Skip to main content
ARS Home » Northeast Area » University Park, Pennsylvania » Pasture Systems & Watershed Management Research » Research » Publications at this Location » Publication #181877
Title: ON THE ANALYSIS OF LONG-TERM EXPERIMENTS

Author
item LOUGHIN, T - KANSAS STATE UNIV
item ROEDIGER, M - UNIV OF MINNESOTA
item MILLIKEN, G - KANSAS STATE UNIV
item Schmidt, John

Submitted to: Journal of the Royal Statistical Society, Series A (Statistics in Society)
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/3/2006
Publication Date: 1/9/2007
Citation: Loughin, T.M., Roediger, M., Milliken, G., Schmidt, J.P. 2007. On the analysis of long-term experiments. Journal of the Royal Statistical Society, Series A (Statistics in Society). 170(1):29-42.

Interpretive Summary: Long-term experiments are commonly used tools in agronomy, soil science, and other disciplines for comparing the effects of different treatment regimes over an extended length of time. Periodic measurements, typically annual, are taken on experimental units and are often analyzed using customary statistical tools and models for repeated (in time) measurements. An appropriate statistical model attempts to account for the variability introduced by every possible influence on the observed response. The currently used repeated measurements’ models contain nothing that accounts for the random environmental variations that typically affect all experimental units simultaneously and can alter treatment effects. This added variability can dominate variability from all other sources, and can adversely influence the results of a statistical analysis and interfere with its interpretation. The impact this has on the standard repeated measures analysis can be quantified using an alternative model that allows for random variations over time. This model, however, is not useful for analysis because the random effects are confounded with fixed effects already in the repeated measures model. Possible solutions are reviewed and recommendations are made for improving statistical analysis and interpretation in the presence of these extra random variations.

Technical Abstract: Long-term experiments are commonly used tools in agronomy, soil science, and other disciplines for comparing the effects of different treatment regimes over an extended length of time. Periodic measurements, typically annual, are taken on experimental units and are often analyzed using customary tools and models for repeated measures. These models contain nothing that accounts for the random environmental variations that typically affect all experimental units simultaneously and can alter treatment effects. This added variability can dominate that from all other sources, and can adversely influence the results of a statistical analysis and interfere with its interpretation. The impact this has on the standard repeated measures analysis are quantified using an alternative model that allows for random variations over time. This model, however, is not useful for analysis because the random effects are confounded with fixed effects already in the repeated measures model. Possible solutions are reviewed and recommendations are made for improving statistical analysis and interpretation in the presence of these extra random variations.