Abstract
From plant and animal breeding studies to industrial applications, the intraclass correlation coefficient (p) is used to measure the proportion of the total variation in the responses that may be attributed to a particular source. Confidence intervals for p are used to determine the optimal allocation of experimental material in one-way random effects models. Assuming the sample size is fixed, the authors investigate the number of groups and the number of observations per group required to minimize the expected length of confidence intervals. Examples are used to illustrate the selection of the best design. Both asymptotic and exact results suggest that practitioners should allocate no more than four experimental units per group.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Recommended Citation
Burch, Brent D. and Harris, Ian R.
(2002).
"ESTIMATING INTRACLASS CORRELATION: OPTIMAL RESULTS USING LIMITED RESOURCES,"
Conference on Applied Statistics in Agriculture.
https://doi.org/10.4148/2475-7772.1199
ESTIMATING INTRACLASS CORRELATION: OPTIMAL RESULTS USING LIMITED RESOURCES
From plant and animal breeding studies to industrial applications, the intraclass correlation coefficient (p) is used to measure the proportion of the total variation in the responses that may be attributed to a particular source. Confidence intervals for p are used to determine the optimal allocation of experimental material in one-way random effects models. Assuming the sample size is fixed, the authors investigate the number of groups and the number of observations per group required to minimize the expected length of confidence intervals. Examples are used to illustrate the selection of the best design. Both asymptotic and exact results suggest that practitioners should allocate no more than four experimental units per group.
