The Incest Study
In any study which needs to secure quantities of data from human subjects, there is no way except to win their voluntary cooperation through the establishment of that intangible thing known as rapport.
-- ALFRED KINSEY, WARDELL POMEROY, and CLYDE MARTIN, 1948.
Many of the findings reported and discussed in this book are unprecedented, shocking, and deeply disturbing. Others are less dramatic but interesting nonetheless. What makes them of particular value is the methodology by which they were obtained. The value of a study depends greatly on the soundness of its methodology. It is true that a study's methodology can be excellent and its findings of little significance. But it is hardly possible for significant findings to result from a study with poor methodology. It is the methodology of our survey that sets it apart from all previous studies and gives such weight and significance to its findings.
The fact that most prior research and writing on incest is based on highly unrepresentative samples has already been mentioned. However, a few other studies that use probability samples have been undertaken in the past few years. Unfortunately, they rarely separate incestuous abuse from extrafamilial child sexual abuse. Nevertheless, these surveys will be discussed at some length in chapter 4. One survey that does differentiate incestuous abuse from nonincestuous abuse will be discussed here, because it provides a dramatic example of why sound methodology is so crucial, particularly when researching a taboo topic like incest. It also shows that employing a random sample does not guarantee sound findings.
In 1983 Dean Kilpatrick and Angelynne Amick had Louis Harris and Associates conduct telephone interviews with a random household sample of 2,004 adult women in Charleston County, South Carolina ( 1984). Their