Measurement Error Models

Spring 2019

Prerequisites: SURV623 or SURV630; or have equivalent survey research experience. And must have completed a basic statistics course in regression modeling

Description: Measurement error in survey data can significantly distort analyses of substantive interest. Means, totals, and proportions will be off if the average answer people give is inaccurate. However, measurement error distorts not only estimates of means but can also severely bias apparent relationships, conditional probabilities, means differences, and other regression-type analyses. To remove such biases it is therefore essential to estimate the extent of measurement error in survey variables. This can be done using a gold standard or, in the absence of such a standard, modeling the error. This course introduces the latter and trains students to perform regression analyses without the influence of measurement error.