DIGITALNA ARHIVA ŠUMARSKOG LISTA
prilagođeno pretraživanje po punom tekstu
| ŠUMARSKI LIST 11-12/2011 str. 25 <-- 25 --> PDF |
Š. Pezdevšek Malovrh, D. G. Hodges, B. Marić, M.Avdibegović: PRIVATE FOREST OWNER EXPECTATIONS ... Šumarski list br. 11–12, CXXXV (2011), 557-566 – Because the majority of private forest owners in Slovenia and Bosnia-Herzegovina neither play an important role in national forest policy processes nor have a strong economic interests in managing their forests, the target population consisted of all private forest owners, not only the so called “active” ones. – Personal data about private forest owners (name, address, attributes of their property, etc.) were identified from the Land and Property Register in Slovenia and from local forest authorities in Bosnia-Herzegovina. – Sample size was adequate to achieve a 5%sampling error at the 95%confidence level. Asystematic random sample was developed for the entire private forest owner population of Slovenia, although the primary focus was on selected forest ma na gement units of the Slovenian Forest Service where private forest owner associations exist. In order to identify 690 owners for interviews, owners were divided in strata according to property size (up to 0.99 ha, from 1 to 4.99 ha, from 5 to 9.99 ha, from 10 to 29.99 ha, and more than 30 ha). The sample within strata was dis proportionate to the property structure of Slovenian private forests (Forest management plans for regional units 2000–2010). Within each stratum, owners were divided into two groups of equal size: owners who were members of interest associations and those who were not. The questionnaire was pre-tested in 2007 and the survey was conducted from May 2008 through May 2009.The sample used in the analyses consisted of 322 owners, or a response rate of 46,6 %. (Pezdevšek Malovrh,2010bibid). The random sample for the door-to-door survey in Bosnia-Herzegovina was drawn from overlapping areas with the highest percentage of forest areas and the highest share of private forests.This ensured that the bulk of private forest owners were included.All municipalities in Bosnia-Herzegovina were ranked by these two criteria, so the most representative municipalities (five in the Republic of Srpska and four in the Federation of Bosnia-Herzegovina) were identified.In order to identify 350 respondents to be interviewed, the list of all local communities (settlements), within 9 selec ted municipalities was created. In total, 35 settlements were randomly chosenfromthe list of all local communities (settlements) within the 9 aforementioned municipalities. We cooperated closely with the public forest administration and forest guards in the field to create lists of all private forest owners in each settlement, as well as to contact 10 per settlement once they were randomly selected from the lists. The questionnaire was pre-tested in March 2008 and the survey was conducted in May and June 2008. All data were summarized by descriptive statistics, frequency distributions, and selected measures of location and dispersion (mean and standard deviation). Multivariate logistic regression (logit models) (Hosmer and Lemashow, 2000) was used to assess how certain forest owners and management characteristics were related to their expectations of interest associations, by means of the Enter algorithm (Field, 2009).The following expectations of private forest owners from their interest associations were evaluated as dependent variables: (i) silvicultural advice, (ii) harvesting advice, (iii) information about timber markets, (iv) information about legal regulations, (v) information about strengthening entrepreneurship, (vi) support of forest road construction/ maintenance and (vii) forest management training. Each dependent variable had binary (dichotomous) responses (1= owners expect service from interest association; 0= owners do not expect service from interest association). Separate logistic regression analyses were conducted for each dependent variable. The following independent variables were evaluated in logit models: (i) gender, (ii) age (less than 50, more than 50), (iii)education (primary school or less, high school, university or more), (iv) forest property size (less than 1 ha, more than 1 ha), (v) fragmentation (consolidate, fragmented), (vi) form of ownership (sole holder, joint ownership) and (vii) harvesting activities (harvest, do not harvest). Before running the analysis, we assessed the data for multicollinearity, using variation inflation factors (VIFs), which ensure that no high correlations exist when one independent variable is regressed on the other. The results of the diagnostics revealed that collinearity was not significant (noVIF exceeded 10). RESULTS AND DISCUSSION – Rezultati i diskusija Expectations of private forest owners of interest associations Očekivanja privatnih šumoposjednika od interesnih asocijacija The results show that private forest owners in both countries have very clear expectations of their interest associations. The expectations vary by country, however (Graph 1 and 2), not only because of differences in private forest owner preferences but also due to the different type of question used in country-specific questionnaires. Slovenian owners were asked ”From which institutions do you expect the services/advice (e.g., silvicultural and harvesting advice, training etc.)”, while private forest owners in Bosnia-Herzegovina were asked “Which service/advice do you expect from private forest owner interest associations”. In general, this has led to the lower absolute value of the results in Slovenia compared to Bosnia-Herzegovina. In both cases, multiple answers were obtained. |