The Inherent Aggravation of Aggregation: An Example with Elk Aerial Survey Data
| Autor: | Kang-Tsung Chang, Peter Zager, J. Michael Scott, Edward O. Garton, Leona K. Svancara, Michael W. Gratson |
|---|---|
| Rok vydání: | 2002 |
| Předmět: |
Variables
Geographic information system Ecology Aerial survey business.industry media_common.quotation_subject Spatial distribution Regression Modifiable areal unit problem Variable (computer science) Geography Statistics General Earth and Planetary Sciences Survey data collection business Ecology Evolution Behavior and Systematics Nature and Landscape Conservation General Environmental Science media_common |
| Zdroj: | The Journal of Wildlife Management. 66:776 |
| ISSN: | 0022-541X |
| DOI: | 10.2307/3803142 |
| Popis: | One of the primary challenges in spatial analysis is the sensitivity of statistical results to the definition of spatial units over which data are collected. In the geographical literature, this sensitivity is known as the modifiable areal unit problem (MAUP), and results have been shown to vary with both the level and the configuration of aggregation. We explored the effect of the MAUP on statistical relationships between elk (Certus elaphus) recruitment and 3 independent variables when game management units were aggregated to 3 different levels in 3 configurations. Changes in variances, correlation coefficients, regression parameters, and model fit were inconsistent across aggregations and as dependent on the particular variable of interest as on the level or configuration of aggregation. While this appears to be I of the first research efforts demonstrating effects of the MAUP in wildlife survey data, more studies are likely to be challenged by the MAUP due to the significant increase in broad-scale research and regional management. We pose fundamental questions that must be considered before conducting spatial analyses of aggregated data. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|