Data analysis for repeated measures apply to a set of measurements of the same variable taken on the same experimental unit over time, where the observation periods are set by the researcher. This data structure usually causes the variances and correlations are different in time, and for this reason it is not appropriate to analyze the data using classical statistical analysis. This article explains the proper methodology using data of bacterial growth in turkey ham, using a factorial model with time-dependent random errors through a covariance structure model ANTE (1) chosen by the AIC criteria, BIC and AICC. The data processing was performed using PROC MIXED of SAS, using as a method of estimating the parameters Maximum Residual Likelihood (REML) by default, employed for mixed effects models because modeling supports the covariance structure of the data. Apply this to variable growth methodology, took us to the conclusion that the use of essential oils without film promotes the conservation of turkey ham in time.