Resumen

KIM, Beob G  y  KIM, Taemin. A program for making completely balanced Latin Square designs employing a systemic method. Rev Colom Cienc Pecua [online]. 2010, vol.23, n.3, pp. 277-282. ISSN 0120-0690.

Animal scientists employ Latin square designs to reduce the required number of animals for detecting statistical differences in animal experiments. Randomization procedures do not balance residual effects that possibly exist in Latin square experiments. A spreadsheet-based program is available for making Latin square designs balanced for the first-order residual effects. The balance of remote residual effects may also be very important to consider when a relatively long latent period exists after a treatment. We have developed a spreadsheet-based program, the Completely Balanced Latin Square Designer (CBLSD), to facilitate the generation of Latin squares (n = prime number – 1) balanced for immediate and remote carryover effects. The program allows a user to choose the number of treatments that is equal to the number of animals and periods in a square. A user may also input the number of squares. Then, the CBLSD instantly generates a Latin square balanced for the multi-order as well as first-order carryover effects. Also provided is a schedule table for an experiment sorted by period and animal. The CBLSD enables animal experimenters to quickly and accurately make Latin squares balanced for the immediate and remote carryover effects. The program is freely available upon request.

Palabras llave : animal experiment; Latin square design; remote carryover effect.