Person-vocation and same organization bridge employment intention: Testing the mediating and moderating roles of person-organization fit

Abstract

BACKGROUND:

During the last decades, much research has been accomplished on the antecedents of bridge employment. Nevertheless, there is limited investigation into the influence of person-environment fit levels on retirement intentions.

OBJECTIVE:

This study’s purpose was to examine the interactive effect of person-environment fit levels on the intention to pursue bridge employment. More specifically, the authors investigated the mediator and moderator roles of person-organization fit on the relationship between person-vocation fit and the intention to pursue bridge employment within the same organization.

METHODS:

Data from 246 older executives working in the French banking sector were used to test the hypotheses.

RESULTS:

The results indicated the existence of a positive and significant relationship between person-vocation fit and the intention to pursue bridge employment within the same organization. Further, the findings confirmed the mediating and moderating roles of person-organization fit on this relationship.

CONCLUSIONS:

The results of this study have theoretically and empirically extended person-environment fit theory, continuity, and role theories to shed light on the underlying mechanisms through which person-vocation fit impact bridge employment intention. The findings of this study can be used by career counsellors and managers to design more effective actions to retain older workers.

Keywords

Person-vocation fit person-organization fit bridge employment retirement machine learning

Khaled Lahlouh is an Assistant Professor in Human Resources Management and Organizational Behavior at Rabat Business School, International University of Rabat, Morocco. His research interests include career management, organizational psychology, and retirement behaviors.

Aïcha Oumessaoud is a PhD in Human Resources Management and Organizational Behavior at Aix-Marseille university, France. Her research focuses on the Family-Work interface, organizational stress and social roles.

Raja Sreedharan is an Assistant Professor in Human Resources Management and Organizational Behavior at Rabat Business School, International University of Rabat, Morocco. His research focuses on the smart supply chain, Industry 4.0 and Lean Six Sigma.

1 Introduction

During the last decades, with the increasing number of retirees reintegrating into the workforce, retirement is conceptualized more and more as a late-career stage in which many professional activities can be held [1, 2]. Known as bridge employment, these professional activities ease the transition from career jobs to full retirement [3, 4]. Bridge employment can be pursued in a different field than the career domain [5] or in the same occupation and within the same organization with reduced work hours [6].

A better understanding of bridge employment is essential since it may benefit both individuals and organizations. Indeed, it may help organizations face labor shortages due to the retirement of the baby boom generation [7]. It also helps individuals maintain their social contacts [8] and promotes their health and psychological well-being [9]. This study focuses on bridge employment intentions within the same organization for two main reasons. First, this kind of bridge employment is common among older workers as many of them work once they retire in the same pre-retirement organization [10]. Second, another consideration is that doing so can help managers and career counsellors implement managerial actions designed toward older workers’ maintenance in employment [2, 11].

While previous studies have shown considerable interest in the influence of both individuals (e.g., subjective norms and perceived control: [12]; Health: [13]) and work-related factors (e.g., job satisfaction: [14]; organizational commitment: [3], workplace incivility: [11]) on bridge employment intention, there is surprisingly a little research on the influence of individual-work environment interaction on retirement intentions [15, 16]. Moreover, the review of the few works exploring the influence of Person-Environment (hereafter P-E fit) fit levels on retirement intentions indicates many shortcomings that deserve to be addressed. First, the only study that has considered the multiple levels of P-E fit as predictors of retirement intentions employed an additive approach that considers fit levels as independent antecedents of work outcomes and behaviors [17]. While this approach may be relevant when comparing the influence of the different fit levels, it fails to capture the combined effect of these same dimensions [18]. Second, to our best knowledge, the question of whether P-E. fit levels influence the same organization bridge employment remains unanswered since previous research has explored the relationship of P-E. fit with career bridge employment [15].

Hence, to address the above limitations, this study explores the relationship between Person-Vocation (P-V) fit and the intention to pursue bridge employment within the same organization. More precisely, by exploring the mediator and moderator roles of Person-Organization (hereafter P-O fit) fit, this study explains how and when P-V fit influences the intention to pursue bridge employment within the same organization. Hence, it contributes to the literature in many ways. First, by investigating the direct relationship between P-V. fit and the intention to pursue bridge employment within the same organization, this study aims to extend the findings of previous works [15] that have explored the direct relationship between P-V. fit and the other retirement intentions. Second, responding to the call made by other scholars [15], the second expected contribution of this study is to explain the underlying mechanisms of the P-V fit and retirement intentions relationship. To that end, similarly to previous works in the literature (e.g. [17 , 20]), this study adopts the interactive approach of P-E. fit levels by assuming that different manifestations of fit can interact to predict work outcomes [21]. Third, by exploring for the first time the moderating role that P-O fit may play in the relationship of P-V fit and work outcomes, this study extends the results of previous works that have focused only on the mediating role of P-O fit [17, 19]. Finally, this research aims to help managers consider maintaining older workers as a managerial responsibility that goes far beyond retirement’s legal framework.

The following sections start with a brief review of the theories mobilized in the current study. Then, the effects of P-V fit on the same organization bridge employment intention and P-O fit’s role in this relationship are established. After testing the conceptual model, the results are discussed, considering the theoretical framework and previous works. Furthermore, the study uses Machine learning-based classification to predict the theoretical model and validate it using K-cross validation. Finally, contributions, limitations, and future voices of research are discussed.

2 Theoretical background and hypotheses development

2.1 Person-environment fit, continuity and role theories

Continuity theory [22] highlights the different continuity strategies older workers use when facing adaptive situations. More specifically, it suggests that older workers adopt strategies according to their internal and external structures to avoid the experience of disruptive events while aging. According to Lieberman & Tobin [23], continuity strategies can be achieved at two levels. Internally, when the aging person adopts a pattern of behavior allowing him/her to preserve the stability of the inner structure (e.g., values, emotions). Externally, when the aging person successfully preserves his/her usual physical and social environment. Concerning retirement, it has been assumed in previous works (e.g. [24, 25]) that a good work environment can encourage older workers to hold bridge employment positions to act in continuity with their past experiences.

For its part, role theory [26] emphasizes the importance of role identity during role transition. According to role theory, role transitions are affected by the involvement in a specific role (e.g., organizational role). The more the person is involved, the more difficult it is for him/her to give up the given role. In other words, the centrality of work-related identities (e.g., vocational, organizational) may encourage older workers to pursue professional activities once retired (i.e., bridge employment) to avoid the stress caused by the role-identity loss [27]. Evidence from retirement literature confirms the relevance of role theory as a theoretical framework to study retirement intention and behaviors. Indeed, by mobilizing role theory, it has been shown that older employees who have a strong commitment toward their career or organization [3] or perceive a high level of fit with their work environment would be more inclined to pursue bridge employment.

Additionally, P-E fit refers to the existing compatibility between the individual and his/her work environment [28]. P-E fit can be categorized in two ways. First, the supplementary kind of P-E fit reflects the similarity between the person and the work environment (e.g., values congruence at the vocational level). Second, the complementary P-E fit is related to situations where individual attributes are complemented by the work environment (e.g., individual needs are satisfied by the organizational resources) and vice versa.

In the organizational/vocational psychology literature, it has been shown that the perception of fit with the work environment is positively related to many attitudes and behaviors at the workplace (e.g., job satisfaction: [29]; organizational citizenship behavior: [30]; organizational commitment: [31]). Moreover, some studies have explored the direct effect of P-E fit levels on different retirement intentions [15, 16]. Nevertheless, understanding the impact of the P-E fit levels on retirement intentions is still at its beginning, and many questions regarding the mediating and moderating variables of the relationship are still unanswered [15]. Therefore, based on the limits of the previous studies, and by mobilizing a theoretical framework (i.e., continuity and role theories) that has been widely used in the retirement literature [15, 32], we formulate a set of hypotheses that aim to explain how P-E fit levels interact to explain the intention to pursue bridge employment within the same organization.

2.2 Person-vocation fit and the intention to pursue bridge employment within the same organization

Person-Vocation fit (hereafter P-V fit) is conceptualized as a value congruence, which occurs when the employee perceives a match between his/her values and the vocational ones [33]. At this level, it seems essential to distinguish between the P-V fit and the somewhat related Person-Job fit (hereafter P-J fit). Indeed, contrary to P-J fit, which is associated with the hierarchical position, P-V fit occurs when the professional trajectory fits with the personal interests, aspirations, and values. Here, we suggest that P-V fit is positively related to the intention to pursue bridge employment within the same organization.

Indeed, Feldman [34] suggested that employees who identify themselves with a given profession are more likely to hold bridge employment than to retire fully. Previous works showed that older workers who perceive work as meaningful prefer to postpone retirement [35] and that P-V fit is positively related to career bridge employment [15].

According to both role theory [26] and continuity theory [22], we consider that when employees perceive a high level of fit with their vocation, the corresponding role identity becomes central in the older worker’s global identity. Hence, to ensure the continuity of their internal structure (i.e., valued vocational role), the older workers would be more inclined to adopt roles that allow them to ensure internal continuity. Based on the arguments discussed above, we formulate the following hypothesis:

H1. P-V fit is positively related to the intention to pursue bridge employment within the same organization.

2.3 Person-vocation fit and the intention to pursue bridge employment within the same organization: The mediating and moderator role of person-organization fit

P-O fit occurs when a similarity or a complementary between the individual and the organizational characteristics exists [33]. Indeed, the P-O fit could be complementary or supplementary [28]. Complementary fit occurs when the characteristics of the person and those of the organization complement each other (e.g., Individual needs, Organization supplies) [28]. Supplementary fit reflects the degree of similarity between the person and the environment characteristics (e.g., values, personalities, goals) [19]. Further, several studies have shown that the P-O fit perception reinforces the positive attitudes toward the employing organization (e.g., organizational commitment: [19, 36]; intention to leave: [37])

Evidence from the P-E fit literature shows that the different levels are conceptually distinct and related to the workplace attitudes and behaviors [28, 30]. Nonetheless, the different fit levels may relate to each other to predict the employees’ intention to hold bridge employment once retired. Indeed, according to Feldman & Ng [38], P-V fit is an important antecedent of P-O fit. The rationale is that employees who do not perceive a congruence between their values and the vocational ones may not perceive a high level of fit at the organizational level [19, 39]. For example, a faculty member who does not perceive a value congruence with his/her vocation may struggle to achieve a high level of fit at the organizational levels no matter where he or she works.

Further, P-V fit may affect how people perceive P-O fit [19]. For instance, even if the different needs are satisfied by the organization (i.e., P-O fit Needs-Supplies), it would be hard to imagine the individual persisting in a career path if he or she perceives that he does not fit anymore with its values. Only one study has explored P-V fit as a predictor of P-O fit to our best knowledge [19].

In parallel, in this study, we argue that P-O fit is positively related to the intention to pursue bridge employment in the same organization. Indeed, empirical evidence shows that when employees perceive that their employing organizations satisfy their different needs (e.g., developmental opportunities: [40, 41]), they are more likely to keep working during retirement. In addition to the satisfaction of the different needs, values congruence between the older worker and the organization is taken into account while considering retirement decisions. For example, Boumans, De Jong, & Vanderlinden [42] demonstrate that older workers evolving in organizational contexts marked by stereotyped visions (e.g., lack of performance and flexibility, reluctant to change) are more likely to retire.

From a theoretical point of view, both continuity [22] and role [26] theories support the hypothesized relationships between P-O fit and the intention to pursue bridge employment in the same organization. According to continuity theory [22], bridge employment jobs offer the best continuity strategy to preserve satisfying environments (i.e., Needs-Supplies fit). Furthermore, in line with role theory [26], the match between an individual and a specific aspect of the environment makes the corresponding role identity central in the global identity of older workers; bridge employment hence offers a way to preserve a central and valued role identity [43]. On those premises, we formulate the following hypothesis:

H2. P-O fit mediates the positive relationship between P-V fit and the intention to pursue bridge employment within the same organization.

In the current study, it is assumed that the relationship between P-V fit and the intention to pursue bridge employment in the same organization would be strengthened by the perception of P-O fit. Indeed, previous studies have demonstrated that the perceived fit with the organization may reinforce the employee’s relationship with his/her organization. For instance, it has been demonstrated that P-O fit is positively related to positive organizational attitudes (e.g., organizational commitment: [28, 44]; organizational identification: [45]) and negatively to turnover intention [21].

According to role theory [26], an older worker who already perceives a high level of P-V fit may be more inclined to pursue bridge employment within the same organization when he/she perceives a high level of P-O fit. In fact, in this case, in addition to the vocational identity, the older worker may also value the organizational one and would adopt strategies that allow the maintenance of both identities. Thus, the intention to pursue bridge employment within the same organization would be reinforced as it enables preserving two-valued identities, namely the vocational and the organizational ones.

From a continuity perspective [22], an older worker who perceives only a high level of P-V fit may ensure continuity by pursuing bridge employment within or outside the organization (i.e., career bridge employment). Nevertheless, when the perception of P-V fit is combined with the perception of P-O fit, the intention to pursue bridge employment within the same organization would become a better strategy as it offers the most suitable option to ensure continuity. Based on this discussion, the current study argues that P-O fit plays a moderation role in the relationship between P-V fit and the intention to pursue bridge employment within the same organization. As such, the following hypothesis will be tested:

H3. The relationship between P-V fit and the intention to pursue bridge employment is moderated by P-O fit, such that the positive relationship between them will be stronger for older workers who are high in P-O fit.

The conceptual model is presented in Fig. 1.

Fig. 1

Conceptual model.

3 Research method

This study has investigated the direct and indirect effects of P-V fit on the intention to pursue bridge employment among executives of the French banking sector for many reasons. First, the French economy relies principally on the tertiary sector, which requires a high supervision rate. Indeed, the recent forecast of the principal confederation of executives in France (APEC) indicates an increasing need for executives, passing from 281.300 in 2019 to 302.100 by 2022 1 . Second, the latest figures provided by the French Association of Banks (2020) [46] indicate that 65% of the entire staff is composed of executives 2 and that employees aged 50 and above represent 30% of the personnel in the French bank sector. Therefore, exploring the antecedents of bridge employment in this setting can be of particular interest for managers looking toward maintaining their older workers.

3.1 Sample and procedure

As the retirement anticipation phase starts by 50 [47], data were collected exclusively from executive employees aged 50 and above. This choice is in line with previous research in the retirement literature (e.g. [13 , 48]).

Concerning the sampling strategy, the convenience sample technique has been used. The reason behind this choice is the difficulty to choose randomly from the entire population (i.e., executives aged 50 and above) and the “convenient” character (e.g., easy accessibility, the willingness of the two companies to participate in the study) for the research team to collect data from the banking sector. This technique is widely used in social sciences and business research [49].

For the current research, data was collected from two different banks in France. Before starting data collection, we first introduced the main objectives to the principal HR manager, who subsequently briefed all the potential participants. Further, the data collection was completed in two different waves. At T1, employees were asked about their perception of P.E fit levels and demographic information. Two weeks later (second wave) T2, the employees who filled the survey at T1 were asked about their intention to pursue bridge employment in the same organization. We assigned codes to the participants in the different waves of data collection to match responses over time (T1 and T2). Specifically, we distributed 370 paper-pencil questionnaires to the respondents at T1, and 296 participants responded (the response rate = 80%).

Similarly, after two weeks in T2, we distributed 296 questionnaires, and a total of 258 employees responded (70% of the retention rate from the initial sample). However, 12 responses were deleted due to the missing data and the presence of multivariate outliers. Data was gathered with the permission of top management. However, we have assured anonymity to the participants to obtain their confidence. In the final sample, 73.5% of the participants were aged between 50 and 60 by the time they participated in the study, over 77% were men, and 88% had at least a bachelor’s degree. Further, 88% were married, and 51% declared having more than ten years of tenure within the same company.

3.2 Measures

In this study, existing multi-items adapted from previous research were used. When needed, translation to French has been accomplished according to back-translation technique recommendations [50]. The respondents were asked to rate their agreement level using a Likert scale, with responses ranging from 1 = strongly disagree to 7 = strongly agree. P-V fit was measured with the scales developed by Kennedy [51].

Regarding the P-O fit, we used respectively the measurement scales developed by Kennedy [51] and Cable & DeRue [30] to measure Values congruences’ and Need-Supplies’ dimensions. Finally, the intention to pursue bridge employment within the same organization was measured using Jones & McIntosh’s [3] measurement scales. The internal consistency was estimated using Cronbach’s alpha and AVE (Table 1).

Table 1
Psychometric properties of measurement models

Measurement models and items Mean (S.D.) Stand. Loading AVE C.R α

>0,70 >0,50 >0,60 >0,60

P-V. fit 0,709 0,879 0.804

PVFIT1 5,40 (1,16) 0.896

PVFIT2 5,71 (1,29) 0.815

PVFIT3 4,66 (1,38) 0.812

P-O fit (Needs-Supplies) 0,860 0,949 0,918

PONS1 4,98 (1,58) 0,960

PONS2 4,89 (1,62) 0,963

PONS3 4,93 (1,76) 0,856

P-O. fit (values congruence) 0,945 0,981 0,971

POVC1 4,82 (1,53) 0,960

POVC2 4,94 (1,50) 0,963

POVC3 4,85 (1,52) 0,856

P-O. fit (second order factor) 0.735 0.943 0.927

P.O. fit (Values congruence) – 0.921

PO fit (Needs-Supplies) – 0.882

Same Organization Bridge Employment Intention (SOBEI) 0,946 0,972 0.943

SOBEI1 3,52 (1,96) 0,973

SOBEI2 3,56 (1,96) 0,972

Health Satisfaction 0.721 0.832 0.837

HSA2 4,73 (1,27) 0.997

HSA4 5,43 (1,30) 0.668

Expected adjustment to retirement 0.692 0.901 0.851

RETADJ1 5,69 (1,23) 0.870

RETADJ2 5,39 (1,56) 0.787

RETADJ3 5,74 (1,09) 0.832

RETADJ4 5,94 (1,22) 0.838

Measurement models and items	Mean (S.D.)	Stand. Loading	AVE	C.R	α
		>0,70	>0,50	>0,60	>0,60
P-V. fit			0,709	0,879	0.804
PVFIT1	5,40 (1,16)	0.896
PVFIT2	5,71 (1,29)	0.815
PVFIT3	4,66 (1,38)	0.812
P-O fit (Needs-Supplies)			0,860	0,949	0,918
PONS1	4,98 (1,58)	0,960
PONS2	4,89 (1,62)	0,963
PONS3	4,93 (1,76)	0,856
P-O. fit (values congruence)			0,945	0,981	0,971
POVC1	4,82 (1,53)	0,960
POVC2	4,94 (1,50)	0,963
POVC3	4,85 (1,52)	0,856
P-O. fit (second order factor)			0.735	0.943	0.927
P.O. fit (Values congruence)	–	0.921
PO fit (Needs-Supplies)	–	0.882
Same Organization Bridge Employment Intention (SOBEI)			0,946	0,972	0.943
SOBEI1	3,52 (1,96)	0,973
SOBEI2	3,56 (1,96)	0,972
Health Satisfaction			0.721	0.832	0.837
HSA2	4,73 (1,27)	0.997
HSA4	5,43 (1,30)	0.668
Expected adjustment to retirement			0.692	0.901	0.851
RETADJ1	5,69 (1,23)	0.870
RETADJ2	5,39 (1,56)	0.787
RETADJ3	5,74 (1,09)	0.832
RETADJ4	5,94 (1,22)	0.838

Note: Reliability and validity tests by using SmartPLS^®; P-V fit = Person-Vocation fit; P-O fit = Person-Organization fit; SOBEI = Same Organization Bridge Employment Intention; HSAT = Health Satisfaction: RETADJ = Adjustment to retirement; HSA1 and HSA3 were deleted due to low factor loadings.

Based on previous research, the analyses of this study included many control variables. Adjustment to retirement was measured with the four-item scale developed by Taylor & Shore [52]. Further, health satisfaction was measured with Adams & Beehr’s measurement scale [53]. Lastly, age and salary were also controlled because previous research [54] have shown that such variables are related to the different retirement intentions.

4 Results

4.1 Psychometric properties of measurement scales

Internal consistency reliability, convergent validity, and discriminant validity were assessed to check if the different reliability and validity requirements were met. As shown in Table 1, first, internal consistency of the different measurement models was established with composite reliability (C.R) ranging from 0,832 to 0,981 (equal or above 0,7: [55]) and Cronbach’s alpha ranging from 0,804 to 0,971. Standardized loadings (lambda) were used to assess indicator reliability, and they were higher than the threshold of 0,70 [55]. Moreover, the convergent validity criterion was met since the Average Variance Extracted (AVE) of all the constructs exceeded the threshold of 0,50 [56].

Finally, many criteria have been used to establish discriminant validity. First, following Fornell & Larcker [56] criteria, the results (Table 2) showed that the square root of AVE for each construct is greater than the variance shared with all the other constructs. Second, Table 3 shows the existence of the discriminant validity as loadings of each construct (in bold and italic) are higher than cross-loadings. Third, to confirm our results, we used HTMT criteria by using the most conservative threshold level of 0,85 [57, 58]. As shown in Table 4, discriminant validity criterion is met since none of the values exceeds 0.85. Consequently, the results obtained indicate that the requirements of validity and reliability have been met.

Table 2
Discriminant validity (latent variable correlation and square root of AVE)

Measurement models 1 2 3 4 5

1. P-V. fit 0.842

2. P-O fit 0.525 0.858

3. SOBEI 0.478 0.466 0.973 .

4. HSAT 0.175 0.142 0.199 0.849

5. RETADJ 0.112 –0.010 –0.162 0.078 0,832

Measurement models	1	2	3	4	5
1. P-V. fit	0.842
2. P-O fit	0.525	0.858
3. SOBEI	0.478	0.466	0.973	.
4. HSAT	0.175	0.142	0.199	0.849
5. RETADJ	0.112	–0.010	–0.162	0.078	0,832

Note: AVE, average variance extracted. The values in italics show the square root of AVE, which is greater than the correlation in the latent variable.

Table 3

Loadings and Cross-Loadings

Constructs	Items	PV fit	PONS	POVC	SOBEI	HSAT	RETADJ	INCSAT
P.V. fit	PVCC1	0,896	0,261	0,533	0,327	0,153	0,156	0,099
	PVVC2	0,815	0,337	0,495	0,344	0,131	0,045	0,062
	PVVC3	0,812	0,248	0,454	0,383	0,159	0,083	0,066
PONS	PONS1	0,355	0,960	0,643	0,358	0,085	–0,004	0,184
	PONS2	0,328	0,963	0,625	0,378	0,089	0,001	0,174
	PONS3	0,243	0,856	0,469	0,309	–0,063	–0,030	0,168
POVC	POVC1	0,584	0,609	0,962	0,453	0,177	–0,032	0,070
	POVC2	0,562	0,612	0,974	0,428	0,133	–0,007	0,099
	POVC3	0,568	0,615	0,980	0,451	0,136	0,015	0,070
SOBEI	SOBE1	0,415	0,371	0,454	0,973	0,219	–0,144	–0,102
	SOBE2	0,398	0,362	0,434	0,972	0,167	–0,171	–0,127
HSAT	HSAT2	0,180	0,048	0,156	0,184	0,997	0,089	0,026
	HSAT4	0,170	0,050	0,135	–0,018	0,668	0,176	0,146
RETADJ	RETADJ1	0,139	–0,027	–0,010	–0,099	0,103	0,870	0,200
	RETADJ2	0,075	–0,077	–0,062	–0,133	0,039	0,787	0,186
	RETADJ3	0,081	0,049	0,059	–0,208	0,037	0,832	0,248
	RETADJ4	0,078	0,021	–0,013	–0,098	0,080	0,838	0,205

Table 4

Discriminant Validity (HTMT.85 criterion)

	PV fit	PO fit	SOBEI	HSA	RETADJ
P.V. fit
P.O. fit	0.599
SOBEI	0.483	0,495
HSA	0.231	0,149	0.129
RETADJ	0.142	0.064	0.181	0.169

4.2 Assessment of structural model

As demonstrated in Table 5, the model fit was ensured by examining many criteria. First, in terms of predictive power, the portion of the variance explained in the two endogenous variables: P-O fit and the intention to pursue bridge employment in the same organization was 27,5 and 36,7, respectively. According to many scholars, those values are satisfactory [55, 59]. For example, Cohen [59] noted that the values of 0,02, 0,15, and 0,35 reflect small, medium, and significant effects, respectively. Second, the results showed that the model has good predictive relevance since Stone & Geisser’s Q² values were larger than zero for each reflective endogenous latent variable. More specifically, the results on Table 5 show medium predictive relevance (0,02 = small, 0,15 = medium and 0,35 = immense, [53]). Third, the presence of collinearity issues was assessed through tests for Variance Inflation Factor (VIF). Results indicated that the structural model do not present any multicollinearity since all VIF values range between 1,050 to 1,513, which falls below the recommended threshold of 5 [60]. Finally, we used the Standard Root Mean Square Residual (SRMR) indicator to ensure that the conceptual model fits the data. The results show a significant value of 0,078 (Table 5), which is satisfactory since it is less than 0,08 [61].

Table 5
Structured model results

Construct R² Adj. R² Q² VIF SRMR

SOBEI 0,367 0,348 0,324 0,078

PO fit 0,275 0,272 0,198 1,473

P.V. fit 1,513

PV fit* P.O. fit 1,074

HSAT 1,057

RETADJ 1,050

SALARY 1,056

Age 1,070

Construct	R²	Adj. R²	Q²	VIF	SRMR
SOBEI	0,367	0,348	0,324		0,078
PO fit	0,275	0,272	0,198	1,473
P.V. fit				1,513
PV fit* P.O. fit				1,074
HSAT				1,057
RETADJ				1,050
SALARY				1,056
Age				1,070

Notes: VIF, variance inflation factor; SRMR, standardized root mean square residual.

The hypotheses were tested using 5000 bootstraps. As shown in Table 6, direct and indirect effects estimates, and 95% bias-corrected bootstrapped confidence intervals are provided for path estimates. Hypothesis 1 tested the direct effect of P-V fit on the intention to pursue bridge employment within the same organization. The result of the regression analysis validated the hypothesis (β= 0,253, t = 3,81, [0,127; 0,383], p < 0,001). Concerning the mediation hypothesis, Hypothesis 2 states that P-O fit mediates the relationship between P-V fit and the intention to pursue bridge employment within the same organization. According to the results, the hypothesized relationship was supported (β= 0.176, t = 4,64, 95% CI [0,105, 0,243]). Finally, the impact of P-O fit as moderator on the direct effect of P-V fit on the intention to pursue bridge employment within the same organization was found to be positive and significant (β= 0,110, t = 2,31, 95% CI [0,013, 0,201]) (Hypothesis 3). Figure 2 shows the interaction effect of P-O fit and P-V fit on the intention to pursue bridge employment within the same organization. The graph shows that when P-O fit is high (one standard deviation above the mean), the perception of PV fit has a higher positive effect on the intention to pursue bridge employment within the same organization than when P-O fit is low (one standard deviation below the mean).

Table 6

Direct, indirect, and total effect

	β	Standard error	t-Values	p-Values	95% CI	Decision
Direct effect
PV fit ⟶ SOBE	0,253	0,066	3,81	0.000	[0,127; 0,383]	Supported
Indirect effect
PV fit ⟶ PO fit ⟶ SOBE	0,176	0,049	4,64	0.000	[0,105; 0,253]	Supported
Moderating effect
PO fit*PV fit ⟶ SOBE	0,110	0,048	2,31	0,021	[0,013; 0,201]	Supported

Notes: CI = confidence interval. Bootstrap samples for the bias-corrected interval is 5.000.

Fig. 2

Interaction diagram P-O fit between P-V fit and the intention to pursue bridge employment within the same organization.

4.3 Testing theoretical model using the machine learning algorithms

The theoretical model is tested using the machine learning classifiers to see the robustness of the proposed theoretical model [62]. Using supervised learning methods, we can establish a classifier algorithm that establishes predictions with suitable accuracy. Individual Likert scale responses are combined depending on the model representation by reducing dimensions [63]. A dimensionality reduction helps in combining dimensions without incurring significant information loss. Owing to the ordinal nature of the data (7 points Likert Scale), we employ a correlation analysis based on the maximum likelihood estimate. A polychoric correlation matrix is layered on, and dimensions are further combined by a Principal component Analysis approach.

Further, the study arrived at a point with exactly four predictors. With the target variable categorized into a binary outcome, the study has set for a classification setup based on a threshold. Four unique classifiers were deployed, in the form of a Support Vector Machine (SVM), Gradient Boosting, Random Forest and Naive Bayes are used. A 90:10 test train split is used, and a 10-fold cross-validation is done to select the best fit and quantify the model in Table 7. Based on the different classifiers, SVM exhibited the best accuracy with minimum error values, as shown in Table 7.

Table 7
Precision; recall; F1 score for the SVM classifier

Hypotheses Class Precision Recall F1 score

P-V Fit based SOBE 1 0.923077 1 0.96

2 0.5 0.5 0.5

3 1 0.2 0.333333

4 0.375 0.75 0.5

5 0.769231 1 0.869565

6 0.9 0.692308 0.782609

7 1 0.5 0.666667

Hypotheses	Class	Precision	Recall	F1 score
P-V Fit based SOBE	1	0.923077	1	0.96
	2	0.5	0.5	0.5
	3	1	0.2	0.333333
	4	0.375	0.75	0.5
	5	0.769231	1	0.869565
	6	0.9	0.692308	0.782609
	7	1	0.5	0.666667

For the proposed model, the prediction of hypotheses was made through training and test of the dataset. Further, to test the model’s ability using K-cross Validation [64]. This test is done to generalize that the model is free from overfitting and selection bias. Table 8 shows that the models have an accuracy of 75%, proving that the model has no underfitting and overfitting [64] as is in the prediction.

Table 8

Cross Validation Accuracy

Model	Cross Validation Accuracy k = 10,90:10	Cross Validation Accuracy k = 5,80:20	Model Accuracy
Random Forest	0.7559	0.7336	0.788
SVM	0.588	0.5641	0.75
GBM	0.7687	0.7482	0.801
NB	0.6422	0.6387	0.75

5 Discussion

This study is a pioneering exploration of the effect of P-V fit on the intention to pursue bridge employment within the same organization, with a particular interest in the mediating and moderating role of P-O fit. The findings suggest that when employees perceive a high level of P-V fit, they are more inclined to pursue bridge employment within the same organization (H1). This result extends the findings of previous studies [15] as it establishes, for the first time, the positive effect of P-V fit on the intention to pursue bridge employment within the same organization. In line with our theoretical assumptions, it confirms that the values congruence at the vocational level makes the related identity central and valuable for older workers. Consequently, older workers may adopt different strategies, including holding a bridge employment position, to avoid the stress that may result from role loss [26] and act in continuity with their internal structure [22].

Furthermore, by showing that P-O fit mediates and moderates simultaneously the direct effect of P-V fit on the intention to pursue bridge employment within the same organization (H2 and H3), the results extend the previous findings in the P-E fit literature [17 , 20]. Those results are consistent with Ng and Feldman’s proposition that P-V fit is an important antecedent of P-O fit. They are also consistent with both role [26]and continuity [22] theories. Indeed, when the employee perceives a high P-O fit, bridge employment within the same organization represents an adapted action that allows role maintenance and continuity.

6 Theoretical and methodological implications

This study differs from past research by exploring the interactive effect of P-V fit and P-O fit on the intention to pursue bridge employment within the same organization. First, investigating the interactive effect of the P-E fit levels brings important knowledge on retirement antecedents. Indeed, this study differs from previous works that examined the impact of P-E fit levels separately [15, 16]. By doing so, it is, to our best knowledge, the first study to shed light on the underlying mechanisms of the P-E fit and retirement intentions relationship.

Second, by considering the impact of PV fit and P-O fit on the intention to pursue bridge employment within the same organization, the present study confirms the explanatory power of the P-E fit levels in predicting the different retirement intentions [25]. Moreover, it expands the results of the few studies which have already established the impact of P-E fit levels on different retirement intentions (i.e., career bridge employment intention: [15]; the anticipated timing of retirement: [2, 16]).

Finally, this study is one of the few studies in the organizational behavior field in which machine learning techniques have been used. The theoretical model was tested using machine learning classifiers and confirmed the model fit. The prediction done by the SVM is validated using the K-Cross validation and found that the P-E fit levels have a positive relationship with the intention to pursue bridge employment within the same organization. Also, Table 8 showed that P-O fit was positively related to both P-V fit and intention to pursue bridge employment within the same organization. Using the K-cross validation enhanced the analysis’s robustness and made it free from over-fitting and selection bias.

Beyond the theoretical implications, the findings of this study may be helpful for career counselors and managers to design more effective actions to retain older workers.

7 Practical implications

The present study has practical implications for career counselors and managers. Consistent with previous research [15], the present findings suggest considering P-V fit as an important antecedent of older workers’ retention. For instance, before designing interventions aiming to enhance P-O fit, managers may first identify older workers who still fit with their vocation. Without this, counselors and managers may expect more costs and assistance to be invested in retaining their older workers. Moreover, organizations may be more proactive by arranging vocational reorientation programs to advise and accompany their workers in their professional progress. Such practices would allow employees to reorient their vocational choices whenever they perceive a low level of P-V fit.

Furthermore, this study demonstrates that P-O fit moderates the direct effect of P-V fit on the intention to pursue bridge employment within the same organization. The result implies that managers may retain older workers who still perceive a value congruence with their vocation by enhancing their perception of P-O fit. Their retention can be done by satisfying the different needs that older workers can express at the workplace (e.g., supportive climate: [40], Training opportunities: [42]). Besides, organizations may develop and promote organizational values (e.g., self-transcendence: [2]) that foster bridge employment intention. For example, as noted by Whörmann et al. [2], organizations interested in retaining their older workers may promote some characteristics of their organizational culture that reflect those values (e.g., altruism, relationships).

8 Limitations and future opportunities

Despite its numerous contributions, the present study presents many limitations that need to be addressed. First, due to the convenience sampling used in this study and the relatively small sample size, the generalizability of the findings may be limited [65]. Indeed, many discrepancies exist between the sample and the target population. For example, women represent 56,3% of the total personnel in the French banking sector; in contrast, they only represent 33% in our study sample [46]. Even though such differences between the study sample and the population may be explained by the difficulties women encounter to reach managerial positions, they may also reflect a lack of sample representativeness. To remedy this, future research is encouraged to adopt more elaborated sampling strategies (i.e., probability sampling methods) and replicate the findings in other sectors or cultures.

Second, even though we used a time lag approach while collecting the current study data, the results did not allow us to make strong inferences about the causal relations among the study variables. Future studies are encouraged to use more sophisticated methodological designs (e.g., experimental design in which perception of P-V fit and P-O fit are manipulated).

Third, by measuring different levels of P-E fit at one point in time, we could not provide a dynamic view of fit [19, 66]. Thus, future studies should measure the P-E fit levels at different moments and explore how the change observed over time can be related to retirement intentions because of the changing nature of both environments and persons.

Finally, although we explored the double role of P.O. fit (i.e., moderator and mediator), many other variables can be explored. Based on previous works [19, 20], a possible mediator would be the P-J fit. We recommend that future studies extend our conceptual model by testing multiple mediations or serial mediation models, including all three levels of P-E.

9 Conclusion

The main purpose of this research was to investigate the interactive effect of P-V and P-O fit on the intention to pursue bridge employment within the same organization in the French banking sector. Structural equation modeling was used to test the proposed conceptual model. The study results indicate that P-V fit is positively related to the intention to pursue bridge employment within the same organization. Moreover, it established the P-O fit’s dual role (i.e., mediator and moderator). Nevertheless, further research is needed to generalize findings from the current study.

Footnotes

Acknowledgments

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author contributions

CONCEPTION: Khaled Lahlouh and Aïcha Oumessaoud

INTERPRETATION OR ANALYSIS OF DATA: Khaled Lahlouh and Raja Sreedharan PREPARATION OF THE MANUSCRIPT: Aïcha Oumessaoud

REVISION FOR IMPORTANT INTELLECTUAL CONTENT: Khaled Lahlouh and Raja Sreedharan

SUPERVISION: Khaled Lahlouh

See

References

Kim

, Hall

. Protean career model and retirement. In: The Oxford handbook of retirement. New York, NY, US: Oxford University Press; 2013. pp. 102–16.

Wöhrmann

, Fasbender

, Deller

. Using work values to predict post-retirement work intentions. The Career Development Quarterly. 2016;64(2):98–113.

Jones

, McIntosh

. Organizational and occupational commitment in relation to bridge employment and retirement intentions. Journal of Vocational Behavior. 2010;77(2):290–303.

Weckerle

, Shultz

. Influences on the bridge employment decision among older USA workers. Journal of Occupational and Organizational Psychology. 1999;72(3):317–29.

Shultz

. Bridge employment: Work after retirement. In: Adams GA, Beehr TA, editors. Retirement: Reasons, processes, and results. New York: Springer; 2003. pp. 214–41.

Raymo

, Liang

, Sugisawa

, Kobayashi

, Sugihara

. Work at Older Ages in Japan: Variation by Gender and Employment Status. Journals of Gerontology: Series B. 2004;59(3):S154–63.

Quinn

. Retirement, and the Encore Career: Elders and the Future of the American Workforce. In 2010.

Lancee

, Radl

. Social connectedness and the transition from work to retirement. The journals of gerontology Series B, Psychological sciences and social sciences [Internet]. 2012;67(4):481–90.

Zhan

, Wang

, Liu

, Shultz

. Bridge Employment and Retirees’ Health: A Longitudinal Investigation. In: Journal of occupational health psychology. 2009.

10.

Zhan

, Wang

, Yao

. Domain-specific effects of commitment on bridge employment decisions: The moderating role of economic stress. European Journal of Work and Organizational Psychology. 2013;22(3):362–75.

11.

Peng

, Xu (Violet)

, Jex

, Chen

. The Roles of Job-Related Psychosocial Factors and Work Meaningfulness in Promoting Nurses’ Bridge Employment Intentions. Journal of Career Development. 2019;47(6):701–16.

12.

Peng

, Min

. Older employees’ postretirement work planning: an application of the theory of planned behavior. Career Development International. 2020;25(6):565–79.

13.

Wang

, Zhan

, Liu

, Shultz

. Antecedents of Bridge Employment: A Longitudinal Investigation. Macroeconomics: Employment. 2008.

14.

Davies

EMM

, der Heijden

BIJM

, Flynn

. Job Satisfaction, Retirement Attitude and Intended Retirement Age: A Conditional Process Analysis across Workers’ Level of Household Income. Frontiers in Psychology. 2017;8:891.

15.

Lahlouh

, Lacaze

, Huaman-Ramirez

. Bridge employment and full retirement intentions: the role of person-environment fit. Personnel Review. 2019;48(6):1469–90.

16.

Oakman

, Wells

. Working longer: What is the relationship between person-environment fit and retirement intentions? Asia Pacific Journal of Human Resources. 2016;54(2):207–29.

17.

Astakhova

. Explaining the effects of perceived person-supervisor fit and person-organization fit on organizational commitment in the U.S. and Japan. Journal of Business Research. 2016;69(2):956–63.

18.

I-S

, Guay

, Kim

, Harold

, Lee

J-H

, Heo

C-G

, et al. Fit Happens Globally: A Meta-Analytic Comparison of the Relationships of Person-Environment Fit Dimensions with Work Attitudes and Performance Across East Asia, Europe, and North America. Personnel Psychology. 2014;67(1):99–152.

19.

Vogel

, Feldman

. Integrating the levels of person-environment fit: The roles of vocational fit and group fit. Journal of Vocational Behavior. 2009;75(1):68–81.

20.

Abdalla

, Elsetouhi

, Negm

, Abdou

. Perceived person-organization fit and turnover intention in medical centers. Personnel Review. 2018 1;47(4):863–81.

21.

Edwards

, Shipp

. The relationship between person environment fit and outcomes: An integrative theoretical framework. In: Perspectives on organizational fit. Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers; 2007. pp. 209–58. (The organizational frontiers series.).

22.

Atchley

. A Continuity Theory of Normal Aging. The Gerontologist [Internet]. 1989;29(2):183–90.

23.

Lieberman

. The experience of old Age-Stress, coping, survival. New York: Basic Books Inc; 1983.

24.

Schmidt

, Lee

. Voluntary retirement and organizational turnover intentions: The differential associations with work and non-work commitment constructs. Journal of Business and Psychology. 2008;22(4):297–309.

25.

Gobeski

, Beehr

. How retirees work: Predictors of different types of bridge employment. Journal of Organizational Behavior. 2009;30(3):401–25.

26.

Ashforth

. Role transitions in organizational life: An identity-based perspective. Role transitions in organizational life: An identity-based perspective. Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers; 2001.

27.

Barness-Farrell

. Beyond health and wealth: attitudinal and other influences on retirement decision making. In: Adams G, Beehr T, editors. Retirement: Reasons, Processes and Results. New York: Springer publishing company; 2003. pp. 159–213.

28.

Kristof-Brown

, Zimmerman

, Johnson

. Consequences of Individuals’ Fit at Work: A Meta-Analysis of Person-Job, Person-Organization, Person-Group, and Person-Supervisor Fit. Personnel Psychology. 2005;58(2):281–342.

29.

KYT

. Inter-Relationships among Different Types of Person-Environment Fit and Job Satisfaction. Applied Psychology. 2016;65(1):38–65.

30.

Cable

, DeRue

. The convergent and discriminant validity of subjective fit perceptions. The Journal of applied psychology. 2002;875:875–84.

31.

O’Reilly

, Chatman

, Caldwell

. People and organizational culture: A profile comparison approach to assessing person-organization fit. Academy of Management Journal. 1991;34(3):487–516.

32.

Zaniboni

, Sarchielli

, Fraccaroli

. How are psychosocial factors related to retirement intentions? Shultz K, Henkens K, editors. International Journal of Manpower. 2010;31(3):271–85.

33.

Kristof

. Person-organization fit: An integrative review of its conceptualizations, measurement, and implications. Personnel Psychology. 1996;49(1):1–49.

34.

Richardson

, Kilty

. Retirement Intentions Among Black Professionals: Implications for Practice with Older Black Adults1. The Gerontologist. 1992 1;32(1):7–16.

35.

Feldman

. The Decision to Retire Early: A Review and Conceptualization. The Academy of Management Review. 1994;19(2):285–311.

36.

Giauque

, Resenterra

, Siggen

. Antecedents of Job Satisfaction, Organizational Commitment and Stress in a Public Hospital: a P-E Fit Perspective. Public Organization Review. 2014;14(2):201–28.

37.

Wheeler

, Gallagher

, Brouer

, Sablynski

. When person-organization (mis)fit and (dis)satisfaction lead to turnover: The moderating role of perceived job mobility. Journal of Managerial Psychology. 2007;22(2):203–19.

38.

Feldman

, Ng

TWH

. Careers: Mobility, Embeddedness, and Success. Journal of Management. 2007 1;33(3):350–77.

39.

Schneider

. The People Make The Place. Personnel Psychology. 1987;40(3):437–53.

40.

Armstrong-Stassen

. Organisational practices and the post-retirement employment experience of older workers. Human Resource Management Journal. 2008;18(1):36–53.

41.

Herrbach

, Mignonac

, Vandenberghe

, Negrini

. Perceived HRM practices, organizational commitment, and voluntary early retirement among late-career managers. Human Resource Management. 2009;48(6):895–915.

42.

Boumans

NPG

, de

Jong AHJ

, Vanderlinden

. Determinants of early retirement intentions among Belgian nurses. Journal of Advanced Nursing. 2008;631:64–74.

43.

Anderson

, Guo

. The Changing Face of Retirement: Exploring Retirees’ Communicative Construction of Tensions through Bridge Employment. Communication Studies. 2018;69(2):196–212.

44.

Nur Iplik

, Can Kilic

, Yalcin

. The simultaneous effects of person-organization and person-job fit on Turkish hotel managers. International Journal of Contemporary Hospitality Management. 2011 1;23(5):644–61.

45.

Demir

, Demir sen

, Nield

. The relationship between person-organization fit, organizational identification and work outcomes. Journal of Business Economics and Management. 2015 1;16(2):369–86.

46.

Association Française des anques. Profil de branche. 2021, available at: https://www.afb.fr/chiffres-et-etudes/

47.

Ekerdt

. Workplace Norms for the Timing of Retirement. In: Schaie KW, Schooler C, editors. Impact of Work on Older Adults. New York, NY: Springer Publishing Co., Inc.; 1998. pp. 101–23.

48.

Vignoli

, Zaniboni

, Chiesa

, Alcover

C-M

, Guglielmi

, Topa

. Maintaining and engaging older workers at work: the trigger role of personal and psychosocial resources. The International Journal of Human Resource Management. 2021;32(8):1731–53.

49.

Galloway . Non-Probability Samling. Encyclopedia of Social Measurement. 2005;2.

50.

Brislin

. Comparative research methodology: Cross-cultural studies. International Journal of Psychology. 1976;11(3):215–29.

51.

Kennedy

MBA

. An integrative investigation of person vocation fit, person-organization fit, and person-job fit perceptions. 2005.

52.

Taylor

MAW

, Shore

. Predictors of planned retirement age: an application of Beehr’s model. Psychology and Aging. 1995;10 1:76–83.

53.

Adams

, Beehr

. Turnover and retirement: A comparison of their similarities and differences. Personnel Psychology. 1998;51(3):643–65.

54.

Wang

, Shultz

. Employee Retirement: A Review and Recommendations for Future Investigation. Journal of Management. 2009 6;36(1):172–206.

55.

Hair

, Ringle

, Sarstedt

. PLS-SEM: Indeed a Silver Bullet. Journal of Marketing Theory and Practice. 2011 1;19(2):139–52.

56.

Fornell

, Larcker

. Evaluating Structural Equation Models with Unobservable Variables and Measurement Error. Journal of Marketing Research. 1981 1;18(1):39–50.

57.

Henseler

, Ringle

, Sarstedt

. A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science. 2015;43(1):115–35.

58.

Gold

, Malhotra

, Segars

. Knowledge Management: An Organizational Capabilities Perspective. Journal of Management Information Systems. 2001;18(1):185–214.

59.

Cohen

. Statistical Power Analysis for the Behavioral Sciences. New York: Routeldge; 2013.

60.

Chin

. The partial least squares approach for structural equation modeling. In: Modern methods for business research. Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers; 1998. pp. 295–336.

61.

, Bentler

. Fit indices in covariance structure modeling: Sensitivity to underparameterized model misspecification. Psychological Methods. 1998;3(4):424–53.

62.

Devarajan

, Manimuthu

, Sreedharan

. Healthcare Operations and BlackSwan Event forCOVID-19 Pandemic: A Predictive Analytics. IEEE Transactions on Engineering Management. 2021;1-15.

63.

Currie

. Intelligent imaging: Anatomy of machine learning and deep learning. Journal of Nuclear Medicine Technology. 2019;47(4):273–81.

64.

Ali

, Kaur

, Persis

, Saha

, Pattusamy

, Sreedharan

. Developing a hybrid evaluation approach for the low carbon performance on sustainable manufacturing environment. Annals of Operations Research. 2020.

65.

Wang

, Shultz

. Employee Retirement: A Review and Recommendations for Future Investigation. Journal of Management. 2009;36(1):172–206.

66.

Jager

, Putnick

, Bornstein

. More than just convenient: The scientific merits of homogeneous convenience samples. Monographs of the Society for Research in Child Development. 2017;82(2):13–30.

67.

Ostroff

, Shin

, Feinberg

. Skill acquisition and person-environment fit. In: Feldman D, editor. Work careers: A developmental perspective. San Francisco: Jossey-Bass; 2002. pp. 63–90.

Person-vocation and same organization bridge employment intention: Testing the mediating and moderating roles of person-organization fit

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1 Introduction

2 Theoretical background and hypotheses development

2.1 Person-environment fit, continuity and role theories

2.2 Person-vocation fit and the intention to pursue bridge employment within the same organization

2.3 Person-vocation fit and the intention to pursue bridge employment within the same organization: The mediating and moderator role of person-organization fit

3.1 Sample and procedure

3.2 Measures

4.1 Psychometric properties of measurement scales

Table 2 Discriminant validity (latent variable correlation and square root of AVE) Measurement models 1 2 3 4 5 1. P-V. fit 0.842 2. P-O fit 0.525 0.858 3. SOBEI 0.478 0.466 0.973 . 4. HSAT 0.175 0.142 0.199 0.849 5. RETADJ 0.112 –0.010 –0.162 0.078 0,832

Table 5 Structured model results Construct R2 Adj. R2 Q2 VIF SRMR SOBEI 0,367 0,348 0,324 0,078 PO fit 0,275 0,272 0,198 1,473 P.V. fit 1,513 PV fit* P.O. fit 1,074 HSAT 1,057 RETADJ 1,050 SALARY 1,056 Age 1,070

Table 7 Precision; recall; F1 score for the SVM classifier Hypotheses Class Precision Recall F1 score P-V Fit based SOBE 1 0.923077 1 0.96 2 0.5 0.5 0.5 3 1 0.2 0.333333 4 0.375 0.75 0.5 5 0.769231 1 0.869565 6 0.9 0.692308 0.782609 7 1 0.5 0.666667

6 Theoretical and methodological implications

7 Practical implications

8 Limitations and future opportunities

9 Conclusion

Footnotes

Acknowledgments

Author contributions

References

Table 2
Discriminant validity (latent variable correlation and square root of AVE)

Measurement models 1 2 3 4 5

1. P-V. fit 0.842

2. P-O fit 0.525 0.858

3. SOBEI 0.478 0.466 0.973 .

4. HSAT 0.175 0.142 0.199 0.849

5. RETADJ 0.112 –0.010 –0.162 0.078 0,832

Table 5
Structured model results

Construct R² Adj. R² Q² VIF SRMR

SOBEI 0,367 0,348 0,324 0,078

PO fit 0,275 0,272 0,198 1,473

P.V. fit 1,513

PV fit* P.O. fit 1,074

HSAT 1,057

RETADJ 1,050

SALARY 1,056

Age 1,070

Table 7
Precision; recall; F1 score for the SVM classifier

Hypotheses Class Precision Recall F1 score

P-V Fit based SOBE 1 0.923077 1 0.96

2 0.5 0.5 0.5

3 1 0.2 0.333333

4 0.375 0.75 0.5

5 0.769231 1 0.869565

6 0.9 0.692308 0.782609

7 1 0.5 0.666667