80 likes | 129 Views
This study explores methods for analyzing completely censused population data using point pattern analysis. The analysis covers the entire study area or each unit of an array of contiguous sample units such as quadrats, examining global vs local types of point data in 1, 2, and 3 dimensions. Various types of point patterns including random, overdispersed, and underdispersed are investigated using techniques like distance to neighbor sampling and Ripley's K analysis.
E N D
Methods for analyzing completely censused population data • Entire extent of study area or • Each unit of an array of contiguous sample units (e.g. quadrats) • Global vs. Local
Types of Point data • Univariate, Bivariate, Multivariate • 1, 2, and 3 Dimensions (x,y,z)
Point Pattern Analysis • Pattern may change with scale! • Test statistic calculated from data vs. expected value of statistic under CSR (complete spatial randomness)
Types of Point Patterns • Random (CSR) • Overdispersed (spaced or regular) • Underdispersed (clumped or aggregated)
Methods • Distance to neighbor • sample • Refined Nearest Neighbor • randomization • Second-order point pattern analysis
Second-order Point Pattern Analysis: Ripley’s K “Used to analyse the mapped positions of events in the plane… and assume a complete census…”