Working the Good Life: How Personality Traits and Feelings of
Isolation Influence the Relationship between Freelancers’ Work
Locations and Subjective Career Success
Alexandra Blaes (10730583) University of Amsterdam Amsterdam Business School
Executive Programme in Management Studies – Leadership Track Thesis supervisor: Dr. W. van Eerde
Second reader: Dr. S.T. Mol Date: June 30, 2016
Statement of Originality
This document is written by Student Alexandra Barbara Blaes who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
This study investigated the influence of feelings of isolation, gender, extraversion and conscientiousness on the relationship between freelancers' work location and subjective career success. The general hypothesis was that freelancers who work at home more often, would report lower levels of career success. Data was collected from 99 freelancers in the Netherlands. Results indeed show that the assumed relationship between work location and subjective career success exists, with feelings of isolation acting as a mediator. Freelancers who work more at home, tend to have stronger feelings of isolation, which in turn leads to lower perceived levels of work-life balance, productivity, quality and meaningfulness of work and career and life satisfaction. Furthermore, the personality trait conscientiousness appears to play a central role for freelancers: It has a positive significant direct effect on five out of six subjective career success measures. No effects were found for gender and extraversion. Future studies that wish to investigate the relationship between work location and career success could focus in more detail on the psychological significance of specific work locations. For example: working in a low-light, messy and silent attic room at home may increase feelings of isolation, which in turn decreases subjective career success. A second recommendation is to examine possible interaction effects between work location characteristics and personality traits, to uncover whether different combinations lead to different levels of feelings of isolation or career success. The practical outcome of such studies could be the development of more personalised work location advice for
freelancers. Overall, in light of the ever-increasing flexibilisation of the labour market, this study suggests there is no one-size-fits-all model and more academic research is needed to untangle the intricate and optimal relationships between work locations and career success.
Keywords: Work location, freelancers, subjective career success, personality traits, feelings of isolation
1 INTRODUCTION... 1
1.1 THE FREELANCER ERA ... 2
1.2 RESEARCH QUESTION ... 4
1.3 RELEVANCE ... 5
2 THEORY ... 6
2.1 WORK LOCATIONS AND CAREER SUCCESS ... 6
2.1.1 Home ... 8
2.1.2 Public Space ... 9
2.1.3 Professional Shared Workspace ... 10
2.2 CAREER SUCCESS ... 11
2.2.1 Performance and satisfaction... 12
2.2.2 Work-life balance ... 13
2.3 GENDER AND PERSONALITY TRAITS ... 14
2.3.1 Gender ... 14
2.3.2 Conscientiousness and Extraversion ... 15
2.4 FEELINGS OF ISOLATION ... 16
3 HYPOTHESES AND CONCEPTUAL MODEL ... 17
3.1 WORKING OUTSIDE HOME ... 17
3.2 PROFESSIONAL SHARED WORKSPACES ... 18
3.3 CONSCIENTIOUSNESS ... 19
3.4 EXTRAVERSION AND FEELINGS OF ISOLATION ... 19
3.5 RESEARCH MODEL ... 20 4 METHOD ... 21 4.1 SAMPLE ... 21 4.2 DATA COLLECTION ... 22 4.3 MEASUREMENTS ... 23 4.3.1 Dependent variables ... 23
4.3.2 Independent and mediating variables ... 24
4.3.3 Sociodemographics ... 26
4.4 TEST OF VALIDITY AND RELIABILITY ... 26
4.5 ANALYSES ... 31
5 RESULTS ... 31
5.1 WORKING AT HOME ... 31
5.2 PROFESSIONAL WORK SPACES ... 37
5.3 FEELINGS OF ISOLATION AND EXTRAVERSION ... 41
6 DISCUSSION ... 46
6.1 MAIN RESULTS ... 46
6.2 LINK WITH PREVIOUS WORK AND THEORY ... 47
6.3 LIMITATIONS ... 48
6.4 RECOMMENDATIONS ... 49
REFERENCES ... 50
1
1 Introduction
Recent years saw the emergence and increasing popularity of flexible work locations for freelancers. Besides working from home or in coffee shops, a growing number of
freelancers are opting for more professional work locations such as shared office space companies and co-working spaces (De Vries & Van den Besselaar, 2013). Examples in the Netherlands of such locations are WeWork, Regus and Seats2meet. An important and often-heard advantage of working in a professional flexible work location, instead of working at home or in coffee bars, is that it improves productivity and work quality. A second advantage of working at such offices, or even at coffee bars, is that it allegedly improves work-life balance. Private and professional life can be separated.
However, scientific evidence for these supposed benefits is scant. Secondly, most studies examining the effects of flexible work locations on various job related outcomes have not distinguished between the various types of flexible work locations that exist. As the characteristics of flexible work location vary widely (e.g., working from home in peace differs significantly from working in a crowded and noisy public coffee shop), it can be expected to find different effects. Thirdly, although many studies have demonstrated a strong link between personality traits and job performance (Barrick, Stewart & Piotrowski, 2002; Seibert, Crant & Kraimer, 1999), it has also been found that the effects of
personality on performance may depend on specific working situations and contexts (e.g., Barrick & Mount, 1993; Judge & Zapata, 2014). This may thus also be the case for freelancers' work settings. Fourthly, people's attitudes towards jobs and careers have been changing substantially in the last two decades (Shockley, Ureksoy, Burcu Rodopman, Poteat & Dullaghan, 2016). Greenhaus and Kóssek (2014) in this respect note that employees that are pursuing a more contemporary career “are more likely to (...) seek
2
reduced-workload arrangements; telework; make career decisions that accommodate their family or personal circumstances.” (p. 365). These changes emphasize the need to include subjective measures in studies, commonly – and also in this study - referred to as
subjective career success (SCS; see Shockley et al., 2016). Lastly, it is unclear if the supposed benefits for one’s perceived career success are an underlying reason for the
growth in the number of freelancers (CBS, 2014; Fox, 2014; Storey et al., 2015). In theory, freelancers should be better able to benefit from their ‘freedom’ and should therefore be able to score higher on contemporary and subjective career success measures. However, relatively little is known about whether the (positive) effects of working from flexible work locations are found for both work-related outcome variables (such as perceived productivity) and life-related outcome variables (such as work-life balance).
This study aims to address a number of the aforementioned gaps in research. It does so by examining the relationship between the characteristics of freelancers’ work locations and a set of subjective career success measures, while paying attention to the role of personality traits and feelings of isolation.
1.1 The Freelancer Era
The number of freelancers in the Netherlands has been growing steadily, surpassing the 1 million in 2014. Roughly one out of six workers is a freelancer nowadays, compared to one out of nine at the end of the 90s (CBS, 2014). As a response, there is a growing market of companies that offer workspaces for freelancers. Although freelancers’ work locations vary, four general categories can be distinguished.
The first one is working from home. Working at home may be a cheap and easy solution but at the same time it can cause stress because of the risks involved in creating
3
imbalance in the so called ‘work-life balance’, the balance between work and non-work
roles (e.g., Mustafa & Gold, 2013). Working from home could also lead to feelings of isolation, which in turn could lead to a decrease in job performance and increase in strain (Golden, Veigen & Dino, 2008). This relation could even become stronger depending on certain personality traits, such as extraversion. One could imagine that people that show higher levels of extraversion, while working from home as a freelancer, are more likely to report feelings of isolation. This feeling of isolation could in turn influence their level of (self-reported) career success.
Second, working in public settings like coffee shops and libraries is another option. For some, this setting might reduce feelings of isolation, as they are being surrounded with other people even if these people are tourists or otherwise unknown to them. On the other hand, negative factors could be the noise, small tables, too much or too little light facilities, and the fact you are obliged to buy something.
A third possible location is the so-called professional shared workspace. Companies like Tribes in Eindhoven, Seat2Meet and market leader Regus serve the growing
population of freelancers. These locations are often ergonomically furnished and offer services like free Wi-Fi and printing facilities. Here freelancers have lots of opportunities to meet other professionals which possibly will inspire them.
A fourth freelancers’ location is the location of his temporal employer. For example,
corporates like telecom corporation KPN and bank ING allocate a certain amount of their workspaces for their temporarily work forces. Within KPN these work forces are referred to as ‘ander personeel’ (externals) in contrast with ‘eigen personeel’ (own personel).
4
1.2 Research question
Freelancers, and more specifically freelancers that can be categorized as ‘knowledge workers’, have been the topic of various studies. Although freelancers have been referred
to with different names and have been captured sometimes by varying definitions1, this study focuses on freelance knowledge workers (see also Burke & Cowling, 2015; Kitching & Smallbone, 2012; Van den Born & Van Witteloostuijn, 2012). This particular type of freelancers has received considerable attention in literature. Because freelancers have become a substantial part of the existing work force, scholars have become interested in the factors and settings that influence the performance and (perceived) success of freelancers.
In traditional organizational settings, High Performance Work Practices (HPWPs) are generally known to contribute to performance and well-being of employees. HPWSs are seen as a set of excellent HRM practices to motivate employees such that their
performance increases, which is claimed to contribute to a sustained competitive advantage of the organization (Kroon, Van de Voorde & Van Veldhoven, 2009). A large body of literature exists that has examined the effects of HRM practices on various performance related outcomes (for overviews, see Boselie, Dietz & Boon, 2005; Guest, 1997). However, despite their value and theoretical inspiration, the majority of studies does not provide enough insight into performance related outcomes in working situations (i.e., freelancers, self-employed) that lack any form of HRM practice. As a freelancer is not related to a traditional company, or at least does not have a similar role as a regular employee, there will be no impact of standard HPWPs of an organization on his or her career success.
1
Other names are skilled independent workers (Van den Born & Van Witteloostuijn, 2012), freelanced employees (Süss & Kleiner, 2010), non-traditional workers, freelance knowledge workers, enterprising self and self-employed workers.
5
This study therefore is interested in the working conditions for freelancers that may increase success. As the work location is the key characteristic that set freelancers apart from regularly employees, the focus of this study is the configuration of the work location. Besides the setting of the work situation, this study takes into account the possible
moderating and mediation roles of personality traits and feelings of isolation. The overall research question therefore is:
To what extent does the configuration of freelancers’ work locations relate to career success and how do personality traits and feelings of isolation impact that relationship?
1.3 Relevance
Besides the theoretical contributions that this study aims to make (see above), it also has practical and societal relevance. From a practical perspective, it is relevant for the growing population of freelancers and for organizations who are working in a setting Handy (1989) called “Shamrock Organizations”. Handy predicted that organizations of the future would
have three types of workers: employees with permanent work contracts, professional freelancers on project basis (i.e. the freelancers in my study), and contingent workforce doing routine jobs, thus a shamrock with three leaves. More and more Handy’s prediction came true the last decades. Already 25 years ago, he foresaw the substantial impact of technological innovation, the growing importance of networks of specialists, and the shift from production to services. Organizations are searching for sustainable and competitive strategies in a fast changing and dynamic world. Knowledge workers are key in
contributing to stay competitive. Davenport (2005) defines the term knowledge workers as follows: “Knowledge workers have high degrees of expertise, education, or experience,
6
and the primary purpose of their jobs involves the creation, distribution, or application of knowledge” (p. 10). Davenport underlines at the same time the decisive contribution of knowledge workers in the advanced economies by stating that they are the “greatest determinants of the worth of their companies” (p. 4).
In the next chapter, I will describe the body of theory I rely on, together with relevant previous empirical evidence related to my research question.
2 Theory
First, I discuss what academic research has shown about the characteristics of work locations, and more specifically the impact of these configurations on career success related outcomes. I will focus on what is known about three general categories of work locations: home, public spaces and professional shared workspaces. Next I provide an overview of the literature on career success. Then, I will discuss the three variables that possible moderate the relation between the work locations and career success, i.e. gender, conscientiousness and extraversion. Finally, I discuss the possible mediating role of feelings of isolation.
2.1 Work locations and career success
Little is known about the relationship between freelance workers, their work location, and the way the characteristics of the work location influence professional and personal performance. However, recently Vande Vrande and Hynes (2016, under review) conducted a case study on flexible work and show that the characteristics of a specific shared workspace (namely, Seats2meet) can lead to positive business outcomes and
7
personal outcomes. However, their study is limited to one specific co-working space and was therefore not able to distinguish between the defining characteristics of flex working spaces. Still, the study provides relevant insights because the study was conducted at Seats2Meet, a professional shared workspace. Results show that some freelancers benefit from working in a professional shared workspace: It significantly improved business outcomes as well as personal outcomes. But what can we expect of these outcomes when these same professionals work at home or at a public space (instead of a professional space)?
Van den Born and Van Witteloostuijn (2012) conclude that “the external environment in which an individual freelancer operates is the most important factor
determining career success” (p. 24).They underline that their study is one of the first large-scale quantitative studies with freelancers as sample group. Work location, however, was not part of their study. It thus seems to be relevant to include work location as a key predictor variable and examine in what way different work locations influence career success. Career success is measured by Van den Born and Van Witteloostuijn in two ways. Revenue was the main indicator for objective career success. Satisfaction, personal
capability and motivation was used to measure subjective career success. De Been and Beijer (2014) determined whether the type of office environment has an impact on satisfaction with the office environment and productivity support. Their results show that the influence of office type is a significant predictor of satisfaction and perceived
productivity support, although “satisfaction with the organization explains the most variance with regard to satisfaction with the office environment and productivity support” (p. 142).
In this study, freelance work locations are divided into three general categories: Home; Public Space (e.g., Coffee Company, Bagles and Beans); Professional Shared
8
Workspace (e.g., Seats2Meet, Regus, WeWork). These locations have different and common characteristics, as presented in Table 1.
Table 1
Characteristics of the three work locations
Category Social setting Work setting
Home Alone Private
Public Space Group Private
Professional Shared Space Group Professional
2.1.1 Home
The first location is working from home. As shown in Table 1 this is a space where people work alone and private. Working from homeof course has both positive and negative aspects. It can be a cheap solution or even a practical one when one needs to work concentrated without noise or other distractions. On the other hand, it can lead to stress when the private environment is combined with professional demands, which can lead to a problematic work-life balance (Eikhof, 2016; Mustafa & Gold, 2013). Eikhof (2016) suggests that modern ICT does not only facilitate working outside traditional office hours and workplaces but also supposes people to be available 24/7. Eikhofstates: “The ICT-facilitated blurring of boundaries between work and non-work results in conflicts from work to family-interference and family to work-interference” (p. 369).
Working at home can cause also procrastination, as people can become distracted by private demands. Van Eerde (2003) typifies procrastination as ‘avoidance behavior and can be seen as the avoidance of the execution of an intended action’ (p. 422). Farrington (2012)
9
emphasizes the serious impact of procrastination on performance and posit it as follows: “Putting something off while expecting negative consequences for doing so not only
provokes internal suffering, but it also results in missed deadlines and opportunities, suboptimal work, and reduced creativity” (p. 15). I already had some idea about this phenomenon as I discussed my research question with fellow students. They already confirmed that for them procrastination does play a role in studying or working at home. As one student stated: “Oh for me, I get distracted at home, I do the laundry, cleaning of the house and other non-work or study related activities”. It has been suggested that gender, namely being female, could even produce stronger effects (Eikhof, 2016). Another consequence of working at home could be the danger of developing feelings of isolation, which in turn may negatively impact performance or success. Golden, Veiga and Dino (2008) in this respect note that “greater isolation is associated with lower performance” (p. 1416).
2.1.2 Public Space
The second location are public spaces like coffee shops (such as Starbucks and Coffee Company) where everyone can sit down, take a coffee or have lunch. In this space people are in a group but it remains a private space to work in as shown in Table 1. More and more people open their laptop in these spaces and are either connecting for private business, study or working on a professional task. Wi-Fi is available for free; there is no privacy or possibility to work in silence. Although in public spaces one is surrounded by other people, it still remains an individual work environment. Van de Vrande and Hynes (2016) emphasize that feelings of isolation do still exist in public spaces: “public spaces are unlikely to help overcome feelings of isolation due to strong social norms and
10
boundaries: the idea of randomly engaging in conversation with someone at Starbucks about business ideas would be seen as at least strange, and at best presumptive” (p. 4).
2.1.3 Professional Shared Workspace
The third work location category is the professional shared workspace, which can be defined as work locations specially designed for freelancers. Here people are in a group and surrounded by other professionals. Facilities like high-speed internet, printing, lounges and high quality coffee. are offered at various prices. At WeWork in Amsterdam, prices vary between 190 Euro for a desk per month and 350 Euro for an office. In some cases, one can take place without paying any fee or just a small fee for free coffee and lunch like at Seats2Meet. Van de Vrande and Hynes (2016) suggest that “Co-working spaces can be designed to facilitate interaction and networking, thereby reducing feelings of isolation and contributing to the personal well-being of teleworkers” (p. 5). An important motivation for working in shared workspace is that people like to be surrounded by other people and the need to separate work and private lives (Van de Vrande & Hynes, 2016).
The aim of this study is to untangle the relationships between characteristics of flexible work locations (see above) on the one hand and career success on the other. Below, I provide an overview of existing studies that have focused on career success, both from more performance or work related perspective and from a personal or private
perspective. More specifically, I will discuss the theoretical and methodological relevance in terms of defining and measuring career success in objective and subjective perspective. Although both concepts have been the subject of a large number of studies, not all studies or measurements are relevant for the research question in this study. Moreover, the majority of studies that have examined performance, professionally and personally, have been carried out in the context of regular or traditional work locations (the office) amongst
11
regular employees. More recent studies investigated performance in the context of New Ways of Working (NWW), and offer interesting insides for freelancers’ work context. Given the different context (flexible work locations) and population of this study (freelance workers); it is important to critically review existing studies for their relevance. In the next two paragraphs I discuss the main dependent variable of this study, namely subjective career success.
2.2 Career success
Career success in this study will be based on the widely accepted definition of Arthur, Hall and Lawrence (1989), describing career as an unfolding sequence of a person’s work experience over time. Career success is an outcome of a person’s career experiences
(Arthur et al., 2005).
Career success at present is not anymore a one-dimensional concept. Individuals are less bound to just one organization and directing increasingly to a protean career (see Hall, 2004). A protean career is a career in which the person and not the organization, is in charge. The core values of the person influence his career decisions and so the main success criteria are subjectively related (Hall, 2004). The most common way to tap into career success of freelancers, and also the most appropriate way for this thesis, is to use a number of self-rated measures in terms of attitudes and opinions of one’s career. The overall term that is typically used in this respect, is Subjective Career Success (SCS). SCS encompasses career perceptions and attitudes towards career satisfaction, recognition, quality of work, personal life and work-life balance (for an overview, see Shockley et al., 2015). A general distinction within SCS measures can be made between measures that relate to people’s views on their job performance and satisfaction on the one hand (so a
12
professional perspective), and measures that relate to people’s perceived work-life balance (a more personal or private perspective).
2.2.1 Performance and satisfaction
Performance rating in professional work settings is in most cases done by supervisors (Judge, Thoresen, Bono & Patton, 2001). However, as freelancers are at the same time employee and supervisor, consequently most of the measurements of this research will be self-rated measures for professional and private related outcomes. Another relevant factor to take into account, is the variety of the industries freelancers are working in. Although researchers typically use the term freelancer as working in creative and media occupations, it is argued that freelancers in other knowledge-intensive work domains should also be included (Kazi, Yusoff, Kahn & Kazi, 2014). Following Kazi et al., in this study
freelancers are defined as workers who are self-employed, apply for projects and work for different employers. Thus, these different kind of freelance jobs implicate constructing generic performance measurements that apply to freelancers from a broad range of jobs and industries. I will discuss a number of articles that vary in terms of their target group and outcome variables.
Van de Vrande and Hynes (2016) measure a number of performance related outcomes of people working in professional shared workspaces. Questions were posed regarding contacts with co-workers, improvement or creation of products and services and finally finding new customers. Additionally, the relationship between the positive
outcomes in the co-working space and personal characteristics of individuals was examined.
Van den Born and Van Witteloostuijn (2013) studied the drivers of freelance career success. Various indicators of career success (perhaps slightly adjusted) could form
13
important input for the measurement of performance in this study. The authors use self-rated perceived productivity, satisfaction, motivational factors and turnover intentions as their subjective success variables. Similarly, Vischer (2008) found that behavioural or outcome measures common in a work environment include employee satisfaction and employee productivity. Furthermore, Van den Born and Van Witteloostuijn used human capital as a variable. This variable entails the ‘knowing how’, which is considered a determinant of success as well as of entrepreneurial performance. Knowing how reveal skills and job-related knowledge. Human capital was evaluated through total work experience and total freelance experience, educational level and finally the number of training days in core skills, new skills and supporting skills of the freelancer.
Another relevant study was conducted by De Been and Beijer (2014), who measured satisfaction with respondents’ work locations and its impact on productivity, based on the WODI light toolkit (see Volker & Maarleveld, 2007).
2.2.2 Work-life balance
Work-life balance is the second component of subjective career success for this study. Work-life balance can be defined as a balance between work and non-work roles (Boell et al., 2013). During the last 30 years, changes in information and communication technology have had a significant and increasing impact on work and private life (Demerouti, Derks, Ten Brummelhuis & Bakker, 2014). Demerouti et al. conducted a study on the
consequences of new ways of working (NWW), which entails flexible work design and freedom or work location and working hours. In other words, the impact of flexible work locations on work-life balance was the key focus of this study. The authors conclude that the main dimensions of NWW (such as telecommuting and flexible work schedules) have both positive and negative effects for work-family balance. Conditional factors like being
14
single or having a family may influence work-life balance. Johns and Gratton (2013) notice that “Always on” technology can cause more imbalance in employees work and life
balance. Davis, Shevchuk and Strebkov (2013) investigated work-life balance specifically for (Internet) freelancers or as they call them non-standard workers. They conclude that “the more hours a freelancer works, the less he/she was satisfied both in general and with
work-life balance”. (p. 551). Having small children had no effect on work-life balance for men in contrast to women who reported less work-life balance. Furthermore, Davis et al. conclude that high levels of professional mastery influenced positively work-life balance. Finally, results show that the more freelancers earn they more they become anxious about work-life balance. Apparently more income increases the pressure to keep the standard of living.
2.3 Gender and personality traits
In this section, I discuss a number of factors that, together with work location, may impact – or at least have a relationship with - career success.
2.3.1 Gender
Eikhof (2016) states that knowledge work and flexible working can pose considerable unseen obstacles to working women. A growing share of knowledge work is undertaken in non-standard or precarious employment relationships (Cappelli & Kallebeg, cf in Eikhof, 2016) like small business and freelance work. These types of work offer little revenue security and parental leave or work-life balance support tends to be minimal or non-existent. Spending time looking for work is an important part of the job. It entails considerable investment in the development of skills and knowledge but it does not
15
automatically offer security or work (Eikhof, 2016). Due to this insecurity, women tend to provide income in several ways such as teaching. Women tends to use this strategy more than men (Gill, as cited in Eikhof, 2016). All together, these challenges and obstacles are better controlled by men. Obstacles of flexible working are not specifically related to gender. Yet, women will opt more for flexible work in order to be able to combine private and professional life. Going to a shared space gave freelancers the feeling of going to work which supported the need to separate work and private life (Van den Vrande & Hynes, 2016).
2.3.2 Conscientiousness and Extraversion
A large strand of literature exists that has demonstrated a link between personality traits and performance at different levels. More specifically, the personality traits extraversion and conscientiousness have been linked to job performance in a number of studies (e.g., Barrick, Stewart & Piotrowski, 2002; Seibert, Crant & Kraimer, 1999; Seibert, Kraimer & Liden, 2001). Summarizing their review of many years of personality and performance research, Barrick, Mount and Judge (2001) state: "In this study we quantitatively summarize the results of 15 prior meta-analytic studies that have investigated the
relationship between the Five Factor Model (FFM) personality traits and job performance. Results support the previous findings that conscientiousness is a valid predictor across performance measures in all occupations studied." (p 9).
At the same time, Barrick et al. (2001) call for more detailed studies to examine specific effects of personality traits, in specific contexts for specific groups: "The studies upon which these results are based comprise most of the research that has been conducted on this topic in the past century. Consequently, we call for a moratorium on meta-analytic studies of the type reviewed in our study and recommend that researchers embark on a new
16
research agenda designed to further our understanding of personality-performance linkages." (p. 9). Following their general advice, my study will exactly do just that: I will take into account the role work locations have in the relationship between personality and performance (defined as career success in my study), while also adding more
contemporary measures of performance or career success.
Demerouti et al. (2012) too suggest to investigate the role of personality in relation to objective career success like revenue and subjective career success like work-life balance. When people are very disciplined (i.e. conscientious), they will possibly work efficiently at home and be able to plan their workday. Conscientiousness is a dimension of the Big Five Personality Model. If an individual is scoring high on extraversion, this person will miss the contact with other people and this will negatively influence performance and work-life balance. The study of Van Eerde (2003) on procrastination revealed a strong correlation with conscientiousness and self-efficacy, both characteristics needed to perform as good as possible.
2.4 Feelings of isolation
Feelings of isolation is a possible variable mediating the relationship between one’s work location and his productivity. Knowledge workers are able to perform their virtual work anywhere and whenever they want. But workers’ isolation is seen as a side effect of this virtualization. Shared spaces should ameliorate workers’ isolation (Johns & Gratton, 2013). Spinuzzi (2012) reported feelings of isolation, distractions and self-motivation problems under workers working from home. In order to make people feel better, to be surrounded by other people, seems to be an effective solution of this problem (Van de Vrande, 2016).
17
Before the Industrial Revolution, people worked at home or close to home. Since the discovery of the steam engine, work was displaced more and more to the factories. At present we see a reverse phenomenon, there is less need to be present in an office due to telecommuting (Harpaz, 2002). Advantages of communication and technology innovations are for example autonomy, independence and flexible working hours. On the other hand, disadvantages are feelings of isolation, no separation between private life and work at home, need for self-control and over-availability syndrome (Harpaz, 2002). In the next section I will formulate my hypotheses and I present the conceptual model.
3 Hypotheses and conceptual model
Theory suggests that, among the many factors that may impact career success, work location can be an important one. Therefore, I will first formulate the hypotheses on how work location may be related to career success. At the end of this chapter I propose the complete conceptual model. The relation between work location and career success will be influenced by several factors, as discussed in the theoretical part of this study. Based on the theory, it can be expected that - in some cases - gender and conscientiousness influence the relation between work location and SCS. Additionally, it can be assumed that feelings of isolation mediates the relationship between work location and (some aspects of) SCS.
3.1 Working outside home
Demerouti et al. (2014) suggest that the blurring boundaries of work and private life lead to undermining work-family balance. Harpaz (2002) refers to this phenomenon as ‘no
18
telecommuters are expected to report blurred work and family boundaries and more role overload and stress. Furthermore, they emphasize the need to establish boundaries between work and personal/family life. It is therefore interesting to view this recommendation in light of Van de Vrande and Hynes’ (2016) findings, who indeed found that shared spaces gave freelancers the feeling of going to work which supported the need to separate work and private life. I expect that gender will also play a role in this relation, based on the findings of Eikhof (2016) who emphasizes the impact of freelance work on women who have very little support to work-life balance and therefore are better off working outside home. This leads to my first set of hypotheses:
Hypothesis 1A: Working outside home is positively related with SCS.
Hypothesis 1B: The positive effect of working outside home on SCS is stronger for female freelancers
3.2 Professional shared workspaces
Where do freelancers perform at their best? Is there a significant variance between working at home, at a public space or in a professional shared workspace? The study of Van de Vrande and Hynes (2016) show that working at a professional shared workspace improves business outcomes as well as personal outcomes. Moreover, they suggest that professional shared workspaces enable communication with other freelancers, which too will increase professional outcomes. Vischer (2008) underlines the importance of environmental aspects of the workspace, which positively influences outcomes measures. Johns and Gratton (2013) define professional workspaces as community-based and low-cost spaces
specifically designed to connect and inspire people who generally work alone, and at the same time, they support productivity by giving a sense of community. Based on the
19
findings above I suggest that working at a professional shared workspace will have a positive impact on SCS. This leads to my second hypothesis:
Hypothesis 2: Working at a professional shared workspace as opposed to other locations positively impacts SCS
3.3 Conscientiousness
As mentioned before, conscientiousness (strongly related to procrastination) could influence the relationship between the independent variable work location and the dependent variable SCS. Barrick, Mount and Judge (2001) confirm the importance of conscientiousness by stating that conscientiousness can be seen as a valid predictor across performance measures. Conscientiousness implies a person is organized and efficient (Demerouti, 2012). People who are very organized and efficient could suffer less from the negative consequences like feelings of isolation, influences of porous boundaries of work and private life or procrastination when working at home. They are better capable of keeping focus on the work-related activities and keeping the boundaries between work and non-work (Boell et al., 2012). Based on these findings I suggest that working at home will have a negative effect on SCS, the negative effect of working at Home on SCS will be less for people who score high on conscientiousness. This leads to my third hypothesis:
Hypothesis 3: Higher values of conscientiousness diminish the negative effect of working at home on SCS
3.4 Extraversion and feelings of isolation
Working from home has been found to lead to feelings of isolation, which in turn may influence subjective career success (Boel et al., 2013; Golden et al. 2008; Harpaz, 2002;
20
Spinuzzi, 2012; Van de Vrande & Hynes, 2016). If an individual is scoring high on extraversion, this person will miss the contact with other people, and this will negatively influence performance and work-life balance (Demerouti et al., 2013). Higher values of extraversion will reinforce feelings of isolation. This leads to the following hypotheses:
Hypothesis 4A: Working from home has a positive relationship with feelings of isolation, which feeling in turn is negatively related with SCS
Hypothesis 4B: The positive relationship between working at home and feelings of isolation is stronger for freelancers that are more extravert.
3.5 Research model
In the above section four sets of hypotheses were established. Figure 1 depicts the full conceptual model.
Figure 1
21
In the next section, I describe the method, the sample and data collection. I will identify the variables and measures used in the conceptual model.
4 Method
The aim of this study is to find out what the impact is of freelancers' work locations on their subjective career success. Work location will therefore be the key independent
variable, while six subjective career success (SCS) measures are the dependent variables of interest.
4.1 Sample
The research population of this study are freelance knowledge workers. Freelancers have traditionally often been found working in the media industry, but the IT industry has gained a big part of this population (Süss & Kleiner, 2010). For this study I will use a contemporary and common definition of knowledge freelancers: knowledge workers who are offering work that consist of intellectual, innovative and creative talents as key
resources (Darr & Warhurst, 2008). Kazi et al. (2014) argue that researchers should include freelancers working in all kind of knowledge-intensive work domains. Davenport (2005) defines knowledge workers as people who have a high degree of expertise and education and for whom the principal drive, is to create, distribute and applicate knowledge. Thus, the data will be collected among freelance knowledge workers in the Netherlands.
22
4.2 Data collection
The data collection to test the hypotheses was of a quantitative nature and was collected through a cross-sectional survey. The survey was designed in Qualtrics. Most of the questions to be filled in made use of a Likert scale from 1 to 5.
Data was collected through my own network, which consists of freelance teachers, trainers and the professional networks of fellow students, friends and relations.
Furthermore, I started a temporarily blog called ‘www.waarwerkje.nl’ with information about my survey and a direct link to my survey. I visited two times Seats2 Meet in Utrecht, a professional shared workspace where I approached every individual with the request if they were a freelancer and if they were willing to do me a favour. Almost every person reacted positive. To be sure they would fill in my survey I gave them a flyer with some information about the survey and with the link. I gave the link the name
‘bit.do/waarwerkje’ which was easy to remember. I visited a freelance network meeting in ‘Pakhuis de Zwijger’ in Amsterdam where I got in touch with two freelancers with a large
network, who promised to distribute my request on Twitter and to put a link in their newsletter.
Some sociodemographics of this sample show that 57 persons are male and 44 women, the mean age is 45, mean work experience is 21 years and mean freelance
experience is 8 years. While almost 3 out of the 4 respondents indicate to work primarily at home, 1 out of 4 say they work primarily outside home, or at least as much outside home as inside home. Most freelancers (n=25) are working in media, followed by ICT (n=13), creative business (n=11), organization and advice (n=18) and education and training (n=10). Around 25% filled in different occupations. Most freelancers have finished a form of higher education, such as HBO (n=38) and WO (n=56).
23
Out of the 113 respondents that started filling in the survey, 101 respondents
completed the full survey. This results in a completion rate of 89%. In the next section, the measures are described.
4.3 Measurements
4.3.1 Dependent variables
As subjective career success (SCS), the key outcome variable of this study, is a multi-faceted concept, it was measured using six different variables. Four of them originate from Shockley et al.’s (2015) SCS measures, while two other SCS measures (work-life balance and perceived productivity) are adapted from other established measures of career success.
Quality of work, meaningfulness of work, personal life and satisfaction.
Measurements of SCS from Shockley et al. (2015) were used to measure quality of work, meaningfulness of work, personal life and satisfaction. Three questions about quality of work were asked: “I am proud of the quality of the work I have produced”; “I have met the highest standards of quality in my work” and “I have been known for the high quality of my work”. Three questions to measure meaningfulness of work were asked: ”I believe my work has made a difference”; “I think my work has been meaningful” and “The work I have done has contributed to society”. Three questions about personal life were asked: “I
have been able to have a satisfying life outside of work”; “I have been able to spend the amount of time I want with my friends and family and “I have been able to deliver good
work while maintaining quality non-work relationships”. Finally, three questions about satisfaction were questioned: “I am enthusiastic about my career”; “My career is personally satisfying” and “I have found my career quite interesting”.
24
Work-life balance. For work-life balance, the measurements were based on Hill, Hawkins, Ferris and Weitzman (2001): “At home I have sufficient time outside of work to find a proper balance between my work and my family and private life”; “I am able to find a proper balance between the demands of my work and my family and private life”; “When I take a vacation, I am able to separate myself form work and enjoy myself’; “I feel often drained when I go home because of work pressure”; “When working I am being disturbed
regularly with non-work related issues” and “I am able to combine my private demands with work related issues”
Perceived productivity. Perceived productivity, which is the opinion of the freelancer about being productive, was questioned through a six items scale developed by Staples, Hulland and Higgens (1998). Five of them were used: “I am a very effective freelancer”: “I am happy with the quality of my work output”; “I work very efficiently”; “I am a highly productive freelancer” and “Among other freelancers, I would rate my performance in the top quarter”. The sixth item concerned the relationship between the manager and the
employee and was therefore left out.
4.3.2 Independent and mediating variables
Work location. First, respondents were asked to fill in the total amount of work hours per week, after which they could divide these hours into different work locations (in
percentages) in order to find out which work location is the main location. Respondents that indicated to work at least 40% at professional shared space or at the location of their temporary employer (or a combination of both), were categorized as professional space worker (=1), the others were given the value 0. An additional question was used to ask respondents to indicate the extent to which they worked at home or outside their home. A
25
5-point scale was used, ranging from Always at home (=1) to Always outside home (=5) and categorized as working outside home.
Personality traits. In the past years more and more short personality instruments have become available which are related to the Big Five Model (De Vries, 2013). Recently there is evidence for six main personality dimensions referred to as the HEXACO
dimensions of personality. HEXACO is the acronym for Honesty-Humility, Emotionality, eXtraversion, Agreeableness, Conscientiousness, and Openness to experience. In this new personality model the dimension Honesty-Humility is added to the Big Five. I used two dimensions of this 24-item instrument, referred to as Brief HEXACO Inventory (BHI), with should result in relatively low loss of validity (De Vries, 2013), to measure
conscientiousness and extraversion. Conscientiousness was measured by four items on a scale from 1-5 (strongly disagree – strongly agree). Four questions were asked: “I make sure that things are on the right spot”; “I postpose complicated tasks as long as possible”; “I work very precisely” and “I often do things without really thinking”. Extraversion was
measured by four items on a scale from 1-5 (strongly disagree – strongly agree). Four questions were asked: “Nobody likes talking to me”; “I easily approach strangers”; “I like to talk with others” and “I am seldom cheerful”.
Feelings of isolation. Feelings of isolation was measured based on the measurements of Golden, Veiga and Dino (2008). These measurements were assessed on validity through the UCLA Loneliness Scale (Version 3; Russell, 1996). The UCLA Loneliness Scale consists of 20 questions which would be too many questions for this survey. The measurement of Golden et al. consist of seven questions, used for Teleworkers. Five of them I used for this survey: “I feel out of the loop”; “I miss face-to-face contact with colleagues”; “I miss the emotional support of co-workers” and “I miss informal interaction
26
with others”. Two questions of Golden et al. are specifically related to the manager and the meeting/activities at the company of the worker, so I left them out of the measurement.
4.3.3 Sociodemographics
Respondents were asked about their age (year of birth), gender (female = 0 and male = 1), whether they had kids under 18 living with them (yes/no) and whether they lived with a partner (yes/no). Education level was measured using a question about the highest
completed educational level, which is a scale running from primary school to a university degree. Education was then recoded into low/middle = 0 (primary school/secondary school/MBO) and high = 1(HBO, University). Years of work experience as a freelancer was also asked. Asking respondents in which industry they worked, the following categories were available: Media, Creative sector, ICT, Education, Finance, Advice and Other.
4.4 Test of validity and reliability
The first step was to give the variables a clear label and an understandable short name.2 A frequency test was run for all variables, which showed there were no missing data or any errors in the data. The second step was to recode counter indicative items. First I recoded two counter-indicative items of the independent variable (IV) extraversion and two counter-indicative items of the independent variable (IV) conscientiousness. Next I recoded three items of the dependent variable (DV) work-life balance.
The assumption of normality was tested via examination of the unstandardized residuals. Review of the Kolmogorov-Smirnov and Shapiro-Wilk tests for normality (Sig.
2
The original (anonymized) data file and the SPSS syntax file that was used to clean the data file and to perform the analyses can be requested from the author.
27
> 0.05) and skewness and kurtosis between -.1.96 and 1.96 statistics suggested that
normality was a reasonable assumption for most dependent and independent variables. The Q-Q plots, histograms and boxplots supported the assumptions of normality. One
exception is freelance work experience, with a skewness of 1.78 (distribution to the left) and kurtosis of 3.77. Finally, meaningful work showed skewness of -1.05 and a kurtosis of 2.88, which shows a high meaningfulness of the freelancer for the work (s)he at 4 (on a scale of 1 to 5). In the boxplot two extreme outliers were identified in the variable meaningful work. Upon closer examination, it seemed very reasonable that these surveys were not truthfully filled in. Therefore, the cases were removed, which in turn normalized the distribution of this dependent variable. It resulted in a skewness of -.263 and a kurtosis of .341.
To assure internal consistency of measurements I checked reliability of the following variables: extraversion, conscientiousness, feelings of isolation, work-life balance,
perceived productivity, quality of work, meaningfulness of work, personal life and satisfaction. An overview is given in Table 2. All composite variables (except two) are above .70, thus showing high internal consistency, except conscientiousness which has a Cronbach Alpha of 0.56.
28
Table 2
Reliability measures composite variables
Variable Cronbach’s Alpha
Extraversion3 0.71 Conscientiousness 0.56 Feelings of isolation 0.82 Work-life balance 0.73 Perceived productivity 0.77 Satisfaction 0.77 Personal life 0.62 Meaningful work 0.77 Quality of work 0.84
The correlation matrix (Table 3) reports all scale means, standard deviations, correlations and reliabilities (reported along the diagonal). The correlation among the six dependent variables (work-life balance, perceived productivity, quality of work, meaningful work, personal life and satisfaction) reveal a significant correlation, especially satisfaction shows a significant correlation with all five dependent variables, except for work-life balance and perceived productivity, which are not related to quality of work and meaningful work. Quality of work is highly correlated with satisfaction (r=.50, n = 99, p < .001). The extent to which people work outside home is significantly and positively related to extraversion
3 Extraversion was originally measured using 4 questions, leading to a reliability of .65. However, closer analysis of the questions showed that two items may have led to confusion for respondents, possibly because of the counter-indicative nature of the two items (which may have led to confusion as to what the respondent was agreeing to or not). The items were: Nobody likes talking to me and I am seldom cheerful.
29
and significantly and negatively related to feelings of isolation. The variable ‘professional space workers’ show a significant negative relationship with feelings of isolation.
Conscientiousness has a significant positive correlation (except for meaningful work) with all dependent variables. Feelings of isolation shows a negative correlation (except for meaningful work) shows with all dependent variables.
30 Means, Standard Deviations, Correlations and Reliabilities
Variables Items M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1 Gender (f=0, m=1) 1 .56 .49 - 2 Age (years) 1 45.50 11.15 .21* - 3 Education (Low=0,High=1) 1 .93 .23 -.23* -.12 - 4 Children (No=0, Yes = 1) 1 .40 .49 -.13 -.12 .21* - 5 Partner (No=0, Yes=1) 1 .79 .41 .13 .02 .08 .07 - 6 Freelance experience (years) 1 8.27 7.36 .12 .59** .03 -.13 .00 - 7 Extent working outside home 1 3.06 1.00 .16 -.06 -.24* .14 -.14 -.10 - 8 Professional space worker 1 .60 .49 .17 -.03 -.12 .22* -.18 -.02 .73** -
9 Conscientiousness 4 3.49 .59 -.03 .02 -.07 -.16 .16 .11 -.14 -.11 (.56) 10 Extraversion 2 3.95 .61 -.16 -.14 .05 .05 -.20 -.21* .29** .16 -.27 (.71) 11 Feelings of isolation 5 2.39 .77 -.07 -.15 -.01 -.08 -.04 -.14 -.25* -.24* -.36** -.05 (.82) 12 Work-life balance 6 3.55 .65 .07 .21* .04 -.10 -.03 .18 -.04 .08 .25* -.-8 -.39** (.73) 13 Perceived productivity 5 3.80 .53 .10 .05 -.04 .10 .29** .25* .01 .01 40** -.06 -.39** .33** (.77) 14 Quality of work 3 4.06 .46 .03 .03 .00 -.08 .17 .29** .01 .02 .41** -.01 -.22* .16 46** (.84) 15 Meaningful work 2 3.93 .60 .05 .15 .01 -.15 -.08 .27** .01 -.07 .05 .09 -.19 .01 -.01 33** (.77) 16 Personal life 3 3.60 .65 .01 .10 .04 -.10 .02 .12 -.07 .02 .26** -.07 -.23* .63** .28** .37** .05 (.62) 17 Satisfaction 3 3.92 .63 .00 -.02 .02 -.04 .12 .15 .04 -.05 .35** -.05 -.37** .35** .37** .50** .37** .47** (.77)
Note: N = 99. Reliabilities are reported along the diagonal.
31
4.5 Analyses
Hierarchical multiple regressions, including interaction effects for conscientiousness and gender, were used to test hypotheses H1A, H1B and H3. A set of six hierarchical multiple regressions were conducted, using the six SCS variables as dependent variables: work-life balance, perceived productivity, quality of work, meaningful work, personal life and satisfaction and the independent variable working outside home.
For H2 a set of six hierarchical multiple regressions were conducted with the same SCS variables mentioned above, but this time using the variable professional space worker as the key predictor variable.
The SPSS macro of Preacher and Hayes (2016) was used in order to test hypothesis 4A, which explores the mediating role of feelings of isolation between work location and subjective career success (Model 4 in PROCESS), and hypothesis 4B, that concerned the moderating role of extraversion on feelings of isolation (Model 7 in PROCESS).
5 Results
5.1 Working at home
Hypothesis 1A stated that working outside home is positively related with subjective career success (SCS), while hypothesis 1B stated that the positive effect of working outside home on SCS will be stronger for female freelancers. Hypothesis 3 stated that the negative effect on SCS of working at home will be less or even non-existent for people who have higher values of conscientiousness.
In the first step of the regression analysis the control variables education, children and work experience freelancer were entered. In the second step of the regression the independent variable working outside home and the two moderators conscientiousness and
32
gender were added. All three variables were centred, to avoid multicollinearity between those variables and the interaction terms. In the 3rd and final step the interaction variables Conscientiousness * Work location and Gender * Work location, were added. The results for H1A, H1B and 3 are presented in Table 4 and 5.
Work-life balance
The first model was not significant, F(3, 95) = 1.39; p > .05 and explained 4.2 % of
variance in work life balance. The total variance explained by the second model was 9.7%; F(6, 92) = 1.64; p > .05, again not statistically significant. The third and final step with the interactions between working outside home and conscientiousness and working outside home and gender did not support hypothesis 1A,1B and 3 for the dependent variable work-life balance. The total variance explained by this full model was 9.9%; F(8, 90) = 1.24; p > .05. Yet, there is evidence that conscientiousness does influence work-life balance. In the second and third step of the regression conscientiousness revealed a significant ß (.233; p < .05 and .232; p < .05) indicating a positive direct relation between conscientiousness and work-life balance.
Perceived productivity
The model in the first step was statistically significant, F(3, 95) = 2.90; p < .05 and explained 8.4% of variance in perceived productivity. Work experience freelancer was statistically significant with a ß of .26 and a p-value of < .05. The total variance explained by the second model was 25.4%; F(6, 92) = 5.21; p <.000. In this model three out of six predictor variables were statistically significant, with conscientiousness recording a higher Beta value (ß = .41, p < .000) than work experience freelancer (ß = .21; p < .05) and children (ß = .20; p < .05). The final step with the interactions between working outside
33
home and conscientiousness and working outside home and gender did not support hypotheses 1A, 1B and 3 for the dependent variable perceived productivity. The total variance explained by this full model was 25.6%; F(8, 90) = 3.86; p < .01.
Quality of work
The model in the first step was statistically significant, F(3, 95) = 2.96; p < .05 and
explained 8.6% of the variance in quality of work. Here again, work experience freelancer, was statistically significant (ß = .28; p < .05). The total variance explained by the second model was 24.2%; F(6, 92) = 4.88; p < .00). Here two of six predictor variables were statistically significant, with conscientiousness recording a higher Beta value (ß = .40, p < .000) than work experience freelancer (ß = .25; p < .05). The third and final step with the interactions between working outside home and conscientiousness and working outside home and gender did not support hypotheses 1A, 1B and 3 on the dependent variable quality of work. The total variance explained by this full model was 26%; F(8, 90) = 3.09; p < .000).
Meaningful work
The model in the first step was not statistically significant F(3.95) = 3.01; p < .05 and explained 8.7% of the variance in meaningful work. The total variance explained by the second model was 9%; F(6, 92) = 1.52; p >.05). Here one of the six predictor variables, work experience freelancer, was statistically significant (ß = .25; p < .05). The third and final step with the interactions between working outside home and conscientiousness and working outside home and gender did not support hypotheses 1A, 1B and 3 on the dependent variable meaningful work. The total variance explained by this full model was 9.7%; F(8, 90) = 1.21; p > .05.
34 Personal life
The model in the first step was not statistically significant, F(3, 95) = .77; p > .05 and explained 2.4% of the variance in personal life. The total variance explained by the second model was 8.4%; F(6, 92) = 1.40; p > .05. Here one of the six predictor variables,
conscientiousness, was statistically significant (ß = .24; p < .05). The third and final step with the interactions between working outside home and conscientiousness and working outside home and gender did not support hypotheses 1A, 1B and 3 on the dependent variable personal life. The total variance explained by this full model was 9.6%; F(8, 90) = 1.19; p > .05.
Satisfaction
The model in the first step was not statistically significant F(3, 95) = .71; p >.05 and explained 2.2% of the variance in satisfaction. The total variance explained by the second model was 13.7%; F(6, 92) = 2.42; p <.05. Here one of the six predictor variables,
conscientiousness, was statistically significant (ß = .34; p < .01). The third and final step with the interactions between working outside home and conscientiousness and working outside home and gender did not support hypotheses 1A, 1B and 3 on the dependent variable satisfaction. The total variance explained by this full model was 15.8%; F(8, 90) = 2.11; p <.05.
35
Table 4
Hierarchical regression analyses predicting subjective career success (work-life balance, perceived productivity and quality of work) for working outside home
Work life balance Perceived producivity Quality of work B SE ß B SE ß B SE ß
Step 1
Education (low/high) .14 .27 .05 -.18 .22 -.08 .00 .19 .00 Children (yrs < 18) -.11 .13 -.08 .16 .11 .14 -.04 .09 -.04 Work experience freelancer .01 .00 .17 .02 .00 .26* .01 .00 .28* R² .042 .084* .086* Step 2 Education (low/high) .23 .29 .08 -.67 .21 -.03 .09 .19 .05 Children (yrs < 18) -.06 .13 -.05 .22 .10 .20* -.00 .09 -.00 Work experience freelancer .01 .00 .14 .06 .00 .21* .01 .00 .25* Working outside home .01 .06 .02 .02 .05 .03 .04 .04 .09 Conscientiousness .25 .11 .23 .37 .08 .41*** .31 .07 .40*** Gender .09 .13 .07 .11 .10 .10 .00 .08 .00 R² .097 .254*** .24** ∆R² .055 .170 .156 Step 3 Education (low/high) .21 .30 .08 -.05 .22 -.02 .08 .19 .04 Children (yrs <18) -.06 .14 -.05 .22 .10 .20* .00 .09 .00 Work experience freelancer .03 .00 .14 .01 .00 .21* .01 .00 .24* Working outside home .01 .07 .02 .01 .05 .03 .04 .04 .09 Conscientiousness .25 .11 .23 .37 .08 .41*** .32 .07 .41*** Gender .09 .13 .07 .11 .10 .10 .00 .08 .00 Conscientiousness * Work
location -.05 .11 -.05 .04 .08 .04 .07 .07 .08 Gender * Work location .00 .14 .00 -.01 .10 -.01 -.10 .09 -.10 R² .099 .256 .26 ∆R² .002 .002 .018
36
Table 5
Hierarchical regression analyses predicting subjective career success (meaningful work, personal life and satisfaction) for working outside home
Meaningful work Personal life Satisfaction
B SE ß B SE ß B SE ß
Step 1
Education (low/high) .06 .25 .02 .15 .28 .05 .04 .27 .01 Children (yrs < 18) -.15 .12 -.12 -.12 .13 -.09 -.03 .13 -.02 Work experience freelancer .02 .00 .25* .00 .00 .10 .01 .00 .14
R² .087 .024 .022 Step 2 Education (low/high) .11 .27 .04 .19 .29 .07 .10 .27 .04 Children (yrs < 18) -.16 .13 -.13 -.08 .14 -.06 .02 .13 .01 Work experience freelancer .02 .00 .25* .00 .00 .07 .01 .00 .11 Working outside home .03 .06 .06 .00 .07 .00 .01 .06 .02 Conscientiousness .01 .10 .01 .27 .11 .24* .36 .10 .34** Gender .00 .12 .00 .01 .13 .01 .00 .13 .00 R² .090 .084 .137 ∆R² .004 .060 .115** Step 3 Education (low/high) .10 .28 .04 .24 .30 .09 .18 .28 .06 Children (yrs <18) .10 .28 .04 -.08 .14 -.06 .02 .13 .01 Work experience freelancer .02 .00 .25* .00 .00 .07 .01 .00 .11 Working outside home .03 .06 .05 -.01 .07 -.02 .00 .06 .00 Conscientiousness .01 .10 .01 .27 .11 .25* .37 .10 .35** Gender .00 .12 .00 .02 .13 .01 .01 .13 .00 Conscientiousness * Work
location .06 .10 .06 .13 .11 .11 .16 .11 .15 Gender * Work location .07 .13 -0.6 .00 .14 .00 .01 .13 .00
R² .097 .096 .158 ∆R²
.007 .012 .022
37
5.2 Professional work spaces
Hypothesis 2 stated that working at a professional work space in contrast with other locations will have a positive influence on the subjective career successes work-life balance, perceived productivity, quality of work, meaningful work, personal life and satisfaction. In the first step of the regression analysis the control variables education, children and work experience freelancer were entered. In the second and final step of the regression the independent variable professional shared space and one more control variable, conscientiousness, were added. The results are presented in Table 6 and 7.
Work-life balance
The first model of work-life balance was not significant, F(3, 95) = 1.39; p >.05 and explained 4.2% of variance in work life balance. The introduction of conscientiousness in the second model explained additional 6.7% variance in work-life balance (R² Change = .067; F(2, 93) = 3.47; p <.05). Conscientiousness was statistically significant (ß = .23; p < .05). The total variance explained by the whole model was not statistically significant at 10,9%; F(5, 93) = 2.26; p >.05.
Perceived productivity
The first model of perceived productivity was significant F(3, 95) = 2.90; p < .05 and explained 8.4% of variance in perceived productivity. In this first model work experience was statistically significant (ß = .26; p < .01). In the second and final step three out of five predictor variables were now statistically significant, with conscientiousness recording a higher Beta value (ß = .40; p <.001) than work experience (ß = .23; p <.05) and children (ß = .19; p < .05). This explained an additional 15% variance in perceived productivity (∆R²=
38
.15; F 2, 93) = 9.67; p < .001). The total variance explained by the whole model was statistically significant at 24,2%; F(5, 93) = 5.92; p < .01.
Quality of work
The first model was significant F(3, 95) = 2.96; p < .05 and explained 8.6% of variance in perceived productivity. In this first model work experience was statistically significant (ß = .28; p < .01). In the second and final step of the regression, two out of five predictor
variables were now statistically significant, with conscientiousness recording a higher Beta value (ß = .39; p <.001) than work experience (ß = .24; p <.01). This explained an
additional 15% variance in quality of work (∆R²= .15; F 2, 93 = 9.29; p < .001). The total variance explained by the whole model 23.8% and was statistically significant at F(5, 93) = 5.80; p < .001).
Meaningful work
The first model was statistically significant F(3, 95) = 3.01; p < .05 and explained 8.7% of variance in meaningful work. In this first model work experience was statistically
significant (ß = .25; p < .05). In the second and final step of the regression there was no further variance in meaningful work (∆R²= .00; F(2, 93) = .076; p > .05). The total
variance explained by the whole model was 8.8% at not statistically significant at F(5, 93) = 1.80; p > .05.
Personal life
The first model was not significant F(3, 95) = .771; p >.05 and explained 2.4% of variance in personal life. In the second and final step of the regression conscientiousness recorded a significant Beta value (ß = .25; p <.01). This explained an additional 6.6% variance in
