The work values of Portuguese Generation Z in the higher education-to-work transition phase

$ € £¥ social sciences Article The Work Values of Portuguese Generation Z in the Higher Education-to-Work Transition Phase Joaquim Silva 1,* and Ana Carvalho 2   Citation: Silva, Joaquim, and Ana Carvalho. 2021. The Work Values of Portuguese Generation Z in the Higher Education-to-Work Transition Phase. Social Sciences 10: 297. https://doi.org/10.3390/socsci 10080297 Academic Editors: Karen Arnold and Kathy Chau Rohn Received: 29 June 2021 Accepted: 3 August 2021 Published: 5 August 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1Interdisciplinary Centre of Social Sciences (CICS.NOVA.UMinho), School of Economics and Management, University of Minho, 4710-057 Braga, Portugal 2NIPE—Centre for Research in Economics and Management, School of Economics and Management, University of Minho, 4710-057 Braga, Portugal; [email protected] *Correspondence: [email protected] Abstract: The cohort of young people born with or after the Internet has been dubbed Generation Z (Gen Z, or post-millennials). They are now entering the higher education-to-work transition phase, although this is yet to be studied. Previous studies have found that work values and work preferences vary across generations and national cultures, justifying regular and localised examination. However, very little is known to date about the work values of Portuguese Gen Zs. In this study, we describe the work values of Portuguese university students in the higher education-to-work transition phase and examine their influence on salient work-related preferences. We surveyed over 3000 students attending university degrees from eight main Portuguese universities. We find that Portuguese Gen Zs value social values above all, followed by intrinsic and then extrinsic work values, potentially configuring a unique profile. Work values partially explain work preferences such as employer size and salary expectations. Gender differences mark our results, with women expressing higher work values in all three dimensions and lower entry salary expectations. These results can be useful for employers seeking to attract the best university graduates, facilitate their integration and promote their development. Keywords: young adults; Generation Z; work values; higher education-to-work transition; Portugal 1. Introduction Work values are pivotal during the higher education-to-work transition. Expressing people’s preferences and expectations about work and the workplace strongly influence career and job choice decisions central in the school-to-work transition phase (Chow et al. 2017;Judge and Bretz 1992;Kuron et al. 2015;Lukeš et al. 2019). Knowledge about young people’s work values can therefore be very useful for employers trying to devise effective attraction and retention strategies (Cogin 2012) and to inform public policies to tackle youth unemployment and to facilitate the transition from education to employment (Lukeš et al. 2019). Research shows that work values vary across generations (e.g., Lyons et al. 2015; Twenge et al. 2010), life and career stages (Kuron et al. 2015), and national cultures (Cennamo and Gardner 2008; Gallie 2019), justifying the continued study of new generations in different countries. The concept of generation is complex and susceptible to the confounding effects of age, life stage, and career stage (Rudolph et al. 2018;Twenge 2010). We follow Lyons and Kuron (2014, p. S141) in taking “age, period and cohort effects as complementary”, which should be concomitantly analysed. A generation is, therefore, the set of individuals experiencing the same historical and sociocultural context at a specific life stage, similarly influenced by common formative experiences and opportunities (Lyons and Kuron 2014). Young people now facing higher education-to-work transition, born with or after the Internet, have been dubbed “Generation Z” (or Gen Z) (Childers and Boatwright Soc. Sci. 2021,10, 297. https://doi.org/10.3390/socsci10080297 https://www.mdpi.com/journal/socsci Soc. Sci. 2021,10, 297 2 of 16 2020;Pichler et al. 2021). Little academic research has been conducted about them as young adults. Although there is consensus about the environment in which Gen Z grew up that contributes to the development of a specific configuration of preferences and expectations about work and the workplace (Johnson et al. 2012;Pichler et al. 2021), this effect has been scarcely documented. This gap is particularly evident in Portugal, where the literature about the work values and preferences of Portuguese young adults is scarce and dated. Therefore, our research addresses the question “What are the work values of the Gen Z Portuguese university students and how do they influence their job preferences in the higher education-to-work transition?”, aiming to portray the work values of these young people about to enter in the job market and analyse their impact on work-related preferences. We base our empirical research on a survey of over 3000 university students in Portugal. The study contributes to a systematic profiling of the work values and job preferences of a new generation of future employees, adding to the existent knowledge about previous generations. Additionally, we expect to offer guidelines to prospective employers to better target their recruitment efforts and devise communication, socialisation, and development strategies that meet graduates’ expectations and aspirations (Casper et al. 2013;Lukeš et al. 2019). 1.1. Generation Z Generation Z (also known as post-millennials, iGen, Gen 2020, or Gen Zs) is defined as those born with of after the Internet (Pichler et al. 2021). Previous generations include the Silent Generation (born 1925–1945), Baby Boomers (born 1946–1964), Generation X (born 1965–1981), and Generation Y or Millennials (born 1982–1999) (Twenge et al. 2010). The extant research on Gen Z studies them as children or adolescents, taking educational (e.g., Plochocki 2019), cultural (e.g., Carrington et al. 2015), or marketing perspectives (e.g., Childers and Boatwright 2020). Very few academic studies have observed Gen Z as young adults, and much of what is written in the popular press is conjecture and surmise, to the point that contradictory depictions are presented. For example, they are portrayed both as interacting electronically more than personally and finding online communication more comfortable (Turner 2015;Clark 2017), but also as preferring in-person communication with managers and peers (Bridges 2015). They seem both to have very little concern about sharing private and personal matters (Clark 2017) and be more conscious about online privacy, preferring anonymous social media platforms (Williams 2015). Rigorous study is therefore much in need. There is more convergence in the description of the context in which Gen Zs were brought up, particularly in the West, much of which is shared by Portuguese Gen Zs. Socioeconomic context is a key variable in explaining differences among generations (Macky et al. 2008;De Hauw and De Vos 2010). Gen Zs grew up in the “unsafe world” (Turner 2015, p. 105) of terrorism, war, high unemployment, and economic depression of the early 21st century (Rainsford et al. 2019). Physical and financial security is, therefore, a central concern for themselves (Williams and Page 2011;Bridges 2015) and their parents (Tulgan 2013), who raised them in an over-protective environment, justifying the “bubble-wrap generation” epithet (Malone 2007). In Portugal, terrorism and war are distant realities, even if made close by (social) media. However, financial uncertainty is very present, with an economic crisis that culminated in an international bailout in 2011 (Suleman and Figueiredo 2019). Rising unemployment (particularly youth unemployment), generalised wage cuts, and a return to (now also highly skilled) emigration (Pereira and Lains 2011;Cruces et al. 2015;Observatório da Emigração 2015) produced a “brain drain” and general disenchantment among young people (Cerdeira et al. 2016, p. 68). Precarious lower-paying jobs have grown, especially for new entrants, affecting young people’s employment expectations and decisions (Suleman and Figueiredo 2019). The ubiquitous presence of technology (e.g., wireless Internet, smartphones, tracking devices) is another key feature (Pichler et al. 2021). Internet access in family households Soc. Sci. 2021,10, 297 3 of 16 in Portugal rose from 15% in 2002 to 84.5% in 2020, 96.7% of which through broadband (INE 2020), which means most Portuguese youngsters have grown up with easy access to the Internet at home. Gen Zs are innately intimate with technology (Pichler et al. 2021), routinely spending extensive time on social media, where they connect with friends and create a sense of community (Childers and Boatwright 2020). They are emotionally attached to their mobile devices and Internet connection (Turner 2015). They are used to having immediate access to their friends and the information they need online (Pichler et al. 2021), as well as constant social recognition, prompting concerns for the need for instant gratification, no time for reflection or idleness (Turner 2015), increase in narcissistic behaviour (Weiser 2015) and declining emotional health (Pichler et al. 2021). 1.2. Work Values Work values represent beliefs related to desired outcomes or aspects of work (Elizur et al. 1991;Judge and Bretz 1992;Lukeš et al. 2019;Ros et al. 1999), serving as criteria for individuals to evaluate work attributes, activities, and results (Elizur et al. 1991;Cogin 2012). As such, they can be expressed as preferences, goals, and expectations (Elizur et al. 1991;Ros et al. 1999;Cennamo and Gardner 2008). They are associated with a number of relevant work-related attitudes and behaviours, including job choice decisions (Judge and Bretz 1992), career intentions (Hirschi and Fischer 2013), entrepreneurial drive (Lukeš et al. 2019), and career satisfaction in later life (Chow et al. 2017). One of the more established work values frameworks is the three-dimensional configuration that distinguishes among intrinsic (relating to the meaning and interest of the work itself), extrinsic (material benefits such as salary and security), and social values (sense of belonging and social relations with co-workers and supervisors) (Ros et al. 1999;Twenge et al. 2010;De Cooman and Dries 2012). Work values seem to emerge and evolve during adolescence and are fairly stable from early adulthood (Hirschi and Fischer 2013;Papavasileiou and Lyons 2015;Chow et al. 2017). Although in previous generations, students’ work values shifted as they started working, they seemed to remain similar in the school-to-work transition in the case of millennials (Kuron et al. 2015), which means the work values of university students may be a good predictor of later-life work values and related outcomes (Chow et al. 2017). Variation across generations has received particular attention, with several studies reporting significant differences in the work values of successive generations (e.g., Cennamo and Gardner 2008;Cogin 2012;Krahn and Galambos 2014). One central argument to explain generational variation is exposure to different sociocultural environments that shape values and expectations. People in the same generational cohort are socialised within a specific environment marked by economic, social, cultural, technological, and political salient events. The environment evolves and changes over time, exerting diverse influence over different generations (Macky et al. 2008;De Hauw and De Vos 2010), particularly in the formative years (Lyons and Kuron 2014). Johnson et al. (2012) argue that conditions of unemployment and precarious jobs, or employment unrelated to career expectations, can weaken extrinsic and intrinsic values, respectively, and they show that extrinsic values are stronger in early adulthood rather than later in life, which is corroborated by recent research of young Europeans (Rainsford et al. 2019). Rainsford et al. (2019) also find that those who feel overqualified place more value on extrinsic rewards (Rainsford et al. 2019). Appreciation for intrinsic values has varied little across generations (Gallie 2019;Twenge et al. 2010), with a decrease noted in some studies (Cennamo and Gardner 2008). Extrinsic work values, in turn, are reported to be on the rise across previous generations (Krahn and Galambos 2014;Lyons et al. 2015;Twenge and Donnelly 2016). In millennials (the generation prior to Gen Z), social values are often prominent (Cennamo and Gardner 2008;Ng et al. 2010). Gen Zs, however, seem to rely more on electronic communication and spend less time in less in-person social interaction, resulting in underdeveloped social skills (Pichler et al. 2021). Soc. Sci. 2021,10, 297 4 of 16 Very little, if anything, has been published in international journals about the work values of Portuguese individuals. An exception is Chaves et al. (2016), who studied the work values of 1000 graduates of the two main universities in Lisbon. The data were collected in 2011, and 75% of the sample were aged up to 31, in what can be considered a sample of millennials. Only the Intrinsic and Extrinsic dimensions were included. Results indicate Portuguese millennials’ highest priority is to have financial autonomy from parents (extrinsic), closely followed by having an interesting job (intrinsic), one that allows them knowledge acquisition and autonomy. A secure, stable, and high-wage job (extrinsic) comes last (although still highly scored). Based on these results, and considering the effect of the recession mentioned above (Johnson et al. 2012; Rainsford et al. 2019), we expect Portuguese Gen Z to score higher on extrinsic than on intrinsic values. For expectations about social values, we rely on cultural aspects. The Globe project reveals that one of the dominant characteristics of Portuguese culture is a high in-group collectivist orientation with a strong appreciation of social loyalty and group ties (Jesuíno 2007). So, we anticipate Portuguese Gen Z to place social values on top. 1.3. The Impact of Work Values on Work Preferences We focus specifically on the impact of work values on preference for some organisational attributes that are highly salient to university students before entering the job market, namely company size and expected entry salary (Cable and Turban 2001;Demel et al. 2019). The link between work values and these preferences can be explained by classical motivational theories. Herzberg’s two-factor theory, for example, proposes that satisfier factors (or motivators) such as achievement, responsibility, and interesting work meet intrinsic needs, whereas dissatisfaction avoidance (or hygiene) factors such as work conditions, status, and security constitute extrinsic incentives (Herzberg 2003). Hill’s affiliation motivation theory explains how people’s affiliation needs are met by social rewards such as interpersonal closeness and communion or attention and praise (Hill 1987). Previous studies found that young candidates have distinct preferences for employer size (Demel et al. 2019). Larger companies provide higher earnings and stability, more fringe benefits, and better promotion opportunities but allow employees less autonomy (Kalleberg and Buren 1996), therefore ensuring mainly extrinsic rewards. This holds true for Portugal (Suleman and Figueiredo 2019). Organisational climate is also expected to be less sociable in larger organisations (Payne and Mansfield 1973) and perceived by employees as offering lower organisational support and less flexibility in dealing with individual needs (Rhoades and Eisenberger 2002), which may be less socially rewarding. More entrepreneurially inclined people, who value variety over long-term security, prefer smaller, less bureaucratic firms (Sørensen 2007), where their intrinsic needs are more likely to be met. We would, therefore, expect young people with stronger intrinsic and social values to prefer smaller companies and those with stronger extrinsic values to prefer larger firms. Salary is another major factor in job choice decisions (Demel et al. 2019) and was especially valued by the millennial generation (Ng et al. 2010). Valuing high salaries is associated with extrinsic values (Rainsford et al. 2019). Unemployment and insecurity can justify Gen Z’s concern with financial security (Williams and Page 2011) but may also weaken extrinsic values (Johnson et al. 2012). On the other hand, Portugal is characterised by low salaries, aggravated by the bailout intervention (Suleman and Figueiredo 2019). So, we expect Portuguese Gen Z not to expect high wages despite valuing financial security. 2. Methods 2.1. Sample Birth dates proposed to define Gen Z range from 1991 (Malone 2007) to 2000 (Pichler et al. 2021). For our study, we use 1995 as the reference (Twenge et al. 2017). The eldest of this generation was, therefore, 23 years old at the time of data collection. Soc. Sci. 2021,10, 297 5 of 16 University students were chosen as a population of interest as they fit Gen Z’s age range and are in the process of higher-education-to-work transition. To collect data, we conducted an online survey of Portuguese Gen Z’s work values, administered to students of business and technology degrees from the main universities in Portugal (Minho, Porto, Aveiro, Coimbra, Lisbon, Nova Lisbon, Catholic Lisbon, ISCTE) between February and April 2018, using e-mail databases, digital communication platforms and the action of student leaders to mobilise participation. We received 3375 questionnaires from which we eliminated 26 (0.77%) because they were very incomplete or showed evidence that answers were given indiscriminately (with all answers positioned in categories 1, 4, or 7 on the Likert scales) (Malhotra 2020). From the resulting 3349 valid questionnaires, a sample of 300 responses was randomly extracted to perform an exploratory factor analysis, and the remaining sample of 3049 responses was used to conduct a confirmatory factor analysis and further data analysis. 2.2. Variables and Measures We measured three dimensions of work values: Intrinsic (the meaning and interest of the work itself), extrinsic (material benefits such as salary and job security), and social (social relations with co-workers and supervisors). An initial pool of 19 items was built based on Berthon et al.’s (2005) scale of employer attractiveness, which measures the “envisioned benefits” of working for an organisation. This conception is similar to our definition of work values as preferences and expectations about work and the workplace (Judge and Bretz 1992; Ros et al. 1999). The items were subject to a process of translation and back-translation (Behling and Law 2000) and were analysed for adequacy by two top recruitment managers and one HRM academic. A reduced list of 15 items (Table 1) was submitted to a pilot test with 50 students to evaluate adequacy, understandability, and phrasing, resulting in minor adjustments. Table 1. Work values scores. Rank Work Values When Choosing an Employer, How Important Are the Following Aspects? (1—Not at All; 7—Absolutely Important) Mean SD 1 I “Working in an exciting environment” 5.96 0.856 2 S “Having a good relationship with my colleagues” 5.93 0.912 3 S “Feeling accepted and have a sense of belonging” 5.90 1.001 4 I “Working in an enterprising place, with innovative work practices and a vision for the future” 5.88 1.002 5 E “Having good promotion opportunities within the organisation” 5.82 0.944 6 S “Having a good relationship with my superiors” 5.75 0.971 7 S “Having supporting and encouraging colleagues” 5.67 1.078 8 E “Having job security and stability” 5.67 1.024 9 I “Working in an organisation that produces high-quality products and services” 5.57 1.047 10 I “Working in an organisation that values creativity and uses my creative abilities” 5.55 1.099 11 S “Having a fun work environment” 5.49 1.183 12 E “Having an attractive overall compensation package” 5.48 0.960 13 E “Having an above-average basic salary” 5.42 0.922 14 E “Hands-on interdepartmental experience” 15.34 1.020 15 I “Working in an organisation that produces innovative/attractive/exciting products and services” 5.29 1.066 1 (S) Social (5 items) 5.75 0.800 2 (I) Intrinsic (5 items) 5.65 0.777 3 (E) Extrinsic (4 items) 5.60 0.745 Notes: Mean and standard deviation values for CFA sample; N = 3049; 1 item 14 excluded in the exploratory factor analysis stage for cross-loadings. Soc. Sci. 2021,10, 297 6 of 16 In the final questionnaire, students were asked to respond to the question “When choosing an employer, how important are the following aspects?”, assessing the 15 items on a 1 (not at all) to 7 (extremely important) Likert scale. Students were also asked about their monthly salary expectations upon graduation (in euros). They preferred company size for first employer (micro, small, medium, and large, along with the specifications of the European Commission (2003)). Other classification questions included age, gender, area of study (business vs. technology), and expected final grade. Subsequent exploratory and confirmatory factor analysis (described ahead) led to the refinement of the scales. The final three-factor work values structure, therefore, comprises 14 items. 2.3. Data Analysis We used IBM ® SPSS ® 26.0 to conduct the descriptive analysis and to test the structure of the data with a subsample of 300 responses randomly extracted from the total sample (3349), following the recommendations regarding the stability and replicability of structural analyses (Gunuc and Kuzu 2015;Field 2018). The initial item analysis identified two potential problematic items from the extrinsic values factor (“Have an above-average basic salary”; “Having job security and stability.”) with corrected item-to-total correlations and an average item-to-item correlation below the recommended thresholds of 0.50 and 0.30, respectively (Hair et al. 2014). We decided to keep these items for further analysis as they presented acceptable correlations ( ≥ 0.310) with the group of items of the same expected common factor (extrinsic values). The pool of 15 items was subject to exploratory factor analysis (EFA) using principal component analysis (PCA) and varimax rotation. The initial rotated component matrix revealed the expected structure, with the variables presenting high loadings on a common factor. The exception was the item “Hands-on interdepartmental experience” (extrinsic value) that presented significant loadings on two factors, justifying its elimination. The final 14 items solution (Table 2) presented a significant Bartlett’s test of sphericity ( χ2 (91) = 1688,490, p ≤ 0.001), suggesting the appropriateness of using the factor analytic model on this set of data, and a high Kaiser-Meyer-Olkin value (KMO = 0.875) indicative of a strong relationship among variables, which made further analysis acceptable. The social, intrinsic, and extrinsic work values factors presented eigenvalues higher than 1 and together explained 60.62% of the variance, exceeding the suggested 60 percent threshold for a satisfactory solution (Hair et al. 2014). All variables loaded under the expected common factor, the factor loadings of the 14 items ranged from 0.555 to 0.848, and no significant cross-loadings were observed, indicating a well-defined structure (Hair et al. 2014). The internal consistency of the three-factor scale was assessed using Cronbach’s coefficient alpha ( α ). The Cronbach’s alpha for the entire scale ( α = 0.877) and for the social ( α = 0.849), intrinsic ( α = 0.807), and extrinsic ( α = 0.742) work values factors were higher than the 0.70 threshold (Hair et al. 2014), suggesting good internal consistency of the scale and of the three subscales. To confirm the goodness-of-fit of the factor structure to the collected data and the reliability and validity of the measures, we conducted a confirmatory factor analysis (CFA) with the remaining sample of 3049 respondents using IBM ® SPSS ® AMOS TM 26.0. This sample shows very similar characteristics to the EFA sample (Table 3). The maximum likelihood method was used to estimate the population parameters from sample data, as it maximises the probability that the model results from the observed data (Kline 2011). The assumption of normal distribution of the variables required by the maximum likelihood estimation was verified: the highest values of 0.846 for skewness and for 0.835 for kurtosis are far below the thresholds of 2 and 7, respectively (Finney and DiStefano 2006). The model comprising the three work values factors with 14 observable variables presented a good fit to the data ( χ2 (74) = 935.792, p ≤ 0.001; CFI = 0.954; RMSEA = 0.062, p ≤ 0.001, 90% CI (0.058, 0.065); SRMR = 0.047) according to recommended thresholds (Kline 2011; Hair et al. 2014). Soc. Sci. 2021,10, 297 7 of 16 Table 2. Work values factors (PCA with a subsample of 300). Dimensions Values/Items Social Intrinsic Extrinsic Eigenvalues 5.492 1.671 1.323 Variance explained (total: 60.62%) 39.23% 11.94% 9.45% Cronbach’s Alpha (total 14 items: 0.864) 0.849 0.807 0.742 Mean average (total: 5.685) 5.690 5.625 5.754 Variance (total: 0.046) 0.036 0.076 0.038 “Having a good relationship with my colleagues” 0.848 “Having supporting and encouraging colleagues” 0.775 “Feeling accepted and have a sense of belonging” 0.765 “Having a good relationship with my superiors” 0.709 “Having a fun work environment” 0.605 “Working in an enterprising place, with innovative work practices and a vision for the future” 0.789 “Working in an organisation that values creativity and uses my creative abilities” 0.308 0.739 “Working in an exciting environment” 0.691 “Working in an organisation that produces innovative/attractive/exciting products and services” 0.674 0.389 “An organisation that produces high-quality products and services” 0.623 0.305 “Having an above-average basic salary” 0.815 “Having an attractive overall compensation package” 0.779 “Having good promotion opportunities within the organisation” 0.703 “Having job security and stability” 0.555 Notes: N = 300; KMO = 0.875; Bartlett’s test of sphericity: χ2 (66) 1688,490; p ≤ 0.001; extraction method: principal component analysis; rotation method: varimax with Kaiser normalisation; loadings below 0.30 were suppressed. Table 3. Descriptive statistics of respondents and job preferences for EFA and CFA samples. Variable EFA CFA N % Mean SD N % Mean SD Gender 300 100.0 3049 100.0 Female 163 54.3 1658 54.4 Male 137 45.7 1391 45.6 Age 300 100.0 20.91 1.623 3049 100.0 20.89 1.560 Area of studies 300 100.0 3049 100.0 Business 162 54.0 1645 54.0 Technology 138 46.0 1404 46.0 Expected final grade (out of 20) 300 100.0 14.98 1.587 3049 100.0 14.80 1.421 Expected monthly salary (in Euros) 300 100.0 €1097.76 €670.06 3049 100.0 €1071.19 €516.71 Preference for employer size (# employees + sales volume) 300 100.0 3049 100.0 Large (>250 employees; >50 m €) 98 32.7 989 32.5 Medium (50–250 employees; 10–50 m € ) 130 43.3 1315 43.1 Small (10–49 employees; 2–10 m €) 59 19.7 638 20.9 Micro (<10 employees; ≤2 m €) 13 4.3 107 3.5 Notes: The CFA sample (N = 3049) is used as the reference to characterise our respondents; all subsequent analyses were conducted using this sample. The reliability and validity of the three-factor scale attested to the good quality of the measures. The internal consistency was confirmed by the high standardised factor loadings (from 0.604 to 0.831), the Cronbach’s alpha for the entire scale (0.861) and for each factor (from 0.776 to 0.832), and composite reliability (CR) (from 0.776 to 0.861) estimates above the recommended cutoff point of 0.70 (Hair et al. 2014). The factors also evidenced convergent Soc. Sci. 2021,10, 297 8 of 16 validity, as the average variance extracted (AVE) estimates, ranging from 0.467 to 0.555, were close or higher than the advocated threshold of 0.50 (Hair et al. 2014). Additionally, the factors exhibited discriminant validity, as the values of the square root of the AVE exceeded the correlations of each factor to all other factors (Fornell and Larcker 1981), suggesting the presence of distinct and uncorrelated factors (Table 4). Table 4. Work values scale reliability and validity assessment. Work Values Factor Loadings αCR AVE 1 2 3 1. Social 0.683–0.817 0.832 0.861 0.555 0.745 2. Intrinsic 0.640–0.831 0.822 0.836 0.508 0.481 *** 0.712 3. Extrinsic 0.604–0.768 0.776 0.776 0.467 0.419 *** 0.438 *** 0.683 Notes: N = 3049; SD—standard deviation; α —Cronbach’s alpha; CR—composite reliability; AVE -average variance extracted; *** p ≤ 0.001; square root of the AVE in bold on the diagonal compared with the inter-construct correlations to test discriminant validity. After assessing the quality of the scale, we proceeded with the data analysis using the sample of 3049 respondents to evaluate the influence of work values of young people about to enter the job market on work-related preferences. We employed multinomial logistic regression models (Hosmer et al. 2013) for the dependent categorical variable “employer size”, and multiple regression analysis (Hair et al. 2014) for the dependent continuous variable “expected salary”. “Gender” and “area of studies” were transformed into dummy variables and used together with “expected final grade” as control variables in all regression models. We used the nonparametric bootstrapping approach in the regression models to obtain the interval estimates, as it does not require distribution assumptions (Efron and Tibshirani 1993). The regressions models were estimated using bootstrapping with 5000 samples to estimate coverage probabilities for 99% confidence intervals, using the bias-corrected and accelerated method to correct for the bias and skewness of the bootstrap parameter estimates (Efron and Tibshirani 1993). 3. Results Our reference CFA sample (Table 3) comprises 54.4% female students, with ages ranging from 18 to 23 (M = 20.89, SD = 1.560), studying business (54.0%) and technology (46.0%). The expected final grade ranged from 10 to 20, averaging 14.80 out of 20 (SD = 1.421). Prospective monthly salary ranged from € 500 to € 9000, with a mean of € 1071.19 (SD = € 516.71), with female students expecting lower entry salaries (M = € 1007.87, SD = 474.594) than their male colleagues (M = € 1146.66, SD = 553.981). The difference was statistically significant (t(3,047) = − 7.450, p ≤ 0.001) with a medium effect size (d= − 0.271) (Cohen 1988). Most students prefer medium-sized (43.1%) and large companies (32.5%). A smaller group prefers small companies (20.9%) and a minority favours micro-firms (3.5%). The evaluation of work values (Table 1) evidence a bias towards the positive end of the scale, perhaps due to the desirable representation of the items. Table 1shows the mean of the items in each factor, revealing social work values to be the highest (M = 5.75, SD = 0.800), followed by intrinsic (M = 5.65, SD = 0.777) and extrinsic work values (M = 5.60, SD = 0.745). To explain the effect of work values on preferred employer size, we used a multinomial logistic regression main-effects model (Hosmer et al. 2013) with 5000 bootstrap samples. “Large firms” was defined as the baseline category. The model is statistically significant (χ2(18) = 142.139, p≤0.001) (Table 5). Soc. Sci. 2021,10, 297 9 of 16 Table 5. Multinomial logistic regression for preferred employer size. BCa 99% CI 4 Firm Size 1Variables B SE OR pB Micro Intercept 3.481 1.640 0.034 [−0.763, 7.435] Social 0.216 0.160 1.241 0.173 [−0.197, 0.662] Extrinsic −0.731 0.140 0.482 0.000 [−1.073, −0.391] Intrinsic 0.182 0.160 1.199 0.247 [−0.199, 0.611] Gender 20.091 0.219 1.095 0.675 [−0.509, 0.642] Area of study 30.005 0.213 1.005 0.986 [−0.516, 0.654] Expected grade −0.269 0.077 0.764 0.000 [−0.469, −0.078] Small Intercept 4.694 0.744 0.000 [2.781, 6.528] Social 0.158 0.074 1.171 0.038 [−0.035, 0.385] Extrinsic −0.429 0.077 0.651 0.000 [−0.638, −0.218] Intrinsic 0.116 0.077 1.123 0.133 [−0.082, 0.313] Gender 2−0.219 0.106 0.804 0.038 [−0.497, 0.045] Area of study 3−0.182 0.106 0.834 0.093 [−0.473, 0.106] Expected grade −0.277 0.037 0.758 0.000 [−0.374, −0.176] Medium Intercept 3.235 0.619 0.000 [1.641, 4.747] Social 0.135 0.061 1.145 0.030 [−0.033, 0.297] Extrinsic −0.213 0.065 0.808 0.003 [−0.383, −0.031] Intrinsic 0.011 0.063 1.011 0.858 [−0.144, 0.179] Gender 2−0.381 0.088 0.683 0.000 [−0.595, −0.151] Area of study 3−0.165 0.087 0.848 0.047 [−0.414, 0.044] Expected grade −0.156 0.030 0.856 0.000 [−0.233, −0.075] Notes: N = 3049; baseline categories— 1 “large firms”, 2 “female”, 3 “business”; 4 bootstrapping with 5000 samples, 99% confidence interval—bias-corrected and accelerated (BCa); χ2(18) = 143.162, p≤0.001; –2 log likelihood = 6860.992; Nagelkerke R2= 0.051. Holding all other independent variables constant, extrinsic work values, gender, and expected grade partially explain the preference for company size (Nagelkerke R 2 = 0.051). For those who favour extrinsic work values, the odds of preferring larger employers increase with the size of the firm (OR Micro = 0.482, p ≤ 0.001; OR Small = 0.651, p ≤ 0.001; OR Medium = 0.808, p= 0.001), meaning these students are, respectively, 51.8%, 34.9% and 19.2% less likely to prefer micro, small and medium when compared to large companies. Academic achievement in the form of expected final grades also explains students’ preference for larger employers. Students expecting higher grades are 23.6%, 24.2%, and 14.4% less likely to prefer micro (OR Micro = 0.764, p= 0.001), small (OR Small = 0.758, p ≤ 0.001) and medium (OR Medium = 0.856, p ≤ 0.001) rather than large firms. Gender is only relevant in the preference for medium-sized firms: compared to their female colleagues, male students are 31.7% less likely to choose medium-sized rather than large companies (ORMedium = 0.683, p≤0.001). To examine salary expectations, we used multiple linear regression with 5000 samples bootstrap and hierarchical block-wise entry method (Field 2018). The final parsimonious model is significant, explaining 5.2% of the variance (F(6,3041) = 27.831 p ≤ 0.001; R 2 = 0.052) (Table 6). As expected, extrinsic work values are the strongest predictor of anticipated entry salary ( β = 0.142; p ≤ 0.001). Intrinsic values are also positively related to salary expectations ( β = 0.061; p= 0.004). On the contrary, social values have a negative relationship with the expected entry salary, which means that, as students’ preferences for social values increase, their salary expectations decrease (β=−0.137; p≤0.001). Soc. Sci. 2021,10, 297 16 of 16 Schweitzer, Linda, Sean Lyons, Lisa K. J. Kuron, and Eddy S. W. Ng. 2014. The gender gap in pre-career salary expectations: A test of five explanations. Career Development International 19: 404–25. [CrossRef] Sørensen, J. B. 2007. Bureaucracy and entrepreneurship: Workplace effects on entrepreneurial entry. Administrative Science Quarterly 52: 387–412. [CrossRef] Suleman, Fátima, and Maria da Conceição Figueiredo. 2019. Entering the labour market in the context of higher education reform and economic recession: Young bachelor and master graduates in Portugal. Journal of Youth Studies, 1–22. [CrossRef] Terjesen, Siri, Susan Vinnicombe, and Cheryl Freeman. 2007. Attracting Generation Y graduates: Organisational attributes, likelihood to apply and sex differences. Career Development International 12: 504–22. [CrossRef] Tulgan, Bruce. 2013. Meet Generation Z: The Second Generation within the Giant “Millennial Cohort”. Available online: http: //www.rainmakerthinking.com/assets/uploads/2013/10/Gen-Z-Whitepaper.pdf (accessed on 21 July 2017). Turner, Anthony. 2015. Generation Z: Technology and Social Interest. The Journal of Individual Psychology 71: 103–13. [CrossRef] Twenge, Jean M. 2010. A Review of the Empirical Evidence on Generational Differences in Work Attitudes. Journal of Business and Psychology 25: 201–10. [CrossRef] Twenge, Jean M., and Kristin Donnelly. 2016. Generational differences in American students’ reasons for going to college, 1971–2014: The rise of extrinsic motives. The Journal of Social Psychology 156: 620–29. [CrossRef] [PubMed] 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: 1117–42. [CrossRef] Twenge, Jean M., Ryne A. Sherman, and Brooke E. Wells. 2017. Sexual Inactivity during Young Adulthood Is More Common among U.S. Millennials and iGen: Age, Period, and Cohort Effects on Having No Sexual Partners After Age 18. Archives of Sexual Behavior, 1–9. [CrossRef] Wan, Yim King Penny, IpKin Anthony Wong, and Weng Hang Kong. 2014. Student career prospect and industry commitment: The roles of industry attitude, perceived social status, and salary expectations. Tourism Management 40: 1–14. [CrossRef] Weiser, Eric B. 2015. #Me: Narcissism and its facets as predictors of selfie-posting frequency. Personality and Individual Differences 86: 477–81. [CrossRef] Williams, Alex. 2015. Move over, millennials, here comes Generation Z. New York Times. Available online: http://blogs.vsb.bc.ca/ sjames/files/2012/10/Move-Over-Millennials-Here-Comes-Generation-Z-The-New-York-Times.pdf (accessed on 20 July 2018). Williams, Kaylene C., and Robert A. Page. 2011. Marketing to the generations. Journal of Behavioral Studies in Business 3: 1–17.