Naslov
My forest for me is…? Building private forest owners' objectives network

Autori
Žunić, Marijana ; Teslak, Krunoslav

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
Book of abstracts BIOSTAT 2019 - 24th International Scientific Symposium on Biometrics / Jazbec, Anamarija ; Pecina, Marija ; Sonicki, Zdenko ; Šimić, Diana ; Vedriš, Mislav ; Sović, Slavica - Zagreb : Hrvatsko BioMetrijsko društvo, 2019, 29-29

Skup
24th International Scientific Symposium on Biometrics (BIOSTAT 2019)

Mjesto i datum
Zagreb, Hrvatska, 05.06.2019. - 08.06.2019

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Biometrics, statistics, data analysis

Sažetak
The general hypothesis is that private forest owners (PFO) are multiobjective favouring both monetary and amenity values toward their forest property. Hence, this study examined how multiple PFO objectives are interrelated in a network and how they mutually interact. Network analysis offers a different conceptual interpretation of data via direct relationships between observed variables. Firstly, PFO were randomly surveyed in Forestry Advisory Services around Croatia and a sample of 442 responses was achieved. The structured face-to-face interview was chosen as the most suitable method for obtaining the informations about meanings that PFO attach to ownership. They were provided with 22 different objectives for ownership and asked to rate the importance for each item on a scale from 1 (not important at all) to 5 (very important). Secondly, undirected network structure was estimated by computing Gaussian graphical model (GGM) with LASSO regularization. After network visualisation, inference methods like edge weights and centrality indices (betweenness, closeness and strength) were assessed. Also, community analysis examined if the network contains clusters. The network robustness were checked by non-parametric bootstrapping (edge weight bootstrap and subset bootstrap). All statistical analysis were performed in R programming software within the corresponding packages qgraph, bootnet, EGA and lasso. Lastly, results showed a strongly connected network of PFO objectives. A community analysis revealed three larger clusters of strongly related items which were labeled as conservation and recreation, economic interests and family heritage. Also, clusters are connected by bridges implying a strong global connectivity of the PFO objectives network. Bootstrap analysis confirmed that centrality measures for betweenness and closeness are not stable under subsetting cases, while order of nodes strength are interpretable (r > 0.5). Therefore, bearing in mind strength centrality, the most central node is: My forest is an object for nature protection and conservation. This study provides a new insight into a complex nature of PFO objectives representing them within a strongly connected network structure.

Izvorni jezik
Engleski

POVEZANOST RADA