Correlation of Essential Servant Leadership Behaviors, Perceived Organizational Support, and Employee Well-being

This current research follows up on Greenleaf’s oft-quoted best test of servant leadership that calls for employees to be better off financially, emotionally, physically, psychologically, etc. because of the time spent with the servant leader. While oft-quoted, little empirical work exists to see if this is true. In this study, 170 participants provided their perception of their supervisors’ level of servant leadership, their perception of the organization’s support, and their self-report of their general well-being. Gender and age bracket information described the participants, and there were no significant differences between gender or age brackets for participants’ perception of their supervisors’ servant leadership. The analysis showed that there was a moderate positive correlation between servant leadership, perceived organizational support, and general well-being. A modification of an existing general well-being instrument provided a new eight-item general well-being scale with a Cronbach’s alpha of 0.956.


Introduction
said that a test of servant leadership includes the evidence that the employee is/was better off because of his/her working in the organization. Greenleaf stated: Do those served grow as persons? Do they, while being served, become healthier, wiser, freer, more autonomous, more likely themselves to become servants? And, what is the effect on the least privileged in society? Will they benefit or at least not be further deprived? .
To further the literature on servant leaders' impact on employees' well-being, the research question for this study is: What correlation exists among employees' perception of their supervisors' servant leadership behaviors, the employees' perception of organizational support, and employees self-report of general well-being. The following sections on servant leadership, perceived organizational support, and employees' general well-being precede the hypotheses tested in this study. Bass (2000) said that servant leadership focused on the employees, while transformational leadership focused on the organization. Bass' comment aligns well with Greenleaf's (1977) test of servant leadership. Patterson (2003), in her model of servant leadership, commented about the importance of the leader's moral love for the employees. Also, an alignment with Greenleaf's test. Page and Wong (2000) included 'caring for others' in their servant leadership model. Barbuto and Wheeler (2006) included 'emotional healing' in their servant leadership model. All of these dimensions of 4 servant leadership point to a relationship between the leader's level of servant leadership and the employees' general well-being. Page and Wong (2000), Sendjaya and Sarros (2002), van Dierendork and Nuijten (2011), Dennis and Bocarnea (2005), Barbuto and Wheeler (2006), Hale and Fields (2007), along with Liden, Wayne, Zhao, and Henderson (2008) include multiple dimensions, which, according to Winston and Fields (2015) obfuscate the ability to measure that which is essentially servant leadership and not dimensions of other leadership styles.

Essential Servant Leadership Behaviors Instrument
To clarify the measurement of servant leadership Winston and Fields (2015) created an item-pool of servant leadership behaviors from the multi-dimensional servant leadership instruments created in the studies cited above, asked a panel of experts to review the list and vote on keeping or removing each behavior, sent the resultant list of 22 items to 443 working adults and asked them to evaluate their supervisors using the 22 items. After principal component analysis and scale optimization, Winston and Fields produced a 10-item single-scale measure of essential servant leadership behaviors (ESLB) that had a Cronbach's alpha of 0.96. Examples of the items in the scale include: 1. Practices what he/she preaches.
2. Serves people without regard to their nationality, gender, or race.
3. Sees serving as a mission of responsibility to others.

Genuinely interested in employees as people.
Subsequent research studies showed Cronbach alphas ranging from 0.88 to 0.96 (Cancino, 2019;Hargadon, 2018;Hirschy;Phillips, 2018). Participants in this study used the ESLB to provide their perception of their supervisor's servant leadership behaviors. Eisenberger, Huntington, Hutchison, and Sowa (1986), in their effort to understand the antecedents to organizational commitment, developed an instrument to measure employees' perception of organizational support (POS). According to Eisenberger et al., perceived organizational support was the "breadth and consistency of the employee's beliefs concerning the organization's commitment to him or her, and the effects of such perceived organizational support. and of the employee's exchange ideology on absenteeism" (p. 501). Eisenberger et al. believed that the organization's level of concern for the employee's well-being positively correlated with the employee's actual well-being. Eisenberger et al.'s level of concern seems to align with Greenleaf's (1977) test of servant leadership. Eisenberger's focus on the organization is replaced in this current study with the leader's servant leadership behaviors with the underlying premise that the leader's values and behaviors positively impact the organization's culture, which is supported by Hambrick and Mason's (1984) upper echelon theory.

Perceived Organizational Support
Since a possible moderation may exist, this research study includes a regression analysis testing for a moderating effect by POS on the ESLB-General Well-being relationship. Eisenberger, Stinglhamber, Vandenberg, Sucharski, and Rhoades (2002) recommended an eight-item short form of Eisenberger, et al.'s (1986)

General Employee Well-being
Page and Vella-Brodrick (2009) reported that employee well-being consisted of "high-levels of positive impact, low levels of negative impact, and cognitive evaluation of one's satisfaction with their life as a whole" (p. 443). Also, according to Page and Vella-Brodrick, someone with positive well-being is said to be in a state of "positive psychological functioning" (p. 443). Bakker and Demerouti (2007) posited that resources provided by the organization (organizational-support) that help employees do their assigned jobs and tasks might contribute to employees' well-being. These definitions help augment Greenleaf's (1977) test of whether employees are better off because of the servant leader and the organization. Taylor et al. (2003) reviewed the General Employee Well-being (GWB) scale developed by Dupuy (1978) and noted concerns about the lack of empirical testing of the dimensions, and inconsistent results from factor analysis. In addition to Taylor et al.'s concerns, 14 of the 18 items were scored on a six-point rating scale, four items were scored on a 0-10 rating scale, and eight of the 18 items were reversed scored. Noting van Sonderen, Sanderman, and Coyne (2013) concern about using reversed scored items and considering Wen-Chung Wang, Chen, and Jin's (2015) admonition about not using mixed response ratings, I created a modified version of the 18-item General Employee Well-being instrument and test the instrument using principal component analysis and scale reliability analysis. Table 1 shows the original 18 items from the GWB instrument and the revised items for the Modified General Employee Well-being Instrument (Modified GWB). Notes: a = six-point rating scale; b = 11-point rating scale; c = reversed scored Note: response scale for the Modified General Well-being instrument: Never 1 2 3 4 5 6 7 8 9 10 Always.

Materials and Methods
The

Participant Demographics
While the required minimum sample size was 129, 170 participants completed the survey. Of the 170 participants, 113 were female, 55 were male, one indicated 'other', and one preferred to not respond to the question. The participants reported that of the 170: (a) 46 were 18-29 years old; 50 were 30-39 years old; 34 were 40-49 years old; 24 were 50-59 years old; 12 were 60-69 years old; 3 were 70+ years old, and 1 person declined to report his/her age.

Modified General Employee Well-being Instrument Analysis
Factor analysis can make use of a rotation, and the selection of a rotation depends on the level of correlation between the items. A Pearson-r correlation of the 18 items showed moderate to high correlation implies that the data was oblique and that a direct oblimin rotation would be appropriate (Brown, 2009). An initial principal component analysis resulted in a Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) of .951 and Bartlett's Test of Sphericity approximate C 2 of 2643.581, which was significant at p=.000, implying that the data might benefit from factor analysis. All 18 items loaded on two factors with eigenvalues greater than one, but the second factor consisted of one item that cross-loaded on factor 1. 1. I felt calm, rather than anxious, during the past two months.
2. I have felt physically healthy during the past two months.
3. I felt emotionally stable for the past two months.
4. I felt energized for the past two months.
5. I felt good about my health for the past two months.
6. I felt relaxed for the past two months.
7. I felt energetic for the past two months.
8. I felt cheerful for the past two months. Table 2 shows the means, standard deviations, Since the Skewness and Kurtosis scores are between -3.00 and +3.00 the data are parametric (Kim, 2013), thus for correlation tests, the Pearson-r test is appropriate. Table 3 shows the Pearson-r correlations for the three scales. Note: **p=.000

Moderation Analysis
ANOVA analysis of the Modified GWB scale across the two gender groups and the five Age groups did not show any significance, however, in the moderation analysis, age, treated as an ordinal variable, showed significance.

Discussion
The finding that there is no significant difference between men and women's perception of supervisor's servant leadership aligns with Barbuto and Gifford (2010) but contrasts with Duff's (2013) assertion: Social role theory, suggesting that women are socialized towards norms of communal support in a manner in which men are not, lends theoretical support to suggest that as with transformational leadership, female leaders should be more likely to employ a servant leadership style than male leaders. (p. 214) The positive correlation between servant leadership perceived organizational support, and general well-being supports the notion that servant leadership may contribute to employees' well-being. More qualitative and quantitative research should occur to understand if, and how servant leaders affect employees' well-being. The lack of significance in the test for moderation, in the presence of a moderate correlation between perceived organizational support and employees' well-being, implies more research should occur to determine if servant leaders impact organizational support or if organizational support is a separate, unrelated concept to servant leadership.
An additional benefit of this current study is the development of the eight-item scale to measure employees' general well-being. The new Modified General Well-being scale should be helpful for researchers who wish to measure employees' self-report of well-being using a parsimonious reliable instrument.