CycDesigN 6.0 NOW AVAILABLE – scroll down to read a list of new features:
CycDesigN is a computer package for generating optimal or near-optimal experimental designs. Four general classifications of designs can be constructed: these are resolvable, non-resolvable, partially replicated and crossover designs. The first three classes can be set out in blocks or in rows and columns. Resolvable designs can be latinized, t-latinized or partially latinized, while non-resolvable designs can also be unequally replicated. In addition, spatial resolvable and non-resolvable designs, using the linear variance or exponential variance models, can be constructed. Partially replicated designs have test treatments set out in a number of locations where each test treatment appears zero, once or twice in each location; standard treatments can be included with multiple replication at each location.
Crossover designs are used when sequences of treatments are applied to several subjects over a number of time periods. The direct effect of the treatment applied in the current period and the carry-over effects of the treatments applied in one or more previous periods can be modelled in various ways, and numerous options are provided in CycDesigN. Crossover designs are also known as change-over or carry-over designs.
CycAnalysis is a separate module in CycDesigN that allows output from a CycDesigN session to be tailored in a form ready for analysis. For example, a spreadsheet of the design blocking and treatment structures is automatically generated. CycDesigN also lets you generate Genstat and SAS code for the analysis of most designs.
CycDesigN provides the most comprehensive design generation package yet available for experimenters. In particular, resolvable and partially replicated designs are used by experimenters involved in field variety trials. For smaller experiments, say in the glasshouse or laboratory, non-resolvable row-column designs are often employed and can typically include spatial enhancement. Crossover designs are frequently used in areas such as clinical trials, taste and psychological testing and psycho-physical experiments. The authors are leading researchers and, as a result, the algorithms incorporate the most recent developments in the construction of experimental designs.
Please contact support for access to CycDesigN 6.0.