Experimental Design
Design
refers only to aspects of the study's conditions, treatments, groups, randomization, counterbalancing, sampling, control, and oftentimes the sequence of measures being used
Independent Variables
One-way design: having a single independent variable
within-subjects design
between subjects
levels: the different values a variable can take on
Randomization/Random Assignment
it is critical to use random assignment to place individuals in groups, to place conditions in an order, or to map different stimuli onto different conditions
Counterbalanced Designs
a fully-counterbalanced design crosses every level of one factor with every level of another factor, creating n x m total conditions
aka full factorial designs
The Latin Square
a mathematical structure that has every number exactly once in each column and in each row
is the basis for games such as Sudoku
Greco-Latin Square
a partial factorial design
using 16 conditions to do a 4x4x4 design
can examine interactions between interactions between variables
Potential Problems with Research Designs
placebo effects
practice effects
order & sequence effects
confounding variables
ceiling effects
failure to test interactions