Initial Development and Validation of Multicultural Counseling Competencies Scale for Malaysian School Counselors

Multicultural counseling is an important aspect in counseling. The introduction of the multicultural counseling competencies was a major step to improve counseling services among school counselors. However, there is a lack of valid and reliable instrument to measure multicultural counseling competencies in Malaysian context. This initial study focuses on the development of a local instrument to assess multicultural counseling competencies among Malaysian school counselors. This instrument consists of 42 items and it was administered to 212 secondary school counselors in Negeri Sembilan and Wilayah Persekutuan, Kuala Lumpur. Principal component analysis (PCA) was conducted in which items within initial dimensions of multicultural counseling competencies were reshuffled and checked for their psychometric properties. After series of principal component analysis (PCA) and analysis from the experts, three dimensions of multicultural counseling competencies with 28 items were formed. This paper also discusses the findings and its implication to the instrument development and the validation process.


Introduction
The guidance and counseling services became an integral part of the Malaysian education system in the year of 1960 (Kementerian Pendidikan Malaysia, 1993).This service foothold our nation's human capital developments plan.A lot of skills training had been introduced to provide more professional counselors to support the vision and mission of counseling ever since (Kementerian Pelajaran Malaysia, 2010).This service has been through enhancement process in phases and it started out with the post of school counselors in secondary schools in 1996.Meanwhile guidance teachers were provided in primary schools in 2002.
Current statistic (Kementerian Pelajaran Malaysia, 2012) shows the total number of 2.33 million secondary school students' with 1163430 male students and 1163797 female students.In general, Malaysian population comprises many ethnic groups with the Malays at around 50.4%, followed by Chinese and Indians with approximately 23.7% and 7.1% of the population respectively and others 7.8%.There are also other indigenous groups in Sabah and Sarawak at around 11% of the population.It gives the whole picture of various kinds of students that leads to multicultural amongst secondary school students which supposed to be faced by school counselors.
A good quality school counselor from the aspect of personal and professional would affect students' self-identity and nation's capital development.Effective counselors do not have only good counseling skills but they are also different in self-beliefs and personalities (Combs, 1982).Having knowledge and knowing one-self from the aspect of efficiency, values and self-belief, patient and loving, power and intimacy, self-analyses and interpersonal awareness, an ability to be a model and an influence to others, less selfish, flexible, sympathy and empathy are the common traits of an effective counselor (Belkin, 1981;Cormier & Cormier, 1998;Lawrence & Ginger, 1999;Hanbal, 1998).
Based on Theory Multicultural Counseling and Therapy by Sue, Ivey and Pedersen (1996) and Model Multicultural Counseling Competencies by Sue, Arredondo and McDavis (1992), the multicultural counseling competencies involve three main dimensions which include awareness, knowledge and skills.Pedersen (2002), Sodowsky (1996) and Contantine & Ladany (2000) had made many research and writings involving multicultural counseling competencies amongst lecturers, trainee counselors and counselors as well as the perceptions of minorities towards the multicultural counseling competencies from the aspect of training, efficiency, practices and counseling etiquette.
According to Ching and Ng (2010), the future of Malaysian's guidance and counseling services depends on the enhancement and active development of the three important aspects; theoretical orientation such as family therapy, multicultural approach, research and clinical trainings that focuses on new findings.Research in multicultural competencies in Malaysia is still in a lower degree compared to research in western countries (Kamarul & Salleh, 2010).The lack of appropriate instruments that suitable for local culture has blazed a trail for researchers to build an instrument that can measure the multicultural counseling competencies.
Result from the literature reviews and interviews from local counseling experts had suggested four main constructs to define the multicultural counseling competencies in Malaysia.In the early stage, four constructs were named as awareness, knowledge, skills and counselor traits.The purpose of this study is to develop and to validate the multicultural counseling competencies scale for Malaysian school counselors.This study has three research questions.The research questions are: 1) Is the sample size adequate to perform principal component analysis (PCA)?
2) What are the factors extracted from PCA?
3) Are there any overlapping measurements (multicollinearity) among the reshuffle items?

Population and Sampling
This study used purposive sampling to collect the data (Ary et al., 2002;Othman, 2001;Konting, 1993).The samples were school counselors in Negeri Sembilan and Wilayah Persekutuan Kuala Lumpur with the population number of 300 out of 500 school counselors in that states.Two hundred and twenty six (226) school counselors were chosen to meet the sample size determination by Krejcie and Morgan (1970).Samples were chosen and were generalized as the whole population of Malaysian school counselors that had been trained and recognized by the Ministry of Education.However, only 212 questionnaire that were successfully recollected and enough to be analyzed with 42 constructed items (Hair et al., 1998;Hair et al., 1987;Coakes & Steed, 1999).

Instrument
This instrument had been built to identify the multicultural counseling competencies scale amongst secondary school counselors with the considerations of Malaysian culture and society.It was build based on the theory, literature reviews and interviews with experts in counseling.The procedures in developing this instrument had been used by most researchers to collect the quantitative data from research samples (Jazimin, 2008;Bogdan & Biklen, 2003;Sidek, 2005;Shertzer & Linden, 1979).The final questionnaire that had been distributed to research samples consists of 42 items.

Results and Discussion
The aim of the study is to examine the psychometric properties of the developed multicultural counseling competencies instrument.This was established by running the reliability test of the instrument using SPSS 20.0.Principal Component Analysis (PCA) with Varimax rotation was performed to find the underlying dimension of the instrument with minimal loss of information.The extracted factors and factor loadings and component matrix were matched with the conceptual framework derived from theories and literature review.Reliability test and item analysis were also performed to describe the psychometric properties of the instrument.

Sample Adequacy
Sphericity Barlett was conducted to ensure that all variables have the factorability with high Chi Square value (significant 0.05).Meanwhile, Kaiser-Meyer-Olkin (KMO) was conducted to see the sampling adequacy (Hair et al., 1998).As shown in Table 1, the Bartlett Sphericity Test yielded statistically significant intercorrelation χ2 (861) = 6891.799,p = .000with an overall MSA of .905,which exceeds the value of .60,indicating that the study had fulfilled the sampling adequacy requirement for factor analysis.Hair et al. (1998) also noted that PCA requires partial correlation between items to be more than 0.7..9056891.799861 .000 Table 1 shows the KMO to answer the first research question "Is the sample size adequate to perform principle component analysis?"To conduct PCA, the minimum value of KMO must not less than .50 and a significant value of < 0.05 (Field 2005) is required.By mapping these requirements to the output of the analysis, it could be concluded that the multicultural counseling competencies questionnaire met the first requirement for the implementation of PCA.Respectively, the anti-image correlation values of the developed items are greater than 0.7.

Structuring Factors Using Principal Component Analysis
This instrument had been through the factor analysis process to choose exact items for each construct.Varimax Rotation was done to validate the constructed items as well as to see the respective group item for each construct.
Through the analysis, the actual numbers of the construct had been determined.In factor loading, practical significance value of 0.40 is required (Coakes & Steed, 1999;Hair et al., 1998).
Table 2 shows the total variance explained.Only a total initial eigenvalues of above 1.0 or cumulative value above 60% was considered (Field, 2005).For the multicultural counseling competencies questionnaire, there were 8 values greater than 1.0.Thus, researchers conclude that there are 8 constructs in this questionnaire with cumulative value greater than 60% that is 69.25%.Besides total variance explained, the scree plot as shown in Figure 1 can also be used to determine the number of constructs.In this case, the scree plot shows only 3 constructs within the multicultural counseling competencies questionnaire.The scree plot supports the theories used in this study claiming that there are three principal factors that could be used to determine the competencies of multicultural counseling among the school counselors.Figure 1.Determination of total factor using the scree plot

Overlapping Measurement (Multicollinearity)
A series of PCA was carried out to identify items that fit in a specific component.Analysis by rotated component matrix table stipulates that only values which have the capacity or loading of .40 and above will be accepted as items for the respective construct.The first extraction method using principal component analysis by Varimax with Kaiser Normalization rotation method converged in 10 iterations yielding 8 factors.After careful considerations, overlapping items were properly examined where some items were selected for certain constructs and some were discarded.These overlapping items indicated the existence of multicollinearity.
At first, overlapping items which discarded were items no 13, 14, 16, 17, 21 and 27.Meanwhile, item no 25 that did not have the factor loading .40 was discarded too.There were no more overlapping items after three times PCA were done.However, 7 components were still yielded.Three components (component 1, 2 and 3) were apparently represented by specific items.Meanwhile the other four components were represented by one to three items.
Based on the expert judgment, component 6 and 7 which represented by one item was discarded.After that, items in component 4 and 5 were evaluated.Items in component 4 (items no 10, 11 and 12) and component 5 (item no 18 and 19) were actually had been represented by items in component 1, 2 and 3. Based on theories and interviews with the experts, researchers agreed that three components can be determined from the analyses.Thus, items in component 4 and 5 had been discarded.In conclusion seven items were finally discarded.Table 3 and 4 shows the last two results of the analysis which eight original components have been trimmed down to 3 components or factors.
Overlapping items show that the respective items, measure more than one construct.Thus analysis and evaluation of the items were made to determine where the respective items belonged to, in addition to take into account the value of the items that were loaded on both constructs.This process was intended to elicit answer for the third research question-"Is there overlapping measurements (multicollinearity) among the items that measure the multicultural counseling competencies"?Based on the rotated component matrix, it could be concluded that the multicultural counseling competencies instrument contained multicollinear items that were cross loading with other constructs at the beginning.Based on the theory and discussion with the research team, these factors were subsequently named as (i) awareness, (ii) knowledge, and (iii) skills.Constructs of the counselor characteristics that had been proposed in the early discussion had been respectively represented by items in component 1, 2 and 3.This questionnaire also has the three main constructs which parallel with multicultural counseling competencies theory.Table 5 summarizes the total item and items that represent each construct.

Conclusion
As one of the initials research, this study had successfully identified three main domains in multicultural counseling competencies for Malaysian school counselors.The research findings also reveal the suggested aspects that a counselor must have in order to become a competent counselor.According to the experts' judgment and counselors, the multicultural counseling competencies in Malaysian cultural context are the awareness, knowledge and skills.From the findings, in-depth research needs to be done in order to identify constructs that lead to multicultural counseling competencies, based on the local culture.More specific instruments are needed especially the constructed items to measure whether it really represents the constructs of multicultural counseling competencies.The developed questionnaire must be used in larger population and the reliability test should be rerun.This 28 items questionnaire needs to be analyzed again so a high validity instrument could be developed.Latest research analysis method such as Rasch Modeling could be applied for future study.

Table 2 .
Total variation that can be determined

Table 3 .
Structuring factors using rotated component matrix Rotation Method: Varimax with Kaiser Normalization.a. Rotation converged in 7 iterations.

Table 4 .
Structuring factors using rotated component matrix

Table 5 .
Summary of constructs