Michael Herzog - Do we really measure what we believe we are measuring?

Title: Do we really measure what we believe we are measuring? About weak correlations in vision, ageing, and schizophrenia research
Speaker: Michael Herzog
Abstract:
Representative tests are crucial in all empirical sciences, supposed to measure the characteristics of an underlying latent variable. For example, eye doctor’s exams are thought to be representative for visual acuity in general. An implicit assumption is that tests, assumed to measure the same latent variable, such as visual acuity, correlate with each other- otherwise the tests would be useless. However, this is exactly what we found in many areas in vision, healthy aging, and schizophrenia research: “Similar” tests do not correlate with each other. Hence, tests are often less representative than thought. I will discuss what these results imply and what can be done.