Research Article
The SmartPLS Analyzes Approach in Validity and Reliability of Graduate
Marketability Instrument
Arasinah Kamis1, Ramdzan Ali Saibon2, Faizal Amin Nur Yunus3, MohdBekri Rahim4, Lazaro Moreno Herrera5, Pedro Luis Yturria Montenegro6
1,2Universiti Pendidikan Sultan Idris, Tanjong Malim 35900, Perak Malaysia, 3,4
Universiti Tun Hussein Onn Malaysia, Parit Raja 86400, BatuPahat Johor, Malaysia, 5Department of Education, Stockholm University, Stockholm, SE 10691, Sweden,
6Technical Science Faculty, Pinar del Rio University, Marti 271 Pinar del Rio, C.P. 20100, Cuba Article History: Received: 10 November 2020; Revised: 12 January 2021; Accepted: 27 January 2021; Published online: 05 April 2021
Abstract: The purpose of this study is to develop and validate the Graduate Marketability Model (GMM) for Malaysian Vocational Colleges (KV) Business Management graduates. The approach used was quantitative with a survey design involving 243 Business Management graduates from KVs across Malaysia. The sample selection was based on simple random sampling techniques. Subsequent findings of the PLS-SEM analysis of the measurement model inidicated that the CR values obtained for each construct were in the range of 0.889 to 0.990 where the skills contributed the highest (CR) value (0.990) whilst the involvement in development scored the lowest CR value (0.800). Overall, this study shows that the three main constructs are important in contributing to the marketability of graduates, which are skills, Proactive Career Behavioursand Positivity Traits.Implications of the study prove that improvements to the tertiary education system are necessary so that graduates are always able to meet the needs and requirements of industry employers. The next development of GMM will be able to assist more relevant parties, especially the Ministry of Education Malaysia (MOE) and the Technical and Vocational Education Division (BPTV), to be a useful guide at producing marketable graduates.
Keywords: Entrepreneurship education, unemployment, Employability Skills, Marketability, Business Management, Vocational College, Skills
1. Introduction
It is undeniable that Malaysia still faces a number of economic issues that cannot be fully resolved. One of them is the issue of unemployment that has to do with the marketability of graduates. The phenomenon of unemployment and the difficulty of obtaining employment among graduates of tertiary education institutions is at an alarming rate(Dian Indrayani Jambari, Umi Asma’ Mokhtar, Hana Yasmein Ishak, & Mohd Ridzwan Yaakub, 2015; Kee-Cheok Cheong, Hill, Fernandez-Chung, & Yin-Ching Leong, 2016; Mei Chou Chun & Hua Shen Chien, 2015; Norliana Hashim et al., 2016; Suhaili Hanafi, 2015; United Nations General Assembly, 2015; Zaliza Hanapi, Mohd Safarin Nordin, & Ridzwan Che Rus, 2015).
Overall, the unemployment rate in Malaysia in August 2017 was at 3.4 percent, down 0.1 percentage points from the previous month, while year-on-year, the unemployment rate also dropped 0.1 percentage points compared to August 2016 (Department of Statistics Malaysia, 2017). Although the unemployment rate sinkedby 0.1 percentage points from the previous month and the year, it is still showing a growth compared to 2015 and previous years. This situation explains that there is a consistent year-over-year trend of unemployment rate that can be considered high. This illustrates that the issue of unemployment in Malaysia is not only current but a lasting issue.
This situation seems to indicate that no solution can be particularly effective in solving unemployment problems especially among Business Management graduates. This phenomenon of unemployment involves graduates in various fields. Literature and Social Sciences, Science and Technical fields are the top three which contribute to the highest unemployment rate among graduates in Malaysia (Department of Statistics Malaysia, 2017). A study conducted by Atif Aziz &FaizuniahPangil (2017) found that Business Management graduates rank highest in the youth unemployment list in Malaysia. This situation clearly shows that the issue of unemployment among Business Management graduates is at anworrying rate.
In addition, this study conducted an initial survey of 263 Business Management graduates of the Vocational College of Malaysia (KV) who had completed their studies. This preliminary review aims to determine the current status of graduates to determine whether there is a gap between Business Management graduates and job opportunities in the labour market. The questionnaire instrument validated by three experts was used in this initial survey. The survey found that 60.1 percent (158) Business Management graduates had work status, 35.4
percent (93) did not work and 4.6 percent (12) continued their studies. This gives the impression that the number of unemployed KV graduates is high and is alarming.
The findings from this preliminary survey also found that 75.9 percent (120) worked in their field of study and 24.1 percent (38) did the opposite. Besides, the number of graduates employed also found that 48.7 per cent (77) were working full time, 37.3 per cent (59) were in temporary status and 13.9 per cent (22) were contractual.These preliminary findings clearly show that the number of unemployed Business Management graduates is at an upsetting rate, and even for working graduates, the status of temporary workers and contracts does not guarantee their continuous employment in the future. The findingsof this preliminary survey also contradict KV's goal of targeting 70 per cent of graduates for one-on-one employment, 10 per cent entrepreneurs and 20 per cent in education.
Although many studies had been conducted on the issue of unemployment and graduation rates, their focuses were to graduates of institutions of higher learning in the first degree and involving either public or private universities(Saibon, Kamis &Zainol, 2019; Amanuddin Shamsuddin, Khairon Hamizah Mohmad Isa, Muhammad Naim Aziz, Nur Zahidah Nafisah Mohamed Mahfol, & Thevamalar Alagari, 2013; Dian Indrayani Jambari et al., 2015; Jackson, 2016a; Kee-Cheok Cheong et al., 2016; Noor Lela Ahmad & Suraini Mohd Rhouse, 2016; Nooriah Yusof et al., 2013; Sarimah Ismail & Dahiru Sale Mohammed, 2015; Shaharuddin Ahmad et al., 2014; Zaliza Hanapi et al., 2015). However, there are only a few studies that observe the marketability of graduates involving graduate-level KV graduates. This is because KV's first Business Management graduates are graduating in 2017.
Also, eventhough many recent studies have focused on employability skills that have been acquired by graduates yet there are very few studies that focused on marketability of graduates( Saibon & Kamis, 2019; Bishanani Omar et al., 2016; Jackson, 2016a; Kee-Cheok Cheong et al., 2016; McLachlan, Yeomans, & Lim, 2017; Mohd Hasril Amiruddin, Isma Atiqah Ngadiran, Fathin Liyana Zainudin, & Norhayati Ngadiman, 2016; Mohd Hazwan Mohd Puad, 2015; Muhammad Sabri Sahrir et al., 2016; O’Neil, 2014; Ridzwan Che Rus, Ruhizan Mohammad Yasin, & Mohammad Sattar Rasul, 2014; Sarimah Ismail & Dahiru Sale Mohammed, 2015; Zaliza Hanapi & Arasinah Kamis, 2017; Zaliza Hanapi et al., 2015). As such, this study was conducted with the aim of developing the KV Malaysia in Business Management Graduate Marketability Model (GMM).
2. Research Questions
The research objectives are; 1) Do GMM Business Management graduates be explained by the constructs and sub-constructs? and2) Do GMM Business Management graduates meet the requirements of validity and reliability?
3. Methodology
This research utilised the quantitative design by applying the surveying research method. The population consisted of Business Management graduates of Vocational Colleges throughout Malaysia with a population of 1040 graduates. Simple Random Sampling sampling technique was used to select 243 samples of the current study and the current location of the sample scattered throughout Malaysia.The instrument of study was adapted from the previous study (Table 1), items of one to seven Likert scale where the value of one represents the statement 'Strongly Disagree', while value seven represents the statement 'Strongly Agree’.In this study, the variables were ‘continuous’ and not ‘categorical’. Therefore, the data obtained in this study is 'continuous data'. Researchers have used questionnaires seven times since they are consistent with 'continuous' data(Bartlett, Kotrlik, & Higgins, 2001).The instrument in google form was distributed using online platforms which aree-mail and WhatsApp application. Out of the total population of 1040 graduates who were distributed with the survey questionnaire, 243 graduates returned the completed questionnaire.
Table 1. Sources and Number of Items in the Instrument
Constructs Total of items Sources
1. Skill
1.1 Basic Academic Skills
14
(Mohamad Sattar Rasul, Md Yusof Ismail, Napsiah Ismail, Rashid Rajuddin, et al., 2009; Muhammad Sabri Sahrir et al., 2016;
O’Neil, 2014;Michigan Employ- ability Skills Employer Survey, conducted by the
Michigan Employability Skills Task Force (Employability Skills Task Force, 1988,
1989; Mehrens, 1989) 1.2 Employability Skills
1.2.1 Analytical Thinking And Problem Solving Skills
1.2.2 Communication Skills 1.2.3 InformationTechnology Application Skills
1.2.4 Leadership And Management Skills
1.2.5 Teamwork And Interpersonal Skills 1.3 Technical Skills 1.4 Entrepreneurship Skills 12 12 7 13 10 14 6
(Fatihya Mahdi Ahed Ali et al., 2017; Mohamad Shukri Abdul Hamid et al., 2014)
(Jackson & Chapman, 2012a; Zaliza Hanapi & Arasinah Kamis, 2017)
(Fatihya Mahdi Ahmed Ali et al., 2017; Mohamad Shukri Abdul Hamid et al., 2014) 2. Proactive Career Behaviors
2.1 Networking
2.2 Participation in Development 2.3 Job Mobility Preparedness
6 3 9 (Day, 2005) 3. Positivity Traits 3.1 Optimism 3.2 Positive Self-Concept 3.3 Learning Goal Orientation
10
12 (Day, 2005)
4. Internship Experience 8 (Muhammad Sabri Sahrir et al., 2016;
Yunjuan He & Xizhen Qin, 2017)
5. Marketability 9 (Day, 2005; Eby et al., 2003; Fatihya Mahdi
Ahmed Ali et al., 2017) Data Analysis
Data were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM) including Confirmatory Factor Analysis (CFA). Also, Statistical Package for Social Science Version 23 software (SPSS 23) and SmartPLS 3.0 were used in this study. Although measurement and structural models can be tested simultaneously through a one-stage approach, it iscommon to perform SEM analysis using two-stage approach. This approach involves the process of validating the measurement model using factor analysis (CFA) and the Structural Model Verification process (Huck, 2012). Model measurement validation involves four outlooks:
i. Identifying the Goodness of Fit (GOF) values for each latent variable. ii. Identifying GOF values for full model Measurement.
iii. Fixed GOF Measurement model value.
iv. Determining the validity and reliability of the Measurement model.
In the second stage of the SEM analysis, tests of the Structural Model need to be made by evaluating the GOF and testing the proposed hypothesis. Table 2 shows the validity and reliability values of the measurement model.
Table 2.Validity and Reliability Values of the Measurement model
Guidance Acceptance level Descriptions Sources
Reliability of the constructs
Guidance Acceptance level Descriptions Sources Construct
Reliability (CR)
0.7 or more higher Shows good reliability (internal consistency) * under 0.6 - low reliability
(Fornell & Larcker, 1981; Hair et al., 2010; Kline, 2011; Ramayah, Lee, & Mohamad, 2010)
Average
Variance Extracted (AVE)
0.5 and above - (Hair et al., 2010)
Constructs Validity Convergent Validity Standardized factor loading >0.5 or preferably 0.7 AVE > 0.5 CR > 0.7
It shows that the variables studied really reflect the laten constructs that are designed to measure
(Fornell & Larcker, 1981; Hair et al., 2010; Ramayah et al., 2010)
Discriminant Validity
AVEs > squared inter-construct correlation (SIC)
Or
square root of AVE > inter-construct
correlation
It shows that the construct is completely different from the other constructs
(Hair et al., 2010)
(Chiu & Wang, 2008; Fornell & Larcker, 1981; Ramayah et al., 2010)
Findings of PLS-SEM Measurement Model
It was a challenge to choose the fit of statistical models for data analysis. In this study, PLS-SEM analysis was used to obtain the study results for the fourth part. The choice ofthe analysis was made based on the criterion of this study's data which was normally distributed. In addition, the study also used a reflective-formative model and involved a moderator variable which was Continues.The choice made to conduct this PLS-SEM analysiswas also due to several other factors that can predict and explain the target constructs (Hair et al., 2016), and may also explore the relationships between constructs (Chin, 1998). To add, PLS-SEM is also capable of conducting analysis of complex structural equation models consisting of many constructs and indicators (Hair et al., 2016; Urbach & Ahlemann, 2010).
Study Questions i: Do GMM Business Management graduates be explained by the constructs and sub-constructs?
This study used Internal Consistency Reliability analysis to determine whether items used in the study could measure similar constructs in scores (Hair et al., 2016). Internal Consistency Reliabilitywasassessed using composite reliability (CR) values to determine internal consistency reliability. According to J. Nunnally & Bernstein (1994), CR values between 0.6 and 0.7 may be acceptable for exploratory studies, whereas for further studies the values between 0.7 and 0.9 may be considered satisfactory. CR values need> 0.7 to ensure adequate or sufficient internal consistency(Hair et al., 2016; Gefen, Straub, & Boudreau, 2000). Cronbach’s Alpha α> 0.7 is also a measure of the reliability of items measuring a construct (Nunnally & Bernstein, 1994). Table 3 shows the values of Cronbach’s Alpha and CR for each construct in this study.
Table 3.Findings ofInternal Consistency Reliability
Constructs Cronbach's Alpha (α > 0.7 ) Composite Reliability (CR > 0.7 )
Basic Academic 0.971 0.974 Goal Orientation 0.954 0.962 Employability 0.986 0.987 Marketability 0.928 0.940 Skill 0.989 0.990 Entrepreneurship 0.957 0.966 Positive Self-Concept 0.884 0.909 Optimism 0.908 0.926 Internship Experience 0.947 0.956
Constructs Cronbach's Alpha (α > 0.7 ) Composite Reliability (CR > 0.7 )
Participation in Development 0.813 0.889
Job Mobility Preparedness 0.946 0.955
Networking 0.913 0.933
Positivity Traits 0.957 0.961
Technical 0.967 0.971
Proactive Career Behaviors 0.925 0.936
The analysis identified that the CR values obtained for each construct were in the range of 0.889 to 0.990. While Cronbach’s Alpha values ranged from 0.813 to 0.989. This proved that the CR and Cronbach’s Alpha values are satisfactory and acceptable, all of these 15 sub-constructors and formative constructions have high levels of internal consistency reliability and reliability(Gefen et al., 2000; Nunnally & Bernstein, 1994). Composite Reliability CR > 0.7 (Gefen et al., 2000), thus answering the research question that GMM construction is reliable.
Research Question ii: Do GMM Business Management graduates meet the requirements of validity and reliability?
The analysis also used Convergent Validity to see how well an item can measure similar constructs in a study (Hair et al., 2016). Convergent validity analysis can be evaluated through three tests namely; 1) Penilaian Outer Loading; 2) Composite Reliability (CR); and 3) Average Variance Extracted (AVE). First, the Outer Loading value is> 0.708 (Hair et al., 2016), because the value of 0.708 in duplicate is equal to 0.5 which represents the Average Variance Extracted (AVE) value. The AVE value must be greater than 0.5. Therefore, the value of Outer Loading between 0.40 to 0.70 should be considered for elimination, in the event of removal of the item it may increase the value of AVE or CR (Hair et al., 2016). Additionally, the value of Outer Loading> 0.5 (Chin, 1998; Hulland, 1999)can also be taken into account as the item was considered a good consonant. Table 4 shows the findings of Outer Loading, CR and AVE values. In this study, several items were excluded based on the AVE value requirement for each construct that should exceed 0.5 (Bartlett, Kotrlik, & Higgins, 2001)and CR values that greater than 0.7 (Hair et al., 2016). This indicated that all items exceeded the predefined level except items KEK 10, KETM 3, PK 12, PK 2, PK 4, PK 6, PO 10 and PO 3 which have been removed for did not meeting the minimum requirements of Outer Loading.
Table 4.Findings of Convergent Validity 2nd Order Construct AVE (>0.5) CR (>0.7) 1st Order Construct Items Loading (>0.5) AVE (>0.5) CR (>0.7) Marketability K1 0.784 0.635 0.940 K2 0.813 K3 0.848 K4 0.821 K5 0.768 K6 0.818 K7 0.815 K8 0.780 K9 0.721
Skill 0.736 0.917 Basic Academic KA1 0.817 0.728 0.974
KA10 0.861 KA11 0.869 KA12 0.897 KA13 0.860 KA14 0.724 KA2 0.832 KA3 0.826 KA4 0.891 KA5 0.846
2nd Order Construct AVE (>0.5) CR (>0.7) 1st Order Construct Items Loading (>0.5) AVE (>0.5) CR (>0.7) KA6 0.845 KA7 0.890 KA8 0.873 KA9 0.902 Employability KEI1 0.723 0.596 0.987 KEI10 0.746 KEI2 0.776 KEI3 0.716 KEI4 0.828 KEI5 0.830 KEI6 0.832 KEI7 0.847 KEI8 0.783 KEI9 0.806 KEK1 0.761 KEK11 0.804 KEK12 0.813 KEK2 0.826 KEK3 0.811 KEK4 0.813 KEK5 0.747 KEK6 0.720 KEK7 0.696 KEK8 0.677 KEK9 0.724 KEKP1 0.818 KEKP10 0.848 KEKP11 0.657 KEKP12 0.808 KEKP13 0.880 KEKP2 0.821 KEKP3 0.870 KEKP4 0.828 KEKP5 0.863 KEKP6 0.853 KEKP7 0.859 KEKP8 0.826 KEKP9 0.833 KEPM1 0.743 KEPM10 0.729 KEPM11 0.713 KEPM12 0.751 KEPM2 0.786 KEPM3 0.789 KEPM4 0.788 KEPM5 0.803
2nd Order Construct AVE (>0.5) CR (>0.7) 1st Order Construct Items Loading (>0.5) AVE (>0.5) CR (>0.7) KEPM6 0.765 KEPM7 0.788 KEPM8 0.684 KEPM9 0.723 KETM1 0.666 KETM2 0.683 KETM4 0.699 KETM5 0.593 KETM6 0.687 KETM7 0.545 Entrepreneurship KK1 0.929 0.829 0.966 KK2 0.946 KK3 0.939 KK4 0.962 KK5 0.941 KK6 0.725 Technical KT1 0.885 0.706 0.971 KT10 0.736 KT11 0.810 KT12 0.776 KT13 0.713 KT14 0.700 KT2 0.896 KT3 0.899 KT4 0.906 KT5 0.895 KT6 0.913 KT7 0.916 KT8 0.861 KT9 0.813 Proactive Career Behaviors 0.560 0.786 Job Mobility Preparedness KPPK1 0.543 0.708 0.955 KPPK2 0.857 KPPK3 0.894 KPPK4 0.911 KPPK5 0.830 KPPK6 0.864 KPPK7 0.838 KPPK8 0.899 KPPK9 0.877 Participation in Development KPP1 0.855 0.727 0.889 KPP2 0.863 KPP3 0.840 Networking KPN1 0.763 0.699 0.933 KPN2 0.834
2nd Order Construct AVE (>0.5) CR (>0.7) 1st Order Construct Items Loading (>0.5) AVE (>0.5) CR (>0.7) KPN3 0.870 KPN4 0.884 KPN5 0.832 KPN6 0.829 Positivity
Traits 0.792 0.920 Optimism PO1 0.737 0.610 0.926
PO2 0.833 PO4 0.780 PO5 0.749 PO6 0.855 PO7 0.825 PO8 0.707 PO9 0.748 Positive Self-Concept PK1 0.627 0.557 0.909 PK10 0.759 PK11 0.776 PK3 0.678 PK5 0.780 PK7 0.798 PK8 0.856 PK9 0.670 Goal Orientation PM1 0.797 0.762 0.962 PM2 0.862 PM3 0.725 PM4 0.918 PM5 0.900 PM6 0.925 PM7 0.931 PM8 0.905 Internship Experience PL1 0.868 0.730 0.956 PL2 0.877 PL3 0.889 PL4 0.876 PL5 0.775 PL6 0.848 PL7 0.871 PL8 0.823 Discriminant Validity
The discriminant validity analysis was performed to assess how well the constructs tested differed from the other constructs. This analysis can determine how much one construct correlates with another construct and how many items can represent a single construct (Hair et al., 2016). This study performed three test analysed to measure discriminant validity: 1) Cross Loading; 2) Kriteria Fornell & Larcker; 3)Heterotrait-Monotrait Ratio (HTMT).
Cross Loading
The loading value of the construct should be greater than all the loadings in the other constructs (Hair et al., 2016). It is the subjective freedom of each predictor to latent Variable. These criteria can help reduce the presence of multicollinearity among latent variables in indicating that the Average Variance Extracted (AVE) value of the latent variable should be higher than all other variables (Chin, 1998; Larcker, 1988; Vinzi, Chin, Henseler & Wang, 2010).If the value of loadings for other constructs exceeds the loading value for the construct then it indicates a problem with discriminant validity (Hair et al., 2016). The results showed that the cross loading value for an item in a given construct is was greater than the loading value for another construct. The findings showed the value of cross loading that provides evidence of validity for the measurement model construct.
Kriteria Fornell & Larcker
The Fornell-Larcker criterion is an analysis to compare the value of the AVE square root with the construct correlation value showing the highest value in any column or row compared to the highest correlation value of any other construct (Hair et al., 2016). This method is based on the view that latent variables should explain better for the item variant than the variant for other latent variables. Table 5 shows the higher AVE squared values compared to the correlation values for each other construct after some indicators that did not meet the outer loading conditions were eliminated. Based on the findings of the analysis, the Fornell-Larcker criterion validation of the discriminant validity test subsequently answered the listed research question regarding the validity of the constructs measurement model.
Table 5.Findings of Kriteria Fornell-Larcker Analysis
Marketability Skill Positivity Traits Proactive Career Behaviors Marketability 0.797 Skill 0.568 0.858 Positivity Traits 0.671 0.756 0.890 Proactive Career Behaviors 0.570 0.696 0.696 0.748 Heterotrait-Monotrait Ratio (HTMT)
Heterotite-Monotrait Criteria Analysis (HTMT) should meet the requirement in which the HTMT value should be greater than HTMT .85, that is 0.85 (Kline, 2011) or HTMT .90, 0.90 (Gold et al., 2001). As a statistical test he can test the null hypothesis (Ho: HTMT <1) vs (HA: HTMT≥1) (Henseler, Ringle, & Sarstedt, 2015) with HTMT 95% confidence interval containing value 1 (ie HA) then no discriminant validity. Based on Table 6, it is found that the HTMT values of the constructs tested met the analysis criteria by obtaining less than one value. This indicated that the relationship between constructs was vehemently weak that verified with the existence of a discriminant validity for each construct tested.
Table 6. Findings of Heterotrait-Monotrait Criteria Analysis
Constructs Marketability Skill Positivity
Traits Proactive Career Behaviors Marketability Skill 0.576 Positivity Traits 0.702 0.779
Constructs Marketability Skill Positivity Traits Proactive Career Behaviors Proactive Career Behaviors 0.595 0.690 0.736
Note: Diagonals represent the square root of the average variance extracted, while the other entries represent the squared Correlations Coefficients.
Figure 1. Structural Model of GGM 4. Discussion
PLS-SEM through the measurement of Internal Consistency Reliability model was used in this study, as it was able to carry out analysis of complex structural equation models consisting of many constructs and indicators (Hair et al., 2016; Urbach & Ahlemann, 2010). The choice of conducting this PLS-SEM analysis is due to its ability to predict and explain target constructs (Hair et al., 2016). In addition, PLS-SEM also has advantages over other analyses.The results of these finding are related with the reliability used to achieve the consistency of the overall outcomes of the item for the same construct (Hair et al., 2016) to determine whether the item used in the study can measure the same construct in the score value (Hair et al., 2016). Therefore, the findings of this study have taken into account composite reliability (CR) values to determine internal consistency reliability with a CR value of> 0.7 to ensure adequate or adequate internal consistency (Hair et al., 2016; Gefen, Straub, & Boudreau, 2000).In addition, Cronbach’s Alpha α> 0.7 values were also considered to determine the reliability of the items measuring the constructs in this study (Nunnally & Bernstein, 1994).
The findings showed that the CR values obtained for each construct range from 0.889 to 0.990, while Cronbach’s Alpha values range from 0.813 to 0.989. This explained that the CR and Cronbach’s Alpha values were satisfactory and accepted, indicating that all 15 sub constructs and formative constructs in this study had high levels of internal consistency reliability and reliability (Gefen et al., 2000; Nunnally & Bernstein, 1994). This clearly indicated that GMM for Business Management graduates can be explained by the constructs and sub-constructs of the study. Whereas to determine validity and reliability, CFA's Convergent Validity analysis refers to the extent to which an item can measure similar constructs in a study (Hair et al., 2016). Three tests were performed in this convergent validity analysis of i) external loading rating> 0.5, ii) composite reliability (CR) value> 0.7, and iii) average extracted variance (AVE) value> 0.5. This study was based on the Outer Loading> 0.5 values used by Chin (1998) and Hulland (1999) which refered as these items are considered good estimators. Based on the analysis conducted, the findings of the study have eliminated some items based on the AVE value for each construct that must be greater than 0.5 (Bartlett, Kotrlik, & Higgins, 2001)and CR values greater than 0.7 (Hair et al., 2016). This study also found that all items exceeded the specified level except items KEK 10, KETM 3, PK 12, PK 2, PK 4, PK 6, PO 10 and PO 3 which had to be removed due to failure to meet the minimum requirements of Outer Loading. In addition to some items that have been omitted, this finding clearly indicated that the items of this study have validity and reliability that successfully met the test requirements, which measure all elements, sub constructs and research consents.
5. Conclusion
This study used three main constructs namely skills, Proactive Career Behaviours and Positivity Traits. There are four sub skills constructs namely Basic Academic Skills, 'employability' skills, technical skills and Entrepreneurship skills; three sub-constructs of Proactive Career Behaviours namely networking, Participation in Development, and Job Mobility Preparedness; three sub-constructs of Positivity Traits namely Optimism, Positive Self-Concept, and Learning Goal Orientation, Internship Experience as moderator and Marketability constructs. The developed GMM has great implications for various parties, especially the Ministry of Education Malaysia (MOE) in designing new curriculum and syllabus that align with national education policy goals. Improvements and changes regarding educational and training curricula in tertiary education institutions in particular can be implemented with reference to GMM so that the curriculum and training implemented can equip graduates with the current knowledge and skills required by employers in the industrial sector. Therefore, the instrument of this research question is very useful to the next researchers as a guideline in developing a new instrument for related research.
GMM has been developed with the aim of assisting various parties in providing graduates with the knowledge and skills required by employers in the industrial sector. This highly particularised for graduates that wish to be successful in securing a job in the labour market whether working with an employer or working alone as a young entrepreneur. The findings from the PLS-SEM analysis that has shaped this GMM measurement model are expected to help more relevant parties, especially the Ministry of Education Malaysia (MOE) and the Technical and Vocational Education Division (BPTV), to be useful guidelines in producing graduates and marketable. The development of GMM should be a key reference for various parties especially the MOE as an effort to develop graduates who have the knowledge and skills that orienting with the needs and requirements of employers in the industrial sector. The issue of on unemployment experienced by every country especially in Malaysia need to be addressed and deciphered with various efforts which includes the increasementin the level of knowledge and skills of graduates especially in the field of Vocational Technical education (TVET). TVET education should focus on knowledge and skills as well as a number of other factors which is a key ingredient and essential element of each graduate's marketability. Therefore, the application of several key skills along with active pro-career behaviours and the positive nature of graduates are the key factors that need to be taken into account to ensure the first-class level of marketability of TVET graduates in particular. Thus, the role of all parties in particular MOE and BPTV through enhancing TVET programs and curricula can assist in the process of applying proactive career behaviours and positive attitudes among graduates in addition to their existing knowledge and skills. The need to have the knowledge and skills required by employers as well as to have a proactive career attitudes and positive attitudes, along with training experiences are bonus for every graduate to get a decent job in the labour market. Hence, the equilibrium between demand and labour supply can be achieved. Consequently, the problem of unemployment is solved and the country's economic growth could be boosted.
References
6. Amanuddin Shamsuddin, Khairon Hamizah Mohmad Isa, Muhammad Naim Aziz, Nur Zahidah Nafisah Mohamed Mahfol, & Thevamalar Alagari. (2013). Graduate unemployment: The awareness and perception of graduates towards government’s initiatives. International Journal of Business, Economics and Law, 3(1), 15–24.
7. Atif Aziz, & Faizuniah Pangil. (2017). Moderating Effect of Emotional Intelligence on the Relationship between Personality Traits and Employability. Saudi Journal of Humanities and Social Sciences, 2(2), 132–139. https://doi.org/10.21276/sjhss.2017.2.2.4
8. Bartlett, J. E., Kotrlik, J. W., & Higgins, C. C. (2001). Organizational Research : Determining Appropriate Sample Size in Survey Research. Information Technology, Learning, and Performance Journal, 19(1), 43–50. https://doi.org/10.1109/LPT.2009.2020494
9. Bishanani Omar, Sharifah Norul Akmar Syed Zamri, & Norlidah Alias. (2016). Kemahiran Kebolehpekerjaan : Isu Graduan Pengurusan Perniagaan Kolej Vokasional. ResearchGate. Retrieved from https://www.researchgate.net/publication/313517099%0AKemahiran
10. Chin, W. W. (1998). Issues and Opinion on Structural Equation Modeling. MISQ Archivist, 22(1). 11. Chiu, C.-M. M., & Wang, E. T. G. (2008). Understanding Web-based learning continuance intention:
The role of subjective task value. Information and Management, 45(3), 194–201. https://doi.org/10.1016/j.im.2008.02.003
12. Day, R. (2005). An empirical investigation of the predictors of self- and other reported marketability: Looking beyond human capital. University of South Florida. Retrieved from http://scholarcommons.usf.edu/etd
13. Dian Indrayani Jambari, Umi Asma’ Mokhtar, Hana Yasmein Ishak, & Mohd Ridzwan Yaakub. (2015). Information Technology Graduate Employability Enhancement Model: Case Study In The Faculty Of Information Science And Technology, Universiti Kebangsaan Malaysia. Asia-Pacific Journal of Information Technology and Multimedia Jurnal Teknologi Maklumat Dan Multimedia Asia-Pasifik, 4(2), 57–68. Retrieved from http://www.ftsm.ukm.my/apjitm
14. Eby, L. T., Butts, M., & Lockwood, A. (2003). Predictors of success in the era of the boundaryless career. Journal of Organizational Behavior J. Organiz. Behav, 24, 689–708. https://doi.org/10.1002/job.214
15. Fatihya Mahdi Ahmed Ali, Abdul Rahim Othman, & Norzieiriani Ahmad. (2017). Graduate’s Marketable Skills: An Empirical Investigation Of Its Effect On Perceived Marketability Of Yemeni Graduates In The Arab Gulf States. Asian Journal of Science and Technology, 8(11), 6497–6505. Retrieved from http://www.journalajst.com/sites/default/files/5135.pdf
16. Fornell, C., & Larcker, D. F. (1981). Evaluating Structural Equation Models with Unobservable Variables and Measurement Error, 18(1), 39–50.
17. Gefen, D., Straub, D., & Boudreau, M.-C. (2000). Structural Equation Modeling and Regression : Guidelines for Research Practice, 4(October). https://doi.org/10.17705/1CAIS.00407
18. Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis: A global perspective. Pearson Education.
19. Hair Jr, J. F., Hult, G. T. M., Ringle, C., & Sarstedt, M. (2016). A primer on partial least squares structural equation modeling (PLS-SEM). Sage publications.
20. Henseler, J., Ringle, C. M., & Sarstedt, M. (2015). A new criterion for assessing discriminant validity in variance-based structural equation modeling. J. of the Acad. Mark. Sci., (43), 115–135. https://doi.org/10.1007/s11747-014-0403-8
21. Huck, S. W. (2012). Reading statistics and research (6th edt). Boston, MA: Pearson, 276–311.
22. Hulland, J. (1999). Use Of Partial Least Squares ( Pls ) In Strategic Management Research : A Review Of Four Recent Studies. Strategic Management Journal, 204(April 1998), 195–204.
23. Jabatan Perangkaan Malaysia (2017).
https://www.dosm.gov.my/v1/index.php?r=column/pdfPrev&id=VzBiQlBKSVc0S0hRN0x1RDBqR0 9BZz09. Dilayari pada 15 Februari 2018.
24. Jackson, D. (2016). Modelling graduate skill transfer from university to the workplace. Journal of Education and Work, 29(2), 199–231. https://doi.org/10.1080/13639080.2014.907486
25. Jackson, D., & Chapman, E. (2012). Non-technical competencies in undergraduate business degree programs: Australian and UK perspectives. Studies in Higher Education, 37(5), 541–567. https://doi.org/10.1080/03075079.2010.527935
26. Kee-Cheok Cheong, Hill, C., Fernandez-Chung, R., & Yin-Ching Leong. (2016). Employing the ‘unemployable’: employer perceptions of Malaysian graduates. Studies in Higher Education, 41(12), 2253–2270. https://doi.org/10.1080/03075079.2015.1034260
27. Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed.). Principles and practice of structural equation modeling (3rd ed.).
28. Larcker, D. F. (1988). Structure Equation Modes With Unobservable Variables And Measurement Error: Algibra And Statistics.
29. McLachlan, K., Yeomans, L., & Lim, K. Z. G. (2017). A competency development approach to learning for employment (pp. 793–800). https://doi.org/10.4995/HEAD17.2017.5421
30. Mei Chou Chun, & Hua Shen Chien. (2015). Tertiary Students ’ Entrepreneurship Learning Socialization : Factor Analysis and Structural Equation Modeling, 2(3), 32–38.
31. Mohamad Sattar Rasul, Md Yusof Ismail, Napsiah Ismail, Rashid Rajuddin, & Roseamnah Abd. Rauf. (2009). Aspek Kemahiran ‘Employability’ yang Dikehendaki Majikan Industri Pembuatan Masa Kini. Jurnal Pendidikan Malaysia, 34(2), 67–79.
32. Mohamad Shukri Abdul Hamid, Rafikul Islam, & Hazilah Abd Manaf. (2014). Employability Skills Development Approaches: an Application of the Analytic Network Process. Asian Academy of Management Journal, 19(1), 93–111.
33. Mohd Hasril Amiruddin, Isma Atiqah Ngadiran, Fathin Liyana Zainudin, & Norhayati Ngadiman. (2016). Tahap kemahiran generik pelajar Malaysia dalam proses pengajaran dan pembelajaran : Kajian kes pelajar Institut Kemahiran Mara , Johor Baharu Malaysian students ’ l evel of generic skill in teaching and learning process : A case study of Mara Skills Insti. Journal of Society and Space, 3(3), 111–121.
34. Mohd Hazwan Mohd Puad. (2015). The Role Of Employability Skills Training Programs In The Workforce Of Malaysia. Purdue University West Lafayette, Indiana. Retrieved from
35. Muhammad Sabri Sahrir, Taufik Ismail, Nik Hanan Mustapha, Radhwa Abu Bakar, Saupi Man, Muhammad Anwar Ahmad, & Maahad Mokhtar. (2016). An Evaluation of Internship Programme in Improving Graduate Skills and Marketability Among Arabic Language students in IIUM from the Perspective of Malaysian Job Market. Journal of Education and Human Development, 5(1), 206–212. https://doi.org/10.15640/jehd.v5n1a21
36. Noor Lela Ahmad, & Suraini Mohd Rhouse. (2016). Kemahiran Profesional Graduan Perakaunan : Antara Jangkaan Majikan dan Keperluan Pelajar dalam Kurikulum Perakaunan. Pendidikan Malaysia, 41(1), 7–17.
37. Nooriah Yusof, Norain Mat Lazim, & Zakiah Jamaluddin. (2013). Persepsi Pelajar Prasiswazah Terhadap Program Pembangunan Kebolehpasaran Siswazah : Kes Universiti Sains Malaysia [ Usm ]. International Journal of Environment, Society and Space, 1(1), 43–61.
38. Norliana Hashim, Chang Peng Kee, & Mat Pauzi Abd. Rahman. (2016). STOPS : Mengungkai Isu Kebolehpasaran Di Malaysia. Malaysian Journal of Communication, 32(2), 139–164.
39. Nunnally, J., & Bernstein, I. (1994). Psychometric Theory, 3rd edn, 1994. McGraw-Hill, New York. 40. O’Neil, H. F. (2014). Workforce Readiness: Competencies and Assessment. Retrieved from
https://books.google.com/books?hl=en&lr=&id=9mgAAwAAQBAJ&pgis=1
41. Ramayah, T., Lee, J. W. C., & Mohamad, O. (2010). Green product purchase intention: Some insights from a developing country. Resources, Conservation and Recycling, 54(12), 1419–1427. https://doi.org/10.1016/j.resconrec.2010.06.007
42. Ridzwan Che Rus, Ruhizan Mohammad Yasin, & Mohammad Sattar Rasul. (2014). From zero to hero : Becoming an employable knowledge worker ( k-worker ) in Malaysia. TVED@Asia, (3), 1–16. Retrieved from http://www.tvet-online.asia/issue3/che-rus_etal_tvet3.pdf
43. Saibon, R. A., Kamis, A., &Zainol, Z. (2019). Entrepreneurship Education: UnemploymentIses, People’s Well Being and Entrepreneurial Intentions among TVET Graduates in Malaysia, International Journal of Psychosocial Rehabilitation, 23(4), 953-965. DOI: 10.37200/IJPR/V23I4/PR190423
44. Saibon, R. A., & Kamis, A. (2019). Employability Skills in Business Management Graduate and Role of Malaysian Vocational College. International Journal of Academic Research in Business and Social Sciences, 9(2), 1175–1192.
45. Sarimah Ismail, & Dahiru Sale Mohammed. (2015). Employability Skills in TVET Curriculum in Nigeria Federal Universities of Technology. Procedia - Social and Behavioral Sciences, 204(November 2014), 73–80. https://doi.org/10.1016/j.sbspro.2015.08.111
46. Shaharuddin Ahmad, Khaidzir, Azizan, Kadir, Zainul Ariffin, Khairil Anwar, & Wan Mazlina. (2014). The Social And Academic Skills And The Marketability Of UKM’s Graduates. Procedia - Social and Behavioral Sciences, 131, 118–123. https://doi.org/10.1016/j.sbspro.2014.04.089
47. Suhaili Hanafi. (2015). Kesediaan Pelajar Dari Aspek Kemahiran Teknikal Terhadap Pembentukan Kebolehkerjaan Di Kolej Vokasional Wilayah Selatan. Uneversiti Tun Hussein Onn. Retrieved from http://eprints.uthm.edu.my/7117/1/SUHAILI_BINTI_HANAFI_24.pdf
48. United Nations General Assembly. (2015). Transforming our world: The 2030 agenda for sustainable development.
https://sustainabledevelopment.un.org/content/documents/7891Transforming%20Our%20World. pdf. https://doi.org/10.1007/s13398-014-0173-7.2
49. Urbach, N., & Ahlemann, F. (2010). Structural Equation Modeling in Information Systems Research Using Partial Least Squares Structural Equation Modeling in Information Systems Research Using Partial Least Squares. Journal of Information Technology Theory and Application, 11(2), 5–40. 50. Vinzi, V. E., W.W.Chin, J.Henseler, & H.Wang. (2010). Springer Handbooks of Computational
Statistics Series Editors. Retrieved from http://www.springer.com/series/7286
51. Yunjuan He, & Xizhen Qin. (2017). Students’ Perceptions of an Internship Experience in China: A Pilot Study. Foreign Language Annals, 50(1), 57–70. https://doi.org/10.1111/flan.12246
52. Zaliza Hanapi, & Arasinah Kamis. (2017). Analisis Perbandingan di Antara Industri, Pensyarah dan Graduan Terhadap Kepentingan Kemahiran Teknikal dan Kemahiran Employability yang Perlu Dikuasai oleh Graduan Bidang Elektrik di Kolej Komuniti. Sains Humanika, 9(1–5), 67–75.
53. Zaliza Hanapi, Mohd Safarin Nordin, & Ridzwan Che Rus. (2015). Unemployment Problem among Graduates of Technical Field : Competencies of the Graduates and Quality of the Education. Sains Humanika, 2(2014), 53–57.
