ROLE AMBIGUITY AND JOB PERFORMANCE OF EMPLOYEES IN THE SERVICE SECTOR SMEs IN MALAYSIA

Small and medium size enterprises (SMEs) play a crucial role in the economic development of Malaysia, of which the majority are in the service sector. Employees of the service sector SMEs have often been associated with low level of job performance and past research has shown that there are many factors that can contribute to employee poor performance such as role ambiguity. Thus the aim of this study was to examine the relationship between role ambiguity and job performance of employees in the service sector SMEs in Malaysia. 1500 questionnaires were distributed and 300 were returned resulting in a 20% response rate. The result revealed that there was a significant relationship between role ambiguity and job performance of employees.


Introduction
Job performance of employees plays a crucial factor in determining an organisation's performance since highly performing individuals will be able to assist the organisation to achieve its strategic aims thus sustaining the organisation's competitive advantage (Lado & Wilson, 1994;Dessler, 2011).At the same time, knowledge of employee job performance will allow managers to make various kinds of decisions ranging from compensation, promotion, and training as a basis for performance improvement to the extent of recommending termination (Piercy, Cravens & Morgan, 1998).In addition, success in job performance is a gateway for reaping financial and nonfinancial rewards for employees (VanScotter, Motowidlo & Cross, 2000).Nevertheless, Galagan (1997) expressed that given the many challenges that employees have to face due to the dynamism of the workplace, it is rather difficult for employees to maintain their job performance.Moreover, employers tend to have high expectations concerning employee job performance by continuously monitoring the job performance of the employees through various performance management activities (Dessler, 2011).Thus, having a workforce that is well equipped with the right skills and is well prepared will ensure that business will not lose out due to the lack of ability to compete both nationally and internationally (Tomaka, 2001).SMEs in this context would not be alienated from similar circumstances.SMEs are regarded as critical especially when these businesses have been contributing to the growth and promoting the competitiveness of many nations (Caniel & Romijn, 2005).This primary economic contribution made by the SMEs to a country had given rise to the interest for researchers to examine the various obstacles that hinder their progress (Alasadi & Abdelrahim, 2008).

SMEs in Malaysia
SMEs in Malaysia are important in driving the Malaysian economy and its unemployment issues (Che Ros, Kumar & Lim, 2006).Sustainable SMEs would be able to help in economic growth (SME Annual Report 2007), which in turn helps in job creation and generation of income and is always regarded as the backbone of the Malaysian economy (Ramayah & Koay, 2002).While the SMEs sector plays an important role in Malaysia, it also faces many challenges.Although SMEs represent 99.2% of the total business establishments in Malaysia, the SMEs contribute only 32% to the GDP (SME Annual Report, 2007;Sin, 2010).This is lower than the average contribution in other Asian countries such as China and Japan in which the rate is over 50% (Ndubisi, 2008;Osman, Ho & Galang, 2011).Past studies about Malaysian SMEs have highlighted problems faced by Malaysian SMEs in general that have led to the Malaysian SMEs appearing to be less competitive (Saleh & Ndubisi, 2006) and as hindrances that prevent good performance (Moha, 1999;Hall, 2002;Stuti, 2005).In relation to this, skill shortages and productivity of employees have been highlighted as one of the on-going problems that dampens the progress of SMEs (SMIDEC, 2002;Wang, 2003;Ting, 2004;United Parcel Service, 2005;Saleh & Ndubisi, 2006) and found to be a serious problem in the majority of the SME firms surveyed in Malaysia (Yogeesvaran, 2005).Skills shortages is defined as the deficiency within the labour pool (Frogner, 2002) or deficiencies in the skills which employees need to carry out their existing tasks (Green, Machin & Wilkinson, 1998).
With regards to the service sector in Malaysia, the majority of service firms in Malaysia are made up of those from the SME.Thus the service sector SMEs forms the largest sector of the SME establishments with a total of 86.5% of the total SME establishment (Department of Statistics, 2006).It generally includes the services, primary agriculture and information, and communication Technology (ICT).There are 2.2 million people employed in the service sector SMEs as compared to the manufacturing sector (740,438) and the agriculture sector (131,130) (Aris, 2007).Even though the service sector SMEs has the highest employment, the labour productivity of the manufacturing sector overtook the service sector SMEs at RM64, 089 while only RM47, 151 was contributed by the employees from the service sector (SME Annual Report, 2007).This supported the notion that job performance of employees in the service sector SMEs tend to be low due to lack of right skills (Saleh & Ndubisi, 2006).Furthermore almost 72% of employment in the service sector SMEs were made up of those that possessed only Sijil Pelajaran Malaysia (SPM) qualification and below, which may affect the ability of the employees to deliver the expected standard of job performance due to lack of skills (Aris, 2007).
At the same time, employees working in a service setting have experienced role ambiguity (Wener, 1985;Singh & Rhoads, 1991;Knowles, Grove & Pickett, 1992;Price, Arnould & Tierney, 1995;Babin & Boles, 1998;Varca, 2002).In any service setting, employees who are able to understand and are clear about their roles in the organisation are likely to perform well in their jobs (Anderson, 2006).At the same time, past researches related to employees that work in a service setting (e.g.Murkherjee & Maholtra, 2006;Lang, Thomas, Bliese & Adler, 2007) have shown that role ambiguity influenced the employees' job performance.Furthermore, the outcome of past researches revealed either a negative, weak or no relationship between role ambiguities and job performance (e.g.Brief & Aldag, 1976;Jackson & Schuler, 1985;Michaels, Day & Joachimsthaler, 1987;Singh, 1993;Beauchamp, Bray, Eys & Carron, 2005;Hall, 2008).
Although past studies have shown that role ambiguity can affect the job performance of employees in the service setting, most of them were conducted abroad, thus very little evidence exists to understand the job performance of employees in the context of Malaysia.Furthermore, most SME researches in Malaysia focused on productivity and skill shortages (Yogeesvaran, 2005;Hamzah & Ho, 1994;Tan, 1996;Saleh & Ndubisi, 2006), entrepreneurial problems (Abdullah, Hamali, Deen, Saban & Abdurahman, 2009), TQM and organizational performances (Sohail & Hoong, 2003), empowerment (Wyer & Mason, 1999), ICT adoption among SMEs (Alam & Ahsan, 2007), SMEs historical development (Hashim, 2000), staff training and SMEs performance (Jamaludin & Hasun, 2007) and assessment criteria of banks towards small business borrowers (Mahmood & Rahman, 2007).Given such a situation, it necessitates this study to be carried out in view of the importance of the service sector SMEs in Malaysia.Therefore, the aim of this study is to investigate the relationship between role ambiguity and the job performance of employees working in the service sector SMEs in Malaysia.

Job Performance: Conceptual, Theoretical and Empirical Review
Job performance has always been regarded as an important factor in employee management and has commonly been defined as what a person does at work.Different stages of a job as well as the complexity of a job can affect the overall performance of the jobholder (Ackerman, 1997, Murphy, 1989).This could mean that job performance as a construct can be defined in different ways since it can be affected by the stage and complexity of the job (Grubb, 1999).It has been associated with the ability of the individual employees realizing their respective work goals, fulfilling expectations as well as attaining job targets and/or accomplishing standards that are set by their organizations (Eysenck, 1998;Mathis & Jackson, 2000;Bohlander, Snell & Sherman, 2001).The definitions of job performance are often unclear and hardly specific thus making them less useful (Campbell, Gasser & Oswald, 1996).Sarmiento and Beale (2007) referred to job performance as the result of two aspects, which consist of the abilities and skills (natural or acquired) of employees; and the motivation of employees to use their skills and abilities to perform a better job.
According to Campbell, McCloy, Oppler and Sager (1993, p. 40), performance is "what the organization hires one to do, and do well".Campbell et al. (1993) further noted that only action that can be measured could be regarded as performance.Further, action was explained as that which is controllable and that which will be geared towards the organization's goal (Campbell et al., 1993).Even though many attempted to introduce various frameworks of performance, Campbell's definition of performance has been acceptable as the basic definition for performance (Borman, Hanson & Hedge, 1997;Motowidlo, Borman & Schmit, 1997;Schmitt & Chan, 1998).Thus in synthesizing the above mentioned definitions, job performance is hereby defined as what an employee is expected to do in relation to job demand as when he or she is hired and covers only those actions or behaviours that are relevant to the organization's goals and measurable in terms of each individual's proficiency.Campbell's (1990) Theory of Performance makes clear distinctions between performance components, performance determinants, and the antecedents of performance determinants.Performance components refer to the performance dimensions that constitute various parts of the overall job performance.Campbell posited that the performance component is a function of three performance determinants which are the declarative knowledge, procedural and skills knowledge and motivation (Campbell, 1990;Campbell et al.,1993).These are the direct determinants of performance, which is the focus of this paper.In detail, declarative knowledge includes knowledge about facts, principles, goals and self-knowledge, which represent an understanding of the requirements of a given task.Procedural and skills knowledge includes cognitive skills, psychomotor skills, physical skills, self-management skills, and interpersonal skills.Motivation is the combined effect of three choice behaviours: the choice to perform, the level of effort, and the persistence of the effort (Campbell et al., 1993).In other words, in order to perform the behaviours in one of the dimensions, a person needs to know what to do, how to do it and possess the desire to do it.
Past researches on job performance found several factors that can influence the employees' job performance ranging from individual-related factors, organizationallevel factors as well as organizationalenvironmental factors.For instance, employees' commitment has been found to affect job performance (Jaramilloa, Mulki & Marshal, 2005;Al-Ahmadi, 2009).Besides that, job satisfaction was also found to have a significant relationship with employees' job performance (Gu & Chi, 2009).In an attempt to determine the factors that influence the perceived job performance among the shopfloor employees, Sarmiento and Beale (2007) found no significant association between age and education among the employees while at the same time, education and job performance were found to have a negative relationship and this was further supported in a study by A-Ahmadi (2009).On the contrary the study by Ng and Feldman (2009), revealed differing results in which education was found to be positively related to employee job performance.Employees' self-efficacy, competitiveness and effort were also found to have a significant positive effect on the job performance of employees serving as front liners (Karatepe, Uludag, Menevis, Hadzimehmedagic & Baddar, 2006).Selfefficacy and psychological climate were also found to be positively associated with the job performance of hospital employees in a study by D'Amato and Zijlstra (2008).Additionally, Knight, Kim and Crutsinger (2007), in their attempt to examine the causal relation between role stress, customer orientation, selling orientation, and job performance of retail salespeople, revealed that role conflict and role ambiguity affected customer orientation while affecting job performance when mediated by customer orientation.

Role Ambiguity: Conceptual, Theoretical and Empirical Review
Role ambiguity has been the study interest of many past researches related to human performance.Jackson and Schuler (1985) stressed the importance of having a clear understanding of one's role from an individual perspective as it is said to have an influence on one's motivation, satisfaction and performance.Furthermore role ambiguity can have various types of effects ranging from less psychological stress (Lang, Thomas, Bliese & Adler, 2007), greater interests, innovation, self-actualization, autonomy, self-esteem, less tension, less physical stress and lower intention to leave (Ivancevich & Donelly, 1974) and greater satisfaction (Busch & Bush, 1978).
There were many attempts to delineate the meaning of role ambiguity.Ilgen and Hollenback (1991) define roles as the pattern of behaviours that are expected or required by the members of an organization.Kalbers and Cenken (2008) relate role ambiguity to the lack of confidence that an employee perceives of his or her responsibility.In another perspective, role ambiguity is said to occur when an insufficient amount of information is given to an individual to perform a role and is commonly seen as a condition when disagreement happens in the work of the community with little understanding on the employee's side of what are expected of them (Fisher & Gitelson, 1983;Jackson & Schuler, 1985).
In addition, role ambiguity is also defined as the occurrence of insufficient information pertaining to powers, authority and duties to perform one's role (Kahn, Wolfe, Snoek & Rosenthal, 1964).Perhaps one of the famous definitions of role ambiguity after the work of Kahn et al. (1964) was given by Rizzo, House and Lirtzman (1970).They mentioned that role ambiguity occurred when one is not equipped with a good understanding of the job responsibilities and having little knowledge of what is expected in relation to job performance.On the other hand, role clarity occurs when an employee is clear about behavioural expectations, which assist in giving the necessary knowledge of what is considered as acceptable behaviour (Rizzo et al., 1970).Since the opposite end of role ambiguity is role clarity, one can only experience one or the other when work is performed (Rizzo et al., 1970).
In view of the potential effect of role ambiguity towards employee job performance, it had received a prominent position in many empirical researches (King & King, 1990;Tubre & Collins, 2000;Ortqvist & Wincent, 2006).For instance, role ambiguity had been found to decrease job performance, satisfaction and commitment (Chang & Chang, 2007).In addition, role ambiguity had also been linked to educational factors in a number of studies (e.g.Thompson, McNamara & Hoyle, 1997;Wolverton, Wolverton & Gmelch, 1999;Koustelios, Theodorakis & Goulimaris, 2004).Thus, the following section will review the previous studies specifically on the relationship between role ambiguity and the job performance of employees.

Exploring Role Ambiguity and Job Performance Link
Reviews of past studies found limited empirical evidence between role ambiguity and performance (Singh, 1993).A metaanalysis based on the work of Jackson and Schuler (1985) by Tubre and Collins (2000) found that in order for an individual to carry out a task effectively, sufficient information is imperative.This is because when there is a lack of information regarding what is to be achieved, and the most effective work behaviour that can help the top management to achieve that will result in ineffective work performance.Although most research has found negative relationships between role ambiguity and job performance, the strength of association between role ambiguity and job performance varies widely according to the types of occupation and performance measure (Jackson & Schuler, 1985).Abramis (1994) provided evidence that role ambiguity resulted in the reduction of work performance.Besides that, research on job performance of the front-line service employees by Singh and Rhoads (1991) found that there were several types of role ambiguity such as those with regards for their superiors, the company, ethical issues, customers, co-workers, family and other managers.Knight, Kim and Crutsinger (2007), in their attempt to examine the causal relationship between role stress, customer orientation, selling orientation, and job performance of retail salespeople, revealed that role conflict and role ambiguity affected customer orientation while affecting job performance when mediated by customer orientation.
In addition, Behrman, Bigoness and Perreault (1981) revealed that there was a positive relationship between job performance and ambiguity concerning family expectation.On the other hand, the same study revealed that ambiguity regarding sales manager and customer expectations was negatively related to job performance.At the same time, the study also discovered that when there existed ambiguous managerial expectations a lower level of satisfaction was recorded.A similar effect was found by Kahn et al. (1964) as cited by Walker, Churchill and Ford (1975).When an employee experienced so much uncertainty about what was expected in performing a job, a high level of anxiety and tension would develop which in turn reduced job satisfaction and this could affect job performance.
Dubinsky, Michaels, Kotabe, Chae and Hee-Cheol (1992) carried out a study to compare whether role stressors such as role ambiguity and role conflict can influence work outcome among salespeople in the US, Japan and Korea.The findings revealed that role ambiguity was significantly negatively related to the job performance of the employees and there was no difference in the magnitude of the coefficients when comparing among the three sample nations.Based on those studies that found role ambiguity to be related to job performance, the relationship was found to be significantly negative (Kahn et al., 1964;House & Rizzo, 1972;Zeithaml, Berry & Parasuraman, 1988).

Gaps in the Literature
The existing empirical researches provided little support for the expected adverse effects of role ambiguity towards job performance.Though role ambiguity was found to have a negative relationship with job performance (Behrman & Perreault, 1984;Lysonski & Johnson, 1983) , Jackson and Schuler's (1985) meta-analytic studies found that the effect of role ambiguity on job performance was rather weak and this was further supported by similar findings by Fisher and Gitelson (1983) and Berkowitz (1980).In addition, studies by Bagozzi (1978), Hampton, Dubinsky andSkinner (1986), andSzilagyi (1977) found no association between role ambiguity and job performance, which contradicted the result found by Bagozzi (1980) where role ambiguity was found to affect the job performance of employees.Past researches on the relationship between role ambiguity and job performance seem to develop mixed results.Since some studies found association while others found no association between role ambiguity and job performance (Michaels, Day & Joachimsthaler, 1987;Wetzels, Ruyter & Bloemer, 2000), a gap was created to re-examine the role of ambiguity in the context of the service sector SMEs in Malaysia.Therefore, this study hypothesizes that there is a significant relationship between role ambiguity and the job performance of employees.

Methodology
A mail survey was carried out to collect data from employees who are currently working the service sector SMEs who form the unit of analysis for this research.In order to determine the population of this study, reference was made to a sampling frame that was obtained from the SME business directory (SME Business Directory, 2011).The total number of service SMEs available in Malaysia is 5,527 as shown in Table1.Total elements for the entire population based on the sampling frame amounted to 27635 employees (non-owner/manager) which is based on the definition given by the central bank of Malaysia, in which a service SME will have a minimum of 5 to a maximum of 50 employees (Bank Negara Malaysia, 2005).Thus, the minimum required sample size as suggested by Krejcie and Morgan (1970), with a population of N = 27635 is 379 employees.A systematic random sampling was used to select 1,500 respondents (based on the response rate of 25% for mail survey in Malaysia ( Ismail & King, 2005) in order to obtain the required minimum sample size of 379.Under the systematic random sampling technique, a sample is chosen by selecting a random starting point and then picking every K th element in succession from the sampling frame (where K is the sampling interval).300 SMEs in total were chosen under this technique (1500 ÷ 5 employees).For this study, every 18 th firm was automatically selected from the list in the sampling.For example, the sample included the 18 th name, the 36 th , 54 th , 72 nd and so on and so forth.Thus, 300 firms (1,500 employees) were finally selected from the list of 5,527 SMEs.
Each variable was measured using a previously developed instrument with a 7-point Likert scale for all the measurements used ranging from (1) -strongly disagree to (7)-strongly agree.Job performance was adopted from William and Anderson (1991) with 6 items with reported reliability of .91 while role ambiguity was adopted from Rizzo et al. (1970) with 6 items with a reported reliability of between .78 to .81,thus meeting the threshold value of Cronbach's alpha of .70 (Nunnally, 1978).In addition, 7 items were on demographic questions, which used ordinal and nominal scales on gender, age, ethnic, qualification, occupational sector, working experience and tenure of service.Pre-test of the measurements was conducted among 30 respondents working in the service sector SMEs in order to determine the reliability of the adopted measurement.Upon completion of the pre-test, the Cronbach's alpha value for job performance (after deletion of 2 items) was .87 and .74for role ambiguity.

Findings
Out of the 1,500 questionnaires distributed and after the screening process, 300 returned survey were usable for further analysis resulting in a 20% response rate.Since the data collected was less than the actual minimum required sample size of 379, a nonresponse bias test was conducted.This test is to indentify whether there is a possibility that those who did not respond would have answered differently.Thus a non-response bias test was carried out on the 300 usable responses in which 180 responses that were obtained first were regarded as the first wave responses while the remaining 120 late respondents who formed the second wave responses were treated as non-respondents as according to the assumption by Armstrong and Overton (1977).Based on the t-test for equality of means, at p value < .05,no significant difference was found between those responses that were obtained from the first wave and those that were obtained from the second wave.Further data screening was carried out.Through the outlier test, the Mahalanobis D 2 scores suggested five respondents treated as outliers be ignored.The variables were found to be normally distributed and hence 295 respondents were valid to be used for further analysis.
Table 2 shows the distribution of the respondents according to their profiles in Most of them were from education (14.9%), restaurant (14.6%) and professional services (14.2%).Only 11.2% of the respondents had worked for less than 1 year.The majority of the respondents had finished their tertiary education: diploma (21.7%), bachelor degree (30.5%) and postgraduate (9.8%).(continued) Factor analysis was utilized to test the construct validity of the measurements for this study.Suitability of using factor analysis as a tool for testing construct validity was determined through Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and the Barttlet's Test of Sphericity.Therefore, if the KMO values are greater than .60(Coakes, Steed & Ong, 2009), and the Bartlett's test of sphericity is large and significant (p<.05) (Hair, Black, Babin, Anderson & Tatham, 2006), factorability is then considered as possible.Once factor analysis is carried out, items with factor loading that is greater than .30will be used to represent a factor since it is regarded as the threshold to meet the minimal level for interpretation of structure (Hair et al., 2006).Thus, factor analysis was carried out on all the variables used in this study.
Table 3 presents the result of factor analysis for job performance.Items were chosen to identify a factor with loadings greater than .30based on the guideline by Hair et al. (2006).
According to Kline (1994), factor loading that is greater than .60 can be considered as high while any factor loading that is greater than 0.3 is regarded as moderately high.The factor loading were all found to be greater than 0.6 indicating good correlation between the items and the factor grouping they belonged to.Thus, job performance had all four questions loaded onto a single factor with eigenvalue of more than 1.0.The single factor extracted 71.13 % of the total variance in response.At the same time, a scree plot also proposed a single factor solution (see Figure 1).
Role ambiguity had all six questions loaded onto a single factor with eigenvalue of more than 1.0.The single factor extracted 61.84 % of the total variance in response.The result is shown in Table 3.The scree plot for role ambiguity also suggested a single factor solution (see Figure 2).Bougie, 2009).The correlation coefficient (r) indicates the strength of the association between any two metric variables (Hair et al., 2006).The sign (+ or -) indicates the direction of the relationship with a value that ranges between +1.0 to -1.0.The value +1.0 indicates a perfect positive relationship, 0 indicates no relationship and -1.0 indicates a perfect negative or reverse relationship (Hair et al., 2006).Thus in testing the hypotheses, null hypotheses will be rejected if the significance p value is less than .05which indirectly means the alternate hypotheses will be accepted, while the opposite will take place should the significance p value be greater than .05.It is concluded that there is a significant relationship between job performance and role ambiguity (r=-0.686,p<0.01).According to Davis (1971), this relationship is strong when the 'r' ranges from 0.50 to 0.69.Thus, the hypothesis of this study is strongly supported.
13 Malaysian Management Journal Vol. 15, 1-20 (2011) Table 4 presents the inter-correlations among all the variables in this study.The level of significance was at 5% (.05) which is a generally accepted conventional level in social sciences research (Sekaran & Bougie, 2009).The correlation coefficient (r) indicates the strength of the association between any two metric variables (Hair et al., 2006).The sign (+ or -) indicates the direction of the relationship with a value that ranges between +1.0 to -1.0.The value +1.0 indicates a perfect positive relationship, 0 indicates no relationship and -1.0 indicates a perfect negative or reverse relationship (Hair Conclusions This study contributes to the present literature on SMEs in Malaysia specifically in the context of service sector SMEs, which have not received much attention.The knowledge obtained can assist employers in developing strategies on how they can redesign the job of employees in order to assist them to perform better.Thus, this study is timely, as evidence from past research has shown that a factor such as role ambiguity is important in the service sector setting.Role theory, as expounded by Kahn et al. (1964) states that role ambiguity will result in a job holder engaging in a coping behaviour (e.g.searching for more information) in an attempt to solve unclear tasks in a move to overcome stress which in turn will distort the reality of job requirements.This will then lead to a person feeling dissatisfied with his role, experiencing anxiety and thus performing less effectively.
The need to study role ambiguity arises because in any service sector setting, an  , 2006).Thus in testing the hypotheses, null hypotheses will be rejected if the significance p value is less than .05which indirectly means the alternate hypotheses will be accepted, while the opposite will take place should the significance p value be greater than .05.It is concluded that there is a significant relationship between job performance and role ambiguity (r=-0.686,p<0.01).According to Davis (1971), this relationship is strong when the 'r' ranges from 0.50 to 0.69.Thus, the hypothesis of this study is strongly supported.

Conclusions
This study contributes to the present literature on SMEs in Malaysia specifically in the context of service sector SMEs, which have not received much attention.The knowledge obtained can assist employers in developing strategies on how they can redesign the job of employees in order to assist them to perform better.Thus, this study is timely, as evidence from past research has shown that a factor such as role ambiguity is important in the service sector setting.Role theory, as expounded by Kahn et al. (1964) states that role ambiguity will result in a job holder engaging in a coping behaviour (e.g.searching for more information) in an attempt to solve unclear tasks in a move to overcome stress which in turn will distort the reality of job requirements.This will then lead to a person feeling dissatisfied with his role, experiencing anxiety and thus performing less effectively.
The need to study role ambiguity arises because in any service sector setting, an employee who is able to understand what his or her role is in the organization is likely to perform well in his or her job when dealing with customers (Anderson, 2006).The result of this study supports past researches that found role ambiguity to have a significant negative relationship with the job performance of employees (e.g.Kahn et al., 1964;Lysonski & Johnson, 1983;Jackson & Schuler, Behrman & Perreault, 1984;Singh, 1993;Bhuian, Merguc & Borsboom, 2005;Chang & Chang, 2007).In other words, when the level of role ambiguity increases, the level of job performance of employees will reduce as employees will not be able to cope with given tasks and thus perform less effectively.The reason for this finding can be explained by the fact that employees in a service setting are particularly susceptible to role ambiguity especially those in SMEs context as they generally receive little training and work with minimal supervision (Dubinsky & Mattson, 1979).Small firms tend to have less formal human resource management (HRM) practices, thus this tendency was also extended in their training policy as well (Price, 1994).Furthermore, SMEs shorter lines of communication, problems being solved faster and ease of readjustment may also be the reasons behind this informal practice (Price, 1994).
Additionally, researches on SMEs in Malaysia are quite limited to the performance of SME firms rather than looking at the individuals' performance (e.g.Hashim, 2000;Sohail & Hoong, 2003;Alam & Ahsan, 2007;Jamaludin & Hasun, 2007).Therefore, this study will contribute to the dearth of research in relation to the job performance of employees.At the same time, researches on the job performance of employees that work in the service sector in Malaysia were conducted mainly on the public sector employees (e.g.Azmi, 2010) which created an insufficiency of studies on the service sector SMEs.Therefore, through this research, employees especially those in the service sector SMEs were given attention to help in understanding the reasons behind poor performance that affects the level of productivity.This study also suggests future research to cover both the service and the manufacturing sector employees so that comparisons can be done in order to see whether employees in the manufacturing sector experience similar conditions like their counterparts in the services sector.In addition, in order to identify specifically the approaches that can be used to enhance the job performance of those employees, it is suggested that future research to be conducted on specific areas of the business sector.Thus, future research may consider specific business sectors such as education, hotel, retailing, and business services which are under the twelve key areas of the national transformation programme (NKEAs) (PEMANDU, 2011), so that more specific strategies can be designed to meet the needs of the employees of the specific sectors.

Table 1
Number of Service SMEs (including ICT) in Malaysia as of March 2011

Table 2
Background of the Respondents

Table 3
Summary of Factor Loading for Job Performance

Table 3
Summary of Factor Loading for Role Ambiguity

Table 4
presents the inter-correlations among all the variables in this study.The level of significance was at 5% (.05) which is a generally accepted conventional level in social sciences research (Sekaran &

Table 4
Correlation Matrix of the Variables **. Correlation is significant at the 0.01 level (2-tailed).etal.