Loading...
A review on incomplete Latin square design of any order
Author(s)
Date Issued
October 24, 2016
Type
Conference Paper
Abstract
In this paper, the aim was to introduce uninitiated readers extensive research on an incomplete Latin square design and pitfalls in methodology for solving the incomplete Latin square design. The incomplete Latin square design commonly leads to difficulties in the analysis of variance (ANOVA). Accordingly, several techniques were discussed by a number of researchers. It should be noted that the values of sums of squares in the ANOVA table by means of a missing-plot approach without adjusting the bias and an exact approach with a general regression significance test are not identical. This paper suggests to readers that the Latin square design of any order consisting of missing values should be analyzed by means of the exact approach or the missing-plot technique with adjusting the bias. A comprehensive review of the literature provided the gap for further research. It was a lack of definite formula for the ANOVA table with the exact approach for the cases of the incomplete Latin square design. It is thus of interest to provide the ready-made formula of the sums of squares by means of the exact approach in order to reduce the calculation process.
Citation
Aip Conference Proceedings, 1775, 2016