Duration: 4 hours
Description: Assessment is inherently a quantitative activity. Assessment professionals need to have a reasonable grasp of a range of numerical indicators of the quality of tests and questions, and how to use those indicators to plan action.
This session will give attendees an introduction to several such statistics; explaining their bases in test theory and giving practical illustrations of how they are used.
However we take care not to delve into complex mathematics – no drowning in formulas! We are fans of Pareto’s Law – 80% of the benefit can be gained with 20% of the effort, so we work hard to keep it simple and practical. We also focus on the practical uses of the analysis to improve assessments, rather than the technical activity of undertaking the analysis.
- why bother about statistical analysis?
- norm-, cohort- and criterion-referenced assessments
- classical test theory
- other theories; IRT and generalizability
- facility and discrimination
- reliability indices
- standard deviation, variance and error
- a basic toolbox of test performance statistics
- a basic toolbox statistics for managing portfolio-based assessments
Typical Task – practical use of statistics for your awarding organisation