Angie Miller — Ever wondered about the psychometric properties of NSSE? If you are interested in learning more about our research exploring the reliability, validity, and other indicators of data quality for the survey, check out the NSSE Psychometric Portfolio. The portfolio features many studies that assess the quality of our survey and the resulting data. Each study is described in a short technical report format and links to full papers or reports are provided when possible. We repeat some studies annually to provide information for each survey administration, while others are one-time studies based on need and data availability. This information is great for those who are interested in more specific aspects of NSSE data quality.
The Psychometric Portfolio is organized into three sections: Validity, Reliability, and Other Quality Indicators. The studies in the Validity section explore, in a general sense, whether NSSE measures what it is intended to measure. One study of note in this section involves Construct Validity for the Engagement Indicators, using exploratory and confirmatory factor analysis. This particular analysis was part of the expansive testing conducted for the 2013 NSSE update.
Many of the studies included in the Reliability section focus on the consistency of the survey instrument. These measures involve both internal consistency for groups of items and consistency over time. We annually update the Internal Consistency study, which details Cronbach's ? for Engagement Indicators by class, major, and Carnegie classification.
The third section, Other Quality Indicators, covers a broader realm of questions. This section examines procedures, standards, and other evaluations implemented by NSSE to reduce error, bias, and increase the precision and rigor of the data. An example of a one-time study looks at Nonresponse Effects from 2005, relatively early on in the project. Looking at this early work can help provide a better understanding of research that informed NSSE's dedication to continuous improvement, and show how administration and reporting processes developed over time.
We welcome any feedback, questions, or suggestions about the Psychometric Portfolio!