Date of this Version
Designing and conducting effective research for range beef systems involves analysis of intended application of the results, identification of factors affecting variation, and selection of appropriate research methods so that precise inferences can be made. Variances associated with time, location, animal, and error in grazing research can be high. Variation due to treatment × location and treatment × time interactions is reduced by increasing the number of locations and periods tested. Random error is reduced by increasing the total number of observations. Animals, pastures, and weather are significant sources of variation in grazing studies. Factors that influence nutrient requirements or nutrient intake of cows are potential sources of variation. Amount and quality of herbage produced are highly variable within and among years and are closely related to the amount and pattern of precipitation. Vegetative measurements (e.g., cover or standing biomass) should be planned as a step in developing experimental designs and to aid in experimental layout and interpretation of the data. Vegetation sampling should be less intensive and largely descriptive in large study areas when the objectives are to measure a livestock production response and vegetation responses are considered incidental. As the priority of the objectives moves toward emphasizing plant response and the size of the study area declines, the intensity of sampling on a land unit basis increases and the need for precision increases. Generally, multiple years of study are required to address between-year variances. Experimental units and replication are key to effective experimentation. Without replication in space and(or) time, there would be no estimate of experimental error. In supplementation studies on range, experimental units are generally animals, pastures, or ranches. Animal, pasture, and ranch have advantages and disadvantages as experimental units. The advantages and disadvantages are related to hypothesis, objectives, inference, resources, number of animals, and number of treatments. When economic evaluations are part of systems research, economists should be involved in planning the experiment and formulating hypotheses. Hypotheses and interpretation of biological data may be different than for economic data. Costs need to be estimated for correct unit of output, and cost alone may be insufficient to properly rank the economic outcomes of the research.