Work Orientations,Well-Being and Job Content of Self-Employed and Employed Professionals

Abstract

Drawing on psychology-derived theories and methods, a questionnaire survey compared principal kinds of work orientation, job content and mental well-being between self-employed and organisationally employed professional workers. Self-employment was found to be particularly associated with energised well-being in the form of job engagement. The presence in self-employment of greater challenge, such as an enhanced requirement for personal innovation, accounted statistically for self-employed professionals’ greater job engagement, and self-employed professionals more strongly valued personal challenge than did professionals employed in an organisation. However, no between-role differences occurred in respect of supportive job features such as having a comfortable workplace. Differences in well-being, job content and work orientations were found primarily in comparison between self-employees and organisational non-managers. The study emphasises the need to distinguish conceptually and empirically between different forms of work orientation, job content and well-being, and points to the value of incorporating psychological thinking in some sociological research.

Keywords

job engagement job satisfaction person–job fit preferences professional workers self-employment values work orientations

Introduction

Hakim (1991: 113) described how research surveys ‘have repeatedly shown the importance of motivations, values and attitudes as key determinants of labour market behaviour, occupational status and even earnings … These “psychological” variables are too often omitted from [sociological] research, so their importance has been overlooked’. More recently, Brown et al. (2002: 1007) concluded that ‘it is in the understanding of why workers report feeling satisfied (or dissatisfied) with their job that sociology can make a positive contribution’. They advocate ‘an interdisciplinary, mixed-methods approach’ (Brown et al., 2002: 1016), including quantitative surveys, qualitative interviews and observational studies in order to examine individual as well as structural processes. Within that approach, the present article empirically examines aspects of self-employment by applying models and techniques drawn from psychology as well as sociology.

The European Union has around 32 million self-employed workers, some 11 per cent of its workforce (European Commission, 2010), and in the United States some 15 million fall in this category (Hipple, 2010).¹ Recent expansion in the United Kingdom means that almost 15 per cent of British workers – around four and a half million people – work for themselves (Office of National Statistics, 2017), and in Australia and New Zealand unincorporated self-employees make up 12 per cent and 16 per cent of the workforce.

Cross-national research has shown that a high rate of unemployment and the presence of strict legislation to protect permanent jobs are associated with more self-employment (Bögenhold and Staber, 1991; Hevenstone, 2010), but other evidence suggests that working for yourself is not disproportionately a refuge from being unemployed (Tatomir, 2015): the proportion of unemployed workers moving into self-employment across Europe is less than half the proportion moving in from an employed position (European Commission, 2016).

Self-employment can be partly driven by demographic influences on labour-force composition. For example, increases in lifespan and the lengthening of working lives can be followed by more self-employment, since, on average, people working for themselves are older than employees in organisations (e.g. European Commission, 2016). The importance of social position is also illustrated by the partly gendered nature of self-employment. As more women move into paid work, many (for example, motivated by family commitments) favour flexibility in their working hours and the possibility of income-generation through home-based self-employment (e.g. Carr, 1996; Jurik, 1998). In that respect, increases in the number of dual-earner families can encourage self-employment by removing the need for a self-employed income to be sufficiently large on its own to support an entire household. In some cases, part-time self-employment may be additional to working in an organisation.

Self-employment is also encouraged by structural developments such as the decline of industrial jobs and expansion of the services sector. Entry costs into new types of work can be substantially reduced, and potential markets for many products have expanded through increased globalisation and the widespread availability of information technology. It is also possible that societal shifts towards post-materialist personal values (emphasising autonomy and self-actualisation rather than security and possessions) are in some countries encouraging a shift away from traditional preferences in lifestyle and from employed work in an organisation.

The individualisation of jobs through self-employment can be encouraged by a neoliberal government that seeks to modify norms and values to create a national enterprise culture, focusing on active citizenship, personal independence and the creation of a socio-legal-economic framework that shifts attitudes and behaviours toward self-reliance and personal agency (e.g. Heelas and Morris, 1992). Viewing forms of national culture as pervasive socio-historical discourses open to alternative interpretations, Mallett and Wapshott (2015) have illustrated how societal enterprise themes can become embedded in personal identity.

Self-employed workers are more likely than organisational employees to be male, and around three-quarters of self-employees in the UK and USA are sole proprietors operating on their own without taking on staff (Baumberg and Meager, 2015; Dellot, 2014; Tatomir, 2015). Only a minority (17% in D’Arcy and Gardiner’s 2014 UK sample) would prefer to be employed, and in comparison with workers in organisations many self-employed individuals report a lower concern for job security (Warr, 2017). They frequently emphasise agentic themes – a desire for freedom and for working in a personally preferred field (Dawson et al., 2014; Smeaton, 2015). Part-time self-employment is more common among women than men, and self-employed mothers of young children particularly value personally convenient arrangements and time schedules (Craig et al., 2012; Hakim, 1988).

Self-employees on average undertake more weekly hours on the job (e.g. European Commission, 2016) and on average earn less than those in organisational jobs (e.g. Green and Mostafa, 2012). The nature of their work can increase variability in income across time, and it necessarily rules out pension, sick pay, holiday and other benefits from an employer (e.g. Schonfeld and Mazzola, 2015). The heavy workload, job insecurity and lack of benefits in self-employment might be expected to yield greater anxiety and strain than does working in an organisation, but comparative studies have found that levels of reported strain are generally similar in the two roles (Andersson, 2008; Baron et al., 2016; Baumberg and Meager, 2015; Prottas and Thompson, 2006; Tuttle and Garr, 2009). Between-role comparisons of interference from work activities to life at home are also in general statistically non-significant, although patterns differ between family situations, types of work and measures applied (König and Cesinger, 2015; Schieman et al., 2006).

Comparative studies have typically focused on well-being in terms of job satisfaction, finding that self-employed individuals on average are significantly more satisfied in their job than are organisational employees. That difference has been observed, for example, in Finland (Hytti et al., 2013), in Sweden (Andersson, 2008), in the United Kingdom (Baumberg and Meager, 2015; Blanchflower and Oswald, 1998; Smeaton, 2015; Warr, 2017), in the United States (Hundley, 2001; Katz, 1993; Prottas and Thompson, 2006), and across a range of countries by Benz and Frey (2008a, 2008b) and Lange (2012). It is now important to develop this general finding through differentiated research of the following kinds.

Differentiating between types of work and grades of job

Most research has examined self-employed people as a single category rather than looking separately at particular sub-groups. This article will instead distinguish between types of work undertaken and between job grades in organisational employment.

Self-employed activities range widely from unskilled and semi-skilled manual tasks to complex professional work. Around a third of UK self-employed workers (32% in the second quarter of 2015) are officially defined as managerial or professional, and after including associate and technical professionals that becomes 48 per cent (Office of National Statistics, 2015). Other countries’ different categories rule out direct comparisons, but in the USA around two-thirds of self-employed workers are in professional and managerial roles.¹ Activities and socio-educational backgrounds differ considerably between those individuals and other self-employees who undertake less conceptual or more practical work. There is a need for more targeted empirical research, and attention here is directed at self-employed workers who are in professional positions.

Individuals in that sub-set are expected from studies cited above to feel more positively about their job than do professionals employed in organisations. The article’s Hypothesis 1a thus proposes that, in full-sample comparisons between self-employed professionals and those in organisational positions, job well-being will be significantly greater in self-employment.

A second failure to make conceptually fruitful distinctions concerns hierarchical levels in an employing organisation. Studies of organisational workers have shown that both job content and satisfaction differ significantly across grades. Managers have more freedom to decide and act and are more involved in planning and co-ordinating, and they also report significantly greater job satisfaction than non-managerial employees (e.g. Rose, 2003). Given that activities in self-employment replicate in many respects those of employed managers and that managers are more job-satisfied than lower-level employees, the raised satisfaction of self-employed workers may not be significantly different from that of organisational managers. This article proposes that the typically found difference in job satisfaction is largely driven by the inclusion in organisational samples of substantial numbers of workers in lower job grades with associated lower job satisfaction. Based on that, Hypothesis 1b predicts that the full-sample difference in well-being expected in Hypothesis 1a will occur in comparison with organisationally employed individuals who are non-managers but not in comparison with employed managers.

The study also addresses different kinds of well-being. Almost all previous comparisons between well-being in self- and organisational employment have asked merely about job satisfaction, but the psychological literature has also identified and investigated several other well-being forms. For example, the circumplex model introduced by Russell (1980) characterises feelings through the degree to which they are physiologically and/or psychologically energised, being in an elevated ‘state of readiness for action or energy expenditure’ (Russell, 2003: 156). For example, a lower energy level is illustrated in job satisfaction (positive feelings which are relatively passive), but well-being in terms of job engagement involves raised energy and task-involvement (Bakker et al., 2011). Positive feelings at different levels of energy have been studied in terms of, for instance, enthusiasm and elation versus calmness and relaxation (Warr et al., 2014).

The more energised construct of job engagement has been studied in employing organisations but not previously in people who work for themselves. Although the feelings in engagement and satisfaction diverge in terms of energy, the two constructs overlap strongly (e.g. Christian et al., 2011) so that Hypotheses 1a and 1b can be applied to both job satisfaction and job engagement.

Differentiating between job features: Challenges and supports

Research into the nature of self-employed work has looked almost exclusively at personal autonomy, despite evidence from employed samples that a wide range of other job characteristics also contribute to well-being (e.g. Humphrey et al., 2007; Loher et al., 1985; Warr, 2007, 2013). This article will expand the focus beyond job autonomy in self-employment to distinguish between features which are personally challenging and those which are more supportive (e.g. Humphrey et al., 2007; Luchman and González-Morales, 2013).

Job features which challenge a worker include financial and organisational responsibility, competition with others, demanding tasks, difficult decision-making, and the requirement for innovation and personal independence. On the other hand, characteristics which may be described as providing support are illustrated by a comfortable workplace, job security and role clarity. All jobs contain a mix of these two kinds of feature, and the two sets differ in their impact on individuals (Warr and Inceoglu, 2012). The typically studied personal autonomy in self-employment is only one of many challenging aspects of work, and it is essential also to learn about other features in that category; a wider perspective will be applied here. Furthermore, self-employment research has yet to consider levels of supportive job content, and that omission will also be rectified.

It is expected from the nature of self-employed work and from research findings cited above that self-employed jobs as a whole will present greater challenge than will those in organisations. However, there is no reason to expect differences between self- and organisational employment in the many supportive aspects of work. Hypothesis 2a thus builds on earlier models to propose that, in full-sample comparisons between self-employed professional workers and those in organisational positions, self-employees will report the presence of significantly more job challenge but that no differences will be present in supportive aspects. In addition, consistent with previously documented differences between organisational grades, Hypothesis 2b predicts that the full-sample difference in challenging job content expected in Hypothesis 2a will be statistically significant in comparison with organisationally employed workers who are non-managers but not in comparison with employed managers.

These comparisons are important in their own right, but particularly in respect of the impact of job features on motivation and satisfaction. In separate empirical investigations, Benz and Frey (2008b), Hytti et al. (2013) and Lange (2012) observed that differences in task autonomy accounted statistically for the greater job satisfaction of self-employed individuals relative to organisational workers. We now need to learn whether the other challenging aspects of a job also have that effect – beyond merely autonomy. Furthermore, although differences in challenging features are expected to influence well-being comparisons between self-employed and organisational workers, there appears to be no reason to expect that job supports are important in that way. These possibilities will be explored here as a corollary to Hypothesis 2, expecting that other challenging features beyond merely autonomy also make a significant statistical contribution to the between-role difference in well-being, but that supportive aspects of a job (whose presence is predicted not to differ between self- and organisational employment) are not important in that way.

Differentiating between work orientations: Preferences for challenge and support

Goldthorpe et al. (1968: 36) concluded from their investigation of ‘affluent’ workers that job satisfaction ‘cannot be usefully considered except in relation to’ a worker’s orientation to his or her job, and this theme was central to Hakim’s (1991) interpretation of women’s greater job satisfaction despite objectively poor job content. Kalleberg (1977) similarly emphasised the importance for job satisfaction of workers’ orientations, which he presented as personal values in terms of ‘conceptions of what is desirable’. He subsequently concluded from a number of studies that ‘the impacts of particular job rewards on satisfaction are filtered through the subjective lens of one’s work values or the importance that people place on various economic and non-economic job rewards’ (Kalleberg, 2013: 165).

Following Kalleberg and other researchers, the terms ‘orientation’, ‘value’ and ‘preference’ will here be treated as interchangeable. They are in part culturally determined, being associated with individuals’ class setting, family life-stage, and the wider structural conditions linked to demographic characteristics. They can operate at personal, social or ideological levels (e.g. Rokeach, 1973) and take forms ranging in scope from broad world views to specific preferences – here for particular features of a job. Orientations to work became the subject of much debate in industrial sociology during the 1970s and 1980s, as scholars reviewed the principal determinants of social position and behaviour by contrasting structural constraints on a person with subjective meanings and preferences.

More recently, structural constraint and personal orientation have both become accepted as important, and researchers have envisaged their joint operation. Crompton and Harris (1998: 121) noted that workers ‘can choose but are also constrained, a fact which lies at the root of sociological explanations of human behaviour’. Vidal (2007: 271) observed that ‘individual orientations to work appear to be at least as important in determining worker satisfaction as job design’, and Rose (2003: 526) drew attention to orientations’ ‘fundamental influence’ on job satisfaction over and above a range of work and demographic factors. Other sociological investigators reaching similar conclusions include Hebson et al. (2015) (gendered and class-linked orientations in care work) and Zou (2015) (gender differences in job satisfaction associated with variation in work orientation).

However, work orientation research has rarely been extended to cover self-employment, and the few studies in this area have concentrated on the minority of self-employed people identified as entrepreneurs, usually defined as individuals who start and expand at least one business. Entrepreneurs have been shown to particularly value autonomy, self-direction, high financial rewards and personal impact on their income, and they tend to be more comfortable with competitive business activity and place less value on security, conformity and tradition (e.g. Beugelsdijk and Noorderhaven, 2005; Noseleit, 2010). Within the present framework of job characteristics viewed as challenges or supports, research has thus suggested that, relative to organisational workers, entrepreneurs prefer more challenging job features and are less concerned with job aspects here described as supports.

It is important to go beyond entrepreneur-only studies to investigate generally the orientations of workers who are self-employed, whether or not they start and expand companies. Evidence about desire for challenge among entrepreneurs suggests that self-employed people in general may prefer more challenging job characteristics, but that supportive features are likely to be less valued by individuals who work for themselves. Hypothesis 3a thus draws from entrepreneur-only findings to propose that, in full-sample comparisons between self-employed professional workers and those in organisational employment, self-employees will more strongly value job challenges and less strongly value supportive features. In parallel with other cases above, Hypothesis 3b predicts that the between-role difference in work orientation expected in Hypothesis 3a will occur in comparison between self-employed and organisational non-managers but not in comparison with managers.

Research into work orientations in organisations has pointed to the importance for job well-being of the closeness of fit between personal preferences and the characteristics of a job (e.g. Oh et al., 2014; Warr and Inceoglu, 2012). Extrapolating from that research restricted to employees, we would expect that the typically found greater satisfaction of self-employed relative to organisational workers is accompanied by a smaller discrepancy between the job features a person prefers and those which are actually present. Furthermore, conceptual and hypothesised differences (above) between challenge and support suggest that a between-role difference in preferred–actual misfit will occur in respect of challenging characteristics but not in respect of features which provide support.

That differential possibility will be investigated through Hypothesis 4a. This predicts that, in full-sample comparisons between self-employed professional workers and those in organisational employment, the misfit between preferences and job content will be significantly smaller in self-employment for challenging job characteristics but not for those which are supportive. In addition, previous hypotheses imply that the difference in misfit between preference and job content expected between self-employees and organisational workers in Hypothesis 4a will occur only in comparison with employed workers in non-managerial positions and not in comparison with managers. That is Hypothesis 4b.

In overview, this study draws from psychological models, methods and findings to examine issues which are important to the sociology of employment. Focus is on the job well-being and personal values of self-employed workers in professional positions in comparison with professionals employed in organisations.

Research methodology

Nationally representative surveys have to be constructed within tight financial and length constraints, and design priority tends to be given to issues and measures which are already established in a field, so that for empirical research into a new topic specialised samples and newly devised questions are often required. In the present case, no nationally representative surveys have covered the self-employment issues set out above, and future representative surveys are unlikely to provide the considerable interview time which is required to obtain the necessary detailed information. Following procedures that are common in psychology and other disciplines, a fresh targeted sample was therefore obtained.

Sample

Hypotheses were tested through a specialised website that is particularly appropriate for a focus on professional work. Operated by an internationally established personnel consultancy and hosted from the company’s head office in the United Kingdom, the website makes available without charge general guidance and illustrations about typical human-resource procedures to assess key personal attributes in professional and managerial work. Although it is not possible to ascertain response rate through this procedure, the appropriateness of information from the site is supported by the demonstrated validity of other material obtained in this way (e.g. Warr and Inceoglu, 2012).

Individuals accessing the site between the years 2008 and 2011 were invited to take part in this study and were offered entry to a prize draw. They were required to be in a job, and were assured that responses would be anonymous and securely stored. In addition to providing demographic and occupational information, participants completed online questionnaires (below) about their job-feature preferences, job content, satisfaction and engagement. Such a sample may not perfectly represent a national population, but it is excellent for this targeted examination of issues previously absent from all national surveys.

Overall, 4855 individuals took part, of whom 318 were self-employed and 4537 worked for an employer. Lower-level organisational employees have dominated previous samples in this area with only a small proportion of managers, but (deriving as intended from the website’s principal readership) more than 99 per cent of reported job titles were in business functions such as finance, project management, marketing, human resources and general management. A substantial number of managerial employees was thus available for separate examination – 2631 versus 1906 non-managers. Overall, 68 per cent of self-employed participants and 66 per cent of those employed in organisations had received a college or university education.

Principal countries of residence were United Kingdom (67%), Australia (15%), USA (6%) and New Zealand (5%), and all participants confirmed that they used English as their first language. Fifty-two per cent were male, and their average age was 39 years. Principal business sectors were retail, healthcare, finance, computer services and education, and seven per cent were in the public sector.

In comparison with respondents working for an employer, self-employed participants were on average older (means of 43 versus 39 years, p < .001), and a higher proportion were men (60% versus 51%, p < .01). Self-employees had been in their present job for an average of 6.44 years versus organisational workers’ 4.25 years (p < .001), the two groups’ average weekly working hours were 41 and 40 (n.s.), and their educational profile was almost identical.

In addition to comparisons between self-employment and organisational employment in terms of the entire sample of employees, separate analyses were undertaken of employees in four job grades: non-managerial workers (N = 1906), supervisors (N = 759), middle managers (N =1103), and senior managers and directors (N = 769).

Survey questions

Work orientations were examined in terms of preferences for each of 31 job characteristics selected to cover those shown in previous reviews to be linked to job satisfaction (e.g. Humphrey et al., 2007; Loher et al., 1985; Warr, 2007, 2013). For each feature, both preferred and perceived actual levels were examined through the same two items derived from a series of pre-test studies. Items were presented in a random sequence, separately for preferences and work content, and pairs of items were subsequently averaged for each job aspect; mean alpha reliability coefficients were .77 and .83 for the preference and actual scales respectively.

Primary categories were identified through principal component analysis with promax rotation applied to job content scales. Six components were found to meet the conventional criterion for extraction, having eigenvalues above 1.00. These were as follows:

Personal agency, six job features: personal autonomy, responsibility, opportunities for innovation, to express views and to have influence beyond one’s job, and the achievement of a high social position. The scale’s alpha coefficient was .87.

Competitive environment, five features: between-person competition, between-organisation competition, focus on finance, results-based payment, and opportunity to reach higher grades (alpha = .83).

Demanding tasks, five aspects: having difficult goals, a heavy workload, meeting high standards, using one’s expertise, and working long hours (alpha = .80).

Supportive environment, nine aspects covering physical, organisational and social support: a comfortable workplace, good equipment and resources, role clarity, non-conflicting goals, helpful colleagues, high ethical standards, fit with personal values, fair treatment of people, and the opportunity to learn (alpha = .86).

Desirable career, three features: job security, the opportunity to move ahead, and the opportunity for varied roles (alpha = .74).

Social relations, three features: networking opportunities, social interaction, and contributing to others (alpha = .76).

It can be seen that the studied job characteristics fell into two conceptual groups of the kinds introduced above. In factors 1, 2 and 3, covering 16 job features which challenge a worker, component characteristics require the expenditure of mental and/or physical energy, whereas the features in factors 4, 5 and 6 (supports, 15 aspects) are less demanding with a lower requirement for effort expenditure.

Participants first described their preference for each feature in their ideal job. For example, ‘In your ideal job, how much opportunity would you have to try out new ideas or procedures?’ and ‘In your ideal job, to what extent would you work in a competitive market?’. After all the 31 features’ preference levels had been reported, participants rated for the same items how much of each was present in their current job – the perceived actual level. For both preferred and actual levels the same nine response options were provided, ranging from ‘None at all’ to ‘The most possible’ and scored from one to nine. For each feature, the misfit between preferred and actual presence was recorded by subtracting a person’s perceived actual level from the level that he or she would prefer. A positive discrepancy score thus indicates that a preference level is greater than a perceived actual level – a person would like more of that characteristic.

In measuring job satisfaction, the study followed convention in this area and asked ‘Overall how satisfied or dissatisfied are you with your job?’, with seven response options from ‘extremely dissatisfied’ to ‘extremely satisfied’. In a review of survey questions used in organisational research, Fisher et al. (2016) placed this item in their most positive category for measurement reliability, validity and research utility.

Also following current procedures and definition (above), job engagement was investigated in terms of energy (three items, for instance ‘My job makes me feel energised’) and absorption (three items, such as ‘I get carried away by what I’m working on’). Engagement items were presented in a random sequence, with a target period of the past two months and nine response options that ranged from ‘never’ to ‘always’. Inceoglu and Warr (2011) have reported from three samples the presence of a single factor and high internal reliability (.91, .85 and .90); the present alpha coefficient was .92. The job engagement scale has clear content validity, and construct and criterion validity have been demonstrated by Inceoglu and Fleck (2010). As entailed by their conceptual overlap and as found by previous investigators (e.g. Christian et al., 2011), the measures of job engagement and job satisfaction were positively intercorrelated (r = .65).

Study findings

Hypothesis 1: Full-sample differences in job satisfaction and engagement

Findings in respect of Hypothesis 1a are shown to the left of Table 1. The table presents average well-being scores in self- and organisational employment, and then reports the statistical significance of differences after including demographic controls through general linear modelling. Columns 2 and 3 show that, although the full sample of self-employed workers reported greater job satisfaction than did organisational workers, this between-role difference was here not statistically significant. However, the difference in job engagement was larger and highly significant. Hypothesis 1a, comparing well-being between the two full samples, is thus supported for job engagement but not for job satisfaction.

Table 1.

Job engagement and job satisfaction in professional workers: average scores of all self-employed workers versus the full sample of organisational employees and versus employee sub-samples in different organisational grades.

	Self-employed workers (N = 318)	All organisational employees (N = 4537)		Non-managerial organisational employees(N = 1906)		Supervisors (N = 759)		Middle managers (N = 1103)		Senior managers and directors (N = 769)
	Mean	Mean	Controlled significance against Self-employed workers	Mean	Controlled significance against Self-employed workers	Mean	Controlled significance against Self-employed workers	Mean	Controlled significance against Self-employed workers	Mean	Controlled significance against Self-employed workers
Job satisfaction	4.79 (1.45)	4.63 (1.47)		4.43(1.50)	***	4.60(1.48)		4.75(1.48)		4.97(1.45)
Job engagement	6.09 (1.76)	5.57 (1.62)	***	5.22(1.81)	***	5.67(1.62)	***	5.81(1.57)	**	6.02(1.57)

Notes (Tables 1, 2 and 4): Controlled significance of difference against Self-employed workers: ** p < .01, *** p < .001; otherwise non-significant. Controls applied to between-group comparisons are gender, highest level of education, job tenure, years of work experience and average hours per week. Standard deviations are in brackets. Full-sample means are in bold font.

The right-hand columns of Table 1 report findings relevant to Hypothesis 1b. They show that among workers employed in organisations the lowest job satisfaction and engagement occurred at non-managerial levels, and that both forms of well-being became greater with each increase in organisational grade. Self-employees’ job satisfaction was significantly greater in comparison with employed non-managers, but was similar to employed workers in all managerial grades. Job engagement was significantly higher in self-employment than in all grades of organisational work except in comparison with the most senior managers.

Hypothesis 2: The presence of challenging and supportive job features

Previous comparative research into job content in self- versus organisational employment has looked almost entirely at personal autonomy, but the left-hand section of Table 2 compares each of the six job factors between self-employed workers and all employees. Autonomy is here within Factor 1 (Personal agency), and that factor was found to be significantly greater in self-employment than in organisational work. A significant difference was also present for the other two challenging factors – Competitive environment and Demanding tasks. On the other hand, between-role differences were non-significant for the three supportive factors 4, 5 and 6. Hypothesis 2a is thus supported.

Table 2.

Job-content factors in professional work: comparisons between self-employed versus all organisational employees and versus employees in different organisational grades.

	Mean perceived job factors and controlled significance of differences relative to all self-employed workers
	Self-employed workers	All organisational employees	Non-managerial employees	Supervisors	Middle managers	Senior managers and directors
Challenging job factors
Factor 1: Personal agency	6.01	5.34 ***	4.47***	5.56***	5.86	6.56***
Factor 2: Competitive environment	5.74	4.65 ***	4.17***	4.64***	4.91***	5.48**
Factor 3: Demanding tasks	6.51	6.15 **	5.67***	6.23	6.46	6.79
Supportive job factors
Factor 4: Supportive environment	5.84	5.69	5.63	5.66	5.72	5.82
Factor 5: Desirable career	4.76	4.70	4.48	4.87	4.76	4.98
Factor 6: Social relations	6.25	6.39	6.18	6.35	6.51	6.77

Notes: see Table 1.

Hypothesis 2b envisaged that the significant full-sample difference in job content between self- and organisational employees would be restricted to employees without managerial responsibility, but Table 2 indicates that greater challenges in self-employment were widespread across employed job grades. That hypothesis should here be rejected.

Enhanced autonomy has previously been shown in statistical terms to be a probable cause of the self-employed advantage in job satisfaction, and it was suggested above that other challenging features also contribute to that difference whereas supportive aspects of a job are not important in that way. Table 1 has shown that in the present full-sample analyses, job satisfaction was not significantly different between the two roles, so this possibility was instead examined in terms of job engagement. Findings were analysed through the procedure applied by Lange (2012) and Hytti et al. (2013), requiring parallel three-step regressions in which the first two steps are identical and the third step contrasts in separate analyses the presence of specified job factors. The process is illustrated in Table 3, which presents beta weights and overall statistics.

Table 3.

Job engagement in contrasting three-step regressions: beta weights and associated statistics from self-employment (Step 1), added demographic controls (Step 2), and separately added sets of job factors (either challenges in Step 3a or supports in Step 3b).

	Step 1:Self-employed or organisational workers	Step 2:Add demographic controls	Step 3a:Add challenging job factors 1, 2 and 3	Step 3b:Add supportive job factors 4, 5 and 6
Self-employed or organisational workers	−.09***	−.08***	.00	−.05***
Gender		.05**	.07***	.06***
Education level		−.01	.05***	.00
Job tenure		.02	−.02	−.01
Years of work experience		.11***	.03	.11***
Average weekly hours		.12***	.10***	.09***
Challenging job factors
Factor 1: Personal agency			.41***
Factor 2: Competitive environment			.02
Factor 3: Demanding tasks			.34***
Supportive job factors
Factor 4: Supportive environment				.42***
Factor 5: Desirable career				.09***
Factor 6: Social relations				.24***

R ²	.007	.043	.440	.466
Adjusted R²	.007	.041	.444	.465
Significance of R² change	.000	.000	.000	.000

The initial regression step in data-column 1 of Table 3 confirms the significantly greater job engagement of self-employed individuals versus all organisational workers shown in Table 1, and the second step again retains that significance despite the inclusion of demographic controls. Two separate forms of Step 3 are then presented for comparison, inserting either challenging job characteristics (Step 3a) or those which provide support (Step 3b). It can be seen that, as expected, statistical control for the presence of challenging job factors (3a) removed the significant difference in engagement between self-employed and other workers, whereas statistical significance was retained in regression 3b despite controlling for a job’s supportive characteristics. Although greater challenge in a job (factors 1, 2 and 3) statistically accounted for greater job engagement in self-employment as in previous studies of autonomy alone, that was not the case for supportive factors 4, 5 and 6.

This analysis was repeated for each job factor singly, with identical results. Each challenging job factor (1, 2 and 3) on its own accounted for the difference in workers’ engagement between self- and in organisational employment, but none of the supports (4, 5 and 6) did this. In sum, the introduction of job challenges together or singly removed the engagement difference linked to self-employment, but the inclusion of job supports together or singly had no such effect.

Hypothesis 3: Preferences for challenging and supportive job features

Findings in respect of Hypothesis 3a are in the first two data columns of Table 4. Despite predicted contrasting patterns between preferences for challenge versus support, no significant full-sample differences in job-feature preference were found between self- and organisational employment. Hypothesis 3b points to a possible source of this overall non-significance in terms of the sample’s high proportion of employees in managerial positions; perhaps the hypothesised preference differences occur only in comparison with organisational workers who are not managers. Data column 3 of Table 4 shows that, when compared to non-managerial organisational workers, self-employed individuals did indeed significantly more strongly value the three challenging job factors, but that for supportive work content differences in preference did not occur.

Table 4.

Preferences for each job factor in professional work, and misfit between preferred and actual levels: comparisons of self-employed workers against all organisational employees and against employees in different organisational grades.

	Mean job-factor preferences, and controlled significance of a difference relative to all self-employed workers						Mean misfit between factor preference and perceived actual level, and controlled significance relative to all self-employed workers
	Self-employed workers	All organisat-ional employees	Organisational non-managers	Supervisors	Middle managers	Senior managers and directors	Self-employed workers	All organisat-ional employees	Organisational non-managers	Supervisors	Middle managers	Senior managers and directors
Challenging job factors
Factor 1: Personal agency	6.80	6.84	6.53***	6.93	6.99	7.30***	.79	1.49 ***	2.06***	1.37***	1.12***	.74
Factor 2: Competitive environment	6.20	6.09	5.95***	6.11	6.11	6.41	.46	1.44 ***	1.78***	1.47***	1.20***	.93***
Factor 3: Demanding tasks	6.63	6.59	6.47**	6.61	6.61	6.84	.12	.44 ***	.80***	.38	.14	.05
Supportive job factors
Factor 4: Supportive environment	6.95	7.00	7.10	7.06	6.94	6.78	1.11	1.31	1.47**	1.40	1.22	.97
Factor 5: Desirable career	6.80	7.03	7.14	7.16**	6.90	6.80	2.05	2.33	2.66***	2.29	2.13	1.82
Factor 6: Social relations	7.00	7.13	7.04	7.17	7.14	7.30**	.74	.74	.86	.82	.63	.52**

Notes: see Table 1.

Hypothesis 4: Misfit between preferred and actual job content

Hypothesis 4a expected that misfit between preferred and actual job content would in full-sample comparisons be smaller for self-employees than for organisational workers in challenging respects (factors 1, 2 and 3), but that no differences in misfit would occur for job factors which provide support (4, 5 and 6). Consistent with this hypothesis, the right-hand section of Table 4 shows that the preferred–actual misfit was indeed significantly smaller for the full sample of self-employees in respect of challenges but not of supports. Hypothesis 4b proposed that the significant difference in challenge misfit would occur only in comparison with organisational non-managers and be absent in relation to managers. However, the right-hand columns of Table 4 report statistically significant greater misfit scores in several job grades – organisational managers as well as non-managers. Hypothesis 4b must therefore be rejected in this sample.

Discussion

Although labour market behaviours and institutions can depend on workers’ personal values (e.g. Hakim, 1991; Kalleberg, 2013), work orientations in self-employment have so far received inadequate empirical attention. Responding to the need for greater conceptual differentiation, the present study is the first in this area to distinguish between work orientations and job features that are either challenging or supportive. It has shown that professional self-employed work contains greater challenge in several forms but is similar to organisational employment in features identified as supportive. In respect of orientations towards job features of those two kinds, self-employed workers were found to value challenging job features more strongly than did organisational non-managers, and self-employment was accompanied by significantly better fit between preferred and perceived jobs in challenging but not supporting features.

The present specification of different categories of job characteristics and personal values takes this study beyond previous surveys. It is now important to develop the present findings in other samples of non-professional as well as professional workers. Linked to that, a general need is for longitudinal studies of entry into and continuation in self-employment. Although this investigation has pointed to the possible influence on job choice of preferences for challenge, causal evidence is still required about the impact of those orientations as ‘standards that guide conduct’ (Rokeach, 1973: 13). Longitudinal research is needed to learn whether people who more strongly value challenging job content are more likely to take up self-employment, whether those individuals are more likely to succeed in the role, and whether the impact of personal orientations is moderated by structural factors. Research in this area that brings together sociology and psychology can benefit both disciplines.

The study found that full-sample levels of job satisfaction were similar between the two roles, but that self-employed individuals were significantly more engaged in their work. How can this divergence between the two forms of job well-being be explained? It probably arises from the greater energy inherent in feelings of engagement that was described earlier. Job engagement is a more energised form of well-being than is satisfaction, which can come close to personal satiation, so that work tasks that are energised yield particular benefit to well-being which is also energised. Measuring well-being in the more energised form of engagement is thus likely to bring out differences between self-employed and organisational work that may be absent for less energised well-being – as in satisfaction. In general, it is important to move beyond a focus on job satisfaction alone to also examine other aspects of well-being.

Separate analyses of organisational employees at different grades revealed that non-managers in organisations experienced significantly lower well-being than did self-employed workers. Given that lower-grade employees make up the large bulk of most published samples, the typically reported difference in job satisfaction between self-employed and organisational workers as a whole is likely to derive largely from survey comparisons that emphasise lower organisational grades. In the present study, employed managers (with higher satisfaction) were proportionately more numerous than in other publications and thus contributed more to a higher average level of employed satisfaction. General statements about well-being in self-employment thus need to be refined to take account of differing values and psychosocial rewards experienced in different occupational grades. More generally, both psychology and sociology can benefit from consideration of models and findings in the other discipline.

Footnotes

Acknowledgements

The authors would like to thank the editor, anonymous reviewers and colleagues at the Universities of Sheffield and Surrey for their helpful suggestions to improve earlier versions of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Notes

Peter Warr is Emeritus Professor in the Institute of Work Psychology at Sheffield University Management School. His publications extend across many areas of social and applied psychology, with a particular focus on well-being and mental health. Emphasising settings of paid work, he has long been concerned to build on overlaps between psychology and other social science disciplines, for example in respect of unemployment, organisational processes, ageing and self-employment.

Ilke Inceoglu is Reader in Organisational Behaviour and Human Resource Management at Surrey Business School, University of Surrey. Her research focuses on employee well-being, work motivation, age differences at work and leadership. Prior to joining the University of Surrey she worked for SHL Group, a global psychometric testing company, where her research focused on the assessment of motivation, personality and performance in the work context.

References

Andersson

(2008) Happiness and health: Well-being among the self-employed. Journal of Socio-Economics 37: 213–236.

Bakker

Albrecht

Leiter

(2011) Key questions regarding work engagement. European Journal of Work and Organizational Psychology 20: 4–28.

Baron

Franklin

Hmieleski

(2016) Why entrepreneurs often experience low, not high, levels of stress: The joint effects of selection and psychological capital. Journal of Management 42: 742–768.

Baumberg

Meager

(2015) Job quality and the self-employed. In: Felstead

Gallie

Green

(eds) Unequal Britain at Work. Oxford: Oxford University Press, 105–129.

Benz

Frey

(2008a) Being independent is a great thing: Subjective evaluations of self-employment and hierarchy. Economica 75: 362–383.

Benz

Frey

(2008b) The value of doing what you like: Evidence from the self-employed in 23 countries. Journal of Economic Behavior and Organization 68: 445–455.

Beugelsdijk

Noorderhaven

(2005) Personality characteristics of the self-employed: An empirical study. Small Business Economics 24: 159–167.

Blanchflower

Oswald

(1998) What makes an entrepreneur? Journal of Labor Economics 16: 26–60.

Bögenhold

Staber

(1991) The decline and rise of self-employment. Work, Employment and Society 5: 223–239.

10.

Brown

Charlwood

Spencer

(2002) Not all that it might seem: Why job satisfaction is worth studying despite it being a poor measure of job quality. Work, Employment and Society 26: 1007–1018.

11.

Carr

(1996) Two paths to self-employment? Women’s and men’s self-employment in the United States, 1980. Work and Occupations 23: 26–53.

12.

Christian

Garza

Slaughter

(2011) Work engagement: A quantitative review and test of its relations with task and contextual performance. Personnel Psychology 64: 89–136.

13.

Craig

Powell

Cortis

(2012) Self-employment, work-family time and the gender division of labour. Work, Employment and Society 26: 716–724.

14.

Crompton

Harris

(1998) Explaining women’s employment patterns: ‘Orientations to work’ revisited. British Journal of Sociology 49: 118–136.

15.

D’Arcy

Gardiner

(2014) Just the Job – Or a Working Compromise? London: Resolution Foundation.

16.

Dawson

Henley

Latreille

(2014) Individual motives for choosing self-employment in the UK: Does region matter? Regional Studies 48: 804–822.

17.

Dellot

(2014) Salvation in a Start-up? The Origins and Nature of the Self-Employment Boom. London: Royal Society of Arts.

18.

European Commission (2010) Self-Employment in Europe. Brussels: Directorate-General for Employment, Social Affairs and Equal Opportunities.

19.

European Commission (2016) Employment and Social Developments in Europe 2015. Brussels: Directorate-General for Employment, Social Affairs and Inclusion. Available at: http://ec.europa.eu/social/main.jsp?catId=738&langId=en&pubId=7859&furtherPubs=yes (accessed 15 October 2016).

20.

Fisher

Matthews

Gibson

(2016) Developing and investigating the use of single-item measures in organisational research. Journal of Occupational Health Psychology 21: 3–23.

21.

Goldthorpe

Lockwood

Bechhofer

et al . (1968) The Affluent Worker: Industrial Attitudes and Behaviour. Cambridge: Cambridge University Press.

22.

Green

Mostafa

(2012) Trends in Job Quality in Europe. Luxembourg: Publications Office of the European Union.

23.

Hakim

(1988) Self-employment in Britain: Recent trends and current issues. Work, Employment and Society 2: 421–450.

24.

Hakim

(1991) Grateful slaves and self-made women: Fact and fantasy in women’s work orientations. European Sociological Review 7: 101–121.

25.

Hebson

Rubery

Grimshaw

(2015) Rethinking job satisfaction in care work: Looking beyond the care debates. Work, Employment and Society 29: 314–330.

26.

Heelas

Morris

(1992) The Values of the Enterprise Culture. London and New York: Routledge.

27.

Hevenstone

(2010) National context and atypical employment. International Sociology 25: 315–347.

28.

Hipple

(2010) Self-employment in the United States. Monthly Labor Review September: 17–32.

29.

Humphrey

Nahrgang

Morgensen

(2007) Integrating motivational, social, and contextual work design features: A meta-analytic summary and theoretical extension of the work design literature. Journal of Applied Psychology 92: 1332–1356.

30.

Hundley

(2001) Why and when are the self-employed more satisfied with their work? Industrial Relations 40: 293–316.

31.

Hytti

Kautonen

Akola

(2013) Determinants of job satisfaction for salaried and self-employed professionals in Finland. International Journal of Human Resource Management 24: 2034–2052.

32.

Inceoglu

Fleck

(2010) Engagement as a motivational construct. In: Albrecht

(ed.) The Handbook of Employee Engagement: Models, Measures and Practices. Cheltenham: Elgar, 74–86.

33.

Inceoglu

Warr

(2011) Personality and job engagement. Journal of Personnel Psychology 10: 177–181.

34.

Jurik

(1998) Getting away and getting by: The experiences of self-employed home workers. Work and Occupations 25: 7–35.

35.

Kalleberg

(1977) Work values and job rewards: A theory of job satisfaction. American Sociological Review 42: 124–143.

36.

Kalleberg

(2013) Good Jobs, Bad Jobs. New York: Russell Sage Foundation.

37.

Katz

(1993) How satisfied are the self-employed: A secondary analysis approach. Entrepreneurship Theory and Practice 17: 35–51.

38.

König

Cesinger

(2015) Gendered work-family conflict in Germany: Do self-employment and flexibility matter? Work, Employment and Society 29: 531–549.

39.

Lange

(2012) Job satisfaction and self-employment: Autonomy or personality? Small Business Economics 38: 165–177.

40.

Loher

Noe

Moeller

et al . (1985) A meta-analysis of the relation of job characteristics to job satisfaction. Journal of Applied Psychology 79: 280–289.

41.

Luchman

González-Morales

(2013) Demands, control and support: A meta-analytic review of work characteristics interrelationships. Journal of Occupational Health Psychology 18: 37–52.

42.

Mallett

Wapshott

(2015) Making sense of self-employment in late career: Understanding the identity work of olderpreneurs. Work, Employment and Society 29: 250–266.

43.

Noseleit

(2010) The entrepreneurial culture: Guiding principles of the self-employed. In: Freytag

Thurik

(eds) Entrepreneurship and Culture. Berlin: Springer, 41–54.

44.

Office of National Statistics (2015) Table EMP09: Employees and self-employed by occupation. Available at: http://www.ons.gov.uk/ons/rel/lms/labour-market-statistics/october-2015/table-emp09.xls (8 November 2015).

45.

Office of National Statistics (2017) Statistical Bulletin: UK labour market, February. Available at: https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/bulletins/uklabourmarket/feb2017#employment (3 March 2017).

46.

I-S

Guay

Kim

et al . (2014) Fit happens globally. Personnel Psychology 67: 99–152.

47.

Prottas

Thompson

(2006) Stress, satisfaction and the work–family interface: A comparison of self-employed business owners, independents, and organisational employees. Journal of Occupational Health Psychology 11: 366–378.

48.

Rokeach

(1973) The Nature of Human Values. New York: Free Press.

49.

Rose

(2003) Good deal, bad deal? Job satisfaction in occupations. Work, Employment and Society 17: 503–530.

50.

Russell

(1980) A circumplex model of affect. Journal of Personality and Social Psychology 39: 1161–1178.

51.

Russell

(2003) Core affect and the psychological construction of emotion. Psychological Review 110: 145–172.

52.

Schieman

Whitestone

Van Gundy

(2006) The nature of work and the stress of higher status. Journal of Health and Social Behavior 47: 242–257.

53.

Schonfeld

Mazzola

(2015) A qualitative study of stress in individuals self-employed in solo businesses. Journal of Occupational Health Psychology 20: 501–513.

54.

Smeaton

(2015) Self-employment: Calling the shots or hesitant independents? Work, Employment and Society 17: 379–391.

55.

Tatomir

(2015) Self-employment: What can we learn from recent developments? Bank of England Quarterly Bulletin Q1: 56–66.

56.

Tuttle

Garr

(2009) Self-employment, work–family fit, and mental health among female workers. Journal of Family and Economic Issues 30: 282–292.

57.

Vidal

(2007) Lean production, worker empowerment, and job satisfaction: A qualitative analysis and critique. Critical Sociology 33: 247–278.

58.

Warr

(2007) Work, Happiness, and Unhappiness. New York: Routledge.

59.

Warr

(2013) Jobs and job-holders: Two sources of happiness and unhappiness. In: David

Boniwell

Ayers

(eds) The Oxford Handbook of Happiness. Oxford and New York: Oxford University Press, 733–750.

60.

Warr

(2017) Self-employment, personal values, and varieties of happiness-unhappiness. Journal of Occupational Health Psychology (in press).

61.

Warr

Inceoglu

(2012) Job engagement, job satisfaction, and contrasting associations with person-job fit. Journal of Occupational Health Psychology 17: 129–138.

62.

Warr

Bindl

Parker

et al . (2014) Four-quadrant investigation of job-related affects and behaviours. European Journal of Work and Organizational Psychology 23: 342–363.

63.

Zou

(2015) Gender, work orientations and job satisfaction. Work, Employment and Society 29: 3–32.