10th International Conference on Agricultural Statistics

10th International Conference on Agricultural Statistics

A PCA-based Rural Development Typology for Denizli and the Targeting of IPARD Support

Conference

10th International Conference on Agricultural Statistics

Format: CPS Paper - ICAS 2026

Keywords: pca, rural employment

Abstract

We analyze territorial patterns of rural development in Denizli, Turkey and the role of EU-IPARD support per capita using Principal Component Analysis (PCA) on a set of 13 district-level indicators (demographic, agro-structural, and socio-economic). Data were winsorized and z-standardized; PCA was run on the correlation matrix with varimax rotation. Sampling adequacy was acceptable (KMO = 0.562), Bartlett’s test was significant (χ²(78)=124.0, p<.001). Parallel analysis retained two components explaining 53% of variance. PC1 loads on education, industrial/agricultural structure and is negatively associated with IPARD per capita, indicating a compensatory allocation to structurally weaker districts. PC2 captures agricultural capital intensity (irrigation, livestock density, cooperative membership). Using component scores, districts were classified into the standard A(+,+), B(−,−), C(+,−), D(−,+) typology, revealing clear spatial differentiation. A complementary TOPSIS index would be used to corroborate these patterns.