Quantitative literature summary (meta-analysis) is often used to generate a more comprehensive understanding of system behavior than can be obtained from individual experiments. Although every data set is unique and often requires some individualized analysis, most meta-analytical data can be evaluated using weighted, mixed effect, regression in a 9-part procedure, described as follows. 1) Search criteria should be clearly defined. 2) The literature should be searched and all response variables, their standard errors, and all explanatory variables should be recorded. 3) Data should be evaluated for transcription errors and outliers. 4) Missing standard errors should be estimated by error propagation, where possible. 5) Standard errors from fixed-effect regression and mixed-effect regression should be standardized to remove statistical analysis effects, and weights should be calculated from these standardized standard errors. 6) Backward, stepwise regression should be performed, using fixed effects for all explanatory variables of interest, and random effects for study, laboratory, or location, as needed. 7) After a model is identified where all variables included are below a significance cutoff defined by the researchers, the parameters removed from the model should be iteratively re-tested for significance in the final model. This step helps ensure variables were removed for non-significance rather than accidently removed due to model instability. 8) Parameter estimate correlation should be evaluated using variance inflation factors. Variance inflation factors above 10 are acceptable for parameters correlated by calculation but all other parameters should have variance inflation factors below 10. If parameter estimates have excessive correlation, the parameter with the highest variance inflation factor should be removed. 9) Researchers should iterate through steps 6 to 8 until a model is identified where all parameters are statistically significant and have acceptable covariation. Although this procedure might require adjustment for some applications, it provides a general framework for performing meta-regression analysis of animal nutrition literature. The workshop associated with this abstract will walk through this process using an example data set.