Exploring Organizational Factors Related to Extension Employee Burnout

Introduction

Burnout, an emotional strain that can occur within an occupational setting, has been defined as "a psychological and multidimensional syndrome of emotional exhaustion, depersonalization, and reduced personal accomplishment that can occur among individuals who work with other people in some capacity" (Sears, Urizar, & Evans, 2000, p. 57). Research has shown burnout is highly correlated with the organizational attitudes of turnover intentions, job satisfaction, and organizational commitment (Alarcon, 2011). At the individual level, manifestations of burnout include professionals being more often absent or late for work, experiencing deterioration in their quality of work, becoming less flexible and responsive, and either dwelling on or pursuing plans to leave the profession (Schabracq, Winnubst, & Cooper, 2003).

A pattern of staff turnover can emerge when there is failure to address issues of burnout (Ensle, 2005). In Extension, the attrition of Extension agents often results in losses of historical and programmatic knowledge, experience, and relationships that have been built up over time (Bradley, Driscoll, & Bardon, 2012). These losses create a vacuum that a new employee must attempt to fill, adding to the successor's stress level. Employee attrition may also cause substantial financial and time strains on an organization (Kutelik, 2000), resulting in client service disruptions, interruptions in programming, increased resources needed for new agent recruitment and training, and increased workloads for remaining staff (Arnold, 2007; Clark, 1992; Ensle, 2005).

Theoretical Framework & Review of Literature

In 1992, Burke and Litwin proposed a model to explain organizational performance and change. Their model, known commonly as the Burke-Litwin model (1992), identifies factors affecting organizational performance and change, and illustrates the critical relationships between those factors. Essentially, Burke and Litwin (1992) theorized organizational change can be explained by an input-output model (Figure 1), where the input is a change in the external environment (e.g., funding, government regulations, technology) that disrupts the status quo, creates a ripple effect throughout the organization, and ultimately affects the output of individual and organizational performance.

Figure 1.
The Burke-Litwin Model of Organizational Performance and Change (Burke & Litwin, 1992.


Note: The figure illustrates the causal relationships within the Burke-Litwin model.

An environmental change can influence the behaviors of the members of an organization at a systems level. This type of change is transformational and affects organizational culture, mission, strategy, and leadership (Burke & Litwin, 1992; see Table 1 for term definitions). Alternatively, an environmental change may influence "short-term reciprocity among people and groups" (Burke & Litwin, 1992, p. 330). Such changes, described as transactional, are observed within the variables of structure, management practices, systems (policies and procedures), work unit climate, task and individual skills, motivation, and individual needs and values.

Table 1.
Definitions of Key Terms

Factor	Type	Definition
External environment	Input	Outside conditions influencing organizational performance
Leadership	Transformational	Organizational direction provided by administrators and their service as behavioral role models
Mission and strategy	Transformational	What administrators believe is the purpose of the organization; what employees believe is the purpose of the organization; how the organization attends to achieve its purpose
Organizational culture	Transformational	"the way we do things around here" (Burke & Litwin, 1992, p. 532); overt and covert rules, values, and principles guiding organizational behavior
Individual needs and values	Transactional	"specific psychological factors that provide desire and worth for individual actions or thoughts" (Burke & Litwin, 1992, p. 533)
Management practices	Transactional	Specific behaviors used by managers to direct human and material resources to carry out the organization's strategy
Motivation	Transactional	Desire to behave in ways leading to satisfaction and accomplishments
Structure	Transactional	How the people and functions of an organization are organized for implementing the mission and strategy
Systems	Transactional	"standardized policies and mechanisms that facilitate work" (Burke & Litwin, 1992, p. 532), including reward systems, performance appraisal, budgeting, reporting systems, and human resource allocation
Task requirements and individual skills/abilities	Transactional	Competencies required for people to be effective in the jobs to which they are assigned
Work unit climate	Transactional	"impressions, expectations, and feelings" (Burke & Litwin, 1992, p. 532) held by employees in local work units that affect their relationships within their own unit and with other units
Individual and organizational performance	Output	Outcomes, results, and indicators "of effort and achievement (e.g., productivity, customer satisfaction, profit, and quality)" (Burke & Litwin, 1992, p. 533)
Note: Definitions are adapted from Burke and Litwin (1992).

Practically, the Burke-Litwin (1992) model is useful for diagnosing where problems exist within an organization's framework and understanding the nature of those problems, so that appropriate solutions can be developed. Problems associated with transformational factors require massive, revolutionary changes within the system to achieve desirable results. However, problems of a transactional nature generally only require "fine tuning" (Burke & Litwin, 1992, p. 535) to improve processes.

In considering the problem of burnout in Extension, past research has identified several contributing transactional factors. In a study of Extension directors, Clark (1992) found high levels of burnout were associated with diminished feelings of personal accomplishment coupled with high levels of stress and strain from perceived responsibility overload. Although not explicitly examining burnout, Rousan and Henderson (1996) found agents voluntarily left the Ohio Extension system because of "other priorities in their lives, other job offers, insufficient pay for the amount of work performed, family obligations, too many late night meetings, too many work responsibilities, and attraction to more money elsewhere" (p. 56). Ensle (2005) described burnout as a problem caused in part because agents felt pressure from having multiple levels of administrative accountability as well as constant requests to validate the educational value of their programs.

Igodan and Newcomb (1986) suggested all individuals will experience some level of burnout during their time with Extension. Sears et al. (2000) found "a significant proportion" (p. 56) of Extension employees in one state reported symptoms of emotional fatigue, professional ineffectiveness, and depression based on scores obtained using the Maslach Burnout Inventory (Maslach & Jackson, 1986). Arnold and Place (2010) examined burnout further, investigating Florida agent employment decisions at different career stages. Agents' career decisions were negatively affected by many factors, such as lack of direction, personal work management issues, job pressures, mandated work requirements, job performance measures, salary disparities, personal work management issues, career overload, self-induced stress, lack of support, unequal recognition, insufficient pay raises, reporting difficulties, and excessive committees.

Despite a general understanding of the contributing factors to burnout, little to no work has been done to examine the issue of burnout holistically using the Burke-Litwin (1992) model, despite research indicating the factors found to influence Extension agent burnout being similar to factors outlined within the model. An examination of the transformational and transactional variables within the Extension system is needed to provide critical information for identifying where organizational breakdowns are occurring and for selecting the right approaches to reduce agent burnout and increase retention.

Purpose and Objectives

The purpose of the descriptive study reported here was to explore organizational factors related to agent burnout within Colorado State University Extension. Specifically, the objectives were to describe:

1. Extension professionals' perceptions of selected transformational factors related to agent burnout.
2. Extension professionals' perceptions of selected transactional factors related to agent burnout.

Methods

The findings presented here are part of a larger study investigating agent burnout, motivation, and retention in Colorado. The study reported here used an ex post facto design to study the variables of interest. A census was conducted of all Extension professional staff working in county or multi-county offices. At the time of the study, there were 140 potential participants.

The online questionnaire used for the study was developed by the researchers. Statements were derived from several sources, including a focus group of Colorado State University Extension professionals, related research, and the researchers themselves. It should be noted several factors identified by Burke and Litwin (1992) did not align with any of the statements generated by the focus group during survey design. Therefore, mission and strategy, structure, motivation, and management practices were omitted from the study.

The entire questionnaire contained 87 questions, 63 of which related to burnout and were used to address the purpose and objectives of this study. Extension professionals were asked to rate the extent to which they agreed or disagreed with statements related to the selected transactional or transformational factors: external environment, leadership, organization culture, work unit climate, task requirements and individual skills/abilities, systems, and individual needs and values. These factors served as the internal constructs of the survey, which are used to organize similar items. A five point Likert scale (1 = Strongly Agree, 2 = Somewhat Agree, 3 = Neither Agree nor Disagree, 4 = Somewhat Disagree, and 5 = Strongly Disagree) was used. The Likert scale was interpreted as follows: Strongly Agree = 1.00 – 1.50, Somewhat Agree = 1.51 – 2.50, Neither Agree nor Disagree = 2.51 – 3.50, Somewhat Disagree = 3.51 – 4.50, Strongly Disagree = 4.51 – 5.00.

The questionnaire was reviewed for content validity by a panel of experts composed of county Extension agents in Colorado and Florida. A pilot study with a subset of Extension agents in Colorado was conducted to test face validity and establish reliability. Following the expert panel review and pilot study, the wording for several statements was modified, constructs reorganized, and additional statements were added to increase the likelihood of obtaining valid and reliable results. The results of an ex post facto analysis of reliability for the final questionnaire based on Cronbach's alpha coefficients is shown in Table 2. Based on George and Mallery's (2003) interpretations, two constructs (Individual Needs & Values, Task Requirements) had questionable reliability.

Table 2.
Reliability Levels for Internal Constructs

Construct	α
Leadership	.93
Work Unit Climate	.92
Organizational Culture	.86
Systems	.83
External Environment	.83a
Individual Needs & Values	.69
Task Requirements and Individual Skills/Abilities	.65a
Note. aOne item deleted to improve reliability.

Dillman, Smyth, and Christian's Tailored Design Method (2008) was followed to collect data. Potential participants (N = 140) were contacted using the e-mail feature within Qualtrics. All e-mail addresses were valid. There were 115 questionnaires submitted for a final response rate of 82.14%. Four responses were discarded due to missing or incomplete data, reducing the number of usable responses to 111. When response rates exceed 80%, there is generally believed to be minimal threat of non-response bias alleviating the need to check for non-response error (Moore & Tarnai, 2002).

Characteristics of the respondents are presented in Table 3.

Table 3.
Characteristics of Respondents

Characteristic	n
Program Area
4-H	97a
Agriculture	37
Family and Consumer Science	27
Horticulture	27
Nature Resources	30
Administration	26
Other	16
Position
County/Area Director	31
Extension Agents/Associates	99
County 4-H Coordinators/4-H Program Assistants	13
Family Structure
No Children	36
Daycare/pre-school Age Children	12
Primary or Secondary School Age Children	31
Children Not Living With Me	41
Marital Status
Married/Domestic Partner	82
Not Married (includes Single, Widowed, or Divorced)	33
Household Income
Single Income	40
Multiple Income	72
Gender
Female	73
Male	37
Experience
<5 Years	43
6-10 Years	17
11-15 Years	17
16-20 Years	13
21-25 Years	5
26-30 Years	4
>30 Years	2
Note: M Age = 45.3 years. aProfessionals in Colorado often have split appointments between program areas, with 4-H as a common program area included in split appointments.

Descriptive statistics were used to analyze the data for the objectives. Frequencies and percentages were calculated for each Likert item. The mean for each construct variable was reported as well as the standard deviation to illustrate how closely to the mean the majority of responses fell. Small standard deviations indicate less variation in respondents' perceptions of an item.

The study is limited in that it is based on self-perceptions of Extension professionals in one state, and self-reports may present a biased interpretation of reality. Therefore, the data should be interpreted to represent trends. Additional research using different methodology, such as external observation, could be used to affirm the validity of the results.

Findings

Selected Transformational Factors

The first objective was to describe respondents' perceptions of selected transformational factors as they relate to burnout. The factors examined were: external environment (Table 4), leadership (Table 5), and organizational culture (Table 6).

In general, respondents did not share a common perception of their external environment and expressed significantly varying opinions that tended toward the negative end of the scale. The interpretation of the construct average of 3.48 +/- .91 fell within the neither agree/disagree and somewhat disagree ranges. The majority of respondents (64.7%, n = 75) somewhat or strongly disagreed their workload was not adversely affected by the budgetary challenges faced by Colorado State University Extension. They indicated very similar sentiments with regard to the impact of county budget cuts on their workload. Respondents were more likely to somewhat or strongly disagreed their quality of life was not being adversely affected by the budgetary challenges, though these percentages were lower than reported for workload impact (Table 4).

Table 4.
Perceptions of External Environment

Item	Strongly Agree f %	Somewhat Agree f %	Neither A/D f %	Somewhat Disagree f %	Strongly Disagree f %
The budgetary challenges CSU Extension faces are not having adverse effects on my workload	4 3.4	13 11.2	24 20.7	45 38.8	30 25.9
The budgetary challenges faced by my county/area are not having adverse effects on my workload	6 5.2	21 18.1	25 21.6	38 32.8	26 22.4
The budgetary challenges faced by my county/area are not having adverse effects on my quality of life	8 6.9	23 19.8	32 27.6	34 29.3	19 16.4
The budgetary challenges CSU Extension faces are not having adverse effects on my quality of life	6 5.2	16 32	32 27.6	46 39.7	16 13.8
Note: Construct M = 3.48, SD = .91. Scale: 1 = Strongly Agree, 2 = Somewhat Agree, 3 = Neither Agree nor Disagree, 4 = Somewhat Disagree, and 5 = Strongly Disagree.

Agents had somewhat positive views of leadership within the organization, although considerable variance was observed in their responses (Table 5). The interpretation of the construct average of 2.38 +/- 1.15 fell within the strongly agree and somewhat disagree ranges. Agents tended to somewhat or strongly agree with all four leadership items. The majority of agents (66.1 %, n = 76) indicated they believed their supervisors saw themselves as mentors to the people they supervised. Slightly fewer agents (56.9%, n = 66) indicated their supervisors were good mentors to the people they supervised.

Table 5.
Perceptions of Leadership

Item	Strongly Agree f %	Somewhat Agree f %	Neither A/D f %	Somewhat Disagree f %	Strongly Disagree f %
My supervisor sees himself/herself as a mentor to the people he/she supervises	37 32.2	39 33.9	23 20.0	10 8.7	6 5.2
My supervisor possesses good leadership skills	37 32.5	33 28.9	14 12.3	17 14.9	13 11.4
My supervisor possesses good team building skills	34 29.6	33 28.7	22 19.1	11 9.6	15 13.0
My supervisor is a good mentor to the people he/she supervises	28 24.1	38 32.8	28 24.1	14 12.1	8 6.9
Note. Construct M = 2.38, SD = 1.15. Scale: 1 = Strongly Agree, 2 = Somewhat Agree, 3 = Neither Agree nor Disagree, 4 = Somewhat Disagree, and 5 = Strongly Disagree.

Agents expressed widely varying perceptions of organizational culture although they tended toward the positive end of the scale (Table 6). The interpretation of the construct average of 2.76 +/- .94 fell within the somewhat agree and somewhat disagree ranges. Agents tended to somewhat agree that management skills (34.5%, n = 40) and interpersonal skills (39.1%, n = 45) were key considerations during the hiring process of County/Area Directors.

Table 6.
Perceptions of Organizational Culture

Item	Strongly Agree f %	Somewhat Agree f %	Neither A/D f %	Somewhat Disagree f %	Strongly Disagree f %
Management skills are currently a key consideration during the hiring process of County/Area Directors	12 10.3	40 34.5	34 29.3	20 17.2	10 8.6
Interpersonal skills are currently a key consideration during the hiring process of County/Area Directors	11 9.6	45 39.1	35 30.4	19 16.5	5 4.3
Leadership skills are currently a key consideration during the hiring process of County/Area Directors	11 9.6	37 32.2	37 32.2	21 18.3	9 7.8
Note: Construct M = 2.76, SD = .94. Scale: 1 = Strongly Agree, 2 = Somewhat Agree, 3 = Neither Agree nor Disagree, 4 = Somewhat Disagree, and 5 = Strongly Disagree.

Selected Transactional Factors

The second objective of the study was to describe agents' perceptions of selected transactional factors as they relate to agent burnout. The factors examined were: task requirements and individual skills/abilities (Table 7), work unit climate (Table 8), systems (Table 9), and individual needs and values (Table 10).

Overall, agents were ambivalent about the construct of task requirements and individual skills/abilities, demonstrating considerable variance in their responses (Table 7). The interpretation of the construct average of 3.33 +/- .94 fell within the somewhat agree and somewhat disagree ranges. However, two-thirds of the responding agents (66.4%, n = 77) somewhat disagreed or strongly disagreed they tended not to overextend themselves with their workload. This finding is echoed by the third item, which found that only 38% (n = 44) of responding agents indicated they were able to manage a healthy balance between work and their personal lives.

Table 7.
Perceptions of Task Requirements and Individual Skills/Abilities

Item	Strongly Agree f %	Somewhat Agree f %	Neither A/D f %	Somewhat Disagree f %	Strongly Disagree f %
I tend not to overextend myself with my workload	7 6.0	25 21.6	7 6.0	47 40.5	30 25.9
I am able to adjust my workload appropriately when I add additional programs	6 5.2	37 31.9	14 12.1	40 34.5	19 16.4
I am able to manage a healthy balance between work and my personal life	9 7.8	35 30.2	17 14.7	40 34.5	15 12.9
Note: Construct M = 3.33, SD = .94. Scale: 1 = Strongly Agree, 2 = Somewhat Agree, 3 = Neither Agree nor Disagree, 4 = Somewhat Disagree, and 5 = Strongly Disagree.

Overall, agents expressed mixed opinions about work unit climate but tended toward the positive end of the scale (Table 8). The interpretation of the construct average of 2.67 +/- .76 fell within the somewhat agree and neither agree/disagree ranges. However, two competing trends in the data are apparent. Agents tended to express favorable opinions related to their supervisors. Over 88% (n = 103) of agents somewhat or strongly agreed their supervisors do not micro-manage their work. Agents tended to indicate their supervisors supported them and their decisions.

Agents were less affirmative when asked if their Regional Directors and campus-based Extension Administrators supported them. Approximately half (50.9%, n = 58) of the agents somewhat or strongly agreed their Regional Directors supported county Extension staff when issues arose. Comparatively, only 35.9% (n = 41) of agents somewhat or strongly agreed campus-based Extension Administrators supported county Extension staff when issues arose. A similar pattern existed when ask to rate the level of trust between field staff and Regional Directors, and field staff and campus-based Extension Administrators. Only 24.2% (n = 28) of agents somewhat or strongly agreed a healthy level of trust existed between field staff and campus-based Extension Administrators, compared to the trust perceived with direct supervisors (72.4%, n = 84 somewhat or strongly agree) and with Regional Directors (41.4%, n = 48 somewhat or strongly agree). Of the 28 agents who somewhat or strongly agreed a healthy level of trust existed between field staff and campus-based Extension Administrators, only six agents selected the strongly agree response option.

Table 8.
Perceptions of Work Unit Climate

Item	Strongly Agree f %	Somewhat Agree f %	Neither A/D f %	Somewhat Disagree f %	Strongly Disagree f %
My supervisor does not micro-manage my work	82 70.7	21 18.1	4 3.4	7 6.0	2 1.7
My supervisor supports me	66 56.9	26 22.4	12 10.3	6 5.2	6 5.2
My supervisor supports my decisions	64 55.7	33 28.7	5 4.3	8 7.0	5 4.3
My supervisor implements the professional scheduling policy in a way that supports agents/staff	63 54.3	30 25.9	13 11.2	6 5.2	4 3.4
My supervisor provides me with due credit and recognition of my work	60 52.5	23 20.2	15 13.2	11 9.6	5 4.4
My direct supervisor has realistic performance expectations of me	57 49.1	39 33.6	8 6.9	5 4.3	7 6.0
There is a healthy level of trust between my supervisor and me	56 48.3	28 24.1	13 11.2	11 9.5	8 6.9
My supervisor is supportive when I "say no" in an attempt to control my work load	47 40.9	26 22.6	28 24.3	9 7.8	5 4.3
My supervisor possesses good management skills	43 37.7	36 31.6	16 14.0	9 7.8	10 8.8
Reasonable work hours are expected of me by my supervisor	41 35.3	42 36.2	11 9.5	13 11.2	9 7.8
My Regional Director listens to the concerns of agents/staff	25 21.9	36 31.6	17 14.9	22 19.3	12 10.5
My county director helps me to do my job more effectively	23 20.2	27 23.7	17 14.9	8 7.0	10 8.8
There is a healthy level of trust between field staff and our Regional Director	21 18.1	27 23.3	20 17.2	28 24.1	20 17.2
I have someone to turn to for counseling and/or mentoring	19 16.5	72 62.6	16 13.9	7 6.1	1 .9
My Regional Director takes action when he/she hears legitimate concerns of agents/staff	19 16.7	36 31.6	16 14.0	24 21.1	16 14.0
Regional Directors support county Extension staff when issues arise with clientele	18 15.8	40 35.1	19 16.7	17 14.9	13 11.4
Campus based Administrators support county Extension staff when issues arise with clientele	12 10.5	29 25.4	35 30.7	22 19.3	12 10.5
My Area Director helps me to do my job more effectively	12 10.5	16 14.0	26 22.8	5 4.4	3 2.6
My Regional Director helps me to do my job more effectively	9 7.9	30 26.3	37 32.5	13 11.4	21 18.4
There is adequate recognition given county agents and staff for the work they perform	8 7.0	27 23.7	22 19.3	29 25.4	26 22.8
I have the opportunity to provide input regarding decisions to fill vacant positions in my area or region	8 6.9	17 14.7	16 13.8	35 30.2	40 34.5
Campus-based Extension Administrators take action when they hear legitimate concerns of agents/staff	7 6.1	26 22.8	25 21.9	30 26.3	22 19.3
There is a healthy level of trust between field staff and campus-based Extension Administrators	6 5.2	22 19.0	28 24.1	42 36.2	18 15.5
Campus-based Extension Administrators listen to the concerns of agents/staff	5 4.4	30 26.3	20 17.5	41 36.0	14 1.8
The CSU Extension professional scheduling policy is being implemented uniformly across Colorado	4 3.5	7 6.1	48 42.1	29 25.4	26 22.8
Note: Construct M = 2.67, SD = .76. Scale: 1 = Strongly Agree, 2 = Somewhat Agree, 3 = Neither Agree nor Disagree, 4 = Somewhat Disagree, and 5 = Strongly Disagree.

On the whole, agents tended to have negative perceptions of the systems construct, demonstrating less variation in their responses than when compared to the other constructs (Table 9). The interpretation of the construct average of 3.55 +/- .58 fell within the neither agree/disagree and somewhat disagree ranges. The vast majority of agents somewhat or strongly disagreed with the statements asking if the current reporting system (CPRS) met their needs (90.3%, n = 102) and adequately represented their jobs (85.8%, n = 97). Agents also tended to somewhat or strongly disagree with the adequacy of the systems in place for reviewing the performance of the State Extension Director (65.2%, n = 73), Program Directors (64.5%, n = 71), and Regional Directors (58.0%, n = 65). The only systems statements with which at least half of the agents somewhat or strongly agreed were: "The Colorado State University Extension professional scheduling policy is adequate to meet the needs of county agents/staff" (75.5%, n = 86), and "County Extension Directors in Colorado receive sufficient training to be successful" (83.7%, n = 97).

Table 9.
Perceptions of Systems

Item	Strongly Agree f %	Somewhat Agree f %	Neither A/D f %	Somewhat Disagree f %	Strongly Disagree f %
The current reporting system (CPRS) meets the needs of the agents in the field	1 .9	6 5.3	4 3.5	47 41.6	55 48.7
The current reporting system (CPRS) adequately represents the job that I do	2 1.8	7 6.2	7 6.2	44 38.9	53 46.9
There is an adequate system in place to review the performance of the State Extension Director	1 .9	5 4.5	33 29.5	27 24.1	46 41.1
There is an adequate system in place to review the performance of Extension Program Directors (4-H, Operations, Fiscal, Technology, Outreach and Engagement, Federal and Civic Engagement)	2 1.8	8 7.3	29 26.4	25 22.7	46 41.8
There is an adequate system in place to review the performance of Regional Directors	1 .9	7 6.3	39 34.8	25 22.3	40 35.7
Existing new employee orientation format/content is adequate	5 4.3	10 8.6	29 25.0	35 30.2	37 31.9
There is a good system of mentoring/coaching/counseling in place to help me be a productive employee	1 .9	14 12.2	29 25.2	44 38.3	27 23.5
There is an adequate system in place to review the performance of County Extension Directors	11 9.7	26 23.0	28 24.8	24 21.2	24 21.2
The current reporting system (CPRS) meets the needs of Colorado State University Extension	5 4.4	20 17.7	38 33.6	27 23.9	23 20.4
Offering new employee orientation once per year is adequate	4 3.5	15 13.0	36 31.3	40 34.8	20 17.4
I have received adequate training in volunteer management to be successful	10 8.6	31 26.7	20 17.2	35 30.2	20 17.2
There is an adequate system in place to review the performance of Area Directors	5 4.5	19 17.0	49 43.8	19 17.0	20 17.9
I have received adequate training in conflict management to be successful	8 6.9	33 28.4	23 19.8	33 28.4	19 16.4
The duration of the search process to fill county level Extension positions is acceptable	14 12.1	35 30.2	22 19.0	26 22.4	19 16.4
The CSU Extension professional scheduling policy is adequate to meet the needs of county agents/staff	28 24.6	58 50.9	11 9.6	15 13.2	2 1.8
County Extension Directors in Colorado receive sufficient training to be successful	25 21.6	72 62.1	10 8.6	9 7.8	-- --
Note: Construct M = 3.55, SD = .58. Scale: 1 = Strongly Agree, 2 = Somewhat Agree, 3 = Neither Agree nor Disagree, 4 = Somewhat Disagree, and 5 = Strongly Disagree.

Agents expressed varied opinions about individual needs and values that tended toward the negative end of the scale (Table 10). The interpretation of the construct average of 3.44 +/- .89 fell within the neither agree nor disagree and somewhat disagree ranges. The majority of agents somewhat or strongly disagreed with three of the four individual needs and values items. Nearly two-thirds of the agents (65.2%, n = 75) somewhat or strongly disagreed with the statement "The salary I receive is fair compensation for the work I perform."

Table 10.
Perceptions of Individual Needs & Values

Item	Strongly Agree f %	Somewhat Agree f %	Neither A/D f %	Somewhat Disagree f %	Strongly Disagree f %
The salary I receive is fair compensation for the work I perform	4 3.5	25 21.7	11 9.6	37 32.2	38 33.0
I rarely find myself torn between developing my career in Extension and my responsibility to my family	9 7.8	23 19.8	15 12.9	43 37.1	26 22.4
The salary I receive is fair compensation for the education level I have obtained	7 6.1	30 26.1	13 11.3	40 34.8	25 21.7
I am able to have adequate family time as an Extension employee	8 6.9	31 26.7	24 20.7	36 31.0	17 14.7
Note: Construct M = 3.44, SD = .89. Scale: 1 = Strongly Agree, 2 = Somewhat Agree, 3 = Neither Agree nor Disagree, 4 = Somewhat Disagree, and 5 = Strongly Disagree.

Conclusions, Implications, & Recommendations

The purpose of the descriptive study was to explore organizational variables related to agent burnout within Colorado State University Extension. Changes in the external environment—specifically, budget concerns—were perceived by the responding Extension professionals to be negatively affecting their workload and quality of life. Disruptive changes in the external environment can be expected to affect the entire organizational framework (Burke & Litwin, 1992), so this is a significant concern for Colorado State University Extension.

Transactional factors (e.g., systems, individual needs and values) within the organizational framework were observed to be more problematic for Colorado State University Extension than transformational factors (e.g., leadership, organizational culture). This is similar to previous research conducted by Rousan and Henderson (1996) and Arnold and Place (2010). Fortunately, transactional factors are easier to improve than transformational factors (Burke & Litwin, 1992).

Colorado State University Extension can begin to address transactional issues by examining the factor of individual needs and values. Extension professionals feel they are underpaid. The issue of inadequate compensation is not unique to Colorado or new to Extension at large; it is a commonly cited factor leading to agent dissatisfaction (e.g., Arnold & Place, 2010; Rousan & Henderson, 1996). New funding and staffing models may need to be considered to generate the organizational resources necessary to retain quality Extension professionals.

Additionally, Colorado State University Extension professionals feel their jobs compromise their availability to their families. It can be observed from statements within the task requirements and individual skills/abilities construct that on-going professional development in the areas of time management and self-expectations would be beneficial because the majority of agents reported a tendency to overextend themselves and difficulty managing their time. And while the Extension professionals indicated their supervisors had reasonable expectations of hours worked, they also indicated they were not micro-managed. Empowering agents to manage their own time more effectively is one pathway toward decreasing burnout, but the findings indicate an opportunity exists for supervisors play a greater role in mentoring agents. Supervisors who counsel agents about healthy decisions regarding professional and personal priorities may help to alleviate the observed issues.

Within the work unit climate factor, issues of trust were apparent between varying levels of the organization. Hierarchical distance appears to negatively affect trust, with trust decreasing at each ascending level in the organizational hierarchy. Colorado State University Extension administration has already begun the process of improving relationships within the organization; conducting the survey forming the basis of this study is an example of their commitment. More convincingly, the transparent manner that administration has opted to adopt in communicating the results with internal and external audiences, and working with county Extension professionals to develop agent-led action teams and plans based on the results, substantiates a strong commitment to improving the organization. Such transparency and continued collaboration across the organizational hierarchy must continue to be a priority moving forward in order to address the issue of burnout and increase retention.

To date, the action teams have primarily focused on issues within the systems factor, which had the lowest overall mean of all constructs studied. Extension professionals were dissatisfied with every major system or policy about which they were asked. Notably, the greatest concerns related to the reporting system, which was not perceived to adequately reflect the value of the professionals' work, and a perceived inability for county-level professionals to participate in the evaluation of state-level administrators. The former is indicative of a desire to feel valued by the organization, and the latter appears related to the lack of trust between hierarchical levels. However, it may be that if county-level professionals feel they and their work are more valued, they will feel less of a need to evaluate senior-level administrators.

Colorado State University Extension administration has taken the results of this research and applied them to strengthen their organization. Their intention is to conduct a follow-up study in the next 2-3 years to measure progress towards their goals. There may be value in adding a known burnout inventory, such as the Maslach Burnout Inventory (Maslach & Jackson, 1986), to the questionnaire used by the authors to facilitate the comparison of perspectives between employees experiencing symptoms of burnout and those who are not. Further, the purpose of the study reported in this article was descriptive in nature. Future research may benefit from investigating the relationships between the transformational and transactional factors to determine the extent to which changing one factor might be expected to impact other factors within the organization. Finally, the authors opted to use the Burke and Litwin (1992) model to organize the research reported here. It may be useful to conduct an integrative inquiry of prior research within Extension to determine which model(s) are most appropriate for studying burnout.

Other states facing burnout and retention issues may benefit from following a process similar to that of Colorado. Transparency and follow-through are keys to obtaining accurate data and building goodwill between professionals in the field and the administrators who oversee the state Extension program.

References

Arnold, S. K. (2007). Career decisions of Florida agricultural extension agents (Doctoral dissertation, University of Florida). Retrieved from: http://www.joe.org/joe/2012april/tt1.php

Burke, W. W., & Litwin, G. H. (1992). A causal model of organizational performance and change. Journal of Management, 18(3), 532-545.

Clark, R. (1992). Stress and turnover among Extension directors. Journal of Extension [On-line], 30(2) Article 2RIB1. Available at: http://www.joe.org/joe/1992summer/rb1.php

Dillman, D. A., Smyth, J. D., & Christian, L. M. (2008). Internet, mail, and mixed-mode surveys: The tailored design method (3rd ed.). Hoboken, NJ: John Wiley and Sons.

Ensle, K. M. (2005). Burnout: How does Extension balance job and family? Journal of Extension [On-line], 43(3) Article 3FEA5. Available at: http://www.joe.org/joe/2005june/a5.php

George, D., & Mallery, P. (2003). SPSS for Windows step by step: A simple guide and reference. 11.0 update (4th ed.). Boston, MA: Allyn & Bacon.

Igodan, O. C., & Newcomb, L. H. (1986). Are you experiencing burnout. Journal of Extension [On-line], 24(1) Article 1FEA1. Available at: http://www.joe.org/joe/1986spring/a1.php

Kutilek, L. (2000). Learning from those who leave. Journal of Extension [On-line], 38(3) Article 3IAW2. Available at: http://www.joe.org/joe/2000june/iw2.php

Maslach, C., & Jackson, S. E. (1986). The Maslach Burnout Inventory manual (2nd ed.). Palo Alto, CA: Consulting Psychologists Press.

Moore, D., & Tarnai, J. (2002). Evaluating nonresponse in mail surveys. In Groves, R. M., Dillman, D. A., Eltinge, J. L., & Little, R. J. A. (Eds.). Survey nonresponse. New York: Wiley-Interscience.

Rousan, L. M., & Henderson, J. L. (1996). Agent turnover in Ohio State University Extension. Journal of Agricultural Education, 37(2), 56-62.

Schabracq, M. J., Winnubst, J. A. M., & Cooper, G. L. (2003). The handbook of work & health psychology (2nd ed.). West Sussex, England: John Wiley & Sons, Ltd.

Sears, S. F., Urizar, G. G., & Evans, G. D. (2000). Examining a stress-coping model of burnout and depression in extension agents. Journal of Occupational Health Psychology, 5(1), 56-62.