Correlation coefficients between the variables were calculated to verify the hypotheses. In turn, canonical analysis was carried out to answer the research question concerning the potential relationships between all measures of trust and multidimensional employee well-being. The statistical analyses were performed using the SPSS statistical package version 27, JASP 0.14.1, and Statistica 12.0.
3.1. Preliminary analysis
The first step of the analysis involved calculating the correlations coefficients between the studied variables.
Table 1 contains the descriptive statistics, correlations and intercorrelations between the studied variables.
According to
Table 1, the components of the following variables: trust in supervisor, trust in co-workers and organizational transparency and kindness, as well as organizational certainty and ethics correlate positively, averagely and highly, with the components of the employee well-being variable.
Both indicators of trust in co-workers are positively related to all scales of well-being in the workplace. The strongest correlation (r > 0.6) appeared in the context of the Positive interpersonal relationships dimension. This result was expected due to the definitions of both scales. The remaining scales of the Eudemonic Well-Being in the Workplace Questionnaire (EWWQ) are also positively related to trust in co-workers. This makes it possible to confirm Hypothesis 1.
Similarly, both indicators of trust in the direct supervisor are positively related to all scales of well-being in the workplace. In this case, the strongest correlation (r > 0.7) appeared in the context of the Positive organization dimension. The remaining scales of the EWWQ Questionnaire are also positively related to trust in co-workers. This makes it possible to confirm Hypothesis 2.
When analyzing the correlation coefficients between organizational well-being and impersonal trust, two clear patterns can be noticed. Firstly, all dimensions of well-being in the workplace correlate more strongly with the Transparency and organizational kindness scale (even r > 0.8) than with the Organizational certainty and ethics scale. Secondly, the highest correlations of both scales of impersonal trust in the organization were recorded for the Positive organization dimension, a result in line with the scale’s definition. This is because all scales of the EWWQ questionnaire are positively related to impersonal trust in the organization. This makes it possible to confirm Hypothesis 3.
3.2. Canonical analysis
In order to examine the relationship between all dimensions of trust simultaneously with all dimensions of employee well-being within the organization, canonical analysis [
38] was applied, performed using the Statistica 12.0 statistical package. Canonical analysis makes it possible to determine the extent of the simultaneous influence of a set of independent variables, taking into account the relationships between the variables in that set, on a set of dependent variables, taking into account the relationships between the variables also in that latter set. Thus, canonical analysis identifies the multiple determination structure of the individual variables, examined together, pointing to the most significant variables. The dimensions of trust (left set) were treated as independent variables (predictors) for the components of well-being in the workplace (right set).
Table 2 contains the results obtained from the canonical analysis. The total redundancy for the set of dependent variables (dimensions of well-being in the workplace) was 62.26%. This means that 62.26% of variance in terms of well-being in the workplace can be explained with the set of predictors – trust dimensions taken into account (by all canonical variables).
Of the pairs of canonical variables obtained, the first three ((U1,V1), (U2,V2), (U3,V3)) were found to be statistically significant for the linear relationship between the sets of trust and employee well-being dimensions. A factor loading structure was used to determine the size of the contribution of the individual dimensions of trust and well-being to the canonical variables forming the given pair. These loadings are a measure of the correlation of a given dimension with a given canonical variable (“factor”). It was assumed that loadings greater than 0.4 would be indicative of a significant relationship between the given dimension and the canonical variable (factor).
According to the data in
Table 2, on the independent variables (trust dimensions) side, the first pair (U1,V1) includes the canonical variable U1. The following dimensions demonstrate the highest positive correlations with this canonical variable: organizational transparency and kindness (0.987), trust in supervisor: benevolence and integrity (0.814), trust in supervisor: competence and skills (0.812), organizational certainty and ethics (0.689), affect-based trust in co-workers (0.516), and cognition-based trust in co-workers (0.498). It was decided to call this factor, i.e. canonical variable U1, “full trust in organization”.
The pair discussed is formed together with full trust in organization by canonical variable V1, which can be described as “full organizational well-being”. In fact, the positive organization dimension correlates most strongly with this variable. The other dimensions of well-being in the workplace also display strong correlations.
The canonical pair discussed (U1,V1) explains 54.7 per cent of variance in the set of predictors – trust dimensions and 68.9 per cent of variance in the set of workplace well-being dimensions. With regard to the respective canonical pair of full trust in organization – full organizational well-being, the set of predictors, i.e. trust dimensions, makes it possible to explain over 57% of variance in the set of well-being dimensions. The relationship in this canonical pair, i.e. between the linear function of trust dimensions and the linear function of well-being dimensions, is very high and amounts to 0.913.
In summary, it can be concluded that employees experiencing well-being in the organization also have trust in organization, in supervisor, and in co-workers, but trust in organization and in supervisor come to the fore here, being the most significant.
In the second pair (U2,V2) of canonical variables, the relationship between the linear function of trust dimensions and the linear function of well-being dimensions is lower and amounts to 0.563. Canonical variable U2 could be called “trust in co-workers”, as the dimensions that have the highest positive correlations with it are cognition-based trust in co-workers (0.792) and affect-based trust in co-workers (0.729), with a simultaneous weak negative correlation with certainty and ethics (-0.289), which means that trust is put in people on the same level within the organization.
On the other hand, the second canonical variable in this pair, V2, is most strongly correlated with the dimension of positive interpersonal relationships (0.548), indicating good relationships with co-workers, and the dimension of contribution to the organization, i.e. the employee’s belief that they play an important role in the organization as a member thereof (0.425), which makes it possible to call this factor “satisfaction with relationships with co-workers and with the contribution to the organization”. The fit dimension also correlates positively and weakly (0.339) with this factor, indicating that satisfaction with the relationship with co-workers may be related to the sense of fit with the organization.
The canonical pair discussed (U2,V2) explains 20.9% of variance in the set of predictors – trust dimensions, and 14.9% of variance in the set of well-being dimensions. With regard to this canonical pair, using the set of predictors – trust dimensions only makes it possible to explain 4.7% of variance in the set of well-being components.
In the third pair of canonical variables (U3,V3), the relationship between the linear function of trust dimensions and the linear function of well-being dimensions is very low and amounts to 0.097. Canonical variable U3 could be called “trust in the supervisor’s competence and skills”, as it has the highest positive and average correlation with the dimension of trust in supervisor based on their competence and skills (0.492), with a simultaneous weak positive correlation with organizational certainty and ethics (0.315), which means that trust is put in people on a higher level within the organization.
The second canonical variable in this pair, V3, is most strongly and negatively correlated with the contribution to the organization dimension (-0.543), indicating the employee’s belief that they play an important role in the organization as a member thereof. Said variable correlates at the same time weakly and negatively with organizational fit (-0.340), making it possible to call this factor “sense of weak contribution to the organization”.
The canonical pair discussed (U3,V3) explains 7.9% of variance in the set of predictors – trust dimensions, and 11.7% of variance in the set of well-being dimensions. With regard to the respective canonical pair, using a set of predictors – trust dimensions, it is possible to explain very little, i.e. less than 1%, of variance in the set of well-being components. This result means that, in the organizations covered by the research, higher trust may be found in the supervisor’s competence and skills, with the employee’s weaker sense of having competencies or qualities important for the organization.
Taking into account the analysis of all three canonical pairs, the U1–V1 pair was found to best explain the relationships between the groups of variables from the left and right set. On the basis of this canonical pair, the following answer to the research question can be assumed: trust in all its dimensions, including trust in co-workers, in supervisor and impersonal trust in the organization as an institution, promotes employee well-being, also in all its dimensions.