STATISTICAL CONSIDERATIONS IN THE DESIGN AND ANALYSIS OF PAIRED-CHOICE ASSAYS

In paired-choice assays, two treatments are presented simultaneously t o each subject. Data from such arrays should not be considered to be i ndependent, and correct statistical-analysis must account for the corr elation. A statistical test that often is appropriate for these assays is the paired-sample t-test. I present curves showing the extent to w hich statistical power of this test is affected by sample size, effect size (i.e., magnitude of treatment differences), and correlation. For a given effect size and replication, positive correlation between pai red observations substantially improves power of the test, whereas neg ative correlation reduces power. I conducted a literature sun ev of pa ired-choice assays to determine whether there are patterns in effect s izes and correlation that might assist in designing studies or in pred icting minimum sample sizes necessary to achieve reasonable statistica l power; experiments were categorized according to whether they were f eeding or oviposition assays. The review indicated that correlation wa s highly variable and ranged between strongly negative and strongly po sitive values. Oviposition assays showed larger positive correlations than did feeding assays, resulting in larger effect sizes (adjusted fo r correlation); however, feeding assays tended to use larger sample si zes than oviposition assays, hence estimated statistical power was sim ilar between the two types of assays. Oviposition assays often used mu ltiple insects per arena, apparently sacrificing replication, whereas feeding assays tended to use a single insect per arena. Approximately 45% of experiments failed to detect significant treatment effects. The majority of nonsignificant assays had too few replications to detect even a large effect size with a reasonable statistical power. Literatu re examples are presented to show that assay methodology (specifically number of insects per arena, distance between paired choices, and ass ay duration) can affect correlation, effect size, and statistical powe r. Finally, scatter plots of Che data, although rarely presented, are shown to provide insight into methodological, statistical, and biologi cal aspects of paired-choice assays.