A nationally representative online survey of 2,905 respondents between the ages of 18 and 64 was designed by Consumer Reports and fielded by Knowledge Networks in May 2007. The sample excluded U.S. residents covered by Medicare or Medicaid. For the full sample, the sampling error was 2.43 percent at the 95 percent confidence level. In other words, there is a 95 percent probability that the true population values reported for the full sample are within +/- 2.43 percent of the reported survey findings. The data are statistically weighted so that respondents in the survey are demographically and geographically representative of the U.S. population of adults ages 18 to 64.

Knowledge Networks’ nationally representative online panel was sampled for this survey. Knowledge Networks’ panel is recruited using random-digit dialing procedures. Respondents who lack Internet access are provided access by Knowledge Networks to ensure that the panel is representative of the population as a whole.

The Underinsured were defined by Consumer Reports based on responses to individual survey items. The Underinsured are insured respondents who complained in our survey about two or more of the following aspects of their plans: It does not adequately cover prescription drug costs; it does not adequately cover the costs of doctors’ visits; it does not adequately cover the costs of medical tests; it does not adequately cover the costs of surgery or other medical procedures; it does not provide enough coverage for catastrophic medical conditions; the deductible is too high.

These six highly intercorrelated items comprise our Scope of Coverage Scale (Cronbach’s alpha = .82). These items were initially identified through a principal components factor analysis of all 11 survey items measuring possible complaints respondents with health plans may have had about their health plans.

We found that an additive scale of the six items that loaded highly on this factor provided a meaningful application to the understanding of differential experiences among our insured respondents. In brief, at the plateau of two complaints, our research revealed sharply divergent health care experiences among insured respondents.