Jump to ContentJump to Main Navigation
Youth Labor in TransitionInequalities, Mobility, and Policies in Europe$

Jacqueline O'Reilly, Janine Leschke, Renate Ortlieb, Martin Seeleib-Kaiser, and Paola Villa

Print publication date: 2018

Print ISBN-13: 9780190864798

Published to Oxford Scholarship Online: January 2019

DOI: 10.1093/oso/9780190864798.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy).date: 24 May 2019

Are the work values of the younger generations changing?

Are the work values of the younger generations changing?

Chapter:
(p.626) 21 Are the work values of the younger generations changing?
Source:
Youth Labor in Transition
Author(s):

Gábor Hajdu

Endre Sik

Publisher:
Oxford University Press
DOI:10.1093/oso/9780190864798.003.0021

Abstract and Keywords

This chapter analyzes whether work values differ between birth cohorts, age groups, and time periods. Using large cross-national surveys from more than 30 countries, it is shown that the centrality of work is highest in the middle age groups and significantly lower during 2005–2009 than in the 1990s. However, there are no detectable gaps between birth cohorts. Thus, in contemporary Europe, the generations are not divided significantly with regard to their work values so that rather than pointing to generational differences, the lack of them should be emphasized. From a policy standpoint, this means that the generational differences often referred to in public debates and used in political discourses are a myth. The results presented in this chapter imply that if sound European Union policies are implemented to cope with youth unemployment, they will not fail because of generation-specific attitudes.

Keywords:   work values, centrality of work, age–period–cohort analysis, generation, generational differences

21.1. Introduction

A common stereotype emerging in political speeches and everyday intellectual conversations about the younger generations paints them as increasingly less work oriented. Specifically, they are seen to be increasingly less motivated with regard to finding a job and working hard in the interests of developing a career. In comparison to older generations, the value of work as a significant part of personal identity is believed to be declining. It is often assumed that one of the consequences of increased labor market flexibility and precarious employment has been to create weaker incentives to build a career or invest in long-term human capital. The seeming impossibility of achieving what previous generations obtained in terms of career jobs (with attractive benefits and pensions) may generate attempts to reduce cognitive dissonance by rejecting the value of these achievements. It is thought that these attitudinal trends are likely to be exacerbated by the growing obstacles to labor market entry, lengthening spells of unemployment, and/or the spread of precarious work. If these arguments are true, youth-oriented European Union (EU) or national labor market policies will fail because the new entrants to the labor market (and even more so those who cannot enter at all) will in any case not respond positively to them.

In this chapter, instead of testing the existing theories of generational differences, our research aim is exploratory: We test empirically whether work values indeed differ between birth cohorts (with an emphasis on the youngest cohorts), age groups, and time periods. Specifically, we analyze whether the (p.627) centrality of work varies by birth date, age, and time period,1 using large cross-national surveys from more than 30 countries (most of the European countries and some Organization for Economic Co-operation and Development (OECD) countries from the Euro-Atlantic area).

Sections 21.2 and 21.3 describe the conceptual basis of the analysis, the main characteristics of the data, and the methodology applied. In Section 21.4, we first illustrate the trends for attitudes regarding the centrality of work and then test the role of age, time period, and birth cohorts with respect to these trends. Section 21.5 concludes that given the lack of evidence of significant gaps between birth cohorts with regard to relative centrality of work, there is not a generational divide in contemporary Europe with respect to work values.

21.2. Background

21.2.1. Birth cohort versus generation

We decided to use the concept of birth cohort as opposed to generation for our analysis because the latter concept is rife with ambiguities. The term generation refers to individuals born at approximately the same time. From this, it follows that they experience more or less similar historical and life events during their early years. The underlying assumption is that because in their childhood and adolescent periods they are influenced by actors with similar value systems and are exposed to identical events and developments (news, economic or social booms and crises, technological innovation, policy and political influences, etc.), the values they hold will be rather similar, and they will be different from those of all other generations. It is also assumed that this impact is the strongest during an individual’s childhood and adolescence and remains relatively stable from then on (Harpaz and Fu 2002). The stability of such generation-specific values offers a chance for a generation to develop into a social group with a shared loose form of self-consciousness and identity (Diepstraten, Ester, and Vinken 1999).

The consciousness of a generation is a stochastic and dynamic social phenomenon. In other words, if it emerges at all, there should be a significant event such as a war or a revolution, a brand-new technology, or some other major phenomenon to lay the foundation of the new generation. If such an impetus is strong enough to mobilize a group of young people who are in a position to influence their fellows from the same cohort in identifying themselves as an “imagined generational community,” then the nucleus of a generation may appear. If such a feeling of generational community takes hold, then the shared set of values and goals becomes the common denominator of a generation.

The essence of this generation concept is well captured by the concept of generation subculture theory, which is defined by Egri and Ralston (2004) as follows:

(p.628)

Significant macro-level social, political, and economic events that occurred during a birth cohort’s impressionable pre-adult years result in a generational identity comprised of a distinctive set of values, beliefs, expectations, and behaviors that remain relatively stable throughout a generation’s lifetime. . . . A generation’s values orientation becomes more pervasive in a national culture as it becomes the majority in societal positions of power and influence. (p. 210)

Although seemingly concise and elegant, there are several problems with the generation concept. For example,

  • It is much too loosely defined timewise, in that it sometimes covers more than a decade, which might be too long to assume that the members of a generation indeed have similar experiences.

  • The characteristics used to capture the main features of generations are often based on anecdotal evidence or on invalid and unreliable survey data.

  • The assumption that there are global generations (i.e., a generation can be defined by the same characteristics all over the planet) is very likely a myth.2 Even if generations are rather similar across different countries, they can be very different in terms of historical moment: Their periodization depends on a country’s specific timeline of technological, political, and policy development.

Unlike the generational approach, the birth cohort is usually narrowly defined—in demography, for example, usually as a 5-year-wide “mini-generation.” Moreover, the birth cohort does not fluctuate according to vague, quasi-theoretical assumptions usually based on technological–political changes in the United States.

21.2.2. Work values

Work values form a core subset of the general value system (Wuthnow 2008; Jin and Rounds 2012). They have been the target of several large-scale comparative projects since the 1970s and 1980s that use quantitative databases to describe the differences between citizens from various countries with respect to the centrality of work in their lives (Roe and Ester 1999). Most of these studies have treated work-related values (Roe and Ester 1999)

as expressions of more general life values. . . . All definitions treat values as latent constructs that refer to the way in which people evaluate activities or outcomes. . . . Holders of values are not necessarily individuals but may also be collectivities, i.e. the people belonging to a certain occupational (p.629) group, a firm, a subculture, a community, a national category, or a country. (pp. 2–4)

To understand the association between values and other socioeconomic characteristics of society, as well as the relationship between value systems in general and work values in particular, large quantitative data sets have been used since the 1980s for comparative analysis of work values (Wuthnow 2008). Since the late 1990s, a promising new direction in comparative quantitative research on values (cultural economics) has emerged, rephrasing old questions in a new format using large-scale surveys carried out in several countries (e.g., the World Value Survey, the European Values Study, and the International Social Survey Programme) to analyze the high inertia of culture.3

In the course of our analysis, we used “centrality of work” as the dependent variable. This term covers both paid and unpaid work and measures the attitude of the respondent toward work in general—in other words, how important work is for a respondent as a part of his or her life and identity.4 Centrality of work (under various names) is a core concept in organization, business, and management sciences, in which it is considered a crucial aspect of activity in a workplace. From the employees’ viewpoint, it is necessary to achieve higher income and subjective well-being, satisfaction, a career, and so forth; from the employers’ viewpoint, it is the primary source of commitment to hard work, efficiency, informal and on-the-job training, and so on (Hansen and Leuty 2012).

21.2.3. Previous literature

The most widely accepted hypothesis regarding the trend followed by centrality of work is that generations have different attitudes toward work to the extent that (Tolbize 2008)

the perceived decline in work ethic is perhaps one of the major contributors of generational conflicts in the workplace. Generation X for instance, has been labelled the “slacker” generation, and employers complain that younger workers are uncommitted to their jobs and work only the required hours and little more. Conversely, Boomers may be workaholics . . . while “Traditionals” have been characterized as the most hardworking generation. (p. 5)

This hypothesis dominates the discourse despite the fact that a meta-analysis of generation-specific work values (Costanza et al. 2012) found moderate or zero differences between generational membership and work-related attitudes.

Other research combining longitudinal panel data between 1981 and 1993 and a representative survey of the Israeli Jewish labor force in 1993 analyzed how time period, cohort, and life course (in our vocabulary, age group) affect work (p.630) values (primarily the importance of work). The study concluded that in contrast to other developed countries, the centrality of work has strengthened in Israel since the early 1980s (Sharabi and Harpaz 2007, 103–4).

Kowske, Rasch, and Wiley (2010) analyzed the role of time period, age, and cohort on work values (satisfaction with company/job, recognition, career, security, pay, and turnover intentions) among generations of Americans with a special focus on the so-called millennial generation. According to their research,

Work attitudes differed across generations, although effect sizes were relatively small and depended on the work attitude. Compared to Boomers and GenXers, Millennials reported higher levels of overall company and job satisfaction, satisfaction with job security, recognition, and career development and advancement, but reported similar levels of satisfaction with pay and benefits and the work itself, and turnover intentions. (p. 265)

According to these authors, the role of generations is significantly weaker than a set of labor-market sensitive individual factors such as gender, industry, and occupation (p. 273).

Regarding the impact of different generations, Kowske et al. (2010) found curvilinear trends (i.e., U-shaped curves) in the case of all work values. This means that the least satisfied with the various aspects of work were the late baby boomers, whereas the “GI” (born around the time of World War II) and millennial generations were the most satisfied (the latter especially with recognition and career). However, the most important conclusion of their analysis was that contrary to the popular view of the role of generation with respect to the labor market, “generational differences might be re-named ‘generational similarities’ ” (p. 275).

To conclude, we quote a more recent overview in which the authors convincingly summarize the theoretical and methodological state of the art of research on generations (Becton, Walker, and Jones-Farmer 2014):

Considering the extent to which generational stereotypes are commonly accepted, it is surprising that empirical evidence of generational differences is relatively sparse, and the research that exists is somewhat contradictory. . . . There exists a great deal of controversy about whether or not generational differences exist at all with some suggesting that perceived generational differences are a product of popular culture versus social science. Scholars have also noted that observed generational differences may be explained, at least in part, by age, life stage, or career stage effects instead of generation. (pp. 175–76)

(p.631) 21.3. Data and methods

The basic problem in analyzing generations stems from the fact that the effects of age, time period, and birth cohort are closely intertwined. Any change over time can be determined by any of the three effects, as can be illustrated with the following fictional dialogue (based on Suzuki 2012, 452):

ENDRE:

  • I’m very tired, I must be getting old. (Age effect)
  • GÁBOR:

  • You’re no spring chicken indeed, but maybe you’re crawling into bed early every night because life is so stressful nowadays. (Period effect)
  • ENDRE:

  • Could be, but you seem to be tired too. The truth is, you young people are not as fit as we used to be at your age. (Cohort effect)
  • 21.3.1. The problem of decomposing the effects of age, period, and birth cohort

    Because age, time period (year of the survey), and birth cohort (year of birth) are linearly interdependent, their effects cannot be simultaneously estimated using standard regression models (Firebaugh 1997; Yang and Land 2006, 2008). A possible solution to this identification problem is to use a hierarchical age–period–cohort (HAPC) regression model (Yang and Land 2006, 2008).5

    To minimize the effect of multicollinearity between age, birth cohort, and period, we defined fixed and equal-period (year of the survey) clusters.6 In this grouped data, age, period (with 5-year intervals), and birth cohort (year of birth) are not perfectly dependent. In other words, we are no longer able to directly calculate the year of birth from age and period (with 5-year intervals); nonetheless, remarkable multicollinearity still remains.

    Moreover, whereas age is an individual-level variable, period and cohort are macro-level variables.7 This means that we have a multilevel data structure assuming that the attitudes of the individuals in the same birth cohort, or interviewed in the same year, will be more similar than those from other periods or birth cohorts.

    Yang and Land (2006, 2008) propose cross-classified hierarchical models to represent clustering effects in individual survey responses by period and birth cohorts when using repeated cross-sectional data. In this analysis, we use these models where it is assumed that individuals are nested simultaneously within the two second-level variables (period and cohort); thus, we use cross-classified hierarchical regression models.8

    Bell and Jones (2014), however, argue that there is no statistically and mathematically correct solution to the age–period–cohort identification problem in the absence of preliminary theoretical assumptions: “There is no technical (p.632) solution to the identification problem without the imposition of strong (and correct) a priori assumptions” (p. 335). They show with simulations that in several scenarios, the results of the HAPC model are biased: For example, if there is quadratic age effect and linear cohort trend, these effects are estimated as a period trend. In other words, the effects of the three time-related variables might be assigned to each other or be combined by the effects of the other two variables. However, Bell and Jones also show that the model works if there are no trends for periods or cohorts. Given that our results show that the cohort and period effects are quite small, our findings should be “probably justifiable,” according to Bell and Jones (i.e., because the results are not biased by strong cohort and/or period effects, the use of the HAPC model is feasible).

    Twenge (2010) recommends another solution to avoid the identification problem mentioned previously by taking one step backward. She suggests using the time-lag method, in which individuals of the same age at different points in time are compared: “With age held constant, any differences are due to either generation (enduring differences based on birth cohort) or time period (change over time that affects all generations)” (p. 202). Twenge argues that because the impact of period is often the weakest, a time-lag design should be able to isolate generational differences.

    Here, we first provide a descriptive analysis in which we separately model age, period, and birth cohort effects on work values to illustrate the main trends. Some of these descriptive analyses are equivalent to Twenge’s (2010) time-lag method; however, the results might be biased by omitted variables because they are based on bivariate relationships in which the sociodemographic characteristics of the respondents are uncontrolled. In the second step, we develop HAPC regression models to avoid problems stemming from the linear dependency of these three dimensions of time (Yang and Land 2006, 2008). As part of this exercise, we also separately run models for the youngest respondents (aged 18–40 years) so as to meet the requirements of the time-lag method recommended by Twenge; in other words, individuals with more homogeneous ages are compared.

    21.3.2. Data

    Given that our strategy of analyzing the changing (or unchanging) attitudes of generations toward work was based on secondary analysis of existing large, repeated cross-sectional, cross-national databases, we first had to select those precious few questions that were asked either similarly in these surveys or could be made identical by recoding and therefore be used as proxies of work values.9

    Questions about the importance of work and other aspects of life were asked in the questionnaires of the World Values Survey/European Values Study (WVS/EVS). Respondents were asked to answer the question, “How important is [life aspect] in your life?” on a 4-point scale.10 We used four variables: importance of (p.633) work, importance of family, importance of friends, and importance of leisure time. We calculated the relative importance of work as the difference between the importance of work and the average importance of the other three life aspects (i.e., family, friends, and leisure time). Thus, positive values of the variable indicate that work is more important in the respondent’s life than other life aspects, whereas negative values indicate that it is less important than other life aspects—in other words, that work plays a relatively minor role in the respondent’s life.

    Our analysis covers most of the European countries and some OECD countries from the Euro-Atlantic area. Table A21.1 in Appendix 2 contains the list of countries (arranged into three groups: post-socialist, EU15, and other OECD countries) included in the various waves.

    Because the question was not asked in the first wave of WVS/EVS and the number of observations between 2000 and 2004 was relatively low, we only have data from three periods.11 Because our analysis is extremely time sensitive to the year of the information the analysis is based on, we decided to use the year of the fieldwork country by country.12

    The number of observations and the means of the variable of relative centrality of work by period are shown in Table 21.1. The aggregate value of relative centrality of work is highest in 1990–1994, somewhat lower in 1995–1999, and lowest in the mid-2000s. This means that compared to the importance of other aspects of life, work was more important in the 1990s and became less so in the second half of the 2000s.

    Table 21.1 Number of observations and average relative centrality of work by period

    Period

    N

    Mean

    SD

    Min

    Max

    1990–1994

    36,370

    0.050

    0.805

    –3

    3

    1995–1999

    64,407

    0.023

    0.810

    –3

    3

    2005–2009

    65,563

    –0.105

    0.832

    –3

    3

    Total

    166,340

    –0.022

    0.821

    –3

    3

    In the descriptive analysis, the period, the age of the respondent, and the birth cohorts were coded into 5-year intervals, which are conventional in age–period–cohort analyses (Yang, Fu, and Land 2008) and significantly shorter than those used in the sociological or management literature on generations. The result of this operation was 12 age groups (from 18–22 to 73+ years), 3 period groups (1990–1994, 1995–1999, and 2005–2009), and 16 cohort groups (from –1916 to 1987–1991).

    In the multivariate models, age was allowed to have a nonlinear (curvilinear) effect (squared age is also included in the models), cohorts were included as birth year, and periods (year of the survey) were grouped into 5-year intervals as in the descriptive analysis.

    (p.634) To control for the changing composition along the basic socioeconomic characteristics of subsequent generations in our multivariate models, we used the following control variables:

    • Gender (binary variable, 1 = female)

    • Education (binary variable, 1 = more than secondary education)

    • Marital status (married/living with partner, divorced/separated, widowed or never married)

    • Labor force status (binary variable, 1 = respondent has a job; i.e., his or her employment status is “working”)

    • Type of settlement (binary variable, 1 = respondent lives in a city (with population >100,000))

    In addition, every model contained country-fixed effects in order to control for time-invariant country characteristics.

    21.4. Results

    21.4.1. The cumulated impact of age and period on the relative centrality of work

    Table 21.2 displays the mean relative centrality of work by age group and period. The last column (age effect) shows that the centrality of work increases until age 43–52 years and then decreases continuously. In other words, people slowly “learn” the importance of work, but this (centrality of work) holds only as long as they are in their active years. If we focus on the bottom row, we find an aggregate decrease in the mean relative centrality of work (period effect) between 1995–1999 and 2005–2009. This can be interpreted as indicating that work in general is losing its importance.

    Table 21.2 Means of relative centrality of work by age group and period (cohort uncontrolled)

    Age group (years)

    Period

    1990–1994

    1995–1999

    2005–2009

    Total

    18–22

    –0.095

    –0.115

    –0.236

    –0.155

    23–27

    –0.028

    –0.006

    –0.090

    –0.042

    28–32

    –0.008

    0.028

    –0.035

    –0.004

    33–37

    0.054

    0.062

    –0.027

    0.028

    38–42

    0.120

    0.128

    0.037

    0.091

    43–47

    0.189

    0.165

    0.043

    0.122

    48–52

    0.181

    0.170

    0.049

    0.123

    53–57

    0.176

    0.100

    –0.018

    0.066

    58–62

    0.111

    0.009

    –0.113

    –0.018

    63–67

    0.036

    –0.077

    –0.312

    –0.144

    68–72

    –0.038

    –0.116

    –0.334

    –0.186

    73+

    –0.152

    –0.245

    –0.409

    –0.308

    Total

    0.050

    0.023

    –0.105

    –0.022

    The differences by age groups and birth cohorts (the final column in Table 21.3) show that work seems to be relatively most important in the birth cohorts 1947–1961 and less important for the earlier and later cohorts.

    Table 21.3 Means of relative centrality of work by birth cohort and age group (period uncontrolled)

    Cohort

    Age

    Total

    18–22

    23–27

    28–32

    33–37

    38–42

    43–47

    48–52

    53–57

    58–62

    63–67

    68–72

    73+

    –1916

    –0.225

    –0.225

    1917–1921

    –0.107

    –0.250

    –0.190

    1922–1926

    –0.004

    –0.082

    –0.312

    –0.173

    1927–1931

    0.056

    –0.073

    –0.081

    –0.429

    –0.138

    1932–1936

    0.151

    0.064

    –0.034

    –0.408

    –0.339

    –0.081

    1937–1941

    0.154

    0.118

    0.032

    –0.403

    –0.300

    –0.056

    1942–1946

    0.168

    0.169

    0.121

    –0.201

    –0.266

    0.006

    1947–1951

    0.107

    0.168

    0.185

    –0.106

    –0.073

    0.074

    1952–1956

    0.046

    0.113

    0.178

    0.009

    0.023

    0.087

    1957–1961

    –0.028

    0.059

    0.141

    0.036

    0.068

    0.069

    1962–1966

    –0.035

    0.009

    0.066

    0.014

    0.046

    0.028

    1967–1971

    –0.097

    –0.042

    0.042

    –0.057

    0.048

    –0.004

    1972–1976

    –0.114

    0.012

    –0.032

    –0.012

    –0.021

    1977–1981

    –0.112

    –0.095

    –0.036

    –0.079

    1982–1986

    –0.168

    –0.088

    –0.114

    1987–1991

    –0.268

    –0.268

    Total

    –0.155

    –0.042

    –0.004

    0.028

    0.091

    0.122

    0.123

    0.066

    –0.018

    –0.144

    –0.186

    –0.308

    –0.022

    To visualize the main differences and similarities of the trends between age and period, we designed two closely related figures (Figure 21.1a and Figure 21.1b). Figure 21.1a shows the trend of the relative centrality of work by age, controlling for period. The general pattern (the inverted U-curve) is rather similar in the three periods, but the highest level of the centrality of work lasts longer (from age 43–47 to age 53–57 years) in the first period than in the second or the third period. For every age group, the importance of work is lowest in the final period (2005–2009). Among those aged older than 53 years, work is relatively more important in the first period (1990–1994), whereas among the younger age groups, there is no real difference between the first two periods. (p.635)

    Are the work values of the younger generations changing?

    Figure 21.1 Means of centrality of work by (a) age in the three periods and (b) period in seven age groups.

    Figure 21.1b focuses on the trend for the relative centrality of work by period in six age groups.13 Although the general trend is a slight decrease between the first two periods and a steeper decrease after 1995–1999, centrality of work declines sharply after 1990–1994 in the two oldest age groups. In the middle age groups, the trend is similar to the average, and they have the highest level of relative centrality of work throughout all periods. As for the youngest age groups, there is a slight increase between 1990–1994 and 1995–1999 in the group aged 23–27 years, whereas subsequently the decrease for both age groups is less sharp than in general.

    21.4.2. The HAPC models of the relative centrality of work

    The HAPC regression models (Table 21.4) contain the three time-related and all control variables. Whereas age and squared age are included as individual-level variables, period (year of the survey) and cohort (year of birth) are second-level predictors. Random period and cohort intercepts allow level 1 intercepts to vary randomly by cohorts and periods; that is, they allow variation from the mean for each cohort and period. The models in columns 0–5 show results from the entire sample, whereas the model in column 6 covers the young (age 18–40 years) individuals only.14

    Table 21.4 HAPC models of centrality of work

    Individual effect

    (0)

    (1)

    (2)

    (3)

    (4)

    (5)

    (6)

    All

    All

    All

    All

    All

    All

    Youth (18–40 years)

    B

    SE

    B

    SE

    B

    SE

    B

    SE

    B

    SE

    B

    SE

    B

    SE

    Age

    –0.0011***

    (0.000)

    –0.0008***

    (0.000)

    0.0024***

    (0.000)

    0.0021***

    (0.000)

    0.0026***

    (0.000)

    –0.0013

    (0.003)

    Age squared

    –0.0004***

    (0.000)

    –0.0004***

    (0.000)

    –0.0004***

    (0.000)

    –0.0003***

    (0.000)

    –0.0003***

    (0.000)

    –0.0003***

    (0.000)

    Female

    –0.0659***

    (0.004)

    –0.0646***

    (0.004)

    –0.0687***

    (0.006)

    Education: More than secondary

    –0.0710***

    (0.005)

    –0.0339***

    (0.005)

    –0.0676***

    (0.006)

    Employment status: Working

    0.1831***

    (0.005)

    0.2064***

    (0.005)

    0.1160***

    (0.006)

    Type of settlement: City

    –0.0607***

    (0.004)

    –0.0491***

    (0.004)

    –0.0518***

    (0.006)

    Marital status

    Single

    Ref.

    Ref.

    Ref.

    Married/living with partner

    0.0122**

    (0.006)

    –0.0007

    (0.006)

    0.0150**

    (0.007)

    Divorced/separated

    0.0603***

    (0.009)

    0.0769***

    (0.009)

    0.1002***

    (0.013)

    Widowed

    0.0051

    (0.010)

    –0.0381***

    (0.010)

    0.0623*

    (0.036)

    Intercept

    –0.0233***

    (0.002)

    0.0899***

    (0.003)

    0.0987**

    (0.040)

    0.1294**

    (0.051)

    0.0540

    (0.046)

    0.0236

    (0.060)

    0.0180

    (0.044)

    Variance component

    Variance

    SE

    Variance

    SE

    Variance

    SE

    Variance

    SE

    Variance

    SE

    Variance

    SE

    Variance

    SE

    Individual

    0.6822***

    (0.001)

    0.6661***

    (0.001)

    0.6613***

    (0.001)

    0.6600***

    (0.001)

    0.6510***

    (0.001)

    0.6175***

    (0.001)

    0.0028***

    (0.001)

    Period

    0.0047***

    (0.002)

    0.0076***

    (0.003)

    0.0061***

    (0.003)

    0.0074***

    (0.003)

    0.0224***

    (0.003)

    Cohort

    0.0055***

    (0.002)

    0.0030***

    (0.001)

    0.0024***

    (0.000)

    0.0004***

    (0.000)

    Country

    0.0381***

    (0.004)

    0.5059***

    (0.001)

    N

    166,340

    166,340

    166,340

    166,340

    166,340

    166,340

    70,664

    AIC

    408,443.9

    404,466.7

    403,287.3

    403,187.4

    400,861.9

    392,271.8

    152,622.9

    Deviance (df)

    408,439.9(2)

    404,458.7(4)

    403,277.3(5)

    403,175.4(6)

    400,835.9(13)

    392,243.8(14)

    152,594.9(14)

    (*) p < 0.10.

    (**) p < 0.05.

    (***) p < 0.01.

    Comparing the six models, the sign and the size of the coefficients are fairly stable. Age differences become smaller with the inclusion of the other variables, (p.636) (p.637) given that there is collinearity between age and other variables (e.g., labor force status or marital status). Focusing on the role of the three time variables, we find that although they have a significant impact on the centrality of work, this is small compared to the impact of the non-age individual variables and the country-fixed effects.

    The visualized results (Figure 21.2) show that—controlling for period, birth cohort, and relevant sociodemographic characteristics—the centrality of work increases from age 18 years, reaches a peak at approximately age 50 years, and decreases thereafter. This result is similar to that of the uncontrolled inverted (p.638) (p.639) (p.640) U-curve (Figure 21.1a) and is in accordance with a life course concept of economic activity: Because younger people are not yet involved and older people are no longer involved in income-generating activities, it makes sense that their attitude toward the importance of work should be lower compared to that of people (p.641) for whom work plays the central role in their identity (i.e., career-oriented, human capital investing, etc., individuals in (early) middle age), who are in their active household and labor market cycles (i.e., entering the labor market, becoming adults, establishing a family, having children, etc.).

    Are the work values of the younger generations changing?

    Figure 21.2 Age, period, and birth cohort effects on relative centrality of work in total sample and in young (aged 18–40 years) cohorts (HAPC regression model).

    The results shown in the second panel in Figure 21.2 confirm that—controlling for age, cohort, and relevant sociodemographic characteristics—the centrality of work is significantly lower in 2005–2009 than in the 1990s. However, period accounts for only 1.17% of the variance of the centrality of work; that is, the effect size is rather small.

    Finally, after controlling for age and period and relevant sociodemographic characteristics, work is somewhat less important for birth cohorts born in the mid-20th century compared to the earlier and later-born cohorts. This result may be interpreted as a generational effect: For those who entered the labor market in approximately 1968, the centrality of work has temporarily decreased. However, because the effect size is quite small (cohort accounts for only 0.38% of the variance in the centrality of work), we are better to conclude that there is no generational effect.

    The cohort and period differences among the youngest group (aged 18–40 years) are even smaller compared to those of the full sample. Period differences are slightly smaller than in the whole sample, suggesting that relative importance of work seems to have decreased less among the younger generation. However, in general, it seems that our findings regarding the full sample are valid in the case of the young subsample as well.

    21.4.3. Gender differences

    To test whether the determinants of the relative centrality of work differ by gender, we ran the HAPC models for men and women separately. The results (the detailed results in Table 21.5 and the visualized effects of the three time-related variables in Figure 21.3) show that the differences by gender are very small.15 The effect of the three time-related variables does not differ between men and women, whereas cohort differences are somewhat larger among women, although the effect size is very small.

    Table 21.5 HAPC models of centrality of work among men and women

    Individual effect

    Men

    Women

    B

    SE

    B

    SE

    Age

    0.0018***

    (0.000)

    0.0013***

    (0.000)

    Age squared

    –0.0003***

    (0.000)

    –0.0003***

    (0.000)

    Education: More than secondary

    –0.0305***

    (0.007)

    –0.0388***

    (0.007)

    Employment status: Working

    0.2313***

    (0.007)

    0.1926***

    (0.006)

    Type of settlement: City

    –0.0392***

    (0.007)

    –0.0577***

    (0.006)

    Marital status

    Single

    Ref.

    Ref.

    Married/living with partner

    0.0144

    (0.009)

    –0.0277***

    (0.008)

    Divorced/separated

    0.0857***

    (0.014)

    0.0625***

    (0.012)

    Widowed

    –0.0440***

    (0.017)

    –0.0518***

    (0.012)

    Intercept

    –0.0350

    (0.051)

    –0.0349

    (0.062)

    Variance component

    Variance

    SE

    Variance

    SE

    Period

    0.0055***

    (0.002)

    0.0071***

    (0.003)

    Cohort

    0.0270***

    (0.003)

    0.0509***

    (0.006)

    Country

    0.0001***

    (0.000)

    0.0009***

    (0.000)

    Individual

    0.6063***

    (0.002)

    0.6248***

    (0.001)

    N

    76,477

    89,863

    AIC

    178,982.0

    213,068.7

    Deviance (df)

    178,956.0(13)

    213,042.7(13)

    * p < 0.10.

    ** p < 0.05.

    (***) p < 0.01.

    Are the work values of the younger generations changing?

    Figure 21.3 Age, period, and birth-cohort effect on centrality of work among men and women (HAPC regression model).

    There are, however, other significant gender differences, such as the following:

    • Being married or living with a partner has a positive but insignificant effect on the centrality of work among men, but it has a negative and significant effect among women.

    • The effect of employment status is larger among men than among women.

    • Work is more important for an average man than for an average woman.16

    These findings might be explained by gender norms, such as the traditional prescription that a man has to work more and has to be the main earner in the family.

    (p.642) 21.4.4. Regional differences

    We compared the impact of age, period, and birth cohort in two subgroups of European countries:17 post-socialist and EU15 countries.18 We hypothesized that because state socialism as a “natural experiment” influenced post-socialist countries for five decades in terms of their state-induced work-oriented ideology, we might detect path-dependent, cohort-specific characteristics for the value of work. For instance, the work values at least at the beginning of the post-socialist period might be stronger than those of people living in EU15 countries—that is, in societies without this socialist heritage.

    The results of two HAPC models for the two groups of countries (the detailed results in Table 21.6 and the visualized effects of the three time-related variables in Figure 21.4) show that the coefficients of the control variables are mostly similar: The centrality of work is significantly higher for men, for divorced people (p.643) (p.644) (compared to single individuals), and for working people, whereas it is lower for city dwellers and for more highly educated people in both country groups. However, there are system-specific differences as well, including the following:

    • In the EU15 countries, the overall level of centrality of work is lower.

    • In the EU15 countries, high education has a more negative effect on the centrality of work.

    • Being widowed has a negative effect in the EU15 countries but no effect in the post-socialist countries.

    • The signs of gender and higher education effects are the same in the two groups, but the sizes of the coefficients are twice as large in the EU15 countries compared to the post-socialist countries.

    • The effect of being married or living with a partner is negative in the EU15 countries, whereas it is positive in the post-socialist countries.

    (p.645)

    (p.646) If we compare the effects of the three time-related variables, we can see that age differences are smaller in EU15 countries, whereas the curve is more similar to an inverted U-shape for post-socialist countries. However, the effect size is notable in EU15 countries as well: Work is 0.15 points less important for an 18-year-old individual than for an individual in his or her fifties. This effect size is close to that of working people and higher than the effect of education. The period trends are similar in the two country groups, but the centrality of work declines somewhat more in the post-socialist countries than in the EU15 countries. Period accounts for 0.7% and 2.2% of the variance in the centrality of work in the EU15 and the post-socialist countries, respectively. Finally, centrality of work falls and remains very low among those born in the 1940s in post-socialist countries and starts increasing thereafter. In the EU15 countries, however, there are no real differences between cohorts. Cohort accounts for only 0.1% of the variance in the centrality of work in the EU15 countries and for 3.5% in the post-socialist countries. It is worth noting that cohort differences in post-socialist countries might not be detectable if we analyze single countries, but a more detailed analysis goes beyond the scope of this chapter. However, some analyses in the working paper version of this chapter suggest that cohort differences within single countries do not exist (Hajdu and Sik 2015).

    Table 21.6 HAPC models of centrality of work in EU15 and post-socialist countries

    Individual effect

    EU15

    Post-socialist

    B

    SE

    B

    SE

    Age

    0.0010***

    (0.000)

    0.0081***

    (0.001)

    Age squared

    –0.0002***

    (0.000)

    –0.0005***

    (0.000)

    Female

    –0.0761***

    (0.006)

    –0.0438***

    (0.006)

    Education: More than secondary

    –0.0534***

    (0.008)

    –0.0196***

    (0.007)

    Employment status: Working

    0.1918***

    (0.007)

    0.1966***

    (0.007)

    Type of settlement: City

    –0.0316***

    (0.007)

    –0.0736***

    (0.006)

    Marital status

    Single

    Ref.

    Ref.

    Married/living with partner

    –0.0194**

    (0.009)

    0.0263***

    (0.009)

    Divorced/separated

    0.0343**

    (0.014)

    0.1154***

    (0.013)

    Widowed

    –0.0863***

    (0.016)

    0.0101

    (0.014)

    Intercept

    –0.0699

    (0.067)

    0.2159***

    (0.075)

    Variance component

    Variance

    SE

    Variance

    SE

    Period

    0.0045***

    (0.002)

    0.0141***

    (0.006)

    Cohort

    0.0004***

    (0.000)

    0.0221***

    (0.003)

    Country

    0.0436***

    (0.008)

    0.0085***

    (0.002)

    Individual

    0.0436***

    (0.008)

    0.0085***

    (0.002)

    N

    66,400

    77,405

    AIC

    157,627.8

    179,739.0

    Deviance (df)

    157,599.8(14)

    179,711.0(14)

    * p < 0.10.

    (**) p < 0.05.

    (***) p < 0.01.

    Are the work values of the younger generations changing?

    Figure 21.4 Age, period, and birth cohort effect on centrality of work in EU15 and post-socialist countries (HAPC regression model).

    21.5. Conclusions

    We did not find significant gaps between birth cohorts with respect to relative centrality of work and thus claim that in contemporary Europe, the generations are not divided significantly with regard to their work values. In this respect, our findings reinforce the results of Clark (2010), Kowske et al. (2010), Costanza et al. (2012), Jin and Rounds (2012), and Becton et al. (2014): Rather than pointing to generational differences, we should instead emphasize the lack of them.

    There are, however, different trends in the centrality of work by age and birth cohort. The effect of the former is close to an inverted U-shaped curve—the centrality of work is higher in the middle age groups than among the younger or older groups—whereas the effect of the latter is closer to a curvilinear curve—the centrality of work is higher in the earlier and in the later-born cohorts. However, it is worth noting that although this effect can be regarded as statistically significant, the effect size is rather small. Regarding the impact of period, it is characterized by a linear and slightly decreasing trend.

    The interpretation of the inverted U-shape of the relative centrality of work by age is rather straightforward: Because younger people are not yet involved and older people are no longer involved in income-generating activities, it is logical to find that work is less central for both of these groups compared to individuals in their active household and labor market cycles. The decreasing linear trend of the centrality of work by period fits well into what the literature proposes: It (p.647) indicates a shift from modernity toward postmodernity (Egri and Ralston 2004; Twenge et al. 2010).

    The U-curve for the centrality of work by birth cohort might mean that work is less central for the cohort born between 1940 and 1959 compared to the earlier and later-born cohorts. This result may be interpreted as a rather weak generational effect in the sense that for those who entered the education system and the labor market in the 1960s and 1970s, intrinsic values became more important than work (or other extrinsic aspects of life). However, this change was quite quickly reversed, and the values of those who entered the labor market after the mid-1970s became more extrinsically oriented again.

    The first conclusion from a policy standpoint is that we could not identify any relevant gap between the birth cohorts. From this follows that the generational differences often referred to in public debates and used in political discourses are a myth. Kowske et al. (2010) quite rightly summarized their findings as indicating that instead of generational differences, we should speak about “generational similarities.” Our results imply that in contemporary Europe, generations follow a similar age trend: As the younger generations become older, their work values change similarly. Of course, this does not mean that within a country (and especially in smaller social units such as a region, a settlement, or a workplace) generational effects could not emerge, but these do not add up in our aggregated analysis as a generational trend.

    If there are no significant differences between the generations, for policymakers this means that those social and economic efforts made in the interest of decreasing youth unemployment will not be hindered by changing generational attitudes toward work.

    In summary, our assumption that younger generations are increasingly less work oriented, have less faith that they will achieve a career, and are less optimistic about getting a job and making ends meet on the basis of a salary turned out to be wrong. Therefore, if there sound EU policies are implemented to cope with youth unemployment, they will not fail because of generation-specific attitudes. Moreover, if the proposition of the management literature is correct that work values have a significant impact on values in general as well as on behavior in the workplace and on the labor market, then the unchanging nature of work values provides policymakers with firm ground to act.

    The second conclusion is based on the fact that although birth cohort does not have a strong impact on work values, we did detect differences in work values by both age and time period. Thus, we should be aware that generational stability does not mean full-scale similarity. For example, the slow but steady decrease in the centrality of work by period might call for the development of policies that relax the association between life and work for future generations. It seems likely that instead of having work as the central social phenomenon that gives meaning to life, multiple centrality (having work as one important life aspect) is becoming increasingly more common among Europeans.

    (p.650)

    References

    Bibliography references:

    Alesina, Alberto, and Paola Giuliano. 2010. “The Power of the Family.” Journal of Economic Growth 15 (2): 93–125.

    Becton, John Bret, Harvell Jack Walker, and Allison Jones-Farmer. 2014. “Generational Differences in Workplace Behavior.” Journal of Applied Social Psychology 44 (3): 175–89. doi10.1111/jasp.12208.

    Bell, Andrew, and Kelvyn Jones. 2014. “Another ‘Futile Quest’? A Simulation Study of Yang and Land’s Hierarchical Age–Period–Cohort Model.” Demographic Research 30 (February): 333–60. doi:10.4054/DemRes.2014.30.11.

    Brief, Arthur P., and Ramon J. Aldag. 1977. “The Intrinsic–Extrinsic Dichotomy: Toward Conceptual Clarity.” Academy of Management Review 2 (3): 496–500. doi:10.5465/AMR.1977.4281861.

    Brügger, Beatrix, Rafael Lalive, and Josef Zweimüller. 2009. “Does Culture Affect Unemployment? Evidence from the Röstigraben.” CEPR Discussion Paper 7405. London: Centre for Economic Policy Research. https://ideas.repec.org/p/cpr/ceprdp/7405.html

    (p.651) Clark, Andrew E. 2010. “Work, Jobs, and Well-Being Across the Millennium.” In International Differences in Well-Being, edited by Ed Diener, John F. Helliwell, and Daniel Kahneman, 436–68. Oxford: Oxford University Press.

    Costanza, David P., Jessica M. Badger, Rebecca L. Fraser, Jamie B. Severt, and Paul A. Gade. 2012. “Generational Differences in Work-Related Attitudes: A Meta-Analysis.” Journal of Business and Psychology 27 (4): 375–94. doi:10.1007/s10869-012-9259-4.

    Den Dulk, Laura, Sandra Groeneveld, Ariane Ollier-Malaterre, and Monique Valcour. 2013. “National Context in Work–Life Research: A Multi-Level Cross-National Analysis of the Adoption of Workplace Work–Life Arrangements in Europe.” European Management Journal 31 (5): 478–94. doi:10.1016/j.emj.2013.04.010.

    Diepstraten, Isabelle, Peter Ester, and Henk Vinken. 1999. “Talkin’ ‘Bout My Generation: Ego and Alter Images of Generations in the Netherlands.” The Netherlands Journal of Social Sciences 35 (2): 91–109.

    Dinesen, Peter Thisted. 2013. “Where You Come from or Where You Live? Examining the Cultural and Institutional Explanation of Generalized Trust Using Migration as a Natural Experiment.” European Sociological Review 29 (1): 114–28. doi:10.1093/esr/jcr044.

    Down, Ian, and Carole J. Wilson. 2013. “A Rising Generation of Europeans? Life-Cycle and Cohort Effects on Support for ‘Europe.’” European Journal of Political Research 52 (4): 431–56. doi:10.1111/1475-6765.12001.

    Egri, Carolyn P., and David A. Ralston. 2004. “Generation Cohorts and Personal Values: A Comparison of China and the United States.” Organization Science 15 (2): 210–20. doi:10.1287/orsc.1030.0048.

    Fernández, Raquel, and Alessandra Fogli. 2009. “Culture: An Empirical Investigation of Beliefs, Work, and Fertility.” American Economic Journal: Macroeconomics 1 (1): 146–77. doi:10.1257/mac.1.1.146.

    Firebaugh, Glenn. 1997. Analyzing Repeated Surveys. Sage University Paper 115. Thousand Oaks, CA: Sage.

    Hajdu, Gábor, and Tamás Hajdu. 2016. “The Impact of Culture on Well-Being: Evidence from a Natural Experiment.” Journal of Happiness Studies 17 (3): 1089–110. doi:10.1007/s10902-015-9633-9.

    Hajdu, Gábor, and Endre Sik. 2015. “Searching for Gaps: Are Work Values of the Younger Generations Changing?” STYLE Working Paper 9.1. Brighton, UK: CROME, University of Brighton. http://www.style-research.eu/publications/working-papers

    Hansen, Jo-Ida C., and Melanie E. Leuty. 2012. “Work Values Across Generations.” Journal of Career Assessment 20 (1): 34–52. doi:10.1177/1069072711417163.

    Harpaz, Itzhak, and Xuanning Fu. 2002. “The Structure of the Meaning of Work: A Relative Stability Amidst Change.” Human Relations 55 (6): 639–67. doi:10.1177/0018726702556002.

    (p.652) Jin, Jing, and James Rounds. 2012. “Stability and Change in Work Values: A Meta-Analysis of Longitudinal Studies.” Journal of Vocational Behavior 80 (2): 326–39. doi:10.1016/j.jvb.2011.10.007.

    Kowske, Brenda J., Rena Rasch, and Jack Wiley. 2010. “Millennials’ (Lack of) Attitude Problem: An Empirical Examination of Generational Effects on Work Attitudes.” Journal of Business and Psychology 25 (2): 265–79. doi:10.1007/s10869-010-9171-8.

    Luttmer, Erzo F. P., and Monica Singhal. 2011. “Culture, Context, and the Taste for Redistribution.” American Economic Journal: Economic Policy 3 (1): 157–79. doi:10.1257/pol.3.1.157.

    Roe, Robert A., and Peter Ester. 1999. “Values and Work: Empirical Findings and Theoretical Perspective.” Applied Psychology 48 (1): 1–21. doi:10.1111/j.1464-0597.1999.tb00046.x.

    Ros, Maria, Shalom H. Schwartz, and Shoshana Surkiss. 1999. “Basic Individual Values, Work Values, and the Meaning of Work.” Applied Psychology 48 (1): 49–71. doi:10.1111/j.1464-0597.1999.tb00048.x.

    Schwadel, Philip. 2014. “Birth Cohort Changes in the Association Between College Education and Religious Non-Affiliation.” Social Forces (August): 1–28. doi:10.1093/sf/sou080.

    Senik, Claudia. 2014. “The French Unhappiness Puzzle: The Cultural Dimension of Happiness.” Journal of Economic Behavior & Organization 106 (October): 379–401. doi:10.1016/j.jebo.2014.05.010.

    Sharabi, Moshe, and Itzhak Harpaz. 2007. “Changes in Work Centrality and Other Life Areas in Israel: A Longitudinal Study.” Journal of Human Values 13 (2): 95–106. doi:10.1177/097168580701300203.

    Suzuki, Etsuji. 2012. “Time Changes, So Do People.” Social Science & Medicine 75 (3): 452–56. doi:10.1016/j.socscimed.2012.03.036.

    Tolbize, Anick. 2008. Generational Differences in the Workplace. Minneapolis: Research and Training Center on Community Living, University of Minnesota.

    Twenge, Jean M. 2010. “A Review of the Empirical Evidence on Generational Differences in Work Attitudes.” Journal of Business and Psychology 25 (2): 201–10. doi:10.1007/s10869-010-9165-6.

    Twenge, Jean M., Stacy M. Campbell, Brian J. Hoffman, and Charles E. Lance. 2010. “Generational Differences in Work Values: Leisure and Extrinsic Values Increasing, Social and Intrinsic Values Decreasing.” Journal of Management 36 (5): 1117–42. doi:10.1177/0149206309352246.

    Wollack, Stephen, James G. Goodale, Jan P. Wijting, and Patricia C. Smith. 1971. “Development of the Survey of Work Values.” Journal of Applied Psychology 55 (4): 331–38. doi:10.1037/h0031531.

    Wuthnow, Robert. 2008. “The Sociological Study of Values.” Sociological Forum 23 (2): 333–43. doi:10.1111/j.1573-7861.2008.00063.x.

    (p.653) Yang, Yang, Wenjiang Fu, and Kenneth C. Land. 2008. “The Intrinsic Estimator for Age–Period–Cohort Analysis: What It Is and How to Use It.” American Journal of Sociology 113 (6): 1697–736.

    Yang, Yang, and Kenneth C. Land. 2006. “A Mixed Models Approach to the Age–Period–Cohort Analysis of Repeated Cross-Section Surveys, with an Application to Data on Trends in Verbal Test Scores.” Sociological Methodology 36 (1): 75–97. doi:10.1111/j.1467-9531.2006.00175.x.

    Yang, Yang, and Kenneth C. Land. 2008. “Age–Period–Cohort Analysis of Repeated Cross-Section Surveys: Fixed or Random Effects?” Sociological Methods & Research 36 (3): 297–326. doi:10.1177/0049124106292360.

    We use cross-classified hierarchical regression models. The level 1 model is as follows:

    Yijk=β0jk+β1AGEijk+β2AGEijk2+β3Xijk+

    The level 2 model is

    β0jk=γ0+u0j+v0k

    The combined model is

    Yijk=γ0+β1AGEijk+β2AGEijk2+β3Xijk+u0j+v0k+eijk

    where, within each cohort j and period k, respondents’ work attitude is a function of their age, squared age, and other individual characteristics (vector of X). This model allows level 1 intercepts to vary randomly by cohorts and periods. β‎0jk is the mean of the work-attitude variable of individuals in cohort j and period k (cell mean); β‎1, β‎2, and β‎3 are the level 1 fixed effects; eijk is the random individual variation, which is assumed to be normally distributed with mean 0 and within-cell variance σ‎2; γ‎0 is the grand mean (across all cohorts and periods) or the model intercept; u0j is the residual random effect of cohort j; and v0j is the residual random effect of period k. Both u0j and v0j are assumed to be normally distributed with mean 0 and variance τ‎u and τ‎v, respectively.

    Table A21.1 Number of observations of relative centrality of work by country and year of fieldwork

    Country

    1990

    1991

    1992

    1993

    1995

    1996

    1997

    1998

    1999

    2005

    2006

    2007

    2008

    2009

    Total

    EU15

    AT

    1,395

    0

    0

    0

    0

    0

    0

    0

    1,495

    0

    0

    0

    1,505

    0

    4,395

    BE

    2,578

    0

    0

    0

    0

    0

    0

    0

    1,785

    0

    0

    0

    0

    1,490

    5,853

    DE-W

    3,276

    0

    0

    0

    0

    0

    1,954

    0

    1,990

    0

    1,908

    0

    1,999

    0

    11,127

    DK

    994

    0

    0

    0

    0

    0

    0

    0

    998

    0

    0

    0

    1,386

    0

    3,378

    ES

    3,404

    0

    0

    0

    1,202

    0

    0

    0

    1,193

    0

    0

    1,175

    1,483

    0

    8,457

    FI

    48

    0

    0

    0

    0

    901

    0

    0

    0

    973

    0

    0

    0

    1,061

    2,983

    FR

    786

    0

    0

    0

    0

    0

    0

    0

    1,541

    0

    963

    0

    1,484

    0

    4,774

    GB

    1,405

    0

    0

    0

    0

    0

    0

    0

    847

    0

    918

    0

    0

    895

    4,065

    EL

    0

    0

    0

    0

    0

    0

    0

    0

    1,039

    0

    0

    0

    1,489

    0

    2,528

    IE

    991

    0

    0

    0

    0

    0

    0

    0

    902

    0

    0

    0

    541

    0

    2,434

    IT

    1,960

    0

    0

    0

    0

    0

    0

    0

    1,970

    978

    0

    0

    0

    1,409

    6,318

    LU

    0

    0

    0

    0

    0

    0

    0

    0

    1,107

    0

    0

    0

    1,592

    0

    2,699

    NL

    976

    0

    0

    0

    0

    0

    0

    0

    959

    0

    950

    0

    1,533

    0

    4,418

    PT

    1,088

    0

    0

    0

    0

    0

    0

    0

    995

    0

    0

    0

    1,527

    0

    3,610

    SE

    909

    0

    0

    0

    0

    990

    0

    0

    740

    0

    984

    0

    0

    987

    4,610

    Post-socialist

    BA

    0

    0

    0

    0

    0

    0

    0

    1,178

    0

    0

    0

    0

    1,356

    0

    2,534

    BG

    942

    0

    0

    0

    0

    0

    986

    0

    974

    0

    963

    0

    1,397

    0

    5,262

    CS

    0

    0

    0

    0

    0

    1,455

    0

    0

    0

    0

    0

    0

    0

    0

    1,455

    CZ

    770

    2,082

    0

    0

    0

    0

    0

    1,084

    1,879

    0

    0

    0

    1,696

    0

    7,512

    EE

    960

    0

    0

    0

    0

    1,000

    0

    0

    989

    0

    0

    0

    1,502

    0

    4,452

    HR

    0

    0

    0

    0

    0

    0

    0

    0

    933

    0

    0

    0

    1,410

    0

    2,343

    HU

    0

    981

    0

    0

    0

    0

    0

    630

    975

    0

    0

    0

    1,506

    0

    4,093

    LT

    0

    0

    0

    0

    0

    0

    969

    0

    993

    0

    0

    0

    1,462

    0

    3,425

    LV

    813

    0

    0

    0

    0

    1,160

    0

    0

    984

    0

    0

    0

    1,488

    0

    4,445

    PL

    960

    0

    0

    0

    0

    0

    0

    0

    1,082

    979

    0

    0

    1,448

    0

    4,469

    RO

    0

    0

    0

    1,077

    0

    0

    0

    1,226

    1,131

    1,709

    0

    0

    1,430

    0

    6,573

    RU

    1,000

    0

    0

    0

    2,007

    0

    0

    0

    2,454

    0

    1,865

    0

    1,442

    0

    8,769

    SI

    0

    0

    948

    0

    970

    0

    0

    0

    987

    1,024

    0

    0

    1,337

    0

    5,266

    SK

    381

    1,104

    0

    0

    0

    0

    0

    1,037

    1,323

    0

    0

    0

    1,493

    0

    5,337

    UA

    0

    0

    0

    0

    0

    2,662

    0

    0

    1,142

    0

    967

    0

    1,478

    0

    6,249

    Other

    AU

    0

    0

    0

    0

    1,857

    0

    0

    0

    0

    1,216

    0

    0

    0

    0

    3,073

    CA

    1,675

    0

    0

    0

    0

    0

    0

    0

    0

    0

    2,015

    0

    0

    0

    3,690

    CH

    0

    0

    0

    0

    0

    1,149

    0

    0

    0

    0

    0

    0

    1,228

    0

    2,377

    IS

    0

    0

    0

    0

    0

    0

    0

    0

    930

    0

    0

    0

    0

    0

    930

    MT

    0

    0

    0

    0

    0

    0

    0

    0

    988

    0

    0

    0

    685

    0

    1,673

    NO

    1,139

    0

    0

    0

    0

    1,114

    0

    0

    0

    0

    0

    0

    2,096

    0

    4,349

    NZ

    0

    0

    0

    0

    0

    0

    0

    1,025

    0

    0

    0

    0

    0

    0

    1,025

    US

    1,662

    0

    0

    0

    1,379

    0

    0

    0

    1,184

    0

    1,163

    0

    0

    0

    5,388

    Total

    30,115

    4,168

    948

    1,077

    7,416

    10,432

    3,910

    6,181

    36,513

    6,879

    12,694

    1,175

    38,991

    5,842

    16,6340

    Source: World Values Survey/European Values Study.

    Table A21.2 Mean of relative centrality of work by country and year of fieldwork

    Country

    1990

    1991

    1992

    1993

    1995

    1996

    1997

    1998

    1999

    2005

    2006

    2007

    2008

    2009

    Total

    EU15

    AT

    0.126

    0.082

    –0.117

    0.028

    BE

    0.001

    0.027

    –0.114

    –0.020

    DE-W

    –0.195

    –0.148

    –0.223

    –0.140

    –0.234

    –0.189

    DK

    –0.173

    –0.358

    –0.309

    –0.283

    ES

    0.116

    0.051

    0.161

    –0.198

    0.016

    0.052

    FI

    0.069

    –0.196

    –0.345

    –0.435

    –0.325

    FR

    0.170

    0.109

    0.072

    0.102

    0.110

    GB

    –0.339

    –0.542

    –0.580

    –0.625

    –0.499

    EL

    0.044

    0.045

    0.045

    IE

    0.033

    –0.255

    –0.349

    –0.159

    IT

    0.136

    0.151

    0.074

    0.168

    0.138

    LU

    –0.075

    0.125

    0.043

    NL

    –0.158

    –0.284

    –0.506

    –0.331

    –0.320

    PT

    0.105

    0.166

    0.131

    0.133

    SE

    –0.042

    –0.090

    –0.258

    –0.280

    –0.334

    –0.200

    Post-socialist

    BA

    0.104

    –0.005

    0.046

    BG

    0.151

    0.065

    0.186

    –0.107

    0.070

    0.073

    CS

    0.126

    0.126

    CZ

    0.135

    0.293

    –0.095

    0.141

    –0.187

    0.075

    EE

    –0.019

    0.248

    0.159

    –0.016

    0.082

    HR

    0.041

    –0.017

    0.006

    HU

    0.199

    0.008

    0.067

    –0.048

    0.047

    LT

    0.157

    0.216

    0.049

    0.128

    LV

    0.041

    0.296

    0.508

    0.192

    0.261

    PL

    0.285

    0.362

    –0.001

    –0.027

    0.140

    RO

    0.401

    0.293

    0.418

    0.183

    0.124

    0.267

    RU

    –0.019

    –0.003

    0.168

    –0.115

    –0.090

    0.005

    SI

    0.409

    0.100

    0.176

    –0.147

    0.031

    0.104

    SK

    0.203

    0.260

    –0.052

    0.147

    0.042

    0.106

    UA

    –0.032

    0.086

    –0.264

    –0.069

    –0.055

    Other

    AU

    –0.263

    –0.463

    –0.342

    CA

    –0.130

    –0.295

    –0.220

    CH

    –0.203

    –0.038

    –0.118

    IS

    –0.021

    –0.021

    MT

    0.233

    0.118

    0.186

    NO

    0.068

    –0.068

    –0.183

    –0.088

    NZ

    –0.271

    –0.271

    US

    –0.115

    –0.265

    –0.222

    –0.534

    –0.268

    Total

    –0.003

    0.262

    0.409

    0.401

    –0.095

    0.011

    –0.019

    0.008

    0.058

    –0.097

    –0.258

    –0.198

    –0.037

    –0.220

    –0.022

    Source: World Values Survey/European Values Study.

    (p.659)

    Notes:

    (1) In the course of our analysis, we use “centrality of work” as the dependent variable because it refers to work in the widest sense (i.e., work as a basic human activity). As we note in Section 21.2.2, the working paper version of this chapter covers other variables of work values as well, such as employment commitment and extrinsic/intrinsic values (Hajdu and Sik 2015).

    (2) Usually, political/economic/technological periodizations relating to the United States are the basis of these global generational definitions, as defined, for example, by Twenge et al. (2010):

    • () Baby boomers by the civil rights and women’s movements, the Vietnam War, and the assassinations of John F. Kennedy and Martin Luther King

    • () GenX by the AIDS epidemic, economic uncertainty, and the fall of the Soviet Union

    • () GenY by being “wired” and “tech savvy,” liking “informality,” learning quickly, and embracing “diversity”

    On the other hand, Diepstraten et al. (1999), for example, identified “prewar,” “silent,” “protest,” “lost,” and “pragmatic” generations for the Netherlands on the basis of an entirely different national “story.”

    (3) For example, on redistribution, see Luttmer and Singhal (2011); on trust, see Dinesen (2013); on subjective well-being, see Senik (2014) and Hajdu and Hajdu (2016); and on female labor force participation, see Fernández and Fogli (2009) and Alesina and Giuliano (2010). The most notable example of illustrating the impact of ethnicity on work values is the analysis of the role of an ethnic border (the so-called Röstigraben) in Switzerland (Brügger, Lalive, and Zweimüller 2009).

    (4) In the working paper version of this chapter, employment commitment and extrinsic/intrinsic work values were used as dependent variables as well. Employment commitment—that is, paid work only—was considered as a more restricted form of the centrality of work. From this viewpoint, work is conceptualized as the source of income, and the question is whether the respondents consider paid work as a standard economic resource (and therefore work only until its aggregate return does not start to decrease) or not (i.e., they do paid work for its own sake). Extrinsic/intrinsic work values are widely used in the organization, business, and management literature. An extrinsic work value is “dependent on a source external to the immediate task-person situation . . . such as status, respect, power, influence, high salary.” An intrinsic value, on the contrary, is “derived from the task per se; that is, from outcomes which are not mediated by a source external to the task–person situation. Such a state of motivation can be characterized as a (p.649) self-fulfilling experience” (Brief and Aldag 1977, 497–98). In the working paper, we used three extrinsic work values (good income, security, and flexibility) and two intrinsic values (interesting job and having a job that is useful to society) that are considered important by the respondents in evaluating a job (Hajdu and Sik 2015).

    (5) Hierarchical age–period–cohort (HAPC) regression models have been used to analyze repeated cross-sectional data by Yang and Land (2006, 2008) in examining verbal test scores; by Schwadel (2014) in examining the changing association between higher education and reporting no religious affiliation in the United States; by Down and Wilson (2013) in examining life cycle and cohort effects on support for the EU; and by Kowske et al. (2010) in examining the effect of generation on job satisfaction and on satisfaction with other job aspects.

    (6) This can only be done artificially, so it is ultimately a subjective decision by the researcher. However, we grouped our data by taking account of waves of surveys so that data from each wave were grouped together into 5-year intervals, which can be considered the most “natural” (i.e., “theory-blind”) grouping principle.

    (7) Yang and Land (2008) argue that whereas the age variable is related to the biological process of individual aging, period and cohort effects reflect the influences of external (political, technological, economic, etc.) forces; thus, the latter two variables can be treated as level 2 (or macro-level) variables. Suzuki (2012, 453) shows a data structure in which individuals are nested simultaneously within periods and birth cohorts, whereas age is an attribute of individuals rather than a random sample of age categories from a population of age groupings.

    (8) Detailed descriptions of the models are provided in Appendix 1.

    (9) Other researchers using these variables created complex scales (Wollack et al. 1971; Ros, Schwartz, and Surkiss 1999; Den Dulk et al. 2013), but we wanted to keep our variables simple so as to ensure that they are understood identically by respondents in subsequent surveys and different cultures.

    (10) The coding was as follows: 1 (very important), 2 (quite important), 3 (not important), and 4 (not important at all).

    (11) Although the second wave of WVS/EVS was conducted between 1989 and 1993, the date of the fieldwork was between 1990 and 1993 in all but two of the participating countries. We excluded from this wave two countries (Poland and Switzerland)—where the year of the fieldwork was 1989—in order to avoid a small sample size for this year (or in the period 1985–1989) and also to avoid results driven by only two countries. Moreover, because the number of observations between 2000 and 2004 is relatively low, given that the fourth (1999–2004) wave of WVS/EVS was conducted in most countries in 1999, we excluded this period from the analysis as well.

    (12) The same applies to defining the age of the respondent: It was calculated as the difference between the year of the fieldwork and the respondent’s birth year.

    (13) We show only six age groups (two of the youngest groups, two from the middle-aged groups, and two of the oldest groups) in order to have a less cluttered table.

    (14) Because we analyze respondents of similar age, this model can be conceptualized as a special form of the time-lag method recommended by Twenge (2010).

    (15) This lack of differences between men and women has also been found by other authors examining various work values (e.g., Clark 2010).

    (16) An “average man” is a man who has average characteristics (average values of the control variables among the men), and an “average woman” is a woman who has average characteristics (average values of the control variables among the women).

    (17) As Table A21.2 in Appendix 2 shows, the relative centrality of work differs significantly across countries. However, because a comparative analysis of the trend for relative centrality of work at the country level would require a separate paper, we restrict ourselves to a regional (i.e., semi-aggregated version of country-specific) comparative analysis.

    (18) Germany is split into two parts: federal states from the former West Germany as an EU15 country and federal states from the former East Germany as a post-socialist country.