Clinicians and researchers are often interested in studying outcomes which may be hard to measure (e.g. patient satisfaction, quality of life, depression, stress, etc.). However, these constructs are often impossible to measure directly. As a result, statisticians have developed methods for utilizing multiple measurable items (such as questions on a survey) to assess these difficult to measure constructs (also known as latent variables). These methods are known as “factor analysis”. The lecture will discuss Exploratory Factor Analysis, Confirmatory Factor Analysis, Internal Consistency and an introduction to the concept of Structural Equation Modeling. The lecture will focus on the concepts of latent variables but will also provide practical advice about how to develop questionnaires and measures for assessing latent variables. A basic understanding of statistics is required but researchers and clinicians of all levels are welcome to attend.

Instructor:

Mark Butler, Clinical Data Analyst – Center for Healthful Behavior Change