Chapter Overview
Chapter three describes the research procedure and methods and shows research framework. Research techniques, data collecting strategy, and data analysis are explained in detail. Moreover, interviews and pilot test results are shown at the end of this section; they clarify how the main research instrument – the questionnaire, was developed. The outcome of pilot study explains why and how the actual research questionnaire was amended.
Research Procedure
The progress of this study is shown in figure 3.1 and researcher uses two methods for data collecting. The field of this study - mandarin training centers in Taiwan – is too broad and variant. Every school may have different approaches to international students learning and may utilize different teaching methods. This study does not aim to encompass them all;
on the contrary, the researched decided to focus on one language center only in order to provide more precise and applicable results.
Even within Mandarin Training Center of National Taiwan Normal University, there are many topics to be studied. However, scale of this paper was narrowed and the focus is kept on international students‟ performance and satisfaction. Research questions were set and the researcher started to search literature sources for such field. As was mentioned in chapter two, not many studies about language centers in general were conducted; however, the literature review was quite extensive.
Furthermore, NTNU Student Counseling Center was found as another rich source of information and therefore, two interviews with psychology counselors were conducted in order to learn more about the topic. Another three interviews were conducted with MTC
32
students. When information needed was collected, the researcher started to design a questionnaire. No existing questionnaire was used for this study, so reliability and validity tests have to be included. For reliability test, researcher used Cronbach‟s alpha, “a coefficient of consistency that measures how well a set of variables or items measures a single, unidimensional latent construct.” (Wikipedia Foundation Inc., 2008) For validity test, revision by two MTC teachers and peers was done after the pilot study analysis.
After data was collected, all information was analyzed using SPSS 16 (Statistical Package for the Social Sciences) software. Research outcomes and discussion are provided in chapter IV. (p. 41); chapter V. (p. 61) includes conclusion and recommendations.
33 order to collect data for actual research
34
Research Framework
Figure 3.2 shows research framework, which illustrates all hypotheses and how they will be tested. For purpose of this study, the author created four factors:
1) Hygiene factor 2) Situational factor 3) Adaptability factor 4) Region of origin
The researcher‟s decision to reduce variables into three factors (region of origin comprises only of one item) was initiated by the large number of items, which cannot be further scaled down. The author also hopes that such reduction will ease the data analysis.
This study examines whether or not the established factors affect international students‟
performance and satisfaction; also, variables are tested individually in order to see if they are significant when sample is divided according to length of stay in Taiwan and according to region of origin; further test will show if there is any correlation between international students‟ satisfaction and their performance.
35
Region of origin Hygiene factor Situational factor Adaptability factor
International students’
performance H1
H5 H2 H3 H4
H6
Figure 3.2. Study model Source: see Research framework
36
Research Methods
The actual research used quantitative method for data gathering. A quantitative questionnaire was designed by researcher and used to collect needed information. Five-point Likert scale was utilized in the questionnaire, which is a commonly used psychometric bipolar scale method that measures positive or negative response to a statement. The format of the scale is:
1 strongly disagree 2 disagree
3 neutral 4 agree
5 strongly agree
The respondents were asked to refer to their experience of studying Chinese language at the National Taiwan Normal University, Mandarin Training Center. They were instructed to answer all parts of questionnaire about their performance, factors that may affect their learning results and level of satisfaction within and also outside the school. They were also invited and encouraged to write down their own performance motivators, if they found it uncovered by the questionnaire.
The respondents were also asked to provide a few demographic data – to indicate gender, their region/continent of origin and if they are scholarship recipients or not. In case they answer positively, they were further asked what kinds of scholarship are they awarded (Taiwan government scholarship, their country government scholarship or some other kind).
A pilot test was conducted with 21 MTC students. Pilot study analysis allowed the researcher to amend the questionnaire so that it can search its purpose as well as possible.
37
Population and Sampling
Population of this study is restricted to NTNU Mandarin Training Center international students. The average size of population is 2000 students for each semester. This number is volatile; therefore only approximate figure is given. There are no limitations about country/region of origin. The only condition that must be accomplished is the time of being involved in mandarin studies, which is at least one semester (MTC has four semesters per year; each of them lasts for three months.), so that the information about students‟
performance is averaged from longer time.
This research is using snowball sampling method, which is recognized by the researcher as the most satisfying mean of data collection. By utilizing snowball method, the researcher expects enough variety of received data and thus rule out the possibility that majority of respondents will be from one region/continent only. Questionnaire was originally only in English; further, Chinese translation was added. The printed questionnaires were first given to selected teachers of MTC, who spread them to their students as well as their colleagues. The researcher herself also addressed MTC students directly in order to ask them to fill in the questionnaire. Students had three to five days to fill in the questionnaire, which were be collected by the researcher from MTC teachers.
Instrumentation
This research used two tools for data collection. In addition to the referred interviews, a questionnaire was designed by the researcher in order to get information that will answer the research question. It is comprised of 26 questions that are to be evaluated with five points Likert scale; further, two questions are open questions and last two items are closed questions.
The reliability was be measured by Cronbach‟s alpha (coefficient of internal consistency) and the validity was enhanced by two MTC teachers‟ review, peer review and
38
the pilot test. SPSS 16 software was used for computing Cronbach‟s alpha, which is defined as
“-where where N is the number of components (items or testlets), is the variance of the
observed total test scores, and is the variance of component i.”(Wikipedia Foundation Inc., 2008
Interview
Five interviews were conducted in order to obtain an in-look into problems that international students might face. The collected information was later used during the designing of this research main instrument – the questionnaire. To ensure the objectivity of the obtained data, the researcher interviewed two psychology counselors and three students of the Mandarin Training Center. First counselor was an American lady, who is a former employee of NTNU Student Counseling Center. Thanks to the advantage of English language fluency, she promoted the NTNU Student Counseling Center to the MTC international students. Her work was exclusively focused on foreign students and therefore, the information she provided to the researcher are of high value.
The second counselor is a Taiwanese lady, a current counselor of the NTNU Student Counseling Center, who also works with international students. Thanks to the ability to deliver counseling in two languages, she was able to compare difficulties that international and local students face.
Three students, who were interviewed, are all European females. Two of them are recent MTC students; one of them is currently enrolled in the MTC. Interviewees were
39
chosen according to their experience if foreign country. Therefore, first lady has experience with living in foreign, Chinese speaking country and the second one has no experience with living in foreign country, however, is used to travel. The last student interviewee has no experience neither with living in a foreign country or with traveling.
Questionnaire
Questionnaire was designed exclusively for this research by its author, without using any existing sample. Questionnaire is divided into five small sections for the sake of researcher‟s overview during the data analysis. The set of variables is a result of careful literature review, researcher‟s observation (during the years 2004 and from 2006 up to now) as well as outcomes of the abovementioned interview.
Some minor changes may be done in the questionnaire after the data from pilot test will be analyzed. As was mentioned before, snowball sampling method will be used for this research; therefore a printed version of questionnaire will be spread among potential respondents. This sampling also ensures respondents anonymity and any extra data (for example respondents „email addresses) will be required.
Data Analysis
The collected data was analyzed by SPSS software version 16. In order to see if there is significant relationship between international students‟ performance and their satisfaction, correlation, which indicates the strength and direction of a linear relationship between two variables, is used. In order to get result needed to answer hypotheses H2-H6, linear regression was used. To see what the strength of prediction of the outcomes was r2 (or coefficient of determination) was used as a statistic tool. The r2 showed how well the outcomes are predicted by the research framework.
40
For the purpose of this study, 300 questionnaires were spread among the MTC international students. The researcher collected 135 filled questionnaires. After careful examination, only 85 questionnaires were found applicable to this study. Collected data did not provide any response from students from Africa. The SPSS 16 statistical software does not consider samples less than 2 (n < 2). As the number of students from Australia and South America was less than two, therefore these samples had to be removed and the final sample number stabilized on 83.
Interviews results
For purpose of this study, the researcher decided to conduct semi-structured interviews with few topics to explore. Questions that the psychology counselors‟ interviewees were asked follows:
1) Did you work with MTC international students during your activity at NTNU Student Counseling Center?
2) How many foreigners seek for counseling services during the time you worked there?
3) Could you tell me, what were the major topics of your sessions with international students?
4) Were these problems related to school or were they behavioral problems or mental disorders?
For interviewees recruited from MTC international students, questions mostly identical to the questionnaire were used. Interviewer provided the students with enough space to share any opinion they had.
41
The results of all interviews are shown below as well as data obtained from the NTNU Student Counseling Center. It is a very limiting fact that only data from the fall semester of 2006 are available for the researcher. By that time, individual counseling as well as group counseling was accessible. Data available for group counseling are as follows:
Table 3.1.
Division of group sessions topics
GROUP TOPIC SESSION HOURS
MTC Support Group 68
Dream work Group 16
Total 84
Source: L.Cornberg, personal communication, October 25, 2008.
The following table summarizes the primary individual counseling topics among international student clients.
Table 3.2.
Divisions of international students’ problems
TYPE OF PROBLEM HOURS
Family relationship 19
Interpersonal relationship 8
Intimate relationship 11
Self-exploration 47
Career plan 30
Crisis 3
Mental disorder 5
Total 123
Source: L.Cornberg, personal communication, October 25, 2008.
42
According to the former psychology counselor of the NTNU Student Counseling Center, the most frequent problems that international students face include: basic health concerns, transitional issues, behaviors related to stress, interpersonal and intimate relationships, identity issues and career concern. (Figure 4.1)
Figure 3.3. Major concerns of MTC international students Source: Lynn Cornberg, personal communication, October 25, 2008.
Basic health concerns involve Sleep (too little or too much) Diet
Eating disorder behaviors Body image
Identity issues
Behaviors related to
stress
Basic health concerns
The most frequent issues
Transitional issues Career
Interpersonal &
intimate relationships
43 Finding satisfying and healthy foods
Behaviors related to stress involve
Behaviors related to alcohol and/or substance abuse Management of previously diagnosed mental disorders Homesickness
According to the American counselor, the age group from 24 -35 years, faces major transitional issues of living in a foreign country (alone or with problematic roommates) than the groups of other ages. They become stressed about long term relationships pressures, career and educational decisions, living in foreign society while at the same time becoming more confused and mixed up about their „own‟. Along with transitional issues, how to eat well and sleep well in modern life, tended to be a goal aimed in most therapy sessions. The counselor also stressed that “Many clients came and kept coming in order to focused non-judgmental and non-comparative conversations – they enjoyed having someone listen with real interests to their experiences as a foreigner and as a student. “ (Lynn Cornberg, personal communication, October 25, 2008)
The other interviewee, who is currently employed by the NTNU Student Counseling Center and counsels both MTC and NTNU students, provided the following information:
especially for students from developing countries, education systems differences, the loading and number of requirements on students becomes a significant stressor
specifically for western students (European and North American students) restrictions within the educational system in Taiwan seems stressing and they demand students‟
rights comparable to their countries (e.g., students opinions are not being considered
44
by faculty or school authorities, complains and demands for improvement are being swept from the table without further discussion, etc.)
language barrier problems
issues with family relationships (parents versus offspring perceptions and points of view) and relationships in general
unclearness or confusion about future career
The data obtained from all interviews helped the researcher to develop questionnaire for the actual study. Many issues were mentioned by interviewees. Therefore, the author summarized the problems, which were brought out by majority of interviewees, in the below figure. In total, there were five interviewees; two psychology counselors and three students.
The MTC students were individuals and they provided only their personal opinion. The two NTNU counselors, on the other hand, furnished the researcher with summaries of their work with international students. Therefore, first counselor represents approximately 100 cases, the other one around 80 cases.
45
Figure 3.4. Interviews summary – items with multiple occurrence
Source: personal conversation with NTNU Student Counseling Center psychology counselors, February-April, 2009)
0 0.5 1 1.5 2 2.5 3 3.5 4
Series 1
46
Pilot Study Results
The questionnaire used in this research was developed by the author and therefore needed to be proved as valid and reliable. For this reason, a pilot study was conducted. 21 questionnaires were collected and obtained data were analyzed in the SPSS 16 statistical software. Cronbach‟s alpha (coefficient of internal consistency) was used in order to run a reliability test. The validity of questionnaire was procured by additional interviews with NTNU Counseling Center psychology counselors and two teachers of the Mandarin Training Center.
Cronbach‟s alpha was run separately for each set of variables – hygiene factor, situational factor and adaptability factor. For hygiene factor, 18 items were processed. The Cronbach‟s alpha was 0.796. Generally, Cronbach‟s Alpha of 0.7 is considered to be a good reliability value. The researcher also used the item-total correlation, which tests each item‟s correlation with the test total in order to ensure that each item is correlated to the total and therefore is contributing to the test. „A high total correlation indicated that the item is closely related to the content that the scale represents. A low correlation between an item and the scale suggest that the item measures something conceptually different from the scale.‟ (Miles
& Gilbert, 197-198) When the item shows correlation coefficient that is lower than 0.3, it should be removed from the scale. Therefore, the researcher extracted two items from the hygiene factor. In consequence, Cronbach‟s Alpha also rose to 0.812.
Situational factor data, obtained in the pilot test, was run in the same way. However, the reliability of the original 19 questions was very low with Cronbach‟s Alpha only 0.433.
Therefore, the researcher ran the item-total correlation test to see, which items are not conceptually related to situational factor. According to the results, 13 questions were deleted.
With the six questions left, the reliability strengthen to α=0.731.
Last group of variables – adaptability factor, only comprised of three questions. The reliability test showed α=0.874 and item correlation coefficient indicated that all items are
47
contributive to the adaptability factor. Therefore, this section remained unchanged.
One question was added in demographic data in order to specify the length of respondent‟s stay in Taiwan. Other change, that emerged as necessary after the pilot test, was adding of Chinese translation. Due to the fact that many NTNU Mandarin Training Center students do not speak English, the researcher decided to provide Chinese mutation in order to augment the number of population who was able to answer the questionnaire. Also, recommendation for translation was given by a MTC teacher who reviewed the pilot test questionnaire.
The accuracy of translation of the questionnaire was checked by two peers‟ review and by professor of International Human Development department of NTNU. The translation intended to use rather simple language so that even students who did not reach higher level of Chinese could answer the questionnaire.
Chapter Summary
Chapter three provided a detailed in-look into the research methodology. The logic of research procedures was explained and illustrated. Interviews results justify the development of major research instrument – the questionnaire; pilot test proves its reliability and validity.
48
49