|
October 2003
|
       |
Research in BriefA Longitudinal Study of the Evolution of Organizational Values of Ohio State University Extension EducatorsR. Dale Safrit Nikki L. Conklin Jo M. Jones IntroductionThe mission of Ohio State University (OSU) Extension is "To help people improve their lives through an educational process using scientific knowledge focused on identified issues and needs" (O.S.U. Extension Annual Report, 1995, n.p.). County-based professionals conduct educational programs in agriculture and natural resources, community development, family and consumer sciences, and 4-H youth development. The last decade of the 20th century has proven both transformational and turbulent for the Cooperative Extension system. As Cooperative Extension entered the 21st century, Jimmerson (1989, p. 16) suggested that "meeting the challenges of the information age will require attention to the values and beliefs that guide us as we work to provide our clients with information and help them solve problems." A value is "an enduring belief that a specific mode of conduct or end-state of existence is personally or socially preferable to an opposite or converse mode of conduct or end-state of existence" (Rokeach, 1973, p. 5). Values play important roles in determining how we function as individuals, family members, and members of work teams, and are a product of our individual experiences. The enduring nature of values and value systems arises from the fact that they are neither completely stable nor unstable, but rather, are evolving continuously according to our changing physical, social, and emotional surroundings. Hitt (1988) suggested that every profession or work organization is guided by certain beliefs or values. "These values communicate 'what we stand for' and 'what is important to us'...values are the soul of the organization" (p. 86). One sign of a healthy, productive organization is agreement between the organization's values and the daily behaviors of its members. Vaill (1990, p. 59) emphasized "how management and leadership in organizational contexts may be viewed as a process of ongoing values clarification. That is the most important business. That is the key job that needs doing, and that is the job whose significance we keep underestimating." Barker (1994) concluded that "a thorough knowledge of the values held by the Minnesota Extension Service will facilitate the building of a foundation which will then enhance . . . the organization as a whole" (p. 8). Although each of us may have unique personal value systems, we function best within organizations and professions where we share values with our colleagues. "An organizational value is any concept or idea that is held in high esteem by the members of an organization and that shapes the organization's philosophy, processes, and goals" (Jones, Safrit, & Conklin, 1991, p.1). Research conducted in 1991 with OSU Extension program personnel (Safrit, Jones, & Conklin, 1995) identified 12 organizational values (Table 1). For almost a decade, these identified values have been used by OSU Extension administrators as an important basis for both managerial decision making and organizational policy development. Furthermore, three potential organizational values, "Racial/ethnic diversity among employees," "Racial/ethnic diversity among clientele," and "OSU Extension as a leader in overall outreach and engagement at OSU" were not valued by a majority of study respondents (47%, 46%, and 58%, respectively.) Subsequently, OSU Extension administrators invested enormous resources into the organization to provide training and continuing professional education opportunities that emphasized the concepts within OSU Extension's mission.
Purpose and MethodologyThe research described here replicated the authors' 1991 study in order to investigate the evolution of OSU Extension organizational values a decade later. The study used a census of 797 OSU Extension program personnel who were active at their assigned professional responsibilities as of April 1, 2001. The researchers utilized a modified version of the 62-item Values Questionnaire used in the 1991 study, organized into two sections. Section 1 contained 52 items using a Likert-type response scale to obtain information on the respondents' organizational values. Response choices ranged from 1 to 5, with 1 representing "not valued" and 5 representing "extremely valued." In Section 2, respondents provided basic background information used to categorize them including: year of birth (i.e., age), marital status, gender, race, job tenure within OSU Extension, job tenure within other Cooperative Extension Services, job classification, major program area responsibility, highest level of formal education, and area of most advanced degree. The researchers established face validity of the instrument with OSU Extension Administrative Cabinet members; a Cronbach's alpha of .87 was computed for Section 1 as a measure of internal validity and indicator of reliability. The final response rate was 75%. Organizational values for OSU Extension were identified by comparing the calculated frequencies of defined groupings of item responses with a predetermined 79% level of agreement that defined an item's acceptance as an organizational value. This was the same procedure used in the 1991 study. Findings and ConclusionsThe researchers identified 10 of the 12 original organizational values (identified in the 1991 study) as current OSU Extension organizational values (Table 1). One additional organizational value was identified in 2001 that was not included in 1991: "Flexibility/ adaptability in local programming" (81%). "Teamwork among immediate coworkers" and "Financial support from the local level" had been identified in the 1991 study (79% and 80%, respectively) but were not among the "highly valued" concepts in the 2001 study (77% and 77%, respectively). Furthermore, neither of the three potential values identified for emphasis by administrators in 1991 increased significantly in their 2001 rankings: "Racial/ethnic diversity among employees" (41% in 2001; increased from 34% in 1991); "Racial/ethnic diversity among clientele" (46% in both 1991 and 2001); and "OSU Extension as a leader in overall outreach and engagement at OSU" (58% in both 1991 and 2001). Although surprising (and somewhat disappointing) to the researchers, these findings are congruent with emerging thoughts in the field of organizational management (Dahler-Larsen, 1998; Siehl & Martin, 1990). The researchers offer three possible explanations. Not Enough Time Ten years is not adequate time for an organization's values to change. For example, efforts to enhance OSU Extension's commitment to outreach and engagement through partnerships with a broader range of academic units on campus may be resulting in building broader university commitment to the land grant mission. However, these efforts are not yet a common norm of operation to achieve change in the culture across the holistic Extension organization. Possible Alienation of Personnel In emphasizing targeted organizational values that were not valued by a majority, administrators may actually have alienated program personnel; Dahler-Larsen refers to this counter-intentional phenomenon in organizations. The organization has emphasized the commitment to diversity for more than 10 years without significant changes in the attitudes or demographics of our personnel. Grant programs often have funded new initiatives with diverse clientele, but then falter during times of financial stress when grants end. Does this result in the professionals not taking the commitment to these values seriously? In active dialogue with personnel statewide, the assumption that non-minority personnel cannot work effectively with diverse audiences has been questioned. Yet the personnel profile for our organization does not yet mirror the diversity of Ohio's population. Long-Standing, Dominant Culture By default, an organization's culture may reflect long-standing core values that historically have defined the image of the organization. Although OSU Extension experiences a 20% turnover rate for paraprofessional roles, overall turnover of personnel averages 7%, with a 5% rate for agents (Kutilek, 2000). Though paraprofessionals reflect a more diverse group than the overall base of personnel, their transitional makeup with a higher turnover rate may limit their impact on changing the organizational culture. In addition, many people applying for or being hired in agent roles are attracted to the organization based upon the existing dominant culture, thus proliferating "what is" rather than "what should be." What does this mean in recruiting professionals for the future who do reflect the rapidly changing demographic profile of the state? The true importance of organizational values may lie not in an organization-wide philosophical position, but rather from each individual determining what shared values really mean through their day-to-day practice. For OSU Extension, the strength and stability of its organizational values may be both a source of continuity and stability during times of rapid social and fiscal change, as well as a source of frustration for administrators and leaders seeking to reshape the organization's culture. ReferencesBarker, W. A. (1994). The identification of organizational values in the Minnesota Extension Service. Unpublished doctoral dissertation, University of Minnesota, Duluth. Dahler-Larsen, P. (1998). What 18 case studies of organizational culture tell us about counter-intentional effects of attempts to establish shared values in organizations. In M.A. Rahim, R.T. Golembiewski, & C.C. Lundberg (Eds.), Current topics in management (Volume 3) (pp. 151-173). Hitt, W.D. (1988). The leader-manager: Guidelines for action. Columbus, Ohio: Battelle Press. Jimmerson, R. M. (1989). What values will guide Extension's future? Journal of Extension, [On-line], 27(3). Available at: http://www.joe.org/joe/1989fall/a5.html Jones, J.M., Safrit, R.D., & Conklin, N.L. (1991, October). Organizational values of Ohio Cooperative Extension Service employees. Paper presented at the annual meeting of the American Evaluation Association, Chicago. Kutilek, L. (2000). Learning from those who leave. Journal of Extension [On-line], 38(3). Available at http://www.joe.org/joe/2000june/iw2.html Ohio State University Extension. (1995). Annual report. Columbus: Author. Rokeach, M. (1973). The nature of human values. New York: The Free Press. Safrit, R. D., Conklin, N. L., & Jones, J. M. (1995). Extension's values: A bridge across turbulent times. Journal of Extension [On-line], 33(1). Available at http://www.joe.org/joe/1995february/a1.html Siehl, C., & Martin, J. (1990). Organizational culture: A key to financial performance? In B. Schneider (Ed.), Organizational climate and culture (pp. 241-281). San Francisco: Jossey-Bass. Vaill, P. B. (1990). Managing as a performing art. San Francisco: Jossey-Bass Publishers.
Concordance Among Extension Workers, Researchers, and Professional Arborists in Rating Landscape TreesAlfredo B. Lorenzo Catalino A. Blanche James F. Henson IntroductionThe extent of urban expansion across the U.S. has tripled since 1950 (Dwyer, et al., 2000), and about 75% of the population now lives in urban areas. There are several reasons why this trend is relevant to the role and work of Cooperative Extension professionals. Most notably, the demographics and needs of clientele have changed. In particular, the client base is increasingly urban and interested in a wide range of natural resource topics rather than a narrow agricultural focus (Rodewald, 2002). This increased interest in natural resources has in turn resulted in increased public awareness of the importance of trees in urbanized and developing landscapes. Landscape trees provide a variety of benefits for urban dwellers, including monetary value. The need to know the monetary value of landscape trees is no longer limited to real estate transactions, lawsuits, insurance claims, and tax purposes, but also holds true for tree inventories and tree care investment decisions. Formulas are the most commonly used method for determining the value of trees. Formulas and methods developed over the years are described in the 8th edition of the Guide for Plant Appraisal by the Council of Tree and Landscape Appraisers (CTLA). The most widely used is the trunk formula method developed for trees too large to be physically replaceable. Using the formula requires the determination of size, condition, location, and species rating value based solely upon species characteristics. Watson (2000) compared the trunk formula method with other formulas and concluded that the major differences in appraised values using the trunk formula are attributed to species rating values. Cooperative Extension specialists, particularly foresters and horticulturists, are often asked to provide species rating value. Traditionally, they have been using species values that came with the Guide. These species rating values were subjectively assigned by a group of experts in 1970 (Lewis, 1970). The current species rating values are in five categories of 20% class intervals. The Guide classifies trees according to species and varieties, varying according to different geographical areas in the country. Problems with Current Species Rating ProceduresThe traditional process of polling experts to arrive at species rating values has inherent problems. First, the likelihood exists for high incidences of missing values due to differences in level of familiarity of the experts with the species in question. Second, the process allows for a wide range of values assigned to every species. This could be attributed to the differences in training and background of the experts and the wide variability of landscapes across a locale, state, or region. Both problems could result in weak agreement or lack of agreement among professionals and thus compromise the utility and credibility of species ratings generated through such a process. Although it is essential that a group of experts representative of end users of the species rating values be formed to determine species rating values that are acceptable and useful, it is equally important that there is strong agreement among them on the resultant species rating values. In an attempt to address these problems, Blanche, Guidry, and Wefel (1997) developed a more objective approach to establishing a single value for a given species using tree characteristics that are quantifiable and inherent, such as: flood tolerance, pollution tolerance, specific gravity, longevity and others. However, their approach remains the exception, and polling judgments and opinions of tree experts remains the most popular approach for determining species rating value. Purpose of Present StudyIn the study described here we used the Delphi method to collect data on species rating and used the Kendall coefficient of concordance to test the strength and quality of agreement of a panel of Extension professionals among themselves and in comparison with other professional groups. The Delphi Method An under-utilized method combining quantitative and qualitative opportunities to poll expert judgments for species rating values is the Delphi method. The Delphi, an iterative process for evoking expert opinion, has three primary features: (1) anonymity, (2) controlled feedback, and (3) statistical group response (Dalkey, 1969). Thus, the Delphi offers several advantages for collecting species rating values from experts with varying background and levels of experience. It reduces the psychological pressures of face-to-face confrontation and the possibility of an especially vocal or powerful individual dominating the final outcome. Also, improved agreement increases with the number of experts and with additional iteration. Last, there is potentially greater acceptance of the group opinion by the individual participants using Delphi than there is with face-to-face procedures. Today, the Delphi is used in a wide range of applications in industry, governmental agencies, and other organizations. Ludwig (1997) described the use of the Delphi by Extension professionals to help clientele or customers determine where programs or applied research should focus. Kendall Coefficient of Concordance (W) The Kendall coefficient of concordance (W) is used to determine the association among k sets of species rating values. W is an index of the divergence of the actual agreement shown in the data from the possible perfect agreement. Values of W can range from 0 to 1, with 0 indicating perfect disagreement, and 1 indicating perfect agreement (Landis & Koch, 1977). Materials and MethodsThe Panel of Experts The study took place in Louisiana through an Urban and Community Forestry grant from the Louisiana State Office of Forestry. Nineteen individuals from each major geographical region of the state were initially identified and asked to independently provide expert opinion on the species rating value of landscape plants in Louisiana. Each individual was considered an "expert" based on their roles and previous experiences, and knowledge of species rating. These experts included six foresters, three professional/consulting arborists, one university horticulture professor, one landscape architect, and eight Cooperative Extension Service personnel. Four of the experts failed to return their respective responses by the due date; thus, the analysis was based on the rating provided by the remaining 15 experts. List of Species A list of 205 plant species was sent to each expert accompanied by a short explanation of the objective of the project and a copy of Chapter 5 of the Guide. Chapter 5 describes the factors to consider in arriving at a species rating value based upon species or cultivar characteristics without regard to condition and location factors. For purposes of this article, results associated with the 63 species in the International Society of Arboriculture (ISA) list of Louisiana trees are shown. The Delphi Method and Test of Agreement During the first round of the Delphi, experts were also invited to comment on their rationale for the rating and add additional species considered to be important and included in the list. The results were compiled, and basic descriptive statistics, including range, average, and median percent rating for each tree species were calculated. During the second round, the experts were asked to review the species rating results from the first round of the Delphi. Experts were asked to reconsider their respective ratings with reference to the ratings of the other experts. The degree of agreement as measured by Kendall coefficient of concordance (W) among the 15 professionals in assigning the plant species among the five classes was determined for the results of each round. Results and DiscussionSpecies Experts Failed to Assign Value Rating The number of species each expert failed to assign a value rating reflects the first problem identified above. Two of the experts, a forester and a horticulturist, rated every species on the entire list. The former works with the NRCS Plant Data Center, and the latter is the oldest expert in the group. Their familiarity with the species could be explained by their work and experiences. Overall, the arborists failed to assign between 10% and 48% of the species; the foresters, between 14% and 31%; the horticulturist, 1% and 49%; and the landscape architect, 11%. This clearly indicates the differences in familiarity of the species by the experts. Species Rating Values Collected Table 1 illustrates problem 2. The assigned species rating values varied substantially among the professional groups. Species values ranged from 40.0% to 95.8% for foresters, 30.8% to 88.6% for horticulturists, and 10.0% to 95.0% for consulting arborists. Six of the 10 highly rated tree species are common to the three professional groups. These six species are Ilex opaca, Quercus alba, Quercus falcata var"pagodaefolia", Quercus shumardii, Quercus virginiana, and Taxodium distichum. In contrast, only two of the 10 tree species which rated the least favorable are common to all three professional groups; these are Salix nigra and Morus rubra.
Agreement Among Professions Each species was assigned to a class based on the rating by each expert. In the Delphi rounds, the degree of agreement among the three professional groups as measured by W (Tables 2 - 5). Overall, W was 0.55. Yet, within groups, the highest degree of agreement was obtained from the horticulturists, (W = 0.77), followed by the foresters (W = 0.60) and arborists/landscape architect (W = 0.48). A highly significant value of W might also imply that the experts (within the same professional groups) applied similar standards in assigning value to the species under study. These standards as listed in the instructions provided them from the CTLA Guidelines.
Another striking aspect is the degree of agreement indicated by the Kappa statistic (K) (Table 5). K measures the quality of agreement among the groups as to the species assigned to each of the five classes. With K= 0.54, there is moderate agreement among the professional groups over the species assigned to Class 1. However, there is only a fair agreement among the groups over the species assigned to Classes 2, 3, and 4, with K = 0.36, K = 0.27, and K = 0.25, respectively. The poor agreement for Class 5 (K = 0.16) reflects strong differences of opinion on what species should be included in this category/class. Testing W The mean difference in W between rounds is positive (0.19), and the probability of the difference occurring by chance is 0.0218, assuming pooled and equal variances, and 0.0271 under unequal variances. The improvement in W between round 1 and round 2 of the Delphi was highly significant (p = 0.0284). Comparison Between Species Rating from Delphi and the Guide Table 6 shows the species ratings of the ISA Louisiana trees by each professional group and the Guide. The mean species rating [2.88, (0.1857)] based on the Guide was compared with the mean species rating obtained from each professional group. The results (Table 7) show that the mean species ratings obtained through our method were significantly lower than those provided in the Guide. This may be a reflection of the appreciation of the local values of species compared with species values at the regional level.
Conclusions and RecommendationsIn this article, our analysis reveals that the Delphi method as a means for developing species value ratings is workable. The good degree of agreement among members of different professional groups in rating individual landscape trees supports the reliability of the method for developing species rating. Committees for developing species rating can be constituted using Cooperative Extension members of the same or different tree or natural resource-related professions without fear that this will unduly influence the outcome of the species ratings. However, in light of our findings we recommend the following to avoid some of the difficulties others might face in developing a species rating for a state:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||