June 2000 // Volume 38 // Number 3 // Ideas at Work // 3IAW2

Previous Article Issue Contents Previous Article

Learning from Those Who Leave

Individual exit interviews have been conducted with employees voluntarily leaving Ohio State University Extension for the past two years. This is an organizational commitment to better understand what motivates staff to leave, with an expectation that the information gathered will improve recruitment efforts, hiring processes, staff development, supervisory practices, worker expectations, job satisfaction, and the organizational environment as it relates to diversity issues. The author shares information about the development of the survey instrument used, the data gathered, and implications of the findings. Information gathered from exit interviews may only lead to the development of more questions; however, they are questions that must be asked if Extension is to continue to recognize and meet the needs of staff members.

Linda M. Kutilek
Assistant Professor
Employee Data Network
Ohio State University Extension
Columbus, Ohio
Internet address: kutilek.1@osu.edu

The primary resource of the Cooperative Extension Service is its staff and volunteers. Because local Extension staff members lead the program process, their loss interrupts program continuity and possibly even creates program voids for periods of time. While some level of staff departure is inevitable, turnover and prolonged vacancies are a financial and time drain for the organization. An Ohio State University Extension study concluded that net costs to Extension are $80,000 each year. This reflects costs to replace all individuals minus salary savings created by the vacancies.

However, an equally serious aspect of the issue is the loss of program effort that most surely accompanies a staff vacancy. To date, no acceptable method exists to determine just how costly the factor of loss is to the clientele, the Extension system, or to staff morale.

In 1995, Rousan studied a group of 67 county agents who had voluntarily left the Ohio State University Extension system. As a result, he made a series of recommendations that challenged OSU Extension to investigate further why staff leave and what similar characteristics they might share.

That work has led to an organizational commitment to gather information to help better understand what motivates staff to leave. Behind the commitment is an expectation that the information will improve recruitment efforts, hiring processes, staff development, supervisory practices, worker expectations, job satisfaction, and the organizational environment as it relates to diversity issues.

Exit Interviews

To further the original work, Ohio State University Extension began in 1997 to conduct exit interviews with all staff who left employment. While Rousan (1995) had studied the agent, the current exit interview process extends to all staff. Agents, specialists, administrators, secretaries, and program assistants are included in the interview process. The expanded analysis enables Extension administrators to gain and examine information from individuals with disparate job responsibilities.

The 1997 group consisted of 84 individuals who voluntarily left the Ohio Extension system; the 1998 group included 87 individuals. All were offered an opportunity for an exit interview; 87% were interviewed in 1997, and 78% in 1998. Interviews were conducted in person or via phone, and the same set of questions was asked each exiting employee.

A series of open-ended questions was asked of the exiting employee. Questions focused on areas of satisfaction about employment with Extension, reasons for leaving, work environment, relationship with supervisor, and opportunities for professional development. Interviews averaged 50 minutes in length, and ranged from 25 minutes to 2 hours. All conversations were recorded in writing and read back to the exiting employee to confirm accuracy.

Key words were then coded for data collection. The key words' code came from a pilot study conducted in 1996. In that pilot study, exiting employees at all levels were interviewed to build the initial data pool that led to the development of key word codes.

All interviews in 1997 and 1998 were conducted by one of two people. Both interviewers independently coded narrative comments to assure some consistency in use of the key codes. Results of the independent activity were then compared to determine how similarly the two used the codes. To address issues of confidentiality, interview forms were coded by a numeric system instead of by name.

The questionnaire included 15 questions covering 5 categories. Categories of questions included positive aspects of Ohio State University Extension; supervision and support; working conditions; benefits and salary; and career development.

The four concluding questions were:

  • "What insights or advice about the work setting or job/position would you give to your successor?";

  • "Explain any complaints or areas of dissatisfaction you may have about the total Ohio State University Extension organization";

  • "Other suggestions you may have to help improve the Ohio State University Extension organization"; and

  • "Additional comments concerning the Ohio State University Extension organization, including previously discussed questions and other things that may not have been discussed."

    After the interview was completed, the interviewer noted the key word codes, and the information was entered into the data pool.

    Questionnaire Analysis

    Because the individuals interviewed had uniquely different position descriptions, in addition to the total summary of results, analysis was reviewed by:

    • Gender,
    • Title,
    • Ethnicity,
    • Program area,
    • Level of appointment (i.e., full-time or part-time), and
    • Administrative regions within the state.

    At this point, extensive analysis of data is difficult because the categories created had relatively small numbers of respondents. However, based upon frequency of recurring responses, some observations are emerging.

    The turnover rate of program assistants is much higher than for all other classification of individuals. They reported that job stress and low pay were the reasons for leaving. Agents also noted issues of stress and low pay, but added the dimension of concern about lack of supervisory support.

    These observations led to an expectation that a larger data pool will provide considerable information for improved organizational policies and priorities. The numbers of individuals within each category will continue to grow as additional interviews are conducted. The database is continually updated and will be reviewed annually to assess organizational trends that could provide insight into virtually every area of organizational life. Additional observations can be made based upon the 1997 and 1998 samples as 1999 data are compiled.


    Rousan's challenges to the organization included the need to study on an ongoing basis the reasons for employee turnover. Ohio has begun to build an ever-increasing database of information that can be applied to strengthening support to the current staff. Turnover rates have remained around 7% for the total Extension staff, 5% for agents. Compared to industry standards, this is a very low rate of turnover and an encouraging indicator of staff stability.

    However, taking the next step in comparing this state's turnover rate to the turnover rate of other Extension systems around the country will be important. Also, making a comparison to other educational institutions may provide insight into the relative significance of a 6-7% turnover rate.

    Reflecting on a 20% turnover rate for the paraprofessional role of program assistant is also important. Is there a clear path of career development for individuals in these roles, or is exiting the system the only opportunity for change? Information gathered from exit interviews may only lead to the development of more questions; however, they are questions that Extension must ask of itself if it is to continue to recognize and meet the needs of staff members.


    Rousan, L. (1995). Agent turnover in Ohio State University Extension. Unpublished doctoral dissertation, Ohio State University, Columbus.