Overview of concepts and techniques
Variance is a measure of spread or scatter in a group of scores. Variance is based on the sizes of the deviations from the mean of each of the scores in the group. When the purpose of the variance calculation is to estimate the variance of a population from a small sample the formula is modified. Most analysis of variance is discussed in terms of estimates of the variance of scores obtained from variances of means. The n is a weight used to scale up the estimate from an estimate of the variance of means to an estimate of the variance of individual scores. Sensitivity is defined as the number of subjects experiencing each experimental condition divided by the variance of scores in the sample. The link with the confidence interval formula referred to means that as sensitivity reduces, the confidence interval widens, indicating that estimates have larger margins of error.