Rank-Based Test Procedures for Interaction in the Two-Way Layout With One Observation Per Cell
ABSTRACT New aligned-rank test procedures for the composite null hypothesis of no interaction effects (without placing restrictions on the two main e#ects) against appropriate composite general alternatives are developed for the standard two-way layout with a single observation per cell. Relative power performances of the two new aligned-rank procedures and existing tests due to Tukey (1949) and de Kroon & van der Laan (1981) are examined via Monte Carlo simulation. Extensive power studies conducted on the 56and59two-way layouts with one observation per cell show superior performance of the new procedures for a variety of interaction e#ects. Simulated critical values for the new procedures are provided in settings where the numbers of levels for each of the factors is between 3 and 9, inclusive.
Hartlaub, B. A., Dean, A. M., and Wolfe, D. A. Rank-Based Test Procedures for Interaction in the Two-Way Layout With One Observation Per Cell. The Canadian Journal of Statistics 27, no. 4 (1999): 863-874.
The Canadian Journal of Statistics