Highlight, copy & paste to cite:

Koh, W. L. & Ten, M. C. (1998). Work, Job Satisfaction and Behavioral Intention of Part-Timers in Singapore, Research and Practice in Human Resource Management, 6(1), 35-50.

Work, Job Satisfaction and Behavioral Intention of Part-Timers in Singapore

William L. Koh & May Cheng Ten


Following the framework proposed by Feldman (1990), this paper examined the impact of part-time work characteristics on employee job satisfaction, work motivation and intention to turnover. Data were collected from 362 part-timers from 7 employment agencies and 12 organizations. The results show that salient differences exist between part-timers contrasted on the permanent/temporary and the organization/agency-hired dimensions. Also, part-time work characteristics exerted collective influence on job satisfaction, work motivation and intention to tumover. These results highlight the importance of integrating these work characteristics in strategic human resource policies and practices. In the light of the tight labour market in Singapore, the findings hold important practical implications for HR practitioners, employment agencies, and labour legislators in their effort to entice the economically inactive to regain their workforce identity.


For years part-timers, reflecting their small number in the workplace, anguished as the “missing persons” in organizational research. Up till the late 70s, only five studies dealt with issues related to part-timers (Rotchford & Roberts, 1982). However, both the number and proportion of part-timers in the workplace have swelled over the last twenty years. In America, with the increasing trend toward corporate downsizing, some have predicted that in the not too distant future, half of the workforce — some 60 million Americans — will be working as part-timers (Greenberg & Baron, 1997). Handy (1994) has described the organization of the future as “an association of temporary residents gathered together for mutual convenience.” Mirroring this increased importance of contingent workers in the workplace, researchers have begun to study this segment of the workforce seriously (Ronen, 1984). The increasing stream of scholarly work in this area (e.g., Moberly, 1986; Feldman, 1990; Caudrori, 1994) is a tell tale sign that part-timers will no longer linger as a poor cousin to full time employees in so far as academic research goes.

The trend of increased participation by part-timers has spread from America to Europe and to countries in Asia Pacific like Australia and New Zealand. In Denmark, the United Kingdom, the Netherlands, Norway and Sweden, part-timers comprise more than 20% of the workforce (Murphy, 1996). In Australia, the proportion of part-timers in the labour force ballooned from 10% in 1964 to 24.8% in 1995 (Dawkins, 1995), while in New Zealand it stands at 30% (Statistics New Zealand, 1996).

In Singapore, rapid economic expansion during the past few years aggravated the shortage of manpower caused by the declining birth rates of the last two decades. To alleviate manpower shortage, an often mooted idea is to encourage women and older workers to rejoin and remain in the workforce. In order to do that, employers must be willing to offer part-time employment (The Straits Times, 6 June 1995; 21 March 1996; The Business Times, 6 June, 1995).

Effort in this direction increased the proportion of part-timers in the labour force from 2.8% in 1994 to 3.4% in 1996 (The Singapore Labour Force Survey, 1994; 1996). This proportion would be higher if the Ministry of Labour had defined part-time work as “less than 35 hours” instead of the current “less than 30 hours.” As more firms in Singapore downsize their workforce to cope with increased competition and slower economic growth — in 1996 alone nearly 11,000 workers were retrenched in Singapore (The Straits Times, 3 July 1997) — more part-timers will be needed to alleviate worker shortage. This would lead to a further increase of part-timers in the labour force and hence it is important that human resource practitioners and policy makers alike understand the job attitudes of this work group.

Research Framework

This research, based on a modified model of part-time work suggested by Feldman (1990), examined the job satisfaction, work motivation and intention to turnover, of different types of part-time workers. Depicted by the bold arrows in Figure 1, this study also examined the impact of part-time work arrangements on work context issues. The collective impact of work arrangement, employee demographics, and work context factors on job attitudes and behavior was also examined.

Types of Part-Time Work Arrangements

Feldman (1990) delineated five ways in which part-time work could be differentiated, two of which are examined in this research. Part time work can be permanent or temporary; and, sourced directly from an organization or indirectly through an employment agency. Permanent part-timers work fewer than 35 hours per week on a continuous basis in the same organization, while temporary workers are hired for limited periods of time to deal with fluctuating worldoads or short-term personnel shortages (Howe, 1986; Moberly, 1986; Nollen, 1978). Part-timers can source their work directly from an organization and be paid by the same organization. Or, they could get their lob assignments through an employment agency, in which case they will be paid by the agency (Howe, 1986; Moberly, 1986).

Figure 1
Proposed Framework for Research
Proposed Framework for Research

Job attitudes and behaviors among part-timers differ across different part-time work arrangements (Feldman, 1990; Feldman & Doerpinghaus, 1992a, b). Generally, temporary part-timers display the poorest job attitudes and lowest commitment to their organizations (Hom, 1979; Logan, O’Reilly & Roberts, 1973). On the other hand, because of the sustained contributions they make to their employers, permanent part-timers are viewed as more valuable. They tend to get higher wages, more fringe benefits and greater job challenge and hence display better job attitudes (Feldman & Doerpinghaus, 1992a, b). Therefore, we predicted that permanent part-timers experience higher levels of job satisfaction, work motivation, and lower level of intention to turnover compared to temporary part-timers.

Feldman (1990) suggested that it is unlikely for employment agencies to invest in training for part-timers who are on their payroll, while organizations are likely to do so. The short term nature of their job assignments makes it uneconomical for agencies to invest in training part-timers. On the other hand, organization-hired part-timers generally stay with the same employee for a longer period. It therefore makes sense for them to invest in training to improve the part-timers’ skills and help them integrate into the work group. Feldman’s suggestion regarding the contrasting attitudes of agencies and organizations toward training of part-timers certainly holds true in the Singapore context (The Business Times, 10 Apr. 1995). As part-timers are likely to view training as beneficial to their career development, we hypothesized that organization-hired part-timers experience higher levels of job satisfaction, work motivation and lower level of intention to turnover compared to agency-hired part-timers.

Work Context Factors

Part-time work arrangements may also affect work context factors which are deemed to be important to such employees (Feldman & Doerpinghaus, 1992a). Five such factors — scheduling flexibility, relation with co-workers, relation with supervisors, pay and fringe benefits packages — are examined in this research.

The ability to integrate work life with personal demands — such as family commitment, leisure activities, school and friends — is a key concern for part-timers. As such, part-timers who are given freedom in work scheduling are likely to have more positive job attitudes and are less likely to change their jobs (Ronen, 1984; Levine, 1987; Feldman & Doerpinghaus, 1992a). In this regard, it is interesting that permanent part-timers experience greater job scheduling conflict than temporary part-timers (Feldman & Doerpinghaus, 1992a). It could be that in order to secure the more permanent temporary jobs — which typically offer higher pay, better fringe benefits and more challenging assignments — permanent part-timers have to trade-off some schedule flexibility. For instance, they may have to work on undesirable schedules which conflict with family demands. Hence we predicted that permanent part-timers experience greater conflict in job scheduling than temporary part-timers.

Agency-hired part-timers are likely to experience more scheduling conflict than organization-hired ones due to restrictions imposed by personal demands (Ronen, 1984; Feldman, 1990). Hence we predicted that agency-hired part-timers experience more conflict in job scheduling than organization-hired part-timers.

Feldman (1990) proposed that permanent and organization-hired part-timers are most likely to receive adequate training opportunities. These training opportunities provide for social interaction, which translates into greater integration in the work groups. On the other hand, temporary and agency hired part-timers, realizing that they are usually the first to be displaced in an economic and organization downturn, are less likely to maintain close bonding with their colleagues (The Straits Times, 15 Sept. 1995). Hence we predicted that compared to their temporary and agency-hired counterparts, permanent and organization-hired part-timers enjoy better relation with co-workers.

The nature of their employment enable the permanent and organization hired part-timers to make a greater sustained contribution to the firm, thus enhancing their value to the organization. Recognizing this fact, supervisors are more likely to support and interact with these part-time employees (Feldman & Doerpinghaus, 1992b). It has also been suggested that the supervisors’ perception regarding the peripherality of employees influence the type of work assigned. For instance, if an employee is perceived to be marginal, less desirable tasks may be assigned (Ronen, 1984). Both these factors would lead one to hypothesize that permanent and organization-hired part-timers should enjoy better work relations with supervisor compared to temporary and agency-hired part-timers.

Research indicated that permanent part-timers receive higher pay and fringe benefits than temporary part-timers (Feldman and Doerpinghaus, 1992a). This suggests that organization-hired part-timers may also receive better pay and fringe benefits compared to agency-hired part-timers. Temporary employment holds great appeal for young workers who want flexible schedules to juggle between school and work. These young part-timers also tend to turn to employment agencies for their job placement to decrease their job search costs (Howe, 1986). As such, agency-hired part-timers, who usually are equipped with minimal work experience, are likely to receive lower pay and fringe benefits. Therefore, permanent part-time employees are likely to receive higher pay and more fringe benefits than temporary part-timers; likewise organization-hired part-timers are likely to receive higher pay and fringe benefits than their agency counterparts.

Impact of Employment Characteristics

Feldman and Doerpinghaus (1992b) found that from a set of employment characteristics (types of work arrangements, demographic factors, work context factors, frame of reference), work arrangements and demographic factors accounted for much of the variance in part-timers’ job attitudes and behavioral intentions. However, they did not provide any theoretical considerations for the order of entry among the independent variables. In order to provide a renewed foundation for the explanation of the dependent variables, this study re-explored the nature of relationships that exist between the variables. Emphasis was placed on deriving a tighter framework for future research.


Sample and Research Design

Organizations which took part in the research were randomly drawn from the Buying Guide of the Singapore Yellow Pages, July 1994 edition. Responses were received from employment agencies, hotel, healthcare, travel, retailing, food and entertainment organizations over a period of time from June to September 1995. The final sample comprised part-timers from 7 employment agencies and 12 organizations.

A strategy of purposeful variation among participants on the organization-hired versus agency-hired dimension ensured that these two sub-groups of part-timers could be adequately represented for statistical analyses. Questionnaires were distributed to the part-timers through the Recruitment/Human Resource Department in the participating employment agencies and organizations. Completed questionnaires were sealed and deposited in a box provided at the Recruitment/Human Resource Department. These were then collected by the researchers after two weeks.

In-depth personal interviews with six part-timers were conducted before the actual survey and feedback from these sessions was used to reduce ambiguity in the questionnaire. The cover letter was specially worded to encourage greater response and, last but not least, respondents were assured of complete and strict confidentiality.

Measurement Scales

Job satisfaction was assessed using a 15-item scale developed by Warr, Cook and Wall (1979). Representative items include, “How satisfied are you with your fellow workers?” and “How satisfied are you with the amount of responsibility you are given?” The 7-point Likert scale has responses ranging from 1 (Extremely Dissatisfied) to 7 (Extremely Satisfied) and reported Cronbach alphas of 0.85 and 0.88. Following Levine (1987), Rotten (1984), and Feldman and Doerpinghaus (1992a), two additional items — satisfaction with the fringe benefits offered and flexibility of their job schedules — were included.

Work motivation was measured by a 6-item scale developed by Hackman and Oldham (1975). Representative questions include, “Most people on this job feel a great sense of personal satisfaction when they do the job well,” and “Most people on this job feel bad or unhappy when they find they have performed the work poorly.” Response options range from 1 (Strongly Disagree) to 7 (Strongly Agree). Coefficient alpha of .79 (Kim & Schuler, 1979) and internal reliability of 0.71 (Hackman, Pearce & Wolfe, 1978) were reported.

Intention to leave was measured using a 3-Item scale derived from Cammann et al. (1979). A representative question from the scale is, “I often think about quitting.” Mean values were not cited in the source publication but the coefficient alpha was reported to be 0.83.

Types of part-time work arrangement were assessed by two close-ended questions. Respondents were asked if their part-time assignment was temporary or permanent in nature and sourced directly from the organization or through an employment agency.

The Work Facilitation-Subordinate Relations sub-scale from the supervisor module of the Michigan Organizational Assessment Questionnaire (Cammann et al., 1979) was used to measure relation with supervisor. A representative item is, “My supervisor keeps subordinates informed about things that go on in the organization.” The 7-items scale has 7 response options ranging from 1 (Strongly Disagree) to 7 (Strongly Agree) and mean values across the items constitute the scale score. On a sample of more than 400 employees, authors of the scale found that the sub-scale yielded alpha coefficient of 0.93.

Relation with co-workers was assessed through the 3-item scale developed by Taylor and Bowers (1972). A representative item is, “Persons in my work group exchange opinions and ideas.” The 7-point Likert scale had responses ranging from 1 (Strongly Disagree) to 7 (Strongly Agree). In the source publication, the alpha coefficient was reported at 0.90.

Pay, fringe benefits, and scheduling flexibility were assessed through single-item measures. A list of fringe benefits was provided and respondents were asked to indicate the benefits that they currently received. Some of the fringe benefits listed were CPF contribution, paid sick leave, and paid holiday leave. Scheduling flexibility was assessed through the question: “How often do you experience outside-life conflict (e.g. family, friends, school, leisure) in working at your current part-time job?”.


Descriptive Statistics

Of the 745 questionnaires distributed, 379 part-time workers from 7 employment agencies and 12 organizations responded; 17 questionnaires were found to be incomplete and thus excluded, bringing the number of usable questionnaires to 362. The effective response rate turned out to be 48.6%.

Table 1 shows that the sample comprised 68.1% females and 31.9% males. These respondents were predominantly single (66.6%), Chinese (85.8%), students (56.5%). More than half of them attained secondary school qualification and below, with 63.2% between 15-24 years of age. The sample characteristics are somewhat different from the national sample which has more married (63.4%) and older (more than 50% between the ages of 30 to 49) part-timers. Two events could have accounted for this occurrence. First of all, this research was carried out during the June school holidays which is a popular period for school students to Join the part-time workforce. Hence more of them were surveyed. Secondly, more organizations which participated in the research were from service rather than the manufacturing sector. As the service sector tend to employ younger part-timers, the current sample ended up with a bigger proportion of younger part-timers. Conclusions drawn from this research should be seen in the light of these factors.

Table 1
Profile of Respondents
Variable Valid Percentage (%)
Male 31.9 %
Female 68.1 %
Marital Status
Single 66.6 %
Married 32.3 %
Others 1.1 %
Chinese 85.8 %
Malay 8.9 %
Indian/Others 5.3 %
PSLE level/‘N’ Level 23.8 %
‘0’ level/NTC 45.6 %
‘A’ level 12.9%
Diploma and above 17.8 %
15 - 24 years 63.2 %
25 - 34 years 10.3%
35 - 44 years 14.6%
45 - 54 years 7.8 %
55 years and above 4.1%
Student status
Yes 56.5 %
No 43.5%
$4 and below 14.9%
$4 - $4.99 16.6%
$5 - $5.99 40.4%
$6 - $6.99 10.4%
$7 - $7.99 6.2%
$8 and above 11.5%
less than 6 months 56.9 %
6 to 12 months 11.9 %
13 to 24 months 6.3%
2 to 5 years 11.9%
more than 5 years 13.0 %
Types of part-time work arrangements
(i) Organization-hired 55.8 %
Agency-hired 44.2 %
(ii) Permanent 32.7 %
Temporary 67.3 %

Note: N = 362; Missing values have been excluded.

Factor and Reliability Analyses

All multi-item scales were factor analyzed to ascertain their dimensionality. Factor analysis performed on the job satisfaction scale showed that the two newly included items (fringe benefits offered and flexibility of job schedule) both displayed strong factor loadings of 0.60 and above. These results confirmed their inclusion in the job satisfaction scale.

The fourth item on the work motivation scale was discarded as it yielded a factor loading of only 0.19 (well below the cutoff point of 0.50). The reliability of the scale improved significantly from .62 to .76 when this item was deleted. Being a reverse-scored item, it could have caused some confusion to the respondents. Items belonging to the intention to turnover, work relations with supervisor and co-workers scales all loaded into the hypothesized factor structure; all factor loadings were at least .60.

Evidence for reliability of the measurement scales was provided by high Cronbach’s alphas. The Pearson correlation matrix (in Table 2) also shows that correlations between these established measurement scales were significantly high, thus confirming the validity of these constructs. Job satisfaction and work motivation were positively and significantly correlated (r = 0.46, p < .001). Furthermore, intention to turnover was negatively and significantly correlated with these two attitudinal scales (r = -0.46 and r = -0.18 respectively; p < .001).

Table 2
Means, Standard Deviation, Reliabilities and Correlations
Variables Mean SD 1 2 3 4 5
1. Supervisor 4.71 1.07 (.88)
2. Co-workers 4.71 1.13 .44*** (.85)
3. Job Satisfaction 78.68 15.17 .55*** .46*** (.91)
4. Work Motivation 5.20 .90 .33*** .33*** .46*** (.76)
5. Intention to Turnover 3.69 1.56 .27*** .22*** .46*** .18*** (.74)

*** p < .001 N = 358 Figures on the diagonals are the reliabilities
Supervisor: Relation wtih supervisor; Co-workers: Relation with co-workers

Results in Table 3 support the hypothesis that permanent part-timers expressed significantly higher levels of job satisfaction (t = 5.32, p < .001), work motivation (t = 1.76, p < .05), and are less likely to quit or leave their current assignments (t = -6.45, p < .001). Likewise, organization-hired part-timers reported higher levels of job satisfaction (t = 5.94, p < .001), work motivation (t = 3.21, p < .001), and less intent to turnover (t = -8.27, p < .001).

Table 3
Comparison of Job Attitudes and Turnover Intention between Permanent/Temporary and Organization-hired/Agency-hired Part-Timers
Variable Job Satisfaction Work Motivation Intention to Turnover
Permanent 84.54 75.41 5.32***
Temporary 5.32 5.14 1.76
t - Value 2.97 4.06 -6.45***
Organization hired 82.52 73.19 5.94***
Agency hired 5.33 5.03 3.21***
t-Value 3.14 4.40 8.27***

*** p < .001 ** p < .01

Results from Table 4 show that permanent part-timers enjoy better relation with supervisor and fringe benefits than temporary part-timers (t = 4.19, p < .001 and t = 10.89, p < .001, respectively). No significant differences were detected in their response to the other work context factors. Interestingly, part-timers contrasted on source of employment show more significant differences. Results supported most of the hypotheses, suggesting that organization-hired part-timers do experience more favorable work conditions than agency-hired ones. These employees experience less scheduling conflict (t = -1.37, p < .1), enjoy better relations with co-workers, (t = 2.48, p < .01) and supervisor (t 4.68, p < .001), and finally, receive significantly more fringe benefits (t = 12.11, p < .001).

Table 4
Sample Sizers, Means and t-values by Work Context Factors
Work-Context Factors Permanent (N) Temporary (N) t - value Organization-hired (N) Agency-hired (N) t - value
Scheduling Flexibility 4.10 (114) 4.04 (235) 0.61 3.99 (200) 4.13 (160) -1.37*
Co-workers 4.75 (114) 4.68 (235) 0.53 4.84 (202) 4.54 (160) 2.48**
Supervisor 5.04 (114) 4.53 (235) 4.19*** 4.93 (202) 4.42 (160) 4.68***
Pay 3.14 (113) 3.15 (233) -0.02 3.22 (196) 3.08 (160) 0.88
Fringe Benefits 3.07 (114) 1.30 (235) 10.89*** 2.57 (202) 0.97 (160) 12.11***

Tests of Multicollinearity

Tolerance and variance inflation factor (VIF) measures were used to help detect the presence of collinear variables (Tabachnick & Fidell, 1989). DIagnostics accompanying the regression results for intention to turnover show that age and marital status were highly correlated. The low tolerances (age = .24 and marital status = .22) and high VIF’s (age = 4.24 and marital status = 4.56) is expected, since there is a natural relationship between these two variables. Bivariate correlation data revealed that the correlation between these 2 variables is .87 (p < .01). This multicollinearlty problem was eliminated by deleting the marital status variable from the regression analyses.

Stepwise Regression Analyses

Results from the regression analyses, shown in Table 5, revealed that variance in job satisfaction were explained by work context factors and type of work arrangement (R2 = .40, p < .001). The bulk of the variance - 39% out of a total of 40% - were explained by relations with supervisor and co-workers, and fringe benefits. Variance in work motivation was significantly explained by relations with supervisor and co-workers, and age (R2 = .21, p < .001). The most significant predictors, once again, were the work context factors. Variance in intention to turnover were significantly predicted by source of employment, age and relations with supervisor and co-worker accounted significantly for the (R2 = .24, p < .001). The first three variables contributed most to the variance in intention to turnover, with the subsequent addition of co-worker relation contributing merely an additional 1% (p < .01).

Table 5
Statistical Stepwise Regression Analysis of Employment Characteristics with Job Satisfaction, Work Motivation and Intention to Turnover
Dependent Variables Independent Variables R2 R2 Change
Job Satisfaction Supervisor .29*** .29***
Supervisor, Co-workers .35*** .06***
Supervisor, Co-workers, Fringe Benefits .39*** .04***
Supervisor, Co-workers, Benefits, Work Arrangement .40*** .01**
Work Motivation Co-worker .11*** .11***
Co-worker, Age .15*** .04***
Co-worker, Age, Supervisor .18*** .03**
Intention to Turnover Employment source .16*** .16***
Employment source, Age .19*** .03***
Employment source, Age, Supervisor .23*** .04*
Employment source, Age, Supervisor, Co-worker .24*** .01*

Note: N = 361; ***p < .001 ** p < .01 * p < .05


The results of this research highlight the importance of part-time work arrangements, demographic, and work context factors in understanding the job attitudes of part-timers. In line with Feldman and Doerpinghaus (1992b), the findings show that permanent and organization-hired part-timers expressed significantly greater job satisfaction, work motivation and less intention to turnover compared to their temporary and agency-hired counterparts.

Contrary to Feldman and Doerpinghaus (1992a), there is no significant difference in scheduling conflict between permanent and temporary part-timers. On the other hand, agency-hired part-timers did experience significantly more outside-life conflict compared to organization-hired ones. This difference could be accounted for by the fact that agency-hired part-timers, many of whom are in their teens, might have more personal commitments than their older organization-hired counterparts. Besides their family, friends, school and leisure, this group of part-timers must also submit time-sheets on a regular basis to their employment agencies or report to the agencies to collect their pay cheques. Having to integrate such demands into their life probably generate the greater level of conflict. Another contributing factor could be the way companies manage their core workers and “buffer” group of part-timers. Firms usually turn to agency-hired part-timers to fill temporary positions in the undesirable weekend and holiday shifts during peak periods. Being stuck with these work schedules could have affected their ability to integrate their work life with personal demands.

The results indicated that permanent and temporary part-timers do not differ significantly in relation with co-workers. However, organization-hired part-timers do enjoy better work relation with co-workers. Training opportunities offered by the organizations could have provided such part-timers greater interaction opportunities which in turn allow them to be integrated into their work groups. Such opportunities are often not given to agency-hired part-timers. The short-term nature of their work contract means that it does not make economic sense for agencies to train them (Feldman, 1990).

In contrast with Feldman and Doerpinghaus (1992a), this research found that permanent and organization-hired part-timers do enjoy better relation with supervisor compared to temporary and agency-hired part-timers. There seems to be greater work facilitation by supervisors which in turn engenders better supervisor-subordinate relations between the two parties. The supervisors’ investment of time and effort in these categories of part-timers is understandable since they remain in the same company for a longer period of time.

In contrast to Feldman and Doerpinghaus (1992a), this research does not support the claim that permanent part-timers receive higher pay than temporary part-timers. The same result was obtained with reference to part-timers contrasted on the source of employment. However, both permanent and organization-hired part-timers received more fringe benefits than their temporary and agency-hired counterparts. The findings indicate that employers are inclined to use fringe benefits rather than pay as the “carrot” to retain part-timers. As a case in point, extending equitable benefits to both part-time and full-time employees has helped JVC Electronics in Singapore to retain their staff and reduce turnover. (The Sunday Times, 10 Dec. 1995).

The regression analyses show that almost all employment characteristics significantly explained the variances in the part-timers’ job satisfaction, work motivation and intention to turnover, though the list of predictors are not identical to that in Feldman and Doerpinghaus (1992b). In line with extensive literature in this area (for example, see Greenberg & Baron, 1997, p.183), work context factors accounted for most variance in job satisfaction. The permanent/temporary job dimension also contributed significantly to the model, thus suggesting that favourable work conditions accompanying permanent part-time work arrangements (e.g., better work relations with supervisors and fringe benefits given) could have led job incumbents to experience higher level of job satisfaction.

As for work motivation, relation with supervisor accounted for most of the variance. Being motivated usually entail the presence of the three psychological states of knowing results, experiencing responsibility and meaningfulness on the job (Hackman & Lawler, 1971; Hackman & Oldham, 1976). Relation with the supervisor is directly linked to these states. If a supervisor provides support and facilitation on the job, the employee is likely to be assigned greater task responsibilities and accumulate learning experiences. Much of the variance in intention to turnover was accounted for by the organization-hired/agency-hired dimension. The other significant predictors in the regression were age and relations with both supervisor and co-workers. This suggests that work arrangements, demographic and work context factors exert collective effects in influencing the turnover intention of part-timers.

To maintain a high level of performance and ensure desired behavioural outcomes, organizations and employment agencies must integrate these relevant factors into their management plans and policies. Agency-hired part-timers paled in comparison with organization-hired ones. The source of these differences lies mainly in training opportunities offered by organizations. However, the lack of training of part-timers by agencies may become a thing of the past. As competition intensifies in the employment services industry, one may see agencies assuming the entire burden of the training costs in order to entice part-timers (The Straits Times, 2 Oct. 1995). The agencies’ role in the functioning of the part-time work system in Singapore may also change as they consider taking on not just the training but the entire contingent staffing function. This is a step in the positive direction as it ensures that part-timers are well matched with the job requirements in organizations (Caudron, 1994). An example in Singapore is PF1, a partnership between ECCO (an employment agency) and IBM, which assumed the role of the human resource department to take care of IBM’s entire contingent staffing needs. It is a win-win situation for both IBM Singapore and ECCO as PF1 ensures a better alignment of recruitment efforts with organizational needs.

According equal opportunities and benefits, in terms of wage increments, training, etc., to all staff, regardless of employment status is another direction which organization should take. This strategy has helped JVC Electronics to retain staff and curb turnover Singapore (The Sunday TImes, 10 Dec. 1995). Besides attracting and motivating part-timers to stay on their assignments, it negates the possible resentment between full-time and part-time employees (Caudron, 1994).

In short, there is need for employment agencies as well as organizations to provide conducive working conditions to all categories of part-timers in order to instil positive attitudes and curb premature turnover. In this regard, it is important for management to note the paradigm shift in employment patterns and take a proactive approach to integrate it into their human resource plans. As Feldman and Doerpinghaus (1992b) noted, employers who view part-time work as merely full-time work with shorter work hours, may miss the most important human resource issues involved in managing part-time employees.

A methodological trade-off may be present in using self-reported questionnaires. Respondents could be subjected to social desirability effect, though this limitation was partly addressed by the assurance of confidentiality. Only two out of the five relevant job dimensions suggested by Feldman (1990) were examined here. Future research could examine other dimensions such as voluntary/involuntary, sole-employment jobs/moonlighting jobs, and year-round/seasonal part-time employment. Amongst these dimensions, sole-employment/moonlighting jobs have not been the subject of any empirical research. In addition, future studies could also investigate whether the various blocks of independent factors exert direct or indirect effects on part-timers’ job attitudes. Regression analyses employed in this study have provided an initial but useful idea concerning the critical independent variables involved. A final related avenue to pursue is to examine the existence of reciprocal causality among the key variables in Feldman’s (1990) model.


Cammann, C., Fichman, M., Jenkins, D. & Kiesh, J. (1979). The Michigan Organizational Assessment Questionnaire. Unpublished Manuscript, University of Michigan, Ann Arbor, Michigan.

Caudron, S. (1994). Contingent work force spurs HR planning. Personnel Journal, 73 (7), 52-60.

Cle, C. W, & Wall, t D. (1981). Note on some new scales for measuring aspects of psychological well-being. Journal of Occupational Psychology, 52, 221-225.

Cohen, J., & Cohen, p. (1983). Applied multiple regression/correlation analysis for the behavioral sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum.

Dawldns, p. (1996) The Economic Record, Vol 72, No. 28,272-286.

Eberhardt, B. J., & Shani, A. B. (1984). The effects of full-time versus part-time employment status on attitudes toward specific organizational characteristics and overall Job satisfaction. Academy of Management Journal, 27,893-900.

Feldman, D. C. (1990). Reconceptualizing the nature and consequences of part-time work. Academy of Management Review, 15, 103-112.

Feldman, D. C., & Doerpinghaus, H. I. (1992a). Missing persons no longer: Managing part-time workers in the ’90s. Organizational Dynamics, 21(1), 59-72.

Feldman, D. C., & Doerpinghaus, H. I (1992b). Patterns of part-time employment. Journal of Vocational Behavior, 41,282-294.

Feldman, D. C., Doerpinghaus, H. I., & Thrnley, W. H. (1994). Managing temporary workers: A permanent HRM challenge. Organizational Dynamics, 23(2), 49-63.

Hackman, J. R., & Lawler E. E. (1971). Employee reactions to job characteristics. Journal of Applied Psychology, 55, 259-286.

Hackman,J. R., & Oldham, G. R. (1975). Development of the Job Diagnostic Survey. Journal of Applied Psychology, 60, 159-170.

Hackman, J. R., Pearce, J. L., & Wolfe, J. C. (1978). Effects of changes in job characteristics on work attitudes and behaviors: A naturally occurring quasi-experiment. Organizational Behavior and Human Performance, 21, 289-304.

Hom, P. W. (1979). Effects of job peripherality and personal characteristics on the job satisfaction of part-time workers. Academy of Management Journal, 22, 551-565.

Howe, W. J. (1986). Temporary help workers: Who are they, what jobs they hold. Monthly Labor Review, 109(1), 47-49.

Jackofsky E. K, & Peters, L. H. (1987). Part-time versus full-time employment status differences: A replication and extension. Journal of Occupational Behaviour, 8, 1-9.

Kahne, H. (1985). Reconceiving part-time work. Totowa, NJ: Bowman and Allanheld.

Kim, J. S., & Schuler, R. S. (1979). The nature of the task as a moderator of the relationship between extrinsic feedback and employee responses. Academy of Management Journal, 22, 157-162.

Levine, H. Z. (1987). Alternative work schedules: Do they meet workforce needs? Personnel, 64(1), 57-62.

Ministry of Labour (Research and Statistics Department) (1994). Report on the labour force survey of Singapore 1994. Singapore: Ministry of Labour.

Ministry of Labour (Research and Statistics Department) (1996). Report on the labour force survey of Singapore 1996. Singapore: Ministry of Labour.

Moberly, R. B. (1986). Temporary, part-time, and other atypical employment relationships In the United States. Labor Law Journal, 37, 689-696.

Murphy, E. (1996) Flexible Work. Director Books. Great Britain: T.J. Press.

Nollen, S. D. (1978). Permanent part-time employment. New York: Praeger.

Ronen, S. (1984). Alternative work schedules. Homewood, IL: Dow Jones-Irwin.

Rotchford, N. L., & Roberts, K. H. (1982). Part-time workers as missing persons in organizational research. Academy of Management Review, 7, 228-234.

Statistics New Zealand (1996) Census of population 1996, Populations Commentary no 3, Commentary on population characteristics, Internet Website: WWW STATS.GEOVT.NZ.

Tabachnick, B. G., & Fidell, L. S. (1989). Using multivariate statistics. 2nd ed. New YorK: Harper & Row.

Taylor, J. C., & Bowers, D. G. (1972). Survey of Organizations: A Machine Scored Standardized Questionnaire Instrument. Institute for Social Research, University of Michigan, Ann Arbor, Michigan.

The Business Times. (6 June 1995). Government may form unit to help women return to work.

The Straits Times. (18 June 1995). Women who stopped working and returned.

The Straits Times. (15 Sept. 1995). From temporary to permanent jobs.

The Straits Times. (2 Oct. 1995). Tempting the temps.

The Sunday Times. (10 Dec. 1995). Dangle equal rights and benefits to tap this pool.

The Straits Times. (21 March 1996). Move to get more women aged 35-54 in workforce.

The Straits Times. (3 July 1997). More older workers seek help to find jobs.

Warr, P. B., Cook, J., & Wall, T. D. (1979). Scales for the measurement of some work attitudes and aspects of psychological well-being. Journal of Occupational Psychology, 52, 129-148.