Pairs of relatives are used in the study; two brothers, a mother a daughter, opposite sexed twins. They fill out a personality questionnaire, or observer reports are done. The results are studied all together and compared. The variance is determined and the researcher checks to see if variance is due to within family variation or. This is done through a technique called anova. Intraclass correlation coefficient: to figure out how similar relatives are, researchers calculate the proportion of variance due to between-family differences. Researchers often use twins to study the nature vs. nurture debate. When they are comparing the variance in a trait between two non identical twins, non twin siblings, or mother and child, then they double the correlation found by. 2, because they only share 50% of their genes. This doubling is not done with identical twins because they share 100% of their genes.