Amy Arneson



Before starting my graduate study, I was a high school mathematics and statistics teacher near St. Louis, MO.  After completing my Ph.D., I hope to return to the public K-12 context and work directly with educators within school districts.

My research interests are in educational statistics and measurement.  I am currently working on projects that aim to model causal effects.  Particularly, I am concerned with issues of causal inference in the multilevel setting.  Most educational research and evaluation is done with observational studies in a naturally 'nested' structure.  These two characteristics introduce complications when defining or selecting treatment and control groups; in most cases, randomized controlled trials are simply impossible.  Thus, setting up a model and defining treatment and control groups (if applicable) without making unrealistic or untestable assumptions is of great concern when investigating the effect of, for example, a curricular intervention on student learning.  I am currently working on a qualifying paper that investigates the issue of matching in a multi-site (thus, multilevel) observational study.

As the greatest concern of most educators (including myself) is student learning, the way in which learning is measured is also of great concern.  In addition to general issues around statistical modeling, I am especially interested in measurement models and psychometrics as pertaining to K-12 mathematics assessment.  Again, it is important to select a model that reflects the reality of the situation as closely as possible, and as such, I work with models that reflect the phenomenon of 'local dependence' when necessary.  I have worked extensively in modeling item responses for assessments with a testlet/bundle structure.

My ultimate goal is to bring the power of the sophisticated statistical and psychometric research of the past decades to educators.  The largest project that I have had the opportunity to work on during my time at Berkeley is developing a tool that guides and promotes classroom teachers in using sound measurement practice to measure and track student growth (bass.bearapps.org).  I hope to continue this kind of work -- creating and distributing tools that can be used in the most practical of settings -- during and after my graduate study.

Last updated: January 2016