction 25

5.3.2 Descriptive statistics 25

5.3.3 Statistical analyses 27

5.4 Results of the SILL and MBTI combined 28

5.4.1 Introduction 28

5.4.2 Correlation analyses 29

6.

D

6.1 SILL 30

6.2 MBTI 32

6.3Relation between SILL & MBTI 33

6.3.1 Introversion 33

6.3.2 Sensing 34

6.3.3 Intuition 35

6.3.4 Thinking 35

7.

C

3

S

B

G

A

CKNOWLEDGMENTS

First of all, I would like to thank my supervisors Prof. Dr. Kees de Bot and Dr. Margot Edens for guiding me through the process of setting up the experiment and the analysis of the data. Furthermore, I would like to thank prof. Dr. Chuming Wang and Ms. Katherine Zhang as well as Dr. Remco Knooihuizen, for helping me find participants. In addition, I would like to thank Ms. Xiaoyan Xu and Dr. Mik van Es for their help with the statistical part of the study.

Most of all I would like to thank the students at the University of Groningen and Guangzhou University of Foreign Studies who chose to spend their valuable time to participate in this study. You have helped me a lot by providing me with valuable information on your personality type and language learning strategies.

1.

I

NTRODUCTION

As early as the 1960s researchers have been interested in those qualities that make one person learn a language better than another person. The question of what the „good language learner‟ distinguishes from the less efficient learner has led to many theories. Theories on individual differences that might influence language development have provided us with a good idea on the possibilities. Some of these individual differences are very discernible in the learner, others are less easy to distinguish. Attitude and motivation are factors in foreign language learning, that have been proven to be important. Similar results have been found for age, and intelligence has also been associated with successful foreign language development.

Empirical research on second language development started focussing on language learning strategies from the 1970s and become a popular subject in the 1990s. In the research on language learning strategies, four different foci can be distinguished. First, the early studies that focussed on the „good‟ language learner. Then the attention shifted to the correlations between the use of language learning strategies and the achievements in second language learning. Thirdly, the use of language learning strategies was standard in relation to other individual differences, such as age, gender, motivation etc. And more recently, the language strategy use and the cultural backgrounds have become a focus of attention (Yu and Wang, 2009).

The current study: objective, relevance and structure

strategy styles most commonly associated with language learning; cognitive, compensation, memory, metacognitive, affective and social strategies.

Previous research has focused on the relation between personality types and learning strategies, however, the main themes were age, gender, education and profession. Research on personality types and language learning strategies has not yet happened from a comparative cultural perspective.

The main question that will be answered is to what extent the relation between personality type and language learning strategies is affected by cultural differences. In order to answer this, the differences in personality type as well as the differences in language learning strategies between the two groups of participants will be investigated.

2.

B

ACKGROUND

In order to provide a background to the potential differences this study hopes to find between, this section the educational system in the People‟s Republic of China as well as the Dutch system will be described in short.

2.1 English as a foreign language in the People’s Republic of China

In China, English education is recommended from the third year of primary school, which is when children are about 8 or 9 years of age. In more economically developed areas English is taught from grade one. By the end of primary school, the students are required to be able to read simple stories and answer simple questions.

Secondary education exists of three compulsory and three voluntary years. English education is compulsory during both programs. At the junior secondary school the foundation for English is being built. The students receive 3 to 4 hours of English language instruction per week, for the course of 3 or 4 years. In order to enter the senior level, students need to take an entrance exam. In senior secondary school the students are also required to learn a foreign language. The amount of language instruction and the duration of the course are dependent on the level of education as well as the score on the entrance exam (Silver, Hu & Iiona, 2002; Xu, 2010).

needed to be able to take part in effective communication. Especially listening and speaking skills are often not developed enough (Silver, Hu & Iino, 2002; Hu, 2004).

2.1.1 Teaching approaches

During the last decades, several teaching methods have been used in Chinese ELT. For a long time a combination of the traditional grammar translation method and the audio-lingual method was used. The problem with this combined approach was that the skills were acquired through pattern drilling, and failed to be transferred to actual communication. The approach has lead to meaningless learning and language use (Silver, Hu & Iino, 2002). Due to social, economical and media developments a need for communication was created. This lead to the communicative language teaching (CLT), however, due to a lack of qualified English language instructors and the importance of texts, the traditional teaching methods are still popular (Xu, 2010).

According to Hu (2004) the current classroom practice does not help the students to develop conversation skills or autonomous learning. Teaching is not oriented towards communication; there are no occasions for the students to practice English. The classes are exam-oriented, however, since tests only focus on grammar and vocabulary and there are no parts for listening or speaking, assessment methods also do not contribute to the development of communication skills.

The idea that true knowledge is in texts (Silver, Hu & Iino, 2002) explains why the Chinese classroom is very textbook-oriented (Hu, 2004). Learning is reading books. Therefore, a commonly used learning method is the reading and reciting of English texts, with a focus on grammar. Mental activeness is preferred over verbal activeness, meaning a preference for attending to details and memorizing.

2.2 English as a foreign language in the Netherlands

In the Netherlands, EFL is compulsory at the primary school level, usually starting around age 10 or 11, however, schools are allowed to start English education earlier, sometimes as early as children of 4 years old. Lessons usually take place once a week, for less than an hour. The aim of these language lessons is to familiarize the students with English and they are expected to be able to understand simple spoken conversations about common subjects and reading simple texts. However, a large minority fails to meet these requirements.

duration of the classes is dependent on the level of education, with a minimum of 400 hours for the lowest level and a CEFR score of A2.1, ranging to 680 hours for the highest levels, with an expected CEFR score of B2.1. Students are also required to learn another foreign language. Another stream in Dutch education is bilingual education, where a foreign language – usually English - is used to teach certain courses.

Students who continue their education can choose between three levels, of which university is the highest. Unlike the Chinese situation, there are no obligatory tests on English proficiency. English Bachelor and Master-programs are offered at six universities in the Netherlands, as well as it is a subject in several other programs. At the MA-level, English is often used as an instructional language (Edelenbos & De Jong, 2004; Xu, 2010).

2.2.1 Teaching approaches

Language teaching in the Netherlands used to be very teacher-controlled. The traditional teaching methods focused on providing students with insights in grammatical structures. Over the last decades the linguistic teaching methods in Dutch classrooms have changed, due to the introduction of the communicative approach (Willems, 1987). Starting at the primary school level, communicative settings were chosen to teach children English, where settings and roles were determined. Authentic dialogue was the aim of these language classes (Edelenbos & Suhre, 1994). A detailed description of communicative teaching approaches would go too far here, but the general idea is that learners can deduct grammatical patterns from the language.

3.

I

NDIVIDUAL DIFFERENCES

3.1 Individual differences

Second language development (SLD) is a complex process and there are differences among learners that may play a role in the diverse ways people learn languages. Every human being is unique and distinct. These distinctions are visible in many ways; ones height, eye colour or ethnic diversity, but is also applicable in the field of second language development. There is a wide range of factors contributing to a person‟s „uniqueness‟. These individual differences (IDs) are characteristics in which individuals might differ from one another (Dewaele, 2009; Dörnyei, 2005). There is a wide range of such individual differences and these are associated with the success of the development of a second language. Many of these IDs are highly visible in the learner, such as age, gender or prior language learning experience (Sanz, 2005), whereas others need systematic observation of behaviour to be inferred (Dewaele, 2009). Examples of these internal factors that differ from person to person and play a role in SLD are memory, motivation, intelligence and aptitude. However, these different variables are not a guarantee of successful acquisition, but are dependent on the interaction with the context in which they occur (Dewaele, 2009; Sanz, 2005).

In the late seventies research in the field of IDs started to focus on the characteristics of „the good language learner‟ (Dörnyei, 2005). These early studies showed that besides language aptitude and motivation, there was something else that made certain students excel in SLD. Active and creative participation during the process of learning has shown to be the reason why some learners perform better than others (Dörnyei, 2005). Learners vary from one another concerning their learning style or the strategies used in learning (Dörnyei, 2005; Sharp, 2004).

3.2 Language learning

The way languages are learned differs from person to person. This depends among other things on the way the learner prefers to learn and whether the learner is a „good‟ language learner or not. Being a good language learner depends to a certain extent on the learning style used (Oxford, 2003).

mentions four dimensions of learning style associated with language learning. The first dimension includes sensory preferences, which refer to the physical, perceptual learning channels, such as visual stimulation, lectures, movement. The second dimension Oxford (2003) classifies is personality types, which is based upon the dimensions mentioned by Jung (1921,1971) and those used in personality taxonomies. The third learning style dimension that can be distinguished is the desired degree of generality, which contrasts the learner who prefers the big picture with the learner who focuses on the details. Biological differences are the last dimension associated with language learning, such as biorhythms, sustenance and location. The time of the day when a learner performs best, the need for food or drink while learning and the nature of the environment can all influence the learning process. Not only the learning style was found to differ between learners, there is also variation regarding the strategies used in language learning. According to Oxford & Ehrman (1988) learning style affects the choice of language learning strategies; learners tend to use learning strategies that fit with their learning styles.

Language learning strategies are specific actions used by students to improve their own learning, to make learning faster and more effective as well as to make the learning process more enjoyable (Oxford, 1990). Language learning strategies used in the L2 learning situation are “specific behaviours or thought processes that students use to enhance their own L2 learning” (Oxford, 2003, p. 8), and it will help learners to become more independent and autonomous in the learning process (Allwright, 1990; Little, 1991 in Oxford, 2003).

Oxford (1990) classifies a strategy system consisting of three direct and three indirect learning strategies (see appendix I) that support and interact with each other. Direct language learning strategies are those that directly involve the target language, because of mental processing. Cognitive strategies are direct language learning strategies and involve the storage or retrieval of words and phrases. These cognitive strategies include understanding and producing new language, by practicing naturalistically or summarizing and repeating (Cohen, 1996; Oxford, 1990). According to Cohen (1998) all three direct strategies classified by Oxford (1990) can be summarized by cognitive strategies, however, Oxford (1990) also distinguishes compensation strategies. Compensation strategies are those strategies that enable the learner to use the language, despite of possible gaps in their knowledge. Guessing meaning from a context, for example. Finally, memory strategies involve strategies that help the learner store and retrieve information, such as the use of grouping, and linking the verbal with the visual (Oxford, 1990).

Indirect language learning strategies, on the other hand, support and manage language learning without directly involving the target language (Oxford, 1990). Metacognitive strategies deal with pre-assessment and pre-planning, they help learners control learning and coordinate the learning process, as well as post-evaluate the language learning activities. Affective strategies are those strategies that help regulate emotions, motivations and attitudes, such as dealing with anxiety and self-encouragement. And lastly, social strategies include the actions the learner chooses in order to interact with others; both other learners and native speakers (Cohen, 1996; Oxford, 1990). For more information concerning the method to test language learning strategies is given in 4.3.1.

3.2.1 Language learning strategy use

broader range of strategies than less proficient learners. More proficient learners also use strategies in more situations than less proficient learners.

Research shows that there are gender differences in the use of language learning strategies, especially regarding the social strategies, such as interaction with native speakers (Ehrman & Oxford, 1988). Females also showed a more frequent use of formal practice, standard study habits and input elicitation, which are part of the cognitive strategies (Ehrman & Oxford, 1988). However, in Ehrman & Oxford (1990) the reported use of language learning strategies was similar for males and females.

3.3 Personality

One of the most prevalent IDs is personality; the characteristics of a person that “account for consistent patterns of feeling, thinking and behaving” (Pervin & John, 2001, p.4). According to Dörnyei (2005), personality is probably the most individual characteristic of a human being. Personality is one of the most researched themes in psychology. There are many theories that attempt to identify the „consistent patterns‟ or traits and try to systematize them.

As well as theories, there are a number of taxonomies focusing on traits in personality, such as Eysenck‟s three-component construct (Eysenck & Eysenck, 1985 in Dörnyei, 2005) and a more extended version, the „Big Five‟ model (Goldberg, 1992 in Dörnyei, 2005). In this research the Myers Briggs Type Indicator will be used, one of the most frequent used questionnaires based on stable patterns of personal preference.

3.3.1 Jung

3.3.2 Myers Briggs

The Jungian type theory has been further redefined over the years by Isabel Myers and Katharine Briggs. Katharine Briggs was an author who felt that her characters needed more human like traits. In order to reach this goal, she teamed up with her daughter Isabel Myers and developed a psychological type indicator based on the types distinguished by Jung (Briggs Myers et al, 1998). They faced difficulties in interpreting the personality types as described by Jung, in that descriptions sometimes seemed to overlap as well as descriptions that did not cohere in a single factor (McCrae & Costa Jr., 1988). Furthermore, they believed the type indicator needed a fourth dimension and added the Judging – Perceiving scale, which describes the way a person usually deals with the outside world (Briggs Myers et al, 1998).

The basic principle of the theory is that what seems like random variation in behaviour is really rather systematic and stable, because of the basic differences that exist in the way individuals prefer to use their perception and judgment (Briggs Myers, 2001). The Myers Briggs Type Indicator (MBTI) measures both mental functions, such as perception and judgment, as well as attitudes, such as orientation of energy and orientation to the outside world.

The theory behind the MBTI is based on four dichotomies, which are thought to reflect innate psychological or mental dispositions. The Extraversion (E) versus Introversion (I) dimension focuses on where someone gets his or her energy from. Sensing (S) versus Intuition (N) is about how information is taken in, the ways one becomes aware of things. Those who prefer sensing are more focused on the specifics, whereas those with a preference for Intuition are more concerned with the big picture. Decision making is defined through Thinking (T) versus Feeling (F) and means all the ways of coming to conclusions about what has been perceived, either through logical implication or by the impact something has on people. Finally, Judging (J) versus Perceiving (P) are the attitudes to the outer world and how one organizes his or her life. Judgers enjoy closure more, while perceivers find joy in processing. A brief summary of the types is given in appendix II.

The value-neutral scales used in the MBTI show similarities with the scales in the SILL; there is no judgment, the results of a person‟s actions can be the same, the only differences is the way how they act (Dörnyei, 2005).

3.4 Personality in language learning

Much research has been done about personality traits in an educational perspective. Even though the impact of personality on learning is not as big as that of other IDs, such as aptitude and motivation, it has shown to have influence on the outcome of learning (Dörnyei, 2005). According to Moody (1988) personality plays an important role in the learning process, because the traits in personality make a difference in how and what people learn. According to Myers & McCaully (1998) personality can make as much difference in grade point averages, as IQ does.

A number of studies have tried to come up with an answer as to what personality traits are most helpful in learning languages. The results are rather contradictory. Introversion has shown to have a positive effect on academic success in general, because Introverts consolidate learning better, have lower distractibility and have shown better study habits. Extraversion, on the other hand, has shown a negative relationship success in higher education (Dörnyei, 2005).

A study by Moody (1988) at the University of Hawai‟i showed that the personalities of language students differ from other college students in two major ways; Intuitive types outnumber Sensing types and there are more Thinking types, and thus fewer Feeling types. This implies that it will be difficult for students with very different personality types to be successful in these courses, because there is a risk that the lessons and textbooks are directed at a different type of student. This difference is in line with what Jung (1921, 1971) expected; language is the manipulation of words, symbols and abstractions, which suits the Intuitive language learner best, since they take in information by looking for patterns and meanings. Language learning can often be compared to textbook learning, with a focus on rules, it is also no surprise that there are more Thinkers among the language students; applying rules has much to do with logic (Moody, 1988).

In Moody‟s (1988) study, the focus was on the most prevalent types in the language learning classroom. He found that the most frequent type was ENFP (11%), and the least frequent types were ESFP (3%) and ISFP (2%). When focussing on the most popular personality traits, he found more Introverts who study languages than extraverts. Thinking was preferred over feeling among the language learners, as well as Perception over Judging. For the larger college population, Judging is the most frequent trait (Briggs Myers et al, 1998).

3.5 Personality type and language learning strategies

The relation between learner characteristics and language learning strategies has been included in a number of studies. Personality-related variables in the learner have shown to have a relation to learning strategies (Cohen, 1996).

3.5.1 Extraversion – Introversion

they prefer to avoid social contact and work best when they are alone. They show a reluctance to participate in speaking activities, so group work may not be very successful (Sharp, 2005). Sharp (2008) found a significant negative relationship for Introversion and social strategies (p<.011). Ehrman & Oxford (1988) found in their study on Foreign Service officers that searching for and communicating meaning strategies (part of compensation strategies) are more frequently used by introverts. Learners who prefer Introversion also showed an extensive use of metacognitive strategies (Ehrman & Oxford, 1990), a result also found by Sharp (2008).

So, the learning of extraverts will be more effective with concrete experiences, contacts with the outside world and relationships with others are present. Introverts will have a more effective learning process when individual, independent situations are present. They prefer ideas and concepts and are able to focus on the task at hand. They sometimes avoid taking linguistic risks in conversation, because they feel that they have to process ideas before speaking (Reid, 1995).

3.5.2 Sensing - Intuition

Ehrman & Oxford‟s (1990) show that searching for and communicating meaning is used more often by Intuitive than by Sensing types, as well as formal model building, a part of cognitive strategies. Intuitive learners show more effective learning when they encounter meaningful experiences and enjoy abstract thinking and are able to find the most important principles of the topic without step by step instruction (Cohen, 1998). Intuitives also chose authentic language use strategies and affective strategies significantly more often than sensing types (Ehrman & Oxford, 1990). Sensing types showed a strong preference for memory strategies and used cognitive and metacognitive strategies frequently. They also showed a rejection to compensation strategies, whereas Intuitive types reported extensive use of those strategies (Ehrman & Oxford, 1988). According to Reid (1998), sensing types showed a more effective learning when they use reports of observable facts. Ehrman & Oxford (1988) found that learners with a preference for the combination Extraversion-Intuition showed a liking for memory strategies. Those learners who favoured Intuition-Judging used metacognitive strategies more often.

Intuitives prefer general concepts rather than focussing on details and are able to guess from the context. However, inaccuracy can make them miss important details (Reid, 1995).

3.5.3 Thinking - Feeling

Ehrman & Oxford (1990) found a large contrast between Thinkers and Feelers with regard to their preferred language learning strategies. Thinkers show a strong preference for cognitive strategies, whereas Feelers reject these kinds of strategies. Metacognitive strategies are commonly used by Thinkers, while Feelers do not use these at all. The study showed a nearly significant relationship for Feelers and social strategies. According to Reid (1998) feelers favour general study strategies more than Thinkers do. They tend to learn more effectively from personalized circumstances, whereas those who score higher on the thinking dimension learn more effectively from impersonal and logical circumstances.

So, Thinking types learn more effectively from impersonal situations and logical consequences. They are good analyzers, however, anxiety due to their perfectionism can hinder their learning process. The learning process of Feelers benefits from personal situations. They value social circumstances, and prefer good relations with those with whom they learn. However, discouragement is possible when they do not feel appreciated (Reid, 1995).

3.5.4 Judging – Perceiving

According to Reid (1998), searching for and communicating meaning strategies are more often used by Perceivers than Judgers. The learning process of Perceivers is more effective if negotiation, feeling and inductive processes that postpone closure are present. Judgers show a more effective learning when they use reflection, deduction, analysis and processes that involve closure. Judgers also show a preference for general and metacognitive strategies, whereas Perceivers usually rejected those strategies. A large contrast was found concerning the use of compensation strategies, Judgers rejected those, unlike Perceivers, who favoured such strategies. Another opposite found was the use of social strategies, which Judgers liked and Perceivers disliked (Ehrman & Oxford, 1990).

present. They are open, flexible and adaptable to change and new experience, however, their learning process might suffer from their passiveness (Reid, 1995).

3.6 Cultural differences

Many factors contribute to the choice of language learning strategies as has been demonstrated above. Demographic factors are also related to learning strategies (Cohen, 1996). Nationality or ethnicity has shown influence on strategy use (Oxford, 1990). Hofstede (1991) sees culture as patterns of thought, feelings and actions that are learned. These patterns are different from the parts of human functioning that are common to all. Culture can also be described as how and why one thinks, learns, worships, fights, and relaxes Ethnicity is related to cultural characteristics, and is the reason why people from a different ethnic group are bound together, because they share a common culture (Bedell & Oxford, 1996).

The psychological type as described by both Jung and Myers and Myers-Briggs is thought to be of a similar structure as the parts that are universal in human nature. Jung believed that he was describing mental processes that were common to the entire human species. If he would be correct, the differences in types should be consistent across cultures (Myers & McCaulley, 1998).

Even though there has not been much research on the use of the MBTI in different cultural settings, current experience is sufficient to proceed with the hypothesis that cultural differences will not disturb recognition and use making of psychological type explanations. Numerous studies have reported on the high validity of the MBTI in different cultural settings (Kirby & Barger, 1998). A great number of studies on multicultural and international use of the MBTI mentioned by Myers- Briggs et al (1998) also report valid results. However, some experiences point out that the MBTI may not be appropriate in collectivist cultures, where communal identity prevails above the individual. This is especially the case when this centrality of group identity is combined with oppression. Cultural values are likely to affect the ways in which individuals express their type preferences. For example, emotions and feelings are not favoured in Korean culture (Sim & Kim, 1993).

on a similar group of Asian students at several US universities. Huang (1984) and Huang and van Naerssen (1987) studied strategy use of English majors in Mainland China and found that memory strategies were often used, as well as metacognitive strategies. Chang (1990) studied mainland Chinese students who took their degree at the University of Georgia and found a high use of compensation strategies and a low use of affective strategies. Rossi-Le (1989) found that Chinese students at two US community colleges used the widest variety of strategies of all students with different ethnic background. He also found that strategy preference was significantly influenced by the home culture of the participants. Wen and Johnson (1991) created different strategy categories in their SILL-based questionnaire and used this in their research on English majors in Mainland China. They found that the participants showed a high use of ambiguity strategies, such as guessing, and can be counted as compensation strategies.

Bedell (1993, in Oxford & Burry-Stock, 1995) also found cultural differences with regard to personality. Caucasians are predominantly extraverted (60%, p <.05), Intuitive (65%, p< .001) and are evenly distributed between Thinking and Feeling, as Judging and Perceiving. The Sensing-Thinking combination is less frequent than the Intuition – Thinking combination. Of the Chinese participants two-third are Judgers (p<.05), the proportion of Sensing – Judging is about 50% more than the rest of the language sample, which is significant (p<.01). In addition, Sensing – Judgers (p<.01) and Thinking – Judgers (p<.001) dominate both with about 27% of the sample. Other studies found similar results, with even stronger preference for thinking (80%, P<.01). The results of the Chinese participants are partly different from what Moody (1988) found among Western students who preferred Intuition over Sensing. However, there is a similarity between the results regarding the preference for thinking.

Language is closely related to culture, considering that one‟s native language is a reflection as well as an influence on one‟s culture (Bedell & Oxford, 1996). A study by Reid (1987) on ESL students from different cultural backgrounds demonstrated that people from different culture expressed a preference for each of the learning strategies. Reid found that Asian language learners showed a preference for visual learning modalities.

The choice of strategy shows a dependence on the cultural background. Cultural issues, such as attitudes towards authority, beliefs about the difficulty of learning languages, concepts on whether personal expressions are allowed or not, the importance of the text and memorizing it, all affect the use and the choice of language learning strategies (Oxford, 1996). Students might not be used to take actively part in their language learning process, something that is nicely described by Oxford (1996) in the surprise that students showed when they heard that they themselves could actually do something to improve their language learning. This sense of passivity differs from culture to culture; each culture has their own approach to learning, which implies that their approach to learning strategies differs as well. Conditioning by the culture and the educational system, has lead many language students to be accustomed to being spoon-fed and has turned them into passive learners (Reid, 1995). Many introverted Asians prefer to work alone (Harshbarger et al., 1986 in Oxford, 1996). However, learning strategies can be taught; O‟Malley (1987) found that training showed significant effect, especially for listening and speaking proficiency.

T

HE STUDY

4.

M

ETHOD

4.1 Introduction

The aim of the thesis is to show the differences between the two groups of participants with regard to the relation of their personality type and their language learning strategies.

4.2 Subjects

The first group consists of 21 Dutch participants, out of which 5 males and 16 females. The Dutch participants were all born and raised in the Netherlands, with Dutch as their first language. They are all enrolled in the first year of the three-year BA program English Language and Culture at the Rijksuniversiteit Groningen. They have learned English at the primary and high school level, which means that their English language learning history started about ten years ago. All of the Dutch participants are above CEF B2 level, considering they graduated the highest level of high school, prior to studying English. The participants were not controlled for other language learning experience, nor were they controlled for their age.

4.3 Materials

In this study, both groups of participants were asked to fill out two surveys. Both surveys were available online. One survey was the Myers Briggs Type Indicator (MBTI), used to indicate the personality type of the participants, the other the Strategy Inventory Language Learning (SILL), used to discover the language learning strategies used by the participants.

4.3.1 Strategy inventory language learning

Language learning strategies can be identified through observations, interviews, diary-studies or self-report surveys. A summative rating scale is amongst the most prevalent ways to assess language learning strategies (Oxford & Burry-Stock, 1995). Amongst the most often used instruments to assess the frequency of strategies that are present in learning is The Strategy Inventory for Language Learning (SILL) (Oxford & Burry-Stock, 1995). The SILL has been developed by Oxford (1986 – present) and is a structured survey, based on the six language learning strategies mentioned above. The averages of the item scores within a scale make up the six scale scores. Each of the SILL items correspond to a language learning strategy, the questionnaire focuses on the specific strategic behaviours as well as on the frequency of the use of the strategy (with a range between „never‟ and „always‟).

The SILL has been often used as an instrument in research, but the inventory itself has been researched often as well. The inventory has been extensively tested and has proven to be reliable regarding the tendency to falsify responses in order to make a good impression, which has proven to be a risk, especially in Asian countries. There is no evidence of social desirability response bias on the SILL (Oxford, 1990). Studies have also reported that the inventory is highly valid and reliable (Oxford, 1990). The SILL has been used in several countries and cultures and it has shown a high reliability across many cultural groups (Oxford & Burry-Stock, 1995). There are different versions of the SILL, developed for different groups. The one that will be used in this research is the 50-item version, which is specially designed for students of English as a foreign or second language. Reliability of this version of the SILL has been tested extensively and shows a reliability of at least .85 (Oxford & Burry-Stock, 1995), therefore, the SILL can be administered in a foreign language with minimal measurement error.

SILL are shown in factor averages, with a range from high ( 5- 3.5), moderate (3.4 – 2.5) and low (2.4 – 1). Averages are computed for each scale. The scores are not cumulative.

4.3.2 Myers Briggs Type Indicator

The MBTI is an 87-item, normative, self-report questionnaire designed to reveal the preferences regarding ones personality. The questionnaire is divided over three parts. Each question measures a certain dimension; extraversion – introversion, sensing – intuition, thinking – feeling or judging – perception, with scores ranging from 0 to 2. The scores were then added for each of the eight traits. For this research, the raw data was used, i.e. the scores for the eight traits. The most prevalent score of each dimension results in a four-letter-type, which was also used in the research. The MBTI has proven to be reliable, even when it is taken in second language, in this case English. Even though some problems have been reported with regard to some of the words or idioms used in the MBTI, this does not make it less reliable (Wederspahn & Barger, 1988). If they would not understand the meaning of a word, it was allowed to skip the question. Also, few omissions were allowed when the participants had difficulty to decide on a question.

4.4 Procedure

The group of Dutch participants were collected by asking them directly during one of their classes in their BA-program. The experiment was explained and those who wanted to participate left their e-mail address on a form. The group of Chinese participants were informed by one of their teacher. Those who wanted to participate left their e-mail address as well. The participants then received an e-mail with the invitation to fill out one of the surveys. By clicking on the link in the e-mail they went directly to the survey on

www.surveymonkey.com. This was to ensure that only those invited would fill out the survey. Between filling out both surveys, the participants had a couple of days rest. To avoid bias, 50% of the participants started with the SILL, the other half started with the MBTI.

Therefore, the participants were told that it was perfectly alright that they did not understand some words, and in case of the MBTI, it was okay to skip a question. For the SILL, all 50 questions were obligatory.

After the results of both questionnaires were tabulated, the participants received a document with details on their personality preferences and language learning strategies, along with some information on how they could use it to their advantage.

4.5 Design and analysis

For the analysis the average scores on the six SILL strategies were used, as well as the average for the total use of strategies. Instead of the four-letter type that the MBTI could provide, the raw data was used, meaning the scores on both sides of the scales. This was done considering the influence both the preferred side and the not preferred side of the scale might have on the use of language learning strategies.

First of all, the reliability of the SILL and the MBTI has been tested. Second, the data of both variables has been plotted in bar graphs in order to compare both groups on their use of language learning strategies and their personality style. Percentages were used to be able to compare the groups, considering the differences in sample size.

5.

R

ESULTS

5.1 Respondents

21 Chinese and 37 Dutch students of English have been asked to fill out two questionnaires. The total number of participants is 14 for the Chinese participants and 21 for the Dutch participants (see table 5.1). A questionnaire is only valid when he or she has completely filled in both questionnaires. In cases where only one of the questionnaires was filled out, or only partly, the participant was excluded.

Table 5.1 Number of respondents and valid questionnaires

Chinese participants Dutch participants

Number of respondents SILL 15 26

Number of valid SILL 15 25

Number of respondents MBTI 19 25

Number of valid MBTI 14 21

Total number of valid participants 14 21

5.2 Results of the SILL

5.2.1 Introduction

To provide insight in the results it is vital to address an important issue. The SILL 7.0 version used in this study has been designed for students of English as a foreign language. It has been proven to be reliable when taken in a foreign language, however, it is still possible that the participants had difficulty understanding a question. All questions were obligatory however, therefore, it is possible that in case of doubt, guessing has taken place. A reliability analysis has shown that this did not happen, considering the high outcome of the Cronbach Alpha at 0.87.

are used are the independent samples t-test and a correlation analysis. The alpha level is set at 0.05.

5.2.2 Within the groups

First, a correlation analysis was done, to see whether there were differences within the group of Chinese participants and within the group of Dutch participants.

Table 5.2.1 Significant correlations SILL scores of Chinese participants

Memory Cognitive Compensation Metacognitive Affective Social Memory Cognitive Compensation 0.559* Metacognitive 0.639* 0,717** Affective 0,540* 0.682** Social 0.538* 0,643* 0,745**

* significant at the 0.05 level ** significant at the 0.01 level

Table 5.2 shows significant correlations of the language learning strategies used by the Chinese participants. This shows that cognitive strategies are often used in combination with other strategies. Social strategies are used with half of the other strategies, which is frequent as well.

Table 5.2.2 Significant correlations SILL scores of Dutch participants

Memory Cognitive Compensation Metacognitive Affective Social Memory Cognitive Compensation Metacognitive Affective 0,573** 0.532* 0,531* Social 0,607** 0,447* 0,523*

* significant at the 0.05 level ** significant at the 0.01 level

in combination with up to five of the six possible strategies. Also, social strategies are used often by those who use other strategies.

5.2.3 Between the groups

The first graph is a bar graph of the average SILL results of the Chinese and the Dutch participants.

Figure 1 Average SILL scores for both groups of participants

Figure 1 shows that both groups score quite similarly on the possible strategies. The Chinese participants score on average very consistent; they have reported to have a medium use on all of the strategies. Naturally, the total use of strategies is also average.

The Dutch participants on average also show a medium use of strategies, an exception is the low use of memory strategies. The use of affective strategies is medium, however, with a score of 2.5 it is precisely on the border of low and medium use.

An independent samples t-test was done to see whether or not the two groups showed differences regarding their language learning strategy use, in other words; whether or not one group uses strategies more than the other. A Bonferroni correction was used.

SD=0.6). This difference was significant (t(33)=3.2;p=0.003. No other significant differences in the use of language learning strategies were found. Thus, the Chinese and Dutch participants differ only from one another on two strategies.

5.3 The results of the MBTI

5.3.1 Introduction

To provide insight in the results it is necessary to address some important issues. The MBTI has shown highly reliability scores even when administered in a foreign language. Therefore, it was safe to use the English version of the MBTI in this study. However, the participants were told that they were allowed to skip a question when the meaning of an idiom used in a question or answer was unclear to them. They were also told that some omissions were permitted when they felt that it was impossible to choose between either one of the answers. Not allowing the participants to skip a question in such cases, would mean an unreliable outcome. In both groups there were participants who did not answer a question. According to the MBTI manual this does not influence the outcome, and the total of unanswered questions did not exceed past one question per participant. Therefore, this is not taken into account during the analysis.

It is important to note that the participants score on both sides of the personality scales. Knowing the preference is helpful, but the raw scores cannot be overlooked.

The results of the MBTI are structured as follows. An overview of the descriptive statistics is presented first, where both the differences within the group of Chinese participants and the Dutch participants are presented, as well as the differences between the groups. This is followed by a statistical analysis to uncover the differences within and between the two groups.

5.3.2 Descriptive statistics

Personality types

Table 5.3 Personality type for Chinese and Dutch participants

ISTJ ISFJ INFJ INTJ

Chinese N=2 N=1 N=3 N=2

Dutch N=1 N=0 N=1 N=5

ISTP ISFP INFP INTP

Chinese N=1 N=1 N=0 N=0

Dutch N=0 N=2 N=3 N=5

ESTP ESFP ENFP ENTP

Chinese N=0 N=0 N=1 N=0

Dutch N=0 N=0 N=2 N=0

ESTJ ESFJ ENFJ ENTJ

Chinese N=0 N=0 N=1 N=2

Dutch N=1 N=0 N=1 N=0

Due to the small sizes of the samples it is not possible to do an analysis on the preferred combinations of personality traits.

Personality traits

Second, to investigate what the differences are with regard to the participants‟ preference for each of the personality dimensions a bar graph was made. Figure 2 shows the results of the MBTI regarding the preferences of both groups of participants. Due to the differences in sample size, the scores are shown in percentages. Each group has a score of a 100% on each of the four dimensions E-I, S-N, T-F and J-P.

First, to see whether there are differences within the two groups, the graph provides valuable information regarding the distribution for each dimension. For the E-I and S-N dimensions there are large differences for both the Chinese and Dutch participants. A large difference was also found for the Chinese participants regarding the J-P dimension.

Figure 3 Average scores on the MBTI for the Chinese and Dutch participants

Figure 3 shows what the participants on average scored on the MBTI. This provides insight in the preferences shown in figure 2 and illustrates that the differences between the two dimensions are not as large as figure 2 seems to indicate. It shows that even though the preference for Thinking and Feeling is evenly distributed for the Chinese participants, the scores visible in figure 3 indicate that those who scored on Thinking, had a higher preference. Between the E-I dimension and the J-P dimension the difference between the sides are less than is visible in figure 2, this goes for both groups.

5.3.3 Statistical analysis

Within the groups

First, an independent samples t-test was done in order to see whether or not there were differences within the two groups of participants with regard to their personality. The analysis showed that there are significantly more Introverted (M=15.8, SD=4.5) than Extraverted. (M=8.4, SD=5.1) in the Dutch group of participants. This difference is significant (t(40)=-4.1, p=0.000014. A similar result was found for the Chinese participants, in this group there were also more Introverted (M=18.5, SD=5.9) than Extraverted (M=13, SD=5.3). This difference is significant (t(26)=-2.5, p=0.017

The analysis showed that within the Dutch group, there are more Intuitive (M=18.3, SD=2.9) than Sensing (M=8.4, SD=5.1). This difference is significant at (t(40)=-4.9, p=0.0008.

Within the Chinese group, there are more Judgers (M=17.5, SD=4.2) than Perceivers (M=11,1, SD =6.2). This difference is significant at (t(26)=3.1, p=0.004

Thus, the only significant differences within the Dutch group are that there are more Introverted and more Intuitive participants. Within the Chinese group there are more participants who prefer Introversion and Judging.

Between the groups

Then, an independent samples t-test was performed to see whether there were any significant differences between the groups. On average, the Chinese participants showed a higher preference for Judging (M=17.5, SD=4.25) than the Dutch participants (M=11.85, SD=8.2). This difference is significant (t(33)=2.4; p=0.025. No other significant differences in personality preference between the groups were found. Thus, the only difference between the groups regarding their personality preference is that there are more Judgers amongst the Chinese participants than amongst the Dutch participants.

5.4 Results SILL & MBTI combined

5.4.1 Introduction

Due to the small sample sizes it was not possible to conduct an analysis on the combination of the personality preferences and the relation with language learning strategies.

The results of the SILL and MBTI are structured as follows. To see whether or not there are differences between the Chinese and Dutch participants regarding the relation between their SILL and MBTI scores, a correlation analysis was performed on the two groups separately. The results of the statistical analysis will be presented, to uncover the potential differences within and between the groups. First the relation between the two scores is presented for the Chinese participants, then the results of the Dutch participants will be presented.

5.4.2 Correlation analysis

Chinese participants

The Chinese participants showed a negative moderate correlations between Introversion and metacognitive strategies, r(9)=-0.54, p=0.047. The Chinese participants also showed a negative moderate correlation between Thinking and social strategies, r(9)=-.55, p=0.041. For the Chinese participants, a moderate correlation between Sensing and metacognitive strategies was found, r(9)=0.56, p=0.027.

Thus, for the Chinese participants goes that the more introverted a person is, the fewer metacognitive strategies are used. Those Chinese who prefer Sensing more, use more metacognitive strategies. Finally, the more one prefers Thinking, the fewer social strategies are used.

Dutch participants

The correlation analysis showed a moderate negative correlation between Sensing and compensation strategies for the Dutch participants, r(19)=.48, p=0.027. The Dutch participants showed a moderate correlation for Intuition and memory strategies, r(19)=.51, p=0.019. Also for Intuition and the total amount of strategies a moderate correlation was found, r(19)=.46, p=0.034.

6.

D

ISCUSSION

In this section, the results of the SILL will be interpreted and discussed first. This will be followed by the results of the MBTI. Then, the relation between the SILL scores and the MBTI scores for the Chinese and the Dutch group of participants will be analysed.

6.1 SILL

In discussing the results of the SILL it is important to keep in mind that it is not possible to say anything about during what activities the specific language learning strategies are used. Not all language learning activities require the use of many strategies, although it is possible that more than one strategy is used. The results discussed below will therefore only focus on whether participants use only one strategy or if the analysis has shown that more than one strategy is used.

When looking at the results of the Chinese participants‟ SILL scores in figure 1, there are no large differences regarding the frequency of the use of the language learning strategies. On average, all six strategies have a medium use, no exceptions. This means that the strategies are sometimes used by the Chinese participants. At first glance, the average scores of the Dutch participants are quite similar to those of the Chinese participants, except for the use of memory strategies. Unlike the Chinese group, the Dutch group only reported a low use of this strategy, which means that memory strategies are generally not used by the Dutch participants. Although there are differences between the groups regarding the use of the strategies, the total strategy use is similar for both groups.

positive correlation found between social strategies and cognitive strategies is also quite logical. A characteristic of cognitive strategies is practicing in naturalistic settings, which shows some similarities to the social strategy to talk with native-speakers. Those learners who are likely to do either one, are likely to do the other as well. Regarding affective strategies and compensation strategies, their significant correlation cannot be explained.

The correlation analysis of the language learning strategy use of the Dutch participants shows a high correlation between affective strategies and four other strategies. First, the correlation between affective and social strategies can be explained by looking at the social part of language learning. Speaking is an important aspect in the learning process and anxiety for speaking can make it difficult for learners. Those who are used to regulating their anxiety would go out and speak in the target language (with a native-speaker). There is no immediate explanation for the correlation between affective strategies and memory, compensation and metacognitive strategies regarding the typical activities that are involved in these strategies. Evidently, a good regulation of any form of emotional distress can come in handy with every activity, whether it is planning a learning activity, or guessing a word from the context. Other characteristics of affective strategies include rewarding oneself for good performance and positive self-talk (Oxford, 2003), both can be very helpful and might even be used in combination with other language learning strategies.

strategies that directly involve the target language, because of mental processing. Thus, the preference in the language classroom seems to have influenced the strategy choice of the learners. In the Netherlands the preference for mental activeness in the classroom is less than in China. Moreover, verbal activeness seems to be favoured. Communication is a more important goal, which explains why the social strategies are used slightly more often by the Dutch participants.

It is surprising that the Dutch group is using affective strategies often in combination with other strategies, but that the Chinese participants use affective strategies significantly more. This could be because the Dutch participants enjoy using a wide range of strategies, while Chinese learners might be more comfortable with a certain strategy and stick to that strategy. A possible explanation could lie in the proficiency of the participants, however, it is still unclear whether good language learners use many strategies, which is what Ehrman & Oxford (1990) claim, or that less proficient learners use many strategies, as Cohen (1998) and Oxford, Nyikos & Crookall (1987) suggest.

6.2 MBTI

Introversion was the most favoured characteristics in both groups; of the Chinese participants; 71% showed a preference for Introversion. The Dutch participants showed with 80% an even larger preference for Introversion. Although this goes against the findings of Bedell (1993, in Oxford & Burry-Stock, 1995), who found that Caucasians are primarily Extraverted, this is not a surprising find. Moody (1988) found in his study on language students that the most preferred personality trait was Introversion and not Extraversion. Hence, the result found in the present study supports the claim that language students are primarily Introverted.

Regarding the T-F scale, a somewhat different result was found to that of Moody (1988), namely, the Dutch participants showed a slight preference for Thinking, though this difference was not significant. The Chinese participants showed no preference for either one of the scales; exactly half of the participants favoured Sensing, half preferred Thinking. This goes as well against the findings of Bedell (1993, in Oxford & Burry-Stock, 1995), considering his claim that Thinking-Judging made up the largest part of the sample. However, in this study, the Feeling-Judging combination was found for 35% of the Chinese participants, compared to only 28% for the Thinking-Judging. This result is surprising, bearing in mind the logic that is involved with language learning. As Moody (1988) stated, language learning, and especially textbook learning, has much to do with abstract rules and logical analysis. This appeals to Thinkers more than to Feelers, however, this study did not find a large preference for this personality trait.

This study also shows different results than Moody‟s 1988 study regarding the J-P dimension. The Dutch participants seem to show a preference for Perceiving which was similar to that what Moody found, however, the difference between both sides is not significant. This is similar to the even distribution Bedell (1993, in Oxford & Burry-Stock, 1995) found for the Western participants. The Chinese participants in this study prefer Judging over Perceiving, 78% versus 21%. Also, the analysis showed that the Chinese participants showed a significantly higher preference for Judging, compared to the Dutch participants. This is in line with what was found by Bedell (1993, in Oxford & Burry-Stock, 1995); two-third of the Chinese participants was found to favour Judging.

According to Moody (1988) the most common type in the language classroom is ENFP. In contrast, this study found that the most popular type for the entire sample was INFJ (21%) for the Chinese participants and INTJ/INTP (23%/23%) for the Dutch participants. This indicates that the two groups of participants are very alike in their overall preferences. It therefore seems to indicate that culture does not affect personality, or rather, that a preference for languages is more influential in this matter.

6.3 Relation SILL & MBTI

6.3.1 Introversion

learning does seem to be in line with what people who get their energy from solitary activities would want. Introverts are primarily oriented towards concepts and ideas in their inner world (Ehrman & Oxford, 1988), which explains why they would prefer metacognitive strategies in language learning, such as identifying their own learning preferences and needs, as well as monitoring mistakes and evaluation of the learning tasks. Metacognitive strategies are focussed on the inner world of the learner, similar to the characteristic of people who prefer Introversion. A negative correlation is therefore highly surprising. The Dutch participants did not show a significant correlation between Introversion and metacognitive strategies, however, it is still likely that culture has an effect on this relation, considering the results previously found in the U.S.

6.3.2 Sensing

Between metacognitive strategies and Sensing a positive moderate correlation was found for the Chinese group. Typical metacognitive strategies, such as gathering and organizing materials, setting clear goals and linking material to previously acquired knowledge fits those who prefer Sensing. A Sensing preference is that gathering information is based on data that they can get through the five senses and they enjoy it when the learning process is practical and factual. A similar result was not found for the Dutch participants, however this result supports the result found by Ehrman & Oxford (1990), where Sensers showed a high use of metacognitive strategies. Taking into account that Westerners who prefer Sensing also use metacognitive strategies, it is unlikely that the difference found in the present study is due to cultural differences.

6.3.3 Intuition

A moderate correlation was found between Dutch Intuitives and memory strategies. Typical for Intuitive learners is that they like to think in abstract and nonsequential ways. Memory strategies often make us of data organized in an orderly way, such as using acronyms (Oxford, 2003). Moreover, memory strategies are usually multisensory, in that they make use of imagery, physical responses and mechanical tricks, such as colour coding. Nevertheless, there are some aspects in memory strategies that might appeal to people who prefer Intuition. Meanings and relationships are liked by Intuitives. Consequently, creating mental linkages, such as grouping and placing new words in a context can for that reason be very appealing. Despite that this result has not been found before, it gives reason to believe that the relation between Intuition and memory strategies is culturally modulated. Especially given the fact that the Chinese participants reported a significantly higher use of memory strategies, and in addition have an educational system that is more focused on memory than the Dutch system, provides evidence for this claim.

6.3.4 Thinking

A moderate correlation was found between Chinese Thinkers and social strategies, which is highly surprising, considering the impersonal aspects of Thinkers. According to Ehrman & Oxford (1990) Thinkers reported having difficulties with self-esteem issues and mentioned social detachment, such as being impersonal. They prefer books over people. Moreover, their need for control makes social strategies an odd choice, in view of the unpredictability of social interaction. Considering that this correlation did not occur for the Dutch participants and a negative correlation was found in prior studies with U.S. inhabitants, there is reason to believe that the relation between Thinking and social strategies is a culturally modulated relation.

7.

C

ONCLUSION

The current study aimed to get more insight in the relation between language learning strategies and personality style and whether this relation is culturally modulated. In order to answer this, a group of Chinese and a group of Dutch learners of English have been studied. The study was motivated by the fact that this relation has not been investigated before from a cultural perspective, even though much has been researched about this interesting subject. This study has shown that Chinese and Dutch students of English differed with regard to their language learning strategies. The Chinese participants used memory and affective strategies more often than the Dutch participants. Moreover, the groups differed in their use of strategies, whereas the Chinese learners used more direct strategies, used the Dutch learner more indirect strategies. Several reasons are given to explain these differences, most likely these differences are a result of differences between the Dutch and Chinese educational system, and hence this is culturally determined.

The study also found differences between the personality styles of the two groups of participants. Overall, there were minor differences between the preferences of each of the dichotomies, however, only Judging turned out to be a large difference; the Chinese favoured this side of the scale more than the Dutch learners. Considering the selective nature of the sample, solely language learners, it is difficult to generalize these findings to a larger population.

The aim of the study was to find out whether the relation between language learning strategies and personality style was modulated by cultural differences. This study has shown that this is indeed the case. Of the five significant correlations found, four can be attributed to cultural differences. For the Chinese participants the study showed that Introversion and metacognitive strategies were positively correlated, as well as Thinking and social strategies. Within the Dutch group Sensing showed a negative correlation with compensation strategies, while Intuition correlated positively with memory strategies.

Reflections, limitations and suggestions for further research

The aim of this study was not to find out what personality type or language learning strategy makes a good language learner, but to investigate what the influence of culture is on these two factors. However, it would have been interesting to take language proficiency into account. A future study with a well-structured assessment tool would provide us with valuable information as to the use of language learning strategies. Not only will that offer better possibilities to analysis, it would also aid in the debate on whether many or few language learning strategies would make a good language learner.

S

UMMARY

This study investigates the differences in personality styles and how these differences relate to language learning strategies. Furthermore, it is investigated to what extent this relation is modulated by cultural differences. A group of Chinese students of English and a group of Dutch students of English have been assessed on their personality style and their language learning strategies using the Strategy Inventory Language Learning (SILL) and the Myers-Briggs Type Indicator (MBTI).

The outcomes of this study show that there are cultural differences regarding the preference for certain language learning strategies as well as the use of these strategies. The Chinese participants used significantly more memory-related and affective strategies than the Dutch participants. The study also found differences between the two cultural groups regarding their personality styles. There were significantly more Judgers than Perceivers in the Chinese group, in comparison to the Dutch group.