Link Search Menu Expand Document
Publications
  1. METHODS

METHODS

Researchers used a descriptive research design to guide this study. The University of Florida Institutional Review Board (IRB) approved this study in December 2019. The target for data collection was a census of currently employed Florida Extension Agents. At the time of survey, the population consisted of 351 Extension agents.

The Utrecht Work Engagement Scale (UWES) instrument, developed by Schaufeli and Bakker (2004), was used to collect data. Since the introduction of the UWES, several studies have tested the validity of the relationship between burnout and workaholism, identified causes and consequences of engagement, and investigated how work engagement might impact an employee’s health (Salanova et al., 2001; Schaufeli et al., 2002; Montgomery et al., 2003).

The lead researcher created an online version of the instrument managed through Qualtrics. The instrument included three sections. Section A asked participants to identify details of their employment. Participants identified Initiative teams with which they primarily worked (e.g., 4-H Youth Development, Agriculture and Natural Resources, etc.), and selected whether they were a county Extension agent, CED, or a regional or state specialized agent.

Section B consisted of items from the UWES instrument. The UWES instrument presents 17 statements relative to how employees might feel about their work (e.g., At my work I feel bursting with energy). These statements corresponded with one of the three facets of engagement (vigor, dedication, and absorption) (Schaufeli & Bakker, 2004). Participants were asked to rate each statement on a 7-point frequency scale (1 = Never, 2 = Almost never/A few times a year or less, 3 = Rarely/ Once a month or less, 4 = Sometimes/A few times a month, 5 = Often/Once a week, 6 = Very Often/A few times a week, and 7 = Always/Every day) (Schaufeli & Bakker, 2004). Scores were interpreted as follows: 1.00 - 1.49 = Never, 1.50 - 2.49 = Almost Never, 2.5 - 3.49 = Rarely, 3.49 - 4.45 = Sometimes, 4.5 - 5.49 = Often, 5.50 - 6.49 = Very Often, and 6.50 - 7.00 = Always. Section C was one open-ended question which asked, “What do you like most about your job?” Section D addressed demographic questions of gender and age.

The UWES has undergone numerous tests for validity since its creation in 1999, and these tests support the assertion that work engagement is negatively associated with burnout (Schaufeli & Bakker, 2004). The scale has additionally been found to be highly internally consistent (Schaufeli & Bakker, 2004). Cronbach’s alpha values for each measure are equal to or exceed .70, with most measurements ranging between .80 and .90 (Salanova et al., 2001). Likewise, the instrument is relatively stable with coefficients for vigor, dedication, and absorption at .30, .36, and .46 respectively (Schaufeli & Bakker, 2004). In this study using the UWES among Florida Extension agents, Cronbach’s alpha values ranged from .91 to .92. According to George and Mallery (2003), a Cronbach’s alpha value greater than .90 indicates excellent internal reliability.

The distribution of the survey was guided by Dillman et al.’s (2014) Tailored Design Method for web-based surveys. Data were collected in January 2020. A final response rate of 65% (n = 229) was achieved. However, when deleting incomplete and unusable responses, a final usable response rate of 62% (n = 216) was achieved.

The largest group of respondents reported to the 4-H youth Initiative team (n = 71, 33%), followed by the Agriculture and Natural Resources Initiative team (n = 60, 27.78%), and the Individual and Family Resources Initiative team (n = 34, 14.74%). Most respondents reported working as a county Extension agent (n = 166, 76.2%). The remaining 23.8% of the population was comprised of CEDs, regional specialized agents, and state specialized agents. Most respondents reported as female (n = 141, 64.7%) and tended to be 30 - 39 years (n = 51, 23.4%) or 40 - 49 years (n = 53, 24.3%) old. Those reporting any other age comprised 39.1% (n = 85) of the population. Most respondents were white (81.0%, n = 171). Asian was the second largest respondent group comprising 9.0% (n = 19) of the population. Those reporting Hispanic or Latino ethnicity comprised only 6.0% (n = 13) of the population.

Lindner et al. (2001) suggested that non-response issues be addressed any time less than an 85% response rate is achieved. Because this study achieved a 62% response rate, early and late respondents were compared using two-group independent t-tests. Early respondents were defined as the individuals who responded after the initial invitation and late respondents were defined as individuals who responded after the second reminder. Ary et al. (2006) and Miller and Smith (1983) stated that research has shown similarities usually exist between late respondents and non-respondents. The variables of interest used to compare early and late respondents were the three constructs of work engagement: (a) vigor, (b) dedication, and (c) absorption. No significant differences between early and late respondents were found for agents’ levels of (a) vigor, t(155) =.55, p > .05; (b) dedication, t(155) = 1.00, p > .05; or (c) absorption, t(155) = -.78, p > .05. The lack of significant differences between early and late respondents suggests the results can be generalized to the target population (Lindner et al., 2001).

Researchers used descriptive statistics to describe the current levels of work engagement possessed by Florida Extension agents as determined by the UWES. They also calculated frequencies, percentages, means, and standard deviations for overall self-perceived employee engagement as well as each of the three subscales of vigor, dedication, and absorption.

Table of Contents