Transformational leadership and horizontal trust as antecedents of team performance in the healthcare context

El liderazgo transformacional y la confianza como antecedentes del desempeño en equipo en el ámbito sanitario

Juana Olvera¹, Susana Llorens¹, Hedy Acosta² and Marisa Salanova¹

¹ Equipo de Investigación WANT Prevención Psicosocial y Organizaciones Saludables. Universitat Jaume I, Castellón de la Plana (España)
² Universidad Adolfo Ibáñez (Chile)

This study was supported by a grant from Consellería de Educación of the Generalitat Valenciana - Prometeo Program (PROMETEO/2013/025) and the Research Promotion Program of Universitat Jaume I (P11B2014-40).

ABSTRACT

The present study analyzes the mediator role of work-team trust (i.e., horizontal trust) in the relationship between transformational leadership, as a social resource, and team performance (i.e., intra- and extra-role performance), as suggested by the HERO model (HEalthy & Resilient Organizations Model, Salanova, Llorens, Cifre, & Martínez, 2012). The sample corresponds to 388 workers nested in 54 work teams from four organizations in the healthcare sector. Horizontal trust and transformational leadership were assessed by the work teams, and performance was assessed by the supervisors of these teams. Structural Equations models reveal, as expected, that horizontal trust has a fully mediating role between transformational leadership perceived at the team level and team performance assessed by the supervisor.

Key words: Horizontal trust; transformational leadership; team performance.

RESUMEN

En el presente estudio analizamos el rol mediador de la confianza en el equipo de trabajo (i.e., confianza horizontal) entre el liderazgo transformacional, como recurso social, y el desempeño en equipo (i.e., desempeño intra y extra rol), tal como sugiere el Modelo HERO (HEalthy & Resilient Organizations Model; Salanova, Llorens, Cifre, & Martínez, 2012). La muestra corresponde a 388 trabajadores/as agrupados/as en 54 equipos de trabajo pertenecientes a cuatro organizaciones del sector sanitario. La confianza horizontal y el liderazgo transformacional fueron evaluados por los equipos de trabajo y el desempeño fue evaluado por los/as supervisores/as de esos equipos. Los modelos de Ecuaciones Estructurales revelaron, tal como se esperaba, que la confianza horizontal, tiene un papel mediador entre el liderazgo transformacional percibido por el equipo y el desempeño en equipo evaluado por el/la supervisor/a.

Palabras clave: Confianza horizontal; liderazgo transformacional; desempeño en equipo.

The healthy organization concept is currently acquiring special relevance in the framework of Positive Occupational Health Psychology, which intends to address the positive aspects of health in organizations, without ignoring the traditional negative aspects. Healthy organizations can be defined as those that make systematic, planned, and proactive efforts to improve employees' health through good practices related to task improvement, the social environment, and the organization (Salanova, 2008; Salanova, Llorens, Cifre, & Martínez, 2012).

Given the consequences of the current socio-economic crisis on the labor setting, the healthcare organization occupies a relevant space that is important to address due to its impact on society's health and well-being as one of the basic pillars of the Welfare State. Currently, we do not have enough scientific evidence about the healthcare organization as a healthy organization. Some studies can be pointed out that are based on the concept of Magnetic Hospitals. The purpose of these studies was to identify the characteristics that functioned as "magnets" in hospitals to capture highly qualified nursing professionals for patient care (McClure, Poulin, Sovie, & Wandelt, 1983; Salanova, Rodríguez-Sánchez, Del Líbano, & Ventura, 2012). The magnetic hospital concept continues to be discussed today, although, considering that we have an active healthcare system that tends to change (Buchan, 1999), it would be interesting to advance the research by addressing the healthy organization concept in the healthcare organization context.

In this context, it is important to highlight the role played by organizational trust as a key construct that is beneficial for the organization's functioning and for its members (Kramer, 1999). Organizational trust is defined as "an employee's willingness to be vulnerable to the actions of the organization, whose behavior and actions he or she cannot control" (Tan & Lim, 2009, p. 46). Research has shown the relationship existing between organizational trust and performance and leadership effectiveness (i.e., Aryee, Budhwar, & Chen, 2002; Dirks & Ferrin, 2002; Mayer & Gavin, 2005). The interest in the study of organizational trust at the level of teams has grown considerably as organizations have evolved toward flatter structures based on work teams (Costa & Anderson, 2011). However, more research is necessary using collective-level data.

Thus, the present study evaluates the mediator role of organizational trust (i.e., horizontal trust) between transformational leadership and team performance in the healthcare organization context and at the work-team level, based on the HERO Model (HEalthy & Resilient Organizations Model; Salanova et al., 2012).

HEalthy and Resilient Organizations Model (HERO)

The HERO concept (HEalthy & Resilient Organization) is emerging as a way to overcome the crises and adversities in today's organizations. It implies a competitive advantage for companies that believe that taking care of the health of work teams and the organization is fundamental in periods of critical economic and social changes. Healthy and Resilient Organizations make systematic, planned, and proactive efforts to improve the processes and results of both the employees and the organization. They are "resilient" because they know how to draw strength from adverse situations, and in critical circumstances, they are able to maintain their functioning and their results in terms of profitability (Salanova et al., 2012; Acosta, Cruz-Ortiz, Salanova, & Llorens, 2015)

The model assumes that a HERO consists of three interrelated components that include different dimensions evaluated at the collective level: (1) healthy organizational resources and practices to structure and manage the work processes (i.e., conciliation practices, transformational leadership); (2) healthy employees and work teams (i.e., organizational trust, engagement); and (3) healthy organizational results (i.e., performance, company social responsibility). Studies on the HERO model indicate that investments in collective resources (i.e., team work, leadership) and healthy organizational practices (i.e., conciliation, career development) can improve employees' collective well-being (efficacy, engagement, and resilience) and performance levels assessed by the supervisor (intra- and extra-role) (Salanova et al., 2012); engagement in team work (Acosta, Salanova & Llorens, 2011; Acosta, Torrente, Llorens, & Salanova, 2013; Torrente, Salanova, Llorens, & Schaufeli, 2012); organizational trust (Acosta et al., 2011); and health-service quality (performance, commitment, perceived quality) (Hernández, Llorens, & Rodríguez, 2014).

The present study focuses on the three components of the HERO model (Salanova et al., 2012): healthy organizational resources and practices (i.e., transformational leadership); healthy employees (i.e., horizontal trust); and healthy organizational results (i.e., team performance), through data aggregated at the work-team level, combining the shared perceptions of the workers and the supervisors.

Transformational leadership is a leadership style that has continued to create great interest in the research in recent decades. It is defined as a management style where the leader broadens and elevates the interests of his/her employees and creates awareness and acceptance of the group's objectives and mission, putting the collective interests ahead of his/her own interests (Bass, 1998). For Salanova, (2008), the transformational leader knows how to lead others toward a goal that is perceived as shared, and he/she achieves the commitment of the work teams and the organization.

This leadership style implies that the leader: (1) manages to satisfy the collaborators' work needs, (2) knows what the collaborators' needs are, (3) facilitates the collaborators' contribution to organizational performance, and (4) considers the performance of the leader as a member of a work team (Bass & Avolio, 1994; Cruz-Ortiz, Salanova, & Martínez, 2013; Llorens, Salanova, & Losilla, 2009; Nielsen, 2014).

According to Rafferty and Griffin (2004), leadership is defined in five dimensions: (1) vision, expression of an idealized vision of the future based on the organization's values; (2) inspirational communication: positive messages about the organization that build motivation and trust; (3) intellectual stimulation: fostering the employees' interest in thinking about the problem in different ways; (4) support: the leader's concern about his/her employees, taking their needs into account; and (5) personal recognition:, rewarding the worker with praise and recognition for effort and for achieving specific goals (Cruz-Ortiz et al., 2013).

Studies point out the relationship between transformational leadership and performance of workers and work teams (i.e., Bass, Avolio, Jung, & Berson, 2003; Chun, Cho, & Sosik, 2015; Cruz-Ortiz et al., 2013; Pourbarkhordari, Zhou & Pourkarimi, 2016; Rao & Kareem, 2015; Walumbwa, Avolio, & Zhu, 2008). This leadership style is a variable that can be considered a social resource within organizations because, according to the literature, it helps to cope with demands and improves performance levels. The present study aims to provide evidence about the importance of transformational leadership as an antecedent of the performance level of work teams.

Transformational Leadership and Organizational Trust

As Kramer (1999) points out, organizational trust has played a leading role in the research, due to its considerable advantages for the organizational environment at both the individual and collective levels. Tan and Lim (2009) define it as "an employee's willingness to be vulnerable to the actions of the organization, whose behavior and actions he or she cannot control" (p. 46). These authors propose an organizational trust model that focuses on two levels: the co-workers and the organization. In addition, they conceive trust between co-workers (horizontal trust) as "the willingness of a person to be vulnerable to the actions of fellow coworkers whose behavior and actions that person cannot control" (Tan & Lim, 2009, p. 46). At the group level, trust is a collective phenomenon. When interacting within their team, individuals are more likely to develop shared perceptions, expectations, and behavioral norms with their team coworkers, and through these interactions, team members are likely to develop shared perceptions of trust (Costa & Anderson, 2011). In the HERO model (Salanova et al., 2012), organizational trust, a key construct in the "healthy employees" component, includes two dimensions: (1) Vertical trust, the degree to which employees trust the actions of their superiors or the organization where they work; and (2) Horizontal trust, the degree to which employees trust the people they work with, trust what they do, and enjoy being with them.

Trust is one of the most frequently cited constructs in the literature on transformational leadership (i.e., Fulmer, & Gelfand, 2012). However, the research contributes little evidence about the importance of transformational leadership in building organizational trust (i.e., Mishra, 1995; Pod-sakoff, MacKenzie, Moorman, & Fetter, 1990; Zhu, Newman, Miao, & Hooke, 2013). Although interest in the study of organizational trust at the team level has increased considerably, most of the research has focused on vertical trust. Therefore, the present study aims to fill this gap by providing evidence about the relationship between transformational leadership and horizontal trust at the work-team level.

Transformational Leadership, Organizational Trust and Team Performance

Within the research framework, performance is viewed as a significant indictor at the organizational level. Some authors consider that the success of an organization depends on its employees' good performance (Colquitt, LePine, & Wesson, 2010). Work performance is generally conceptualized as "the actions and behaviors that are under the control of the individual and that contribute to the organization's objectives" (Rotundo & Sackett, 2002, p. 66). Moreover, Campbell (1990) states that work performance is a multidimensional construct, an opinion that is shared by other researchers (Borman & Motowidlo, 1993; Goodman & Svyantek, 1999; Johnson, 2009).

Evidence shows that transformational leadership has a strong influence on performance (Gang Wang, Oh, Courtright, & Colbert, 2011). However, there is still little understanding of the processes through which leadership exercises this influence, as few studies have focused on the mediating influence of other strategic variables (Bass, 1999; Yukl, 2006; García-Morales, Llorens-Montes, & Verdú-Jover, 2008; Pourbarkhordari et al., 2016).

Different studies support the relationship between trust and organizational performance (Acosta et al., 2015; Mayer et al., 2005; Brown, Gray, McHardy, & Taylor, 2015; Zawawi & Nasurdin, 2016). In other studies, horizontal trust has been associated with improvements in communication, job satisfaction, commitment, learning, decision-making, team work, and performance (Costa, 2003; Kiffin-Petersen & Cordery, 2003; Lee, Stajkovic, & Cho, 2011; Salanova, Acosta, Llorens, & Le Blanc, 2017). Furthermore, some studies point out the mediator role of organizational trust in the relationship between resources and performance at the work team level (Palanski, Kahai, & Yammarino, 2011; Tan & Lim, 2009; Zhu et al., 2013).

Based on these arguments, the present study aims to provide evidence about the mediator role of horizontal trust between transformational leadership and performance at the work team level.

The Present Study

Based on the HERO model (Salanova et al., 2012), our objective is to evaluate the relationship between the transformational leadership perceived by the work team and the team performance perceived by the supervisor, taking into account the mediator role of the horizontal trust perceived by the work team in the healthcare organization context.

The novel aspects of this study are the following: (1) It analyzes the relationships among variables from the perspective of different key agents in the organization (work teams and supervisors); (2) It analyzes the mediator role of horizontal trust in the relationship between transformational leadership and performance at the work team level; (3) It focuses on the healthcare context, where the work team plays a key role, and where it is important to more closely examine organizational trust as a strategic variable in the development of healthy and resilient healthcare organizations.

Based on the above, we propose the following hypothesis:

Hypothesis 1. We expect that the horizontal trust perceived by the work team will fully mediate the relationship between transformational leadership perceived by the work team and team performance perceived by the supervisor (see Figure 1).

Participants

The participants in the present study are 388 workers grouped in 54 work teams and the 54 supervisors of these teams, from four healthcare centers (three Hospitals and one Primary Attention Center). The universe of the population is the healthcare context of the Valencian Community, and the sampling was non-probabilistic. In the study, 53% of the total number of workers participated, so that it is a representative sample with a margin of error of .03 and 90% reliability.

Of all the workers who participated in the study, 62% are women, and 78% have a permanent contract and a mean of 5.25 years working in the firm (SD = 3.6). Specifically, the workers in Hospitals 1, 2 and 3 have been working for a mean of 4 (SD = 2.2), 10 (SD = 9.8) and 4 (SD = 1.2) years, respectively. The mean number of years working in the Primary Attention center is 3 years (SD = 1.25).

In the case of the supervisors who participated in the study, a response rate of 71% was obtained, so that it is a representative sample with a margin of error of .06 and 90% reliability. Of all the supervisors participating in the study, 70% are women, and 98% have a permanent contract and a mean of 9 years in the firm (SD = 2.97). Specifically, the supervisors of Hospitals 1, 2 and 3 have worked for a mean of 4 (SD = 2.1), 24 (SD = 8.9) and 4.6 (SD = .3) years, respectively. The mean number of years working in the Primary Healthcare center is 3.6 (SD = .6). Finally, the work teams have an average of 7.2 (SD = 6) workers.

A criterion for participating in the study, based on McCarthy (1992), is that the workers had to have worked in the firm for at least six months. This criterion provides guarantees about adaptation to the job and the firm after having overcome the first phases of the organizational socialization process. Furthermore, in the present study, the work team is understood as a group of workers who work together under the same supervisor and collectively share responsibilities related to their performance results (George, 1990).

Procedure

Regarding the procedure, the first contact was made with a key agent of the management team of each healthcare center in order to present the research project and request their participation in it. After various initial interviews, and after approval had been received from the management of each healthcare center, the study was carried out. Informative meetings were held with the workers and supervisors of each healthcare center to present the objectives of the project, ethical aspects, and the procedure to follow.

The workers and supervisors who agreed to participate in the study filled out either a self-report questionnaire provided in paper format or the online version. The questionnaire takes approximately 30 minutes to complete.

Regarding the ethical aspects of the present study, the researchers guaranteed that the applicable norm would be followed at all times. Specifically, as required by Organic Law 15/1999, of December 13^th, on Protection of Data of a Personal Nature, the data were handled with complete confidentiality.

Transformational leadership, considered as the independent variable, was measured with the questionnaire by Rafferty and Griffin (2004), based on the Multifactor Leadership Questionnaire (Bass, 1985; Bass & Avolio, 1990), and adapted to Spanish and included in the HERO-HOS questionnaire by Salanova et al. (2012) in its versions for employees and supervisors in the healthcare context. This measurement instrument contains 15 items grouped in five dimensions: (1) Vision, three items (alpha = .74) (e.g., "Our immediate boss understands perfectly what the group's objectives are"); (2) Inspirational communication, three items (alpha = .88) (e.g., "Our immediate boss encourages us to see changes as situations full of opportunities"); (3) Intellectual stimulation, three items (alpha = .84) (e.g., "Our immediate boss encourages us to reconsider some basic questions that we had accepted about our work"); (4) Support, three items (alpha = .93) (e.g., "Our immediate boss thinks about our needs"); (5) Personal recognition, three items (alpha = .96) (e.g., "Our immediate boss congratulates us personally when we do an excellent job"). All the items on the questionnaire are formulated from the perspective of the work team, and they are measured on a seven-point Likert scale ranging from 0 (Strongly disagree) to 6 (Strongly agree).

Horizontal trust, considered as a mediator variable, was measured through the questionnaire by McAllister (1995), adapted and included in the HERO-HOS questionnaire by Salanova et al. (2012), in its versions for employees and supervisors. This instrument (alpha = .75) contains four items (e.g., "In this organization, we can share our ideas, emotions, and hopes"). The workers answered with their work team in mind, using a seven-point Likert scale ranging from 0 (Strongly disagree) to 6 (Strongly agree).

Supervisors' Measures

Team performance was the dependent variable. It is a multidimensional variable composed of two dimensions: intra-role performance and extra-role performance. Team performance was evaluated through six items that correspond to an adaptation of the performance scale by Goodman and Svyantek (1999), adapted and included in the HERO-HOS questionnaire by Salanova et al. (2012), in its versions for employees and supervisors. It consists of three items about intra-role performance (alpha = .72) and three others about extra-role performance (alpha = .83). An example of an item from the intra-role dimension is "The team I supervise fulfills all the functions and tasks the job requires". An example of an item from the extrarole dimension is 'The team I supervise performs functions that are not required, but improve the organization's image. Supervisors answered with their own work team in mind, using a seven-point Likert scale ranging from 0 (Strongly disagree) to 6 (Strongly agree).

Data Analysis

Using version 22.0 of the IBM-SPSS software, and based on the individual databases, an analysis of the internal consistency of the scales was performed by calculating the Cronbach's Alpha coefficient. Next, and because the study is focused on the work-group level, the agreement indices for the different scales were calculated (Chen, Mathieu, & Bliese, 2004). For this purpose, two complementary methods were used. The first method was to calculate the Intraclass Correlation Coefficients, CCIi and CCI2 (Glick, 1985), an approach based on consistency, where agreement between work teams is indicated when the CCI1 and CCI2 values are above .12 and .60, respectively (Bliese, 2000; Glick, 1985). The second method was to calculate as consensus measures: (1) the Average Deviation Index, AD_mj, where the criterion for a seven-point scale is that the AD_mj must be equal to or less than 1.20 (Burke, Finkelstein, & Dusig,1999; Cohen, Doveh, & Nahum-Shani, 2010), and (2) the Within-Group Interrater Reliability Index, rwgj, with index values equal to or above.70 showing good agreement (James, Demaree, & Wolf, 1993).

Finally, an analysis of variance (ANOVA) was carried out to show significant differences between the work teams on the scales evaluated. In addition, descriptive analyses were performed, as well as analyses of the correlations between the scales at the individual level and at the aggregated data level, that is, the work-team level.

To analyze the relationships among the variables transformational leadership, horizontal trust in work teams, and team performance evaluated by the supervisor, Structural Equations analysis (SEM) was performed on the proposed models. For this purpose, the statistical packet used was version 22.0 of Analysis of Moment Structures (AMOS), created by Arbuckle (2003).

We compared two mediation models in order to verify the study hypothesis: M1, the proposed model, in which horizontal trust (at the work-team level) fully mediates the relationship between transformational leadership (at the work-team level) and team performance (evaluated by the supervisor); M2, partial mediation model, which considers the existence of a direct relationship between transformational leadership (at the work-team level) and team performance (evaluated by the supervisor). Therefore, in this model, horizontal trust (at the work-team level) would partially mediate the relationship between transformational leadership (at the work-team level) and team performance (evaluated by the supervisor).

The method used to estimate the structural models is maximum likelihood. Among the indicators used, the following absolute fit indexes were considered: the Chi-square (X²) index, the Chi-square (X²/g), and the Root Mean Square Error of Approximation, (RMSEA). The relative fit measures we used were: The Normed Fit Index, (NFI); Bentler & Bonett, 1980), the Non-normalized Fit Index (also called the Tucker-Lewis Index, TLI or NNFI), the Comparative Fit Index, (CFI); Bentler, 1990), and the Incremental Fit Index, (IFI); Bollen, 1986). In addition, the Akaike Information Criterion (AIC) was calculated to compare non-nested competitive models. These fit measures were suggested by Marsh, Balla, and Hau (1996). Ap value associated with a Chi-square above .05 indicates good fit; likewise, chi square (X²/g), used to reduce the sensitivity of X² to the sample size, indicates a good fit of the model for values below 2. The RMSEA index responds to how well the model would fit the covariance matrix of the population. RMSEA results below .05 indicate very good fit; values between .05 and .08 are considered acceptable fit; values between .08 and .10 indicate a moderate fit; and values above .10 indicate a poor fit and, therefore, suggest that the model should be rejected (Browne & Cudeck, 1993). Regarding the TLI, IFI, CFI and NFI indices, values above .90 are considered indicators of good fit (Hoyle, 1995). With regard to the AIC, the lower the levels of AIC, the better the fit is supposed to be (Akaike, 1987).

Next, in order to test the proposed mediation model, an analysis was performed following Baron and Kenny's steps (1986). And finally, the Sobel test (1988) was carried out to confirm the indirect effect of transformational leadership perceived by the team of workers on team performance evaluated by the supervisor, through the horizontal trust of the work team.

Results

Descriptive Analyses and Data Aggregation

Table 1 shows the results for the means, standard deviations, correlations, intraclass correlation indices (CCI₁ and CCI₂), and AD_Mj and rwg_j agreement indices for the study variables.

Given that the study focuses on the work-team level, and taking into account the two approaches chosen to calculate the scales' agreement indices used to justify the data aggregation (Chen et al., 2004), the results obtained were: (1) For the consistency indices using Intraclass Correlation Coefficients, CCI₁, ranged from .39 to .94, and the CCI₂ values were maintained in a range from .67 to .98; (2) The agreement indices using the Average Deviation Index, AD_Mj, ranged from .15 to .80, and using the Within-Group Interrater Reliability, rwg_j, the indices met the criterion of being above .70 (James et al., 1993), except in the case of the variable Transformational leadership (TL) Vision, which obtained a result of .60. Therefore, we can conclude that the inter-group agreement fulfills the criteria for aggregating the group members' perceptions at the work-team level.

The results of the ANOVA show significant differences between groups on the scales evaluated, supporting the validity of the measures: TL Vision, F (53.334) = 3.29, p < .001; TL Communication, F (53.334) = 2.41, p < .001; TL Stimulation, F (53.332) = 2.78, p < .001; TL Support, F (53.331) = 2.29, p < .001; TL Recognition, F (53.332) = 2.52,p < .001; Horizontal Trust, F (53.327) = 2.76,p < .001; Intra-role Performance, F (53.81) = 2.87, p < .001 and Extra-role Performance, F (53.82) = 2.04, p = .002.

The results of the analysis of the correlations between the scales show that all the dimensions of transformational leadership (at the work-team level) correlate positively and significantly with each other, with values ranging from .69 to .87 (p < .01);The horizontal trust variable (at the work-team level) correlates positively and significantly with the dimensions of leadership (at the work-team level), with values ranging from .49 to .62 (p < .01), and with the dimension of team performance (evaluated by the supervisor), with values ranging from .31 to .34 (p < .05); The dimensions of team performance evaluated by the supervisor show a positive and significant correlation of .67 (p < .01) with each other. However, the dimension of intra-role performance (evaluated by the supervisor) does not correlate significantly with the vision, communication, or intellectual stimulation dimensions of transformational leadership (at the work-team level), and there is no significant correlation between the extra-role performance and intellectual stimulation dimensions of transformational leadership.

To perform the analysis of the Structural Equations Models (SEM), we used the data aggregated at the team level (N = 54) (see Table 2). The transformational leadership variable has five indicators: vision, communication, intellectual stimulation, support, and recognition; the horizontal trust variable has only one indicator, and the team performance variable has two indicators: intra-role performance and extra-role performance.

The results obtained for the proposed model, M₁, in which horizontal trust (at the team level) fully mediates the relationship between transformational leadership (at the team level) and team performance (evaluated by the supervisor) are the following: X²(19) = 25.28; p = .15; X²/gl = 1.33; TLI = .97; IFI = .98; CFI = .98; NFI = .93 and RMSEA = .07; AIC = 75.27. Regarding M₂, the partial mediation model, which considers a direct relationship between transformational leadership (at the team level) and team performance (evaluated by the supervisor), the following values were obtained: X²(18) = 24.99; p = .12; X²/gl= 1.40; TLI = .96; IFI = .98; CFI = .98; NFI = .93 and RMSEA = .09; AIC = 77.00. The Chi-squared tests between M₂ and M₁ did not show significant differences between the two models, Delta X²(1) = .29, ns. However, the results obtained support the proposed model, M₁, because: (1) It is more parsimonious than M₂, as M₁ showed lower AIC values; (2) In the case of M₂, partial mediation, the relationship between horizontal trust and team performance was no longer significant, β = .34, p = .13; (3) In the partial mediation model, M2, the results show that the relationship between transformational leadership and team performance was not significant, β = .11, p = .59. Therefore, the proposed model, M₁, fits the data better, supporting the hypothesis of full mediation.

Based on model M₁, and to show that the model was not chosen arbitrarily (Kline, 1998), two other alternative models were analyzed (M₃ and M₄) with the following results: (1) In M₃, the team performance variable (evaluated by the supervisor) fully mediates the relationship between transformational leadership (at the team level) and horizontal trust (at the team level), X²(19) = 43.03; p = .01; X²/gl = 2.30; TLI = .89; IFI = .93; CFI = .93; NFI = .88 and RMSA = .15; AIC = 93.03; (2) In the case of M₄, the variable transformational leadership (work team) fully mediates the relationship between horizontal trust (work team) and team performance (evaluated by the supervisor), f²(19) = 27.30; p = .09; X²/gl = 1.44; TLI = .96; IFI = .93; CFI = .97; NFI = .92 and RMSA = .10; AIC = 77,30. The Chi-squared tests performed between the models M₃-M₁, M₃-M₂, M₄-M₁, M₄-M₂ and M₄-M₃ showed non-significant results. Even though M₄ offers a generally good fit (leading to the idea that the relationship between trust and leadership could be longitudinal), the results obtained continue to support model M₁, proposed from the beginning of the study, as this model continues to be more parsimonious and has a lower AIC value.

Next, we evaluated the mediation of Mi, following Baron and Kenny's four steps (1986): (1) Transformational leadership (at the team level) is positively and significantly related to team performance (evaluated by the supervisor) (β = .36, p < .05); (2) Transformational leadership is significantly related to horizontal trust (β = .68, p < .001); (3) Horizontal trust (at the team level) is significantly related to team performance (evaluated by the supervisor) (β = .43, p < .01); (4) The relationship between transformational leadership (at the team level) and team performance (evaluated by the supervisor) is no longer significant when controlled by horizontal trust (at the team level) (β = .11, p = .59). Therefore, the four conditions proposed by Baron and Kenny (1986) are met, showing that the results are consistent with the full mediation hypothesis, proposed in M₁.

The Sobel test (1982) confirms the indirect effect of transformational leadership perceived by the work team on team performance evaluated by the supervisor, through the horizontal trust of the work team (Sobel test = 2.40, p < .01). The results show that this relationship is statistically significant, and that the mediation is full.

Consequently, the results of the SEM analysis provide evidence in favor of M₁. Figure 2 shows the representation of the final proposed model.

All the scales used score significantly in the corresponding latent factors, with factorial weights ranging from .79 to .95. The review of the regression estimators for the proposed model, M₁, reveals that transformational leadership (at the team level) is positively and significantly related to horizontal trust (at the work team level), β = .70, p < .001, and this, in turn, is positively and significantly related to team performance (evaluated by the supervisor), β = .45, p < .01. In addition, transformational leadership explains 47.4% of the variance in horizontal trust, R² = .47, which explains 20% of the variance in team performance, R² = .20 (R²_intra-role = .72, R²_Extra-role = .63).

Discussion

The results of the Structural Equations Models showed, as expected, that horizontal trust perceived by the work team fully mediates the relationship between transformational leadership perceived by the work team and team performance perceived by the supervisor.

In the organizational context, this common perception of transformational leadership, as a social resource of the organization, can cause interaction processes to emerge in workers that increase this shared vision of trust among team members. The workers have a common scenario in which to interact both consciously and unconsciously (Torrente, Salanova, Llorens, & Schaufeli, 2015), in order to reciprocally influence and trigger a positive shared state, as in the case of horizontal trust (Bakker, Van Emmerik, & Euwema, 2006).

The results obtained contribute evidence to the previous research on the importance of horizontal trust as a strategic mediator variable between transformational leadership at the team level and team performance (i.e., Costa & Anderson, 2011; Palanski, et al., 2011; Tan & Lim, 2009).

In addition, the present study provides evidence for the three elements that make up the HERO model (Salanova et al., 2012), which are: healthy organizational resources and practices (i.e., transformational leadership); healthy employees (i.e., horizontal trust); and healthy organizational results (i.e., team performance), through aggregated data at the team level taken from different information sources (i.e., workers and supervisors). The results obtained at the level of work teams and the supervisors of these teams show how resources such as transformational leadership foment horizontal trust, viewed as one of the dimensions that the model interprets as healthy workers, which in turn increases the level of intra- and extra-role performance, dimensions the model regards as healthy results (Salanova, Martínez, & Llorens, 2014).

The present study advances the research, given that: (1) It considers the work-team level; (2) It considers the relationships that can be maintained between the perceptions of different key agents, as in the case of the work team and the supervisor; (3) It analyzes the mediator role of horizontal trust in the relationship between transformational leadership and team performance, whereas most of the research on organizational trust has focused on the analysis of vertical trust; (4) It focuses on the healthcare context because of its specific characteristics, where the work team plays a protagonist role, and due to the importance of further examining organizational trust as a strategic variable in the development of healthy and resilient healthcare organizations.

We can conclude that when teams are the main work structure in a certain organization, the healthcare organization in our case, the promotion of policies oriented toward teams will be the most efficient management behavior. Along these lines, the results of our study show the need for social resources and organizational practices that can trigger positive consequences in the professionals and produce the desired outcomes. Therefore, the results support the hypothesis proposed in this research.

Theoretical and Practical Implications

From a theoretical point of view, the present study extends the current knowledge about the key role of horizontal trust between the social resources of the organization (i.e., transformational leadership perceived by team members) and the supervisor's perception of team performance, taking into account data aggregated at the team level in four healthcare centers in Spain. Moreover, the results obtained support the HERO model, and they provide evidence of its theoretical validity in explaining team-level processes in a socially important sample like the healthcare context.

Limitations and Future Research

Some of the main limitations of our study are the following: (1) We use a non-probabilistic, convenience sample, which could affect the possible generalization of the results. However, it is a heterogeneous sample because it includes different teams from different healthcare firms and different healthcare levels, which allows us to obtain a view of the reality of the organization; (2) The data were obtained through self-report measures. However, we have to consider that the data were considered as shared perceptions at the level of work teams belonging to different firms within the healthcare sector. In addition, two sources of information were used (i.e., workers and supervisors), in order to increase the validity of the scores obtained and reduce the effects of common variance; (3) The design is cross-sectional, and so it is not possible to draw conclusions about causality between the variables included in the model. Therefore, future research should conduct longitudinal studies to discover causality. In fact, through different scientific studies, Salanova, Llorens, Acosta, and Torrente (2013) show the relevance of carrying out positive interventions based on practical research and a professional scientific model.

References

1. Acosta, H., Cruz-Ortiz, V. M., Salanova, M., & Llorens, S. (2015). Organizaciones saludables: Analizando su significado desde el Modelo HERO. Revista de Psicología Social, 13-24. Doi: 10.1080/21711976.2015.1016751.         [ Links ]

2. Acosta, H., Salanova, M., & Llorens, S. (2011). ¿Cómo predicen las prácticas organizacionales el Engagement en el trabajo en equipo? El rol de la confianza organizacional. Ciencia & Trabajo, 41, 125-134.         [ Links ]

3. Acosta, H., Torrente, P., Llorens, S., & Salanova, M. (2013). Prácticas organizacionales saludables: Un análisis estudio de su impacto relativo sobre el engagement con el trabajo. Revista Peruana de Psicología y Trabajo Social, 2, 107-120.         [ Links ]

4. Akaike, H. (1987). Factor analysis and AIC. Psychometrika, 52, 317-332. Doi: 10.1007/BF02294359.         [ Links ]

5. Arbuckle, J. L. (2003). Amos 5.0 update to the Amos user's guide. Marketing Department, SPSS Incorporated.         [ Links ]

6. Aryee, S., Budhwar, P. S., & Chen, Z. X. (2002). Trust as a mediator of the relationship between organizational justice and work outcomes: Test of a social exchange model. Journal of Organisational Behavior, 23, 267-285. Doi: 10.1002/job.138.         [ Links ]

7. Bakker, A.B., Van Emmerik, H., & Euwema, M.C. (2006). Crossover of burnout and engagement in work teams. Work and Occupations, 33, 464-489. Doi: 10.1177/0730888406291310.         [ Links ]

8. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical consideration. Journal of Personality and Social Psychology, 51, 1173-1182. Doi: 10.1037/0022-3514.51.6.1173.         [ Links ]

9. Bass, B. M. (1985). Leadership and Performance beyond Expectations. New York: Free Press.         [ Links ]

10. Bass, B. M., & Avolio, B. J. (1990). Transformational Leadership Development: Manualfor the Multifactor Leadership Questionnaire. Palo Alto, CA: Consulting Psychologists Press.         [ Links ]

11. Bass, B. M., & Avolio, B. J. (1994). Improving Organisational Effectiveness through Transformational Leadership. Sage, Thousand Oaks, CA.         [ Links ]

12. Bass, B. M. (1998). Transformational Leadership: Industry, Military, and Educational Impact. Mahwah, NJ: Erlbaum.         [ Links ]

13. Bass, B. M. (1999). Two decades of research and development in transformational leadership. European Journal of Work and Organizational Psychology, 8, 9-32. Doi: 10.1080/135943299398410.         [ Links ]

14. Bass, B. M., Avolio, B. J., Jung, D. I., & Berson, Y. (2003). Predicting unit performance by assessing transformational and transactional leadership. Journal of Applied Psychology, 88, 207-218. Doi: 10.1037/0021-9010.88.2.207.         [ Links ]

15. Bentler, P.M. (1990). Comparative fit indices in structural models. Psychological Bulletin, 107, 238-246. Doi: 10.1037/0033-2909.107.2.238.         [ Links ]

16. Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88, 588-606. Doi: 10.1037/0033-2909.88.3.588.         [ Links ]

17. Bliese, P. D. (2000). Within-group agreement, non-independence, and reliability. Implications for data aggregation and analysis. In K. J. Klein, & S. W. J. Kozlowski (Eds.), Multilevel theory, Research, and, Methods in Organization (pp. 349-381). San Francisco: Jossey-Bass.         [ Links ]

18. Bollen, K.A. (1989). A new incremental fit index for general structural equations models. Sociological Methods and Research, 17, 303-316. Doi: 10.1177/0049124189017003004.         [ Links ]

19. Borman, W. C., & Motowidlo, S. M. (1993). Expanding the criterion domain to include elements of contextual performance. In N. Schmitt, & W. C. Borman (Eds.), Personnel Selection in Organizations (pp. 71-98). San Francisco: Jossey-Bass.         [ Links ]

20. Borman, W. C., & Motowidlo, S. J. (1997). Task performance and contextual performance: The meaning for personnel selection research. Human Performance, 10, 99-109. Doi: 10.1207/s15327043hup1002_3.         [ Links ]

21. Brown, S., Gray, D., McHardy, J., & Taylor, K. (2015). Employee trust and workplace performance. Journal of Economic Behavior & Organization, 116, 361-378. Doi: 10.1016/j.jebo.2015.05.001.         [ Links ]

22. Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & J. S. Long (Eds.), Testing Structural Equation Models (pp. 136-162). Newbury Park, CA: Sage.         [ Links ]

23. Buchan, J. (1999). Still attractive after all these years? Magnet hospitals in a changing health care environment. Journal of Advanced Nursing, 30, 100-108. Doi: 10.1046/j.1365-2648.1999.01054.x.         [ Links ]

24. Burke, N.J., Finkelstein, L.M., & Dusig, M.S. (1999). On average deviation indices for estimating interrater agreement. Organizational Research Methods, 2, 49-68. Doi: 10.1177/109442819921004.         [ Links ]

25. Campbell, J. P. (1990). Modeling the performance prediction problem in industrial and organisational psychology. Consulting Psychologists Press, Palo Alto, CA.         [ Links ]

26. Chen, G., Mathieu, J. E., & Bliese, P. D. (2004). A framework for conducting multilevel construct validation. In F. J. Yammarino, & F. Dansereau (Eds.), Research in multilevel issues: Vol. 3. Multilevel issues in organizational behavior and processes (pp. 273-303). Amsterdam: Elsevier.         [ Links ]

27. Chun, J., Cho, K., & Sosik, J. (2015). A multilevel study of group-focused and individual-focused transformational leadership, social exchange relationships, and performance in teams. Journal of Organizational Behavior, 37, 374-396. Doi: 10.1002/job.2048.         [ Links ]

28. Cohen, A., Nahum-Shani, I., & Doveh, E. (2010). Further insight and additional inference methods for polynomial regression applied to the analysis of congruence. Multivariate Behavioral Research, 45, 828-852. Doi: 10.1080/00273171.2010.519272.         [ Links ]

29. Colquitt, J., LePine, J. A., & Wesson, M. J. (2010). Organizational Behavior: Essentials for Improving Performance and Commitment. Dongbei University of Finance & Economics Press.         [ Links ]

30. Costa, A. C. (2003). Work team trust and effectiveness. Personnel Review, 32, 605-622. Doi: 10.1108/00483480310488360.         [ Links ]

31. Costa, A. C., & Anderson, N. (2011). Measuring trust in teams: Development and validation of a multifaceted measure of formative and reflective indicators of team trust. European Journal of Work and Organizational Psychology, 20, 119-154. Doi: 10.1080/13594320903272083.         [ Links ]

32. Cruz-Ortiz, V., Salanova, M., & Martínez, I. M. (2013). Liderazgo transformacional y desempeño grupal: unidos por el engagement grupal. Revista de Psicología Social, 28, 183-196. Doi: 10.1174/021347413806196762.         [ Links ]

33. Dirks, K. T., & Ferrin, D. L. (2002). Trust in leadership: Meta-analytic findings and implications for research and practice. Journal of Applied Psychology, 87, 611. Doi: 10.1037/0021-9010.87.4.611.         [ Links ]

34. Fulmer, C., & Gelfand, M. (2012). At What Level (and in Whom) We Trust: Trust Across Multiple Organizational Levels. Journal of Management, 38, 1167-1230. Doi: 10.1177/0149206312439327.         [ Links ]

35. Gang Wang, Oh, I., Courtright, S., & Colbert, A. (2011). Transformational Leadership and Performance Across Criteria and Levels: A Meta-Analytic Review of 25 Years of Research. Group & Organization Management, 36, 223-270. Doi: 10.1177/1059601111401017.         [ Links ]

36. García-Morales, V., Lloréns-Montes, F., & Verdú-Jover, A. (2008). The Effects of Transformational Leadership on Organizational Performance through Knowledge and Innovation. British Journal of Management, 19, 299-319. Doi: 10.1111/j.1467-8551.2007.00547.x.         [ Links ]

37. George, J. M. (1990). Personality, affect, and behaviors in groups. Journal of Applied Psychology, 75, 107-116. Doi: 10.1037/0021-9010.75.2.107.         [ Links ]

38. Glick, W.H. (1985). Conceptualizing and measuring organizational and psychological climate: Pitfalls in multilevel research. Academy of Management Renew, 10, 601-616. Doi: 10.5465/AMR.1985.4279045.         [ Links ]

39. Goodman, S. A., & Svyantek, D. J. (1999). Person-organization fit and contextual performance: Do shared values matter. Journal of Vocational Behavior, 55, 254-275. Doi: 10.1006/jvbe.1998.1682.         [ Links ]

40. Hernández, C.I., Llorens, S. & Rodríguez, A.M. (2014). Empleados saludables y calidad de servicio en el sector sanitario. Anales de Psicología, 30, 247-258. Doi: 10.6018/analesps.30.1.143631.         [ Links ]

41. Hoyle, R. H. (1995). The structural equation modeling approach: Basic concepts and fundamental issues. En R. H. Hoyle (Ed.), Structural Equation Modeling Concepts, Issues and applications (pp. 1-15). Thousand Oaks, Ca: Sage.         [ Links ]

42. James, L. R., Demaree, R. G., & Wolf, G. (1993). rwg_j: An assessment of Within-Group Interrater Agreement. Journal of Applied Psychology, 78, 306-309. Doi: 10.1037/0021-9010.78.2.306.         [ Links ]

43. Johnson, E. C. (2009). A multi-level investigation of overall job performance ratings (Doctoral dissertation). North Carolina State University, United States. ProQuest Dissertations and Theses. Retrieved from http://search.proquest.com/docview/304965718?accountid=14744. (304965718).         [ Links ]

44. Kiffin-Petersen, S., & Cordery, J. (2003). Trust, individualism and job characteristics as predictors of employee preference for teamwork. International Journal of Human Resource Management, 14, 93-116. Doi: 10.1080/09585190210158538.         [ Links ]

45. Kline, R. B. (1998). Principles and Practice of Structural Equation Modeling. New York: Guilford Press.         [ Links ]

46. Kramer, R. M. (1999). Trust and distrust in organizations: Emerging perspectives, enduring questions. Annual Review of Psychology, 50, 569-598. Doi: 10.1146/annurev.psych.50.1.569.         [ Links ]

47. Lee, D., Stajkovic, A. D., & Cho, B. (2011). Interpersonal Trust and Emotion as Antecedents of Cooperation: Evidence From Korea1. Journal of Applied Social Psychology, 41, 1603-1631. Doi: 0.1111/j.1559-1816.2011.00776.x.         [ Links ]

48. Llorens, S., Salanova, M., & Losilla, J. (2009). Liderazgo transformacional y capital psicológico positivo. Un estudio de caso en una empresa de construcción. Directivos de Construcción, 220, 48-56.         [ Links ]

49. Mayer, R. C., & Gavin, M. B. (2005). Trust in management and performance: who minds the shop while the employees watch the boss? Academy of Management Journal, 48, 874-888. Doi: 10.5465/AMJ.2005.18803928.         [ Links ]

50. Marsh, H. W., Balla, J. R., & Hau, K. T. (1996). An evaluation of Incremental Fit Indices: A clarification of mathematical and empirical properties. En G. A. Marcoulides & R. E. Schumacker (Eds.), Advanced structural equation modeling: Issues and techniques (pp. 315-353). Mahwah, NJ: Lawrence Erl-baum Associates.         [ Links ]

51. McAllister, D. J. (1995). Affect-and cognition-based trust as foundations for interpersonal cooperation in organizations. Academy of Management Journal, 38, 24-59. Doi: 10.2307/256727.         [ Links ]

52. McCarthy, J. P. (1992). Focus from the Start: An orientation program designed to value employees from day one. HR Magazine, 37, 77-83.         [ Links ]

53. McClure, M., M., Poulin, M., Sovie, M., & Wandelt, M. (1983). Magnet Hospitals: Attraction and Retention of Professional Nurses. American Academy of Nursing. Task Force on Nursing Practice in Hospitals. Kansas City, MO: American Nurses Association.         [ Links ]

54. Mishra, A. K. (1995). Organizational Responses to Crisis. Trust in organizations: Frontiers of theory and research, 261.         [ Links ]

55. Nielsen, K. (2014). Leadership and Climate in a Psychologically Healthy Workplace. In Day, Kelloway, & Hurrell (Eds.), Workplace Well-Being Building Positive & Psychologically Healthy Workplaces (pp. 226-244). Wiley Publications.         [ Links ]

56. Palanski, M. E., Kahai, S. S., & Yammarino, F. J. (2011). Team virtues and performance: An examination of transparency, behavioral integrity, and trust. Journal of Business Ethics, 99, 201-216. Doi: 10.1007/s10551-010-0650-7.         [ Links ]

57. Podsakoff, P. M., MacKenzie, S. B., Moorman, R. H., & Fetter, R. (1990). Transformational leader behaviors and their effects on followers' trust in leader, satisfaction, and organizational citizenship behaviors. The Leadership Quarterly, 1, 107-142. Doi: 10.1016/1048-9843(90)90009-7.         [ Links ]

58. Pourbarkhordari, A., Zhou, E. H., & Pourkarimi, J. (2016). How Individual-focused Transformational Leadership Enhances Its Influence on Job Performance through Employee Work Engagement. International Journal of Business and Management, 11, 249. Doi: 10.5539/ijbm.v11n2p249.         [ Links ]

59. Rafferty, A. E., & Griffin, M. A. (2004). Dimensions of transformational leadership: Conceptual and empirical extensions. The Leadership Quarterly, 15, 329-354. Doi: 10.1016/j.leaqua.2004.02.009        [ Links ]

60. Rao, A. & Kareem Abdul, W. (2015). Impact of transformational leadership on team performance: an empirical study in UAE. Measuring Business Excellence, 19, 30-56. Doi: 10.1108/mbe-07-2014-0022.         [ Links ]

61. Rotundo, M., & Sackett, P. R. (2002). The relative importance of task, citizenship, and counterproductive performance to global ratings of job performance: A policy capturing approach. Journal of Applied Psychology, 87, 66-80. Doi: 10.1037/0021-9010.87.1.66.         [ Links ]

62. Salanova, M. (2008). Organizaciones saludables: una aproximación desde la psicología positiva. En Vázquez y Hervás (Eds.), Psicología Positiva Aplicada (pp. 403-458). Bilbao, España: Desclée de Brouwer.         [ Links ]

63. Salanova, M., Acosta, H., Llorens, S., & Le Blanc, P. M. (2017). Learn to Trust your Company! A Multilevel-Multireferent Model to explain Organizational and Team Performance Manuscript submitted to publish.         [ Links ]

64. Salanova, M., Llorens, S., Acosta, H., & Torrente, P. (2013). Positive interventions in organizations. Terapia psicológica, 31, 101-113. Doi: 10.4067/S0718-48082013000100010.         [ Links ]

65. Salanova, M., Llorens, S., Cifre, E., & Martínez, I. M. (2012). We need a hero! Toward a validation of the healthy and resilient organization (HERO) model. Group & Organization Management, 37, 785-822. Doi: 10.1177/1059601112470405.         [ Links ]

66. Salanova, M., Martínez, I. M., & Llorens, S. (2014). Una mirada más "positiva" a la salud ocupacional dese la Psicología Organizacional Positiva en tiempos de crisis: Aportaciones desde el equipo de investigación WoNT. Papeles del Psicólogo, 35, 22-30.         [ Links ]

67. Salanova, M., Rodríguez-Sánchez, A. M., Del Líbano, M., & Ventura, M. (2012). Organizaciones Sanitarias Saludables y Resilientes: Marco teórico y conceptual. En S. Llorens y M. Salanova (Dirs.) Organizaciones sanitarias saludables y resilientes: Un estudio de caso en hospitales de Castellón y provincia. Castellón: Fundación Dávalos-Fletcher.         [ Links ]

68. Sobel, M.E. (1988). Direct and indirect effects in linear structural equation models. In J.S. Long (Ed.), Common problems/proper solutions (pp. 46-64). Beverly Hills, CA: Sage.         [ Links ]

69. Tan, H., & Lim, A. (2009). Trust in co-workers and trust in organizations. The Journal of Psychology, 143, 45-66. Doi: 10.3200/JRLP.143.1.45-66.         [ Links ]

70. Torrente, P., Salanova, M., Llorens, S., & Schaufeli, W. B. (2012). Teams make it work: How team work engagement mediates between social resources and performance in teams. Psicothema, 24, 106-112.         [ Links ]

71. Torrente, P., Salanova, M., Llorens, S., & Schaufeli, W. B. (2015). Committing the engaged: A multilevel mediation model of team work engagement and organizational affective commitment. Manuscrito sometido para su publicación.         [ Links ]

72. Walumbwa, F. O., Avolio, B. J., & Zhu, W. (2008). How transformational leadership weaves its influence on individual job performance: The role of identification and efficacy beliefs. Personnel Psychology, 61, 793-825. Doi: 10.1111/j.1744-6570.2008.00131.x.         [ Links ]

73. Yukl, G. (2006). Leadership in organizations (6th Ed.). Upper Saddle River, NJ: Prentice Hall.         [ Links ]

74. Zawawi, A. A., & Nasurdin, A. M. (2016). Championing a team: reviewing the role of team composition, context, and trust on nursing team performance. Australian Journal of Sustainable Business and Society, 2.         [ Links ]

75. Zhu, W., Newman, A., Miao, Q., & Hooke, A. (2013). Revisiting the mediating role of trust in transformational leadership effects: Do different types of trust make a difference? The Leadership Quarterly, 24, 94-105. Doi: 10.1016/j.leaqua.2012.08.004.         [ Links ]

Correspondence:
Juana Olvera
Departamento de Psicología Social,
Universitat Jaume I, ]]> Av. Sos Baynat, s/n. 12071.
Castellón (Spain).
E-mail: juana.olvera@uji.es

Article received: 15-09-2015;
revised: 07-03-2016;
accepted: 06-06-2016