Unemployment Among Malaysian Graduates

Malaysian Journal of Economic Studies 47of Unemployment: the Case of Malaysian Graduates ISSN 1511-4554 Estimating Psychological Impact (1): 33-53, 2010

Estimating Psychological Impact of Unemployment: the Case of Malaysian Graduates
Hock-Eam Lim* Universiti Utara Malaysia
Abstract: The objective of this paper is to estimate the psychological impact of unemployment for a group of 240 Malaysian graduates during their transition from university to labour market. There is evidence of negative psychological impact of unemployment. Results also reveal that treating employment or unemployment as a homogenous state is subject to state aggregation bias. Keywords: Aggregation bias, graduate unemployment, happiness, psychological impact of unemployment JEL classification: J64; Z19

1. Introduction
During the past one decade, despite some disagreements on validity, reliability and comparability of happiness measurement, we have witnessed a growing literature on happiness in economic studies. Ng (1997) suggested happiness is the ultimate objective for most people, if not all. Various determinants of happiness have been identified in the literature. For instance, it is found that income, employment status, age, and marital status are significant determinants (Clark and Oswald 1994; Winkelmann and Winkelmann 1998; Easterlin 2001; Blanchflower and Oswald 2004). One of the most consistent findings in happiness studies is the negative psychological impact of unemployment. This finding is of particular importance because it highlights the cost of unemployment to be much larger due to this non pecuniary cost, in addition to the pecuniary cost. The negative psychological impact of unemployment is found to be greater than some life-change events such as divorce or marital separation (Clark and Oswald 1994), and having bad health (Winkelmann and Winkelmann 1998). Winkelmann and Winkelmann (1998) segregated the cost of unemployment into a pecuniary cost (reduction in household income) and a non pecuniary cost (reduction in life happiness). They found that non pecuniary cost is larger than pecuniary cost. Using cross-section data on Malaysian graduates, Morshidi et al. (2004) observed that the mean scores of negative psychological attributes (such as being sad, feeling worried and thinking negatively) for unemployed graduates are higher than for employed graduates. Frey and Stutzer (2002) classified the happiness determinants into five categories: personality factors, socio-demographic factors, economics factors, contextual and situational factors,

*

College of Arts and Sciences (Economics), Universiti Utara Malaysia, 06010 Sintok, Kedah, Malaysia. Email: lheam@uum.edu.my I would like to thank the anonymous referee whose comments have improved this paper substantially. Thanks are also due to Dr Thi Lip-Sum for his valuable comments on this paper.

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

33

Hock-Eam Lim

and institutional factors. Employment status is one of the main determinants of happiness. Unemployment is suggested to have a negative impact on one’s happiness by Frey and Stutzer (2002). Thus, the negative psychological impact of unemployment is clearly established in the literature. The questions that follow are: What is the psychological impact of unemployment for fresh graduates who are in their transition from university to labour market? Will gaining employment improve one’s life happiness regardless of types of employment? Is there any aggregation bias on estimating the effects of employment status towards happiness? These are the research questions that the present study will attempt to examine. In evaluating the psychological impact of unemployment, binary aggregation of employment status into ‘unemployed’ against ‘employed’ is subject to aggregation bias. It is possible that employment status at a disaggregated level has a different psychological impact. For instance, for those who are economically inactive (those who withdrew from the labour force due to disappointment or discouraged worker effect), the psychological impact might differ from being unemployed. To quote Dockery (2003: 1), “…it is dangerous to treat ‘employment’ as a homogenous, alternative state to unemployment.” Hence, the psychological impact of unemployment ought to be evaluated at a disaggregated level of employment status, such as unemployed, economically inactive, part-time employment, self-employment, and full-time employment that commensurate or does not commensurate with qualification. Indeed, it is imperative to compare the psychological impact of the different employment status. The evaluation that is based only on binary aggregation of ‘unemployed’ and ‘employed’, is subjected to aggregation bias (Edin 1989; Lim 2007). Furthermore, the graduates know that upon completing their studies, they will enter into a phase of unemployment. Their expectation on the duration of unemployment might be different. For example, given two graduates with similar unemployment duration of 8 months, if the first and second graduate expect their unemployment duration to be 2 and 7 months respectively, the negative psychological impact for the first graduate is expected to be higher than the second graduate, ceteris paribus. Thus, expectation may play an important role in determining the psychological impact of unemployment. In addition, happiness is expected to decline with the increase in actual unemployment duration. Empirically, this negative duration dependency is substantiated by past findings (for example, Clark and Osward 1994; Lucas et al. 2004)). In short, the graduates’ observed individual heterogeneities including self-expected and actual unemployment duration, and use of disaggregated employment status, are important considerations in estimating the psychological impact of unemployment. Morshidi et al. (2004) appears to have carried out the only study focusing on the psychological impact of unemployment for Malaysian graduates. However, their binary aggregation of employment status (employed versus unemployed) has subject their findings to aggregation bias. This paper consists of four sections. Section 1 which contains the introduction includes a brief literature review on happiness and aggregation problem. Section 2 presents the data and methodology. The analysis and finding are discussed in Sections 3 and 4. The final section concludes the findings of this paper.

34

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

Estimating Psychological Impact of Unemployment: the Case of Malaysian Graduates

2. Data and Methodology
2.1 Data The present study used panel data that comprise 240 respondents from two surveys. The first survey was implemented from July 2005 to March 2006 targeting final year students from Universiti Utara Malaysia (UUM) and Universiti Tunku Abdul Rahman (UTAR). A total of 430 responses (304 from UUM and 126 from UTAR) were collected. Targeting these 430 respondents, the second survey was implemented from November 2006 to February 2007 which obtained 240 returned questionnaires. The overall life happiness measured during the second survey using one question that asked, ‘In general, how happy are you at present with your life as a whole?’ It was followed by a Likert-like rating scale ranging from ‘1’ being very unhappy to ‘7’ being very happy. This is a typical measurement of life happiness adopted in previous literature (Lim 2008). 2.2 Methodology Following the latent variable framework of Blanchflower and Oswald (2004) which assumed that for each graduate, there is a latent variable which represents his or her underlying happiness. This latent variable is associated with individual characteristics of the graduate which are obtained at first and second surveys (Xi). Let Y* represents this latent variable and assume that Y* is a linear function of Xi , thus Yi* =β Xi + ui where Yi* = underlying change in happiness (unobservable) X = independent variables (first and second survey) The model assumes that the observed happiness (Y) is related to the Y* (which is unobservable) and also the six boundary parameters (or cut-off points), μj, where j=1,2,…,6 and μ1 < μ2 90). For those who are unemployed for less than 61 days, their chances of having a happier life are not
3

Due to low number of observations, we are not able to disaggregate the duration into more disaggregate dummies such as 91-120 days (only 7 observations), 121-150 days (12 observations), and 151-180 days (only 3 observations); 181-210 days (only 2 observations) and above 210 days (only 7 observations).

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

43

Hock-Eam Lim

Figure 2: Impact of employment status on happiness

significantly different from those in FT1. This result clearly illustrates the aggregation bias of treating unemployment as a homogenous state. Relating to FT2 and SEPT, those employed with FT2 and SEPT are not insignificantly different (in terms of their chances of having a happier life) than FT1. This finding is consistent with the finding in Table 4 (insignificant FT2 and SEPT). To gain further insights, the influence of employment status on life happiness is predicted and plotted. These predictions are made by holding the other variables at their mean values respectively. 4.4 Predicted Probabilities of Happiness Figure 2 presents the influence of employment status on a graduate’s life happiness. Since the mid-point of the 7-point rating scale is 4 (Prob4), which is labelled as ‘neither happy nor unhappy’, the probability of obtaining point 1 to 3 (Prob1-3) can be interpreted as ‘probability of being unhappy’. Whereas, probability of obtaining point 5 to 7 (Prob5-7) is interpreted as ‘probability of being happy’. From Figure 2, those who are unemployed have the highest probability of being unhappy. The probabilities are 35.06, 32.88, 23.1, and 19.54 per cent for being unemployed for above 90 days, 61-90 days, 31-60 days and below 31 days respectively. Then, it is followed by those who are employed with FT2 (18.85%), SEPT (17.15%) and FT1 (11.49%). In terms of probability of being happy (Prob5-7), the unemployed graduates have the lowest probability. The probabilities are 35.4, 33.2, 47.19, and 52.5 per cent for being unemployed for above 90 days, 61-90 days, 31-60 days and below 31 days respectively. Then, it is followed by those who are employed with SEPT (53.62%), FT2 (56.46%) and FT1 (67.69%).

44

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

Estimating Psychological Impact of Unemployment: the Case of Malaysian Graduates

Figure 3: Impact of disaggregate unemployment states on probability of being ‘happy’ and ‘unhappy’

Clearly, those who are unemployed for 61 days and above have a significantly lower (higher) probability of being happy (unhappy) than others including those who are unemployed for below 61 days. To examine the effect of unemployment at disaggregated level, the effect of the four unemployment states (by unemployment duration: Below 31 days, 31-60 days, 61-90 days and above 90 days) on graduate’s life happiness is presented (Figure 3). For simplicity of presentation, Prob 5 to 7 are combined as ‘Happy’ and Prob1 to 3 are combined as ‘Unhappy’. In general, the influence of this unemployment duration on the graduate’s life happiness is negative. Figure 3 reveals that increasing duration of unemployment decreases (increases) the probability of being happy (unhappy). Specifically, during the 1 st–60 th day of unemployment, the probability of being happy is substantially higher than the probability of being unhappy. Then, after 60 days of being unemployed and onwards, the probability of being unhappy is approximately equal to the probability of being happy. This indicates that unemployment duration of below two months is not harmful psychologically (in terms of one’s probability of having a happier life). Thus, the effect of unemployment on one’s life happiness varies across different durations of unemployment.

5. Discussion and Conclusion
Descriptive analysis shows that the graduates’ life happiness decreases over the duration of unemployment. Nevertheless, during the first 120 days of being unemployed, the graduates still reported as being “happy” in their overall life happiness. Results of estimated ordered logit model reveal no significant difference in happiness between those who are FT1 employed and those who are unemployed below 61 days. In addition, for those who are unemployed below 61 days, the predicted probability of being happy is found to be substantially higher than the probability of being unhappy.

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

45

Hock-Eam Lim

Thus, the first 60 days of being unemployed brings no harmful impact on happiness. This finding suggests that the duration of ‘frictional’ unemployment is two months (in terms of one’s life happiness) for graduates in Malaysia. It is suggested that existing government programmes to assist unemployed graduate such as re-training courses should focus only on graduates who have at least endured more than two months of unemployment. There is further evidence of negative psychological impact of unemployment. The statistical evidence also illustrates this negative impact varies according to quality of employment. In terms of happiness, FT1 employment is significantly different from those who are unemployed. This highlights the importance of disaggregating the state of ‘being employed’ ranging from less-quality jobs to good-quality jobs, instead of treating employment as a homogenous state. It is suggested that government statistics indicating graduate employment should not aggregate the employed into one homogenous state. Disaggregated statistics on employment status are needed to provide insights and better understanding of graduate unemployment in Malaysia. In addition, unemployment cannot be treated as a homogenous state. In terms of life happiness, effect of unemployment varies across different levels of unemployment duration. State aggregation bias is not only applied to employment (due to quality of employment obtained), it is also applied to unemployment (due to different durations of unemployment). Hence, it is further suggested that the government statistics of graduate unemployment should disaggregate the unemployed into different states based on unemployment duration. However, there are some caveats to the findings of this paper. First, potential endogeneity bias between happiness and employment outcomes cannot be ignored. Nevertheless, due to data limitation, this endogeneity problem cannot be examined in the present paper. Second, the data collected were limited to only two universities in Malaysia. It is suggested that future research include more universities in Malaysia and also investigate this potential endogeneity bias.

References
Blanchflower, G.D. and J.A. Oswald. 2004. Well-being over time in Britain and USA. Journal of Public Economics 88: 1359-1386. Burkam, T. David and E. Valerie Lee. 1998. Effects of monotone and non monotone attrition on parameter estimates in regression models with educational data: demographic effects on achievement, aspirations and attitudes. Journal of Human Resources 33(2): 555-574. Clark, E.A and J.A. Oswald. 1994. Unhappiness and unemployment. Economic Journal 104: 648659. Carroll, N. 2005. Unemployment and Psychological Well-being. Dsicussion Paper No. 492, Centre for Economic Policy Research, Australian National University. Dockery, A.M. 2003. Happiness, Life Satisfaction and the Role of Work: Evidence from Two Australian Surveys. Working Paper 03.10, Curtin Business School, Curtin University of Technology. Dockery, A.M. 2005. The happiness of young Australians: empirical evidence on the role of labour market experience. Economic Record 81(255): 322-335. Easterlin, R. 2001. Income and happiness: towards a unified theory. Economic Journal 111: 465484. Edin, Per-Anders. 1989. Unemployment duration and competing risks: evidence from Sweden. Scandinavian Journal of Economics 91(4): 639-653.

46

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

Estimating Psychological Impact of Unemployment: the Case of Malaysian Graduates

Frey, S.B. and Alois, Stutzer. 2002. Happiness & Economics. USA: Princeton University Press. Jahoda, M. 1982. Employment and Unemployment: a Social-Psychological Analysis. Cambridge: Cambridge University Press. Lim, H.E. 2007. Estimating the employment performance indicator: the case of Universiti Utara Malaysia graduates. Singapore Economic Review 52(1): 73-91. Lim, H.E. 2008. The use of different happiness rating scales: bias and comparison problem? Social Indicators Research 87: 259-267. Long, J.Scott. 1997. Regression Models for Categorical and Limited Dependent Variables. California: Sage Publications. Lucas, E.R., E.A. Clark, Y. Georgellis and E.D. Diener. 2004. Unemployment alters the set point for life satisfaction. Psychological Science 15(1): 8-13. Morshidi Sirat, Abd. Aziz Buang, Abd Majid Mohd Isa, Ambigapathy Pandian, Moha Asri Abdullah, Mohamed Dahlan Ibrahim, Mohd Haflah Piei, Molly N.N. Lee, Munir Shuib, Rosni Bakar, Rujhan Mustafa, Shukran Abdul Rahman, Siti Zubaidah A. Hamid, Susie See Ching Mey and Wan Ahmad Kamil Mahmood. 2004. Masalah Pengangguran di Kalangan Siswazah. USM IPPTN Monograf 2/2004. Penang. Ng, Y-K. 1997. A case for happiness, cardinalism & interpersonal comparability. Economic Journal 107: 1848-1858. Winkelmann, L. and R. Winkelmann, 1998. Why are the unemployed so unhappy? Evidence from panel data. Economica 65: 1-15.

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

47

Hock-Eam Lim

Appendix 1. Definition and measurement of variables Variable abbreviation Employment status Full-time employment commensurate with qualification Full-time employment not commensurate with qualification Self-employment/part-time employment Definition Dummy variable for full-time employment commensurate with qualication (comparison group: unemployed) Dummy variable for full-time employment not commensurate with qualication (comparison group: unemployed) Dummy variable for self-employed or parttime employment (comparison group: unemployed) Self-reported (number of weeks) Number of days unemployed Interaction variable between EXPUNE and UNEDUR Self-perceived (ordinal scale: 1 ‘low’ to 7 ‘high’) Financial difficultes faced while unemployed (ordinal scale: 0 ‘no’ to 6 ‘high’) Dummy variable for Buddhist (comparison group: Islam) Dummy variable for Christian/Catholist (comparison group: Islam) Dummy variable for Hindu/Taoism/others (comparison group: Islam) Dummy variable for UUM Public Mgt and Development Mgt (comparison group: UUM Economics) Dummy variable for UUM Business Admin (UBBA) (comparison group: UUM Economics) Dummy variable for UUM Accounting (UBACC) (comparison group: UUM Economics) Dummy variable for UUM Info Tech (UBIT) (comparison group: UUM Economics) Dummy variable for UUM Others degree: Tourism / Education / Technology Mgt / Decicision Sciences (comparison group: UUM Economics)

Job search related Self-expected unemployment duration (EXPUNE) Unemployment duration (UNEDUR) Interaction between EXPUNE and UNEDUR Self-perceived marketability of degree studied Financial difficulties faced Religion Buddhism Christianity/Catholic Other Religions Types of degree UUM Public/Development Management

UUM Business Administration

UUM Accounting

UUM IT UUM Other degrees

48

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

Estimating Psychological Impact of Unemployment: the Case of Malaysian Graduates

UUM Human Resource/Social Work

Dummy variable for UUM Human Resource / Social Work Mgt (comparison group: UUM Economics) Dummy variable for UUM International Bussiness / Issues Mgt (comparison group: UUM Economics) Dummy variable for UUM Finance / Banking (comparison group: UUM Economics) Dummy variable for UUM Communication (comparison group: UUM Economics) Dummy variable for UTAR Business Admin (comparison group: UUM Economics) Dummy variable for UTAR Accounting (TBACCT) (comparison group: UUM Economics) Dummy variable for UTAR Info System / Info System Engineering / Computer Sciences (comparison group: UUM Economics) Dummy variable for UTAR other degrees: Chinese Studies / Journalism / Public Relations (comparison group: UUM Economics) 1=no formal schooling; 2=do not complete primary; 3=complete primary; 4=do not complete secondary; 5=complete secondary;6=O level or equ; 7=A level & above Number of persons in family Self-perceived proficiency of English (Ordinal scale: 0 ‘non user’ to 12 ‘expert-user’ Cumulative Grade Point Average age in years Dummy variable for being male (comparison group: female) Self-reported health condition (ordinal scale: 0 ‘poor’ to 6 ‘excellent’. Dummy variable for home town in rural (other than big cities or state capital) Dummy variable for having a car driving license The boundary parameters

UUM International Business/Issues Mgt

UUM Finance UUM Communication UTAR Business Administration UTAR Accounting

UTAR IT/Computer Sciences

UTAR Other degrees

Family background Father’s education level

Family size English and academic related English language proficiency level Academic attainment Social-demographic related Age Male Health Home town: rural Car driving license Cut off-points: μ1 - μ6

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

49

50

Appendix 2. Comparison of ordered logit and ordered probit model 3
Ordered logit model Coefficient ystandardised coefficient P-value Coefficient ystandardised coefficient P-value Ordered probit model

Variables

0.9059 0.3560 0.4831 0.2208 0.4960 0.1862 0.1627 0.5130 0.2164 0.1796 0.1546

0.4141

0.0820

0.5091

0.4225

0.0520 0.4110 0.6190
Hock-Eam Lim

Employment status2a Full-time employment commensurate with qualification (FT1) Full-time employment not commensurate with qualification (FT2) Self-employment/ part-time employment (SEPT) -0.0565 -0.0124 0.0025 -0.2317 -0.2757 -1.0159 0.4752 -1.3081 1.2533 1.8471 1.5695 1.2641 1.4579 0.7192 0.9822 0.5729 0.8444 0.7174 0.5779 0.6665 0.3288 0.4490 -0.4644 0.2172 -0.5980 0.0880 0.6750 0.0450 0.1000 0.0070 0.0420 0.0900 0.1190 0.4600 0.2280 -0.0258 -0.0057 0.0011 -0.1059 -0.1261 0.6950 0.0390 0.1670 0.2200 0.0190 -0.0310 -0.0065 0.0014 -0.0845 -0.1392 -0.4849 0.3438 -0.7020 0.6027 0.9666 0.8178 0.6380 0.6940 0.2907 0.4630 -0.0257 -0.0054 0.0012 -0.0701 -0.1155 -0.4025 0.2853 -0.5826 0.5002 0.8023 0.6788 0.5295 0.5760 0.2413 0.3843

Job search related Self-expected unemployment duration (EXPUNE) Unemployment duration (UNEDUR) Interaction between EXPUNE and UNEDUR Self-perceived marketability of degree studied Financial difficulties faced

0.6880 0.0180 0.1040 0.3590 0.0200 0.0990 0.5150 0.0430 0.1600 0.0140 0.0610 0.1220 0.1650 0.5920 0.2940

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

Religion2b Buddhism Christianity/Catholic Other religions

Types of degree2c UUM Public/ Development Management UUM Business Administartion UUM Accounting UUM IT UUM Other degrees UUM Human Resource/ Social Work UUM International Business/ Issues Management

UUM Finance UUM Communication UTAR Business Administration UTAR Accounting UTAR IT/ Computer Sciences UTAR Other degrees 0.1361 0.0526 0.1779 0.4013 0.1828 0.3036 0.0241 0.4910 0.6751 0.0836 0.1388 0.0110 0.2245 0.3086 0.0810 0.4340 0.8960 0.1430 0.0900 0.0932 0.2226 -0.0181 0.2637 0.4063 0.1834 0.5960 0.0660 0.0813 0.1210 0.1098 0.0912 0.0547 0.0774 0.1848 -0.0150 0.2189 0.3372 0.0622 0.0240 0.1570 0.6360 0.0838 0.0223 0.0695 0.0185 0.1090 0.6980 0.0610 0.8620 0.1210 0.2860 0.8530 0.1340 0.0570

0.5271 0.1964 1.5991 1.6849 1.6231 1.7630

0.2409 0.0898 0.7310 0.7702 0.7420 0.8059

0.5200 0.7870 0.0680 0.0560 0.0210 0.0480

0.1889 0.0790 0.7669 0.7834 0.8525 0.8479

0.1568 0.0655 0.6365 0.6502 0.7076 0.7037

0.6710 0.8550 0.1190 0.1010 0.0590 0.0870

Family background Father’s education level Family size

English and academic related English language proficiency level Academic attainment

Estimating Psychological Impact of Unemployment: the Case of Malaysian Graduates

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

Socio-demographic related Age Male Health Home town: rural Car driving license

Notes: 1. Explanation and measurement of variables are presented in Appendix 1. 2. Comparison group of dummy variables of: a. employment status: unemployed b. religion: Islam c. type of degree: UUM Economics 3. Due to different assumption on the value of variance between the logistic and normal distribution, the estimated coefficients are not directly comparable. However, one may compare the standardised coefficients as suggested by Long (1997: 128-129).

51

Hock-Eam Lim

Appendix 3. Ordered logit model: use FT1 as comparison category Variables Employment status3a Unemployed (UNE) Full-time employment not commensurate with qualification (FT2) Self-employment/part-time employment (SEPT) Job search related Self-expected unemployment duration (EXPUNE) Unemployment duration (UNEDUR) Interaction between EXPUNE and UNEDUR Self-perceived marketability of degree studied Financial difficulties faced Religion3b Buddhism Christianity/Catholic Other religions Types of degree3c UUM Public/Development Management UUM Business Administartion UUM Accounting UUM IT UUM Other degrees UUM Human Resource/Social Work UUM International Business/Issues Management UUM Finance UUM Communication UTAR Business Administration UTAR Accounting UTAR IT/Computer Sciences UTAR Other degrees Family background Father’s education level Family size English and academic related English language proficiency level Academic attainment Odds ratio Std error

0.4042 0.5770 0.6552 0.9451 0.9876 1.0025 0.7932 0.7590 0.3621 1.6083 0.2703 3.5020 6.3412 4.8041 3.5400 4.2971 2.0528 2.6704 1.6939 1.2170 4.9483 5.3921 5.0689 5.8297 1.1458 1.0540 1.1947 1.4938

0.2102* 0.2640 0.4532 0.1363 0.0060** 0.0018 0.1499 0.0893** 0.2156* 1.8216 0.1767** 2.6658* 4.3326*** 3.7154** 2.6381* 4.0224 1.9984 2.1765 1.3873 0.8845 4.3404* 4.7526* 3.5708** 5.2089** 0.1102 0.1172 0.1370 1.1307

52

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

Estimating Psychological Impact of Unemployment: the Case of Malaysian Graduates

Socio-demographic related Age Male Health Home town: rural Car driving license

1.2006 1.3547 1.0244 1.6340 1.9642

0.1256* 0.5252 0.1888 0.5475 0.7817*

Notes: 1. *, **, and *** represent significance at 10%, 5% and 1% levels, respectively. 2. Explanation and measurement of variables are presented in Appendix 1. 3. Comparison group of dummy variables of: a. employment status: Full-time employment that commensurate with qualification (FT1) b. religion: Islam c. types of degree: UUM Economics

Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010

53

Copyright of Malaysian Journal of Economic Studies is the property of Malaysian Economic Association & Faculty of Economics & Administration, University of Malay and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder 's express written permission. However, users may print, download, or email articles for individual use.

References: Malaysian Journal of Economic Studies Vol. 47 No. 1, 2010 47