Economics Responses, Questionnaire Example

A key assumption in economics is that people are “rational.” That is, people are able to survey the economic landscape and make consumption and career decisions that integrate this information to maximize personal utility. Objections from Daniel Kahneman aside, rationality is applied in the following context to the “move rate” of physicians over the past three years: Due to a poor economic environment and uncertainty over the future, doctors have “moved” less in 2010-2011 (March to March) compared to 2008-2010; this is interpreted as a symbol of stability in that doctors do not want to either retire or move to another office due to a combination of unknown factors: lower reimbursement rates; new health reform efforts; medical liability.

There are a number of factors that might call this hypothesis into question: 1) is the sample size adequate to reflect the true market for doctors?; 2) Due to a high level of consolidation in the industry over the past three years, are doctors staying put because they are getting a better deal from new owners while keeping the same office? These questions are not answered by the surveyors’ methods.

From the question, I posit that the dependent variable is the retirement rate of physicians and the independent variable is insurance reimbursement rates. With this analytical schema, one of the main questions is how to operationalize both insurance reimbursement rates and the retirement rate of physicians from the data source.

Ideally, one would be able to know which doctor offices accepted which insurance payments (e.g., Medicaid) so that a model could be developed looking at variation among the different reimbursement rates. That is not likely possible; therefore, one could create an index of all reimbursement rates looking at the general direction year-on-year (have they increased or decreased)[1]. Teasing out retirement rates from the data will be harder: If a doctor stopped listing does that mean he/she retired; took a sabbatical; or simply didn’t respond and so we don’t know. Since the data source is not sensitive we would likely have to take the aggregate number of individuals who dropped out of the listing divided by the total number of doctors in the directory. Then we could take a year-by-year snapshot looking at the correlation between the reimbursement rate index and the doctor retirement rate.