DEWEY DEFEATS TRUMAN and Cancer Statistics

Extract

DEWEY DEFEATS TRUMAN was a famous headline in a 1948 issue of the Chicago Daily Tribune ( 1 ). Truman's 4.4-percentage point victory contrasted with preelection polls predicting Dewey by 5–15 percentage points. Several prominent pollsters including Gallup used quota sampling (asking until one gets a certain number of respondents from certain groups) instead of probability sampling (asking people who are randomly chosen from some list). The polling was done by telephone. In 1948, Dewey Republicans were more likely to have a telephone than generally less affluent Truman Democrats. Although quota samples favored Dewey, a probability sample predicted Truman's win. Thus, it is important to be sure that the techniques and methods used to create a sample from a population actually provide an accurate proportional representation of the population parameter under study.

Similarly, deviations from chance selection of a sample can weight the results in favor of cancer survivors or people without cancer when estimating cancer risks in carriers of cancer-related gene mutations. Cancer patients are more likely than people without cancer to die and to be lost to follow-up. This loss may bias the sample even though the sample is large. For example, BRCA1 and BRCA2 mutations increase the risk for early-onset ovarian cancer. According to Surveillance, Epidemiology, and End Results data for the general population, 62% of ovarian cancers are diagnosed after metastasis has occurred and survival is only 28.2% ( 2 ). Early deaths from ovarian cancer in mutation carriers may appear to lower cancer incidence if records from these early deaths are overlooked.