Reliability refers to the consistency or stability of measurement. The reliability evidence presented here assesses the extent to which items within a scale are internally consistent or homogenous and the extent to which results are similar across periods of time or different forms of the NSSE survey. Use of a reliable instrument or scale implies that data and results are reproducible.
Internal consistencyDo the items within a scale correlate well with each other?
Internal consistency is the extent to which a group of items measure the same construct, as evidenced by how well they vary together, or intercorrelate.
- NSSE Engagement Indicators (by class): 2016, 2015, 2014, 2013
- NSSE Engagement Indicators (by sex, major, Carnegie): 2016, 2015, 2014, 2013
- NSSE Benchmark Inter-item correlation tables: 2011, 2010, 2009, 2008
- NSSE Deep Learning Scales: 2011, 2010
- NSSE Gains Scales: 2011, 2010
Temporal stabilityHow stable are the results for institutions and students upon repeated administrations?
Temporal stability, as the name implies, refers to the consistency of scores over time, as evidenced by the correlation of the score on two occasions. This can be measured at the student level where the students complete an instrument twice and achieve consistent scores, or at the institution level where institutions receive aggregate results from one year to the next.
EquivalenceDo results correlate well with those of a similar measure on the same population?
Equivalence reliability is measured by the correlation of scores between different versions of the same instrument, or between instruments that measure the same or similar constructs, such that one instrument can be reproduced by the other.