Download Job Title for Interview questionnaire and more Exercises Computer Science in PDF only on Docsity!
Very preliminary paper. Please do not quote.
Quality of work – concept and measurement
Svenn-Åge Dahl, Torstein Nesheim and Karen M. Olsen
SNF - Institute for Research in Economics and Business AdministrationBergen, Norway June 2007
Introduction
Companies on both sides of the Atlantic, but especially in Europe, started with experiments and policies for improving the quality of working life in the 1960s (for discussion/overview see e.g. Gallie, 2003; Martel and Dupuis, 2006). The most systematic work for improving the quality of work took place in Sweden, and it gained momentum in the 1970s through acts such as the Co-Determination Act (1976) and the Work Environment Act (1978) and other
important initiatives like for example the Working Life Fund (AFL) in 1990. All this inspired the other Scandinavian countries to introduce similar legislation, and it is therefore no surprise that employees in the Scandinavian countries is found to have higher quality of work tasks and better opportunities for participation than employees in other European Union countries (Gallie, 2003).
On the other hand the main policy objective in the European Union during the 1980s and most of the 1990s was to create jobs and reduce unemployment, and the focus was less on job quality. Increasing the employment rate per se was seen as the key to social inclusion. However, at the meeting of the European Commission in Lisbon in the spring of 2000 improving job quality became an explicit objective as part of EU’s aim to become the most advanced economy in the world by 2010. Both the ILO and the OECD have taken up similar themes. Concern with job quality was said to be at the heart of both the European Social Model and the European Employment Strategy. The idea was that what was needed was not only more jobs, but better jobs. 1 The logic is that improved employment rates and quality of work can promote financial self-sufficiency and diminish the pressure on the welfare state. High quality of employment can also boost competitiveness, to the extent that it promotes motivation, productivity and commitment. Effective policies for social integration must take into account not simply the quantity of jobs, but also of the importance of providing jobs that give people the opportunity to extend their skills and to move into better jobs (Gallie, 2003).
The current understanding of both objective and subjective quality of employment is far from being clear and broadly shared. In this paper we will discuss the concept of job quality. First we give a short description of some theoretical approaches to quality of work, then we argue
(^1) To come to grips with what better jobs means the EC has developed its own concept of “quality in work”. This takes into account objective characteristics of the job, subjectiveviews of workers, worker characteristics and the match between the worker and the job.
environment or undesirable working hours, because workers trade off working conditions and benefits for pay.
Research based on different academic fields conceptualizes the quality of work in different ways. In the social sciences there is no comprehensive measure of job quality, and the study of job quality is therefore approached in different ways. Economists tend to focus on aspects of economic compensation such as working hours, hourly wages, annual earnings or fringe benefits, especially health insurance or retirement benefits to measure a jobs’ quality or desirability. Pay is generally regarded as the single most important item of a job, and wages are in general positively correlated with other favourable working conditions. But quality of work is comprised of more than monetary awards (Clark, 2005a), and job security and having an interesting job were considered most important among male as well as female participants (Clark, 2005b). Economists do not have a global measure of jobs’ non-monetary benefits (or costs) (Jencks et al., 1988) and omit potentially important aspects such as job autonomy and satisfaction with certain job facets. The problem is that the effect of non-monetary job characteristics on job ratings is more than twice that of earnings (Jencks et al., 1988). Earnings and wages are important aspects of job quality, but only one of many.
Sociologists generally study occupational prestige or status within a system of social stratification as well as the degree of autonomy and control that workers have over their jobs (Kalleberg and Vaisey, 2005). In the sociological tradition the concept of skill is also very central in the study of quality of work. Skilled work is seen as involving both complex operation and autonomy (influence and discretion over daily work tasks) for the worker. Studies that use occupational status to measure labour-market success often reach conclusions quite different from those of investigations that use earnings to measure success (e.g. Sewell and Hauser, 1975; Jencks et al., 1979). This approach has among other things been criticized
because occupational titles can not tell us much about specific jobs if most of the variance is within occupations as opposed to if the variance is between occupations (Jencks et al., 1988). It is also worth noting that Gallie (1996) has criticized the sociological approach for neglecting the mediating effects of employee aspirations when assessing quality of work.
According to Kalleberg and Vaisey (2005) psychologists often emphasize non-economic aspects of work (such as intrinsically meaningful and challenging work) and assess the variety of psychological sources of job satisfaction. Occupational psychology (implicit theories of human need) have had focus on job satisfaction and well-being, and have confirmed the importance for workers of having discretion and trust in their jobs. This tradition has also focused on the workplace as a social arena, and the relevance to the quality of work life of having good social relations among workers (Green, 2006). The psychological approach can for example be criticized for being too oriented towards subjective measures and for ignoring objective or extrinsic measures such as wages and fringe benefits (Rose, 2003). Quality of work is a much broader conception than job satisfaction – job satisfaction is one of many possible outcomes of quality of work (Sirgy et al., 2001).
Several leading researchers have pointed to the fact that to understand job quality a multi- dimensional approach is needed, and that it is important to measure both monetary and non- monetary job characteristics. The above mentioned perspectives are useful, but only partially, and an approach that takes into account economic as well as non-economic sources of variation in the goodness of jobs is necessary.
Measuring job quality
To measure which jobs are best and worst is not a simple task as jobs are made up of many components, but there are situations when it is useful to talk about their overall quality
The second approach is to ask workers directly to provide a global or general assessment of their jobs. The most frequent example of this is when workers are asked about their degree of job satisfaction. This global approach does not measure all relevant job characteristics, but assumes that the workers are able to balance out the various aspects of job characteristics to come up with an overall assessment of job quality. An obvious disadvantage of this approach is that it does not tell us how good of bad a job is along various dimensions. It is therefore not possible to make an assessment of the relative importance of different job facets in determining the quality of jobs.
There are two major ways of linking global and specific measures of job quality – regression analysis and configurational approaches. Regression analysis is the most common way of relating measures of the overall goodness of jobs to the quality of specific dimensions of work. Studies of job satisfaction often attempt to explain why some workers are more satisfied with their jobs than others by regressing a global measure of satisfaction on a set of indicators of the quality of specific job facets. Jencks et al. (1988) illustrate the use of regression analysis to combine global and specific measures of job quality. They asked workers to rate the perceived desirability of their own jobs, taking everything into account (such as pay, fringe benefits, working conditions and kind of work). Then, they regressed this overall rating of ‘job goodness’ on 48 job characteristics. Individual differences in conceptions of the goodness of jobs were assumed to be reflected in the weights attached to the various job facets. Fourteen of the 48 variables that were most strongly related to the global measure were used to generate predicted scores of job quality (their ‘index of job desirability’).
Configurational approaches facilitate the analyses of jobs as bundles of characteristics of differing quality. Job facets are assumed to form job types or bundles of job characteristics rather than to compete with each other to explain variation. It is therefore not individuals or single variables that are the unit of analysis but rather all the ‘types’ that can be formed by the possible combinations of the components of job quality. Fuzzy-set Qualitative Comparative Analysis is an example of a configurational approach (e.g. Ragin, 1987, 2000), and uses techniques of combinatorial analysis to investigate explicitly how different combinations of job characteristics produce outcomes.
Central dimensions in the measurement of quality of work
According to Gallie (2003) there has been a remarkable convergence in terms of the aspects of work that is considered crucial for well-being. For example what Gallie call the liberal and neo-marxist perspectives, place central emphasis on the scope for initiative in carrying out the job, the variety of work, the opportunities for learning, and the ability to participate in decision-making. Green (2006: 13) also points to the convergence and writes: “There is a broad convergence of the sociological position on the quality of work with application of the ideas of A. Sen (1987, 1993).” This is the perspective Green himself uses in his important new book “Demanding Work. The Paradox of Job Quality in the Affluent Economy.” Here he focuses on skill, effort, personal discretion, wages and risk as key indicators of job quality. High quality jobs generate capabilities that allow workers to achieve well-being and to achieve a range of personal goals. Capabilities are derived through wages and other rewards, future prospects (pensions and security), job control (the ability to choose). A high quality job is defined as: “... one that affords the worker a certain capability – the ability and the flexibility to perform a range of tasks (including the necessary sense of personal control), to draw on the comradeship of others working in cooperation, to choose from and pursue a range
but potential conflicts of interests between workers and employers are underplayed. Green (2006) himself examines the following aspects of job quality: 1) The skill involved in a job is considered because the utilization of skill is an end in itself, with intrinsic value (skill polarization), 2) work effort, 3) participation in workplace decisions, 4) pay, 5) low risks and security, and 6) job satisfaction and affective well-being. Kalleberg and Vaisey (2005) use the following six job characteristics which have been central to discussions of job quality: 1) Economic benefits: satisfaction with earnings , fringe benefits (health insurance and pensions),
- Non-economic benefits: the degree of autonomy and control one has over one’s work and the extent to which one receives intrinsic rewards from the job, and 3) a measure of perceived job security as well as the extent to which the worker is satisfied with the opportunities for advancement. McGovern et al.’s (2004) definition of bad jobs concentrates on the economic nature of the employment relationship, in particular to the level of income from employment, continuity of income from employment, continuity of income while sick or in retirement, and the prospect of increased income through promotion. More specific, they conceptualize bad jobs as those with: a) low pay, b) no sick pay, c) no pension scheme, beyond the basic state scheme; and d) are not part of a recognized career or promotion ladder. As Kalleberg et al. (2000) they use a summative measure of badness – the number of bad job characteristics as the dependent variable for the multivariate analysis. The advantage of such a measure is that it enables us to examine the determinants of (bad) job quality generally as well as those for individual dimensions. This conception of bad jobs implies that the characteristics are objective, manifest, and of equal importance: The presence or absence of any of the characteristics has the same implications as that of any other in the set.
Other researchers have introduced additional aspects. For example Requena (2003, see also Lowe and Schellenberg, 2001) introduce social capital and Clark (2001) measures job quality/ satisfaction as with a focus on job quits.
Based on the overview and discussion above we have decided to focus on six dimensions which we consider as the most important in the measurement of job quality: 1) Pay and fringe benefits, 2) job security, 3) work intensity, 4) intrinsic job rewards, 5) skill and 6) autonomy and control. Below we will discuss these dimensions in more detail.
Pay and fringe benefits Pay is the core dimension in the economic perspective on quality of jobs, and commonly used in sociology. By contrast, psychological perspectives more often emphasize non-economic, intrinsic aspects of work. Objective living conditions, involving dimensions, such as pay was one of the main aspects of the early research on quality of life (Beham et al., 2006). “The level of living approach, defined quality of life in terms of control over resources such as money, property knowledge, security...” (Beham et al., 2006: 6). A low wage has consequences long after the have left employment, but it makes it difficult to build up savings to draw upon when there is no longer and income from employment, and with earnings related pension schemes the disadvantage continues as people enter retirement (Gallie, 2002).
Recent studies on quality of jobs also argue that wages should be included as one of the core dimensions, and an increasing wage is considered as a sign of improving job quality and a declining as an indication of the opposite. Although a raise in wages does not proportionately increase people’s happiness, a minimum level of pay is necessary to uphold a basic level of living (Green, 2006). Low pay has become a problem, and about 15 per cent of the employees in the EU report that they have difficulties of great difficulty making ends meet (Gallie, 2002). The incidence of low-paid work has increased in several countries (OECD, 2006), and there has been a persistent increase of the proportion of working poor in the population and in 2005 8 per cent (15.5 million workers) of the EU-25 workers could be placed in this category
Along with pay, job security was an essential aspect of quality of work, early on (Martel and Dupuis, 2006). Both pay and job security provide a basic source of living. “Being employed and receiving adequate pay to make ones living is consistently ranked as important requirement for individual quality of life” (Beham et al., 2006: 20). Realising one’s potential, such as growth of skill, require time, and jobs of short and uncertain duration therefore normally are of low quality, and the same are jobs where the work itself is pervaded with uncertainty (Green, 2003). Having a job provides people with higher life satisfaction, and an absence of job security leads to greater levels of strain for individuals (Mak and Mueller, 2000), Job insecurity is a major source of ill-health and job dissatisfaction, has long-lasting impact on individuals and their households and creates tensions at home (Burchell, 1994; Burchell et al., 1999; Wichert, 2002). Physical and mental well-being continues to deteriorate the longer the employees remain in a state of insecurity (Burchell et al., 1999).
There is not one single measure of job insecurity, and conceptually, job security involves more than having a job versus no job. As the concept of job quality itself, job insecurity also has a multidimensional character. Job security depends on aspects of the current job as well as the possibility of alternative jobs ( employability ) (Green et al., 2000). Thus, job security involves more than the specific terms in the employment contract. Green (2006: 130) defines job insecurity as the loss of welfare that comes from the uncertainty at work, and this insecurity may come from either economic aspects of a job or from the content of the work itself. Uncertainty about the economic aspects of the job can involve more than just loosing the job, losses can also occur in the current job through wage cuts, missed promotion opportunities etc., but it also involves uncertainty about the income stream in the current or future jobs. Objective measures of job insecurity include separation rates (the rates at which employees leave jobs), redundancy rates (the rates at which employees are forced to leave their jobs), job tenure (time spent in one job), duration of unemployment and impact of job
loss on future pay (Green, 2003). Other measures are perceived job insecurity (fear of involuntary job loss, whether existing wages will be maintained), access to training, promotion opportunities and deployment of workers’ labour. Green (2006) also includes the risk of workplace accidents and diseases in his study of job insecurity.
Perceived job insecurity measured by involuntary job loss varies with the unemployment rate, by industry employment growth, local labour market environment, previous unemployment experiences and type of job contract, exposure to competition and ownership of the organization (Green, 2006).
Trends in job insecurity have been gradual and relatively modest (Green, 2003). Job insecurity is found to vary between countries, with business cycles, and is in general found to be much higher for temporary workers and for men. Job insecurity is associated with the type of job contract, and it is possible to identify insecurity by measures of the contract composition of the work force (Green, 2003). The use of temporary contracts has for example increased substantially in several European countries (Finland, France, Italy, the Netherlands and Spain) (Green, 2003). The cost incurred by the worker in the event that job loss takes place has broadly risen and fallen with the unemployment cycle (Green, 2006). The job insecurity will also be influence by worker protection in general in laws and rules enforced by the government or conditions negotiated by trade unions (e.g. replacement rate and length of unemployment benefits period).
Work intensity On average people work less than they did in the years after WWII, but there has been increased pressure on the time balance between work and non-work and in effort intensity while at work especially in the 1990s (Green, 2006). This has happened in almost all
The literature using the term work intensity is scarce. However, work intensity is closely linked to “job demands”– a core concept within psychological perspectives, and the measurements of these concepts sometimes overlap. Whereas job demands initially entailed psychological stressors (Karasek, 1979), using perceptual measures, it has become to embrace objective measures such as hours worked, overtime, etc. (e.g. Michie and Williams, 2003).
Job demands (and intensity) vary by occupation; when pace and tempo are included, a number of blue-collar jobs will be identified as high demand jobs, whereas when including long hours and overtime, a different picture is found (Kristensen et al., 2004).
Work intensity is sometimes (implicitly) used as change in intensity , for instance working longer hours (than before) (Anderson-Connolly et al., 2002). We argue that one should distinguish between level of work intensity and change in work intensity (work intensification).
Intrinsic job rewards Intrinsic job rewards differ from extrinsic job rewards such as pay and promotion in that the reward is derived from the job experience in itself. In terms of motivation, intrinsic (rather than extrinsic motivation) is based on the expected pleasure of the activity in itself rather than its results; and it is based on self-administered rewards rather than rewards distributed by an external agent (Shamir, 1996). Typically, an intrinsic rewarding job is interesting and challenging, one does a number of things at the job, one is able to use skills and abilities, one is able to learn new things, work independently and being recognized for doing a good job (Kalleberg and Vaisey, 2005; Huang and Vliert, 2002). Several intrinsic job characteristics have been found to be correlated with higher level of job satisfaction and subjective well- being. Examples are meaningfulness of work, clear and identifiable piece of work, task
complexity, whether the task is recognizable, task variety, pace, opportunity for use of initiative (e.g. Warr, 1987; Judge and Watanabe, 1993; Clark, 2005a). In his comparative study of quality of working life Gallie (2003) uses the following indicators of intrinsic job quality: 1) there is a lot of variety in my work, 2) my job require that I keep learning new things, 3) I have a lot to say over what happens in my job, and 4) my job allows me to take part in decisions that affect my work. His conclusion is that there is: “... a disturbingly low proportion of people who give the type of response that is indicative of a job of high quality” (Gallie, 2003: 66). Men and persons aged 35-54 were found to be in better jobs, while non-manual, skilled manual, and particularly non-skilled workers had jobs of poor intrinsic job quality. There are also differences between countries and Denmark and Sweden stands out as the ones with best intrinsic job quality both for men and women, while Portugal and Spain are at the other end.
Using the same data set (Employment in Europe Survey, 1996) Gallie (2002) finds that several groups are deprived in terms of the job characteristics associated with personal development. This is semi- and non-skilled workers who are in jobs with low task complexity and where little is required of qualifications and training, part-time employees who have few opportunities to learn now things and little ability to exercise discretion over the job, and older semi- and non-skilled workers who are a deprived group when it comes to ongoing learning opportunities. Based on the above mentioned criteria for personal development Gallie concludes that four out of ten jobs are of poor quality.
Skill Different academic fields approach the concept and measurement of skill in different ways. In psychology skill refers to the competence to perform specific tasks, while for sociologists the most important indication of skill is the degree of complexity of work, and economists looks
organization. On the other hand this optimistic picture of skill development has been challenged. Keep and Mayhew (1999) state that the economy may generate skills at different levels and of different types, some of which do not require a lot formal education. Whether the positive implications for skills encompass both the lower as well as the upper sections of society is an open question.
Even though a job requires a certain skill this may not match the skill which the person holding the job has. There can be a mismatch as workers are both under- and overqualified for the job. Even though under-qualification has lead to some concern, over-qualification is most discussed in the academic literature (e.g. Dolton and Vignoles, 2000; Brynin, 2002; Green and McIntosh, 2007). Both over- and under-skilling may have potential problems. Under-skilling may be a problem for the person if it leads to stress and lower work quality, and over-skilling can be a source of frustration if the workers can not use their qualification in the job and it may also mean a that they have to accept a lower wage than in a job with a better match. Overqualification may not be a big problem if it is only temporary, but both Dolton and Vignoles (2000) and Frenette (2004) have found that there is a tendency for workers to become stuck in such a state, and Brynin (2002) concludes that there is a structural tendency towards overqualification. Tendencies of both over- and underqualification have been observed for example in Britain (Green, 2006). About a quarter to a third of a nation’s employees tend to work in jobs for which they are overqualified, while the proportion having jobs for which they are under-qualified is somewhat lower (Green and McIntosh, 2007).
There are several ways of measuring workplace skills. Green (2006) has made a list based on an overview of the literature. This list includes the following measures: qualifications, length of education, occupation, scores from literacy and numeracy tests, self assessment and job requirements.
Autonomy and control Traditionally the study of autonomy and control has been the preoccupation of sociologists and psychologists and to a lesser degree a research theme for economists, but this has been changing as it has been shown to increase job satisfaction and productivity (ref.). Autonomy refers to the extent to which an employee is able to exercise discretion and initiative over what happens on the job. The degree of autonomy is an outcome of the manner in which the work is organized, especially the extent of standardization of work processes and whether the execution of work tasks is controlled through rules and procedures or surveillance systems.
This is vital due to the fact that job contracts are imprecise. According to Green (2006), there always remain some part of the planning to which the individual employee contributes. Employees also influence the execution of work. The latitude over the manner of performance of work is a matter of great interest for workers. If there were no discretion left, they might as well be a machine. To retain a certain element of discretion is a manner in which on may retain an element of humanity.
Thus, autonomy is related to control. Gallie et al. (1998) argue that there are three predictions about control. The first, which comes from the HRM movement, argues that there is a transition from strategy of control to strategy of commitment. The commitment strategy tends to deliver a higher level of effort and a lower level of alienation and conflict. The second prediction states that with the diffusion of information technology in both manufacturing and services work at all levels tend to become more cognitive in nature, responsibility is dispersed. Also, differences between managerial and subordinate roles are reduced. A higher level of discretion is related to a high-trust system of work organization. A third, contrasting
