Abstract:
Latin Square Design is a square design where the number of treatments is the same as
the number of rows or columns. If there are more treatments than the number of rows
and columns, the Latin Square Design cannot be used, the Youden Square Design is the
correct design. In the experimental design there are several things that can make a data
outlier, namely the failure of an observation in one of the experimental units or errors in
data collection which can have a separate impact on the analysis process. If there are
outliers, the data must be estimated or repeated trials. In this study, outliers were
overcome by estimating the data and replacing it with new data by removing the outlier
values in the experimental data and treating them as missing data. Then, analysis of
variance was carried out again and the study shows that there is a significant effect of
the treatment observed.
