ABSTRACT: |
| Statistical methods usually used in agricultural
assays of grouping
means present ambiguity of the results. The present work had as objective
to develop a system to analyze and separate means using the Scott-Knott
test, besides Tukey and Duncan tests. The SASM-Agri was developed with Borland
Delphi to platform PC, compatible with Windows 98 or above, and allows to
import data and to export results for other software. The data set also
can be transformed before the analysis. The system was validate comparing
the generated results with results calculated in a manual way. The system
used recursive functions to increase precision and speed of calculations.
When compared to the other methods of mean separation, the Scott- Knott
method facilitates the interpretation of the results. The data were classified
in different groups and there is not overlapping among the found groups.
That overlapping is characteristic of other methods quite used as Duncan’s
and Tukey’s methods. The developed system makes possible to use the
test of Scott-Knott in the analysis of experiments with a great number of
treatments. The system is been used to sugar cane genotype screening relatively
to diseases. A trial version of the software in Portuguese is available
asking the authors or at www.infoagro.uepg.br/~ralthaus/ sasm/avaliacao/sasm-agri.exe. Keywords: multiple comparisons, statistical software, Scott-Knott test. |