Abstract
Greater awareness of potential environmental problems has created the need to monitor total organic carbon (TOC) and extractable phosphorus (P) concentrations at a regional scale. The probability distribution of these soil properties can have a significant effect on the power of statistical tests and the quality of inferences applied to these properties. The objectives of this study were to: (1) evaluate the probability distribution of TOC and extractable P at the regional scale in three Major Land Resource Areas (MLRA), and (2) identify appropriate transformations that will result in a normal distribution. Both TOC and extractable P were non-normally distributed in all three MLRAs. Suggested power transformations did not result in normality, but a natural log and negative binomial transformation did produce distributions that met the assumptions of normality in most cases. Statistical analysis of TOC and extractable P data at the regional scale will need to take into account the non-normal distribution of these properties for accurate and precise estimates.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Recommended Citation
Brejda, John J.; Meek, David W.; and Karlen, Douglas L.
(1999).
"CHARACTERIZING THE STATISTICAL DISTRIBUTION OF ORGANIC CARBON AND EXTRACTABLE PHOSPHORUS AT A REGIONAL SCALE,"
Conference on Applied Statistics in Agriculture.
https://doi.org/10.4148/2475-7772.1261
CHARACTERIZING THE STATISTICAL DISTRIBUTION OF ORGANIC CARBON AND EXTRACTABLE PHOSPHORUS AT A REGIONAL SCALE
Greater awareness of potential environmental problems has created the need to monitor total organic carbon (TOC) and extractable phosphorus (P) concentrations at a regional scale. The probability distribution of these soil properties can have a significant effect on the power of statistical tests and the quality of inferences applied to these properties. The objectives of this study were to: (1) evaluate the probability distribution of TOC and extractable P at the regional scale in three Major Land Resource Areas (MLRA), and (2) identify appropriate transformations that will result in a normal distribution. Both TOC and extractable P were non-normally distributed in all three MLRAs. Suggested power transformations did not result in normality, but a natural log and negative binomial transformation did produce distributions that met the assumptions of normality in most cases. Statistical analysis of TOC and extractable P data at the regional scale will need to take into account the non-normal distribution of these properties for accurate and precise estimates.
