Planning and Collecting Assessment Data

If you have created a curriculum map that thoughtfully and deliberately aligns courses to program learning outcomes, and have developed learning outcomes within these courses, it is time to plan instructional activities and assessments.

Instructional activities, both in and out of class, should reinforce learning outcomes and prepare students for assessments. Assessments are the assignments, exams, performances, reflections, etc. that you will evaluate to identify whether students have achieved the learning outcomes. They are the instruments or vehicles by which to conduct the assessment process.

Selecting or developing appropriate assessment instruments to collect data takes planning.

Tips and Considerations

  • Select methods that:
    • answer specific assessment questions
    • are credible to faculty and others who will use the results
    • provide useful information
  • Include both formative and summative assessments methods.
  • Take steps to improve the reliability and validity of your assessment instrument.
    • Conduct norming sessions among those using a common rubric.
    • Create a table of specifications prior to developing exams.
    • Review an item analysis of multiple choice questions.
  • Use multiple methods when possible. Pairing direct and indirect assessment methods are especially valuable.
    • Direct methods are when students demonstrate through their work how well they have achieved a learning outcome and provide the strongest evidence of learning.
    • Indirect methods are when students report their perceptions of how well they have achieved a learning outcome. The evidence of learning is weaker, but indirect evidence can help explain some of the assessment results seen using direct methods.
  • Establish benchmarks of learning achievement, such as an aspirational percentage of students achieving proficiency.
  • Set priorities. While multiple methods are advisable, it is not productive to try and measure everything, particularly at once.
  • Sample when appropriate. For programs with 40 or more graduates, or courses with multiple sections, random (as best as possible) sampling is probably OK for workload purposes, though be aware that your sample will likely not be large enough to be generalizable to all students. For a program with fewer than 40 graduates per year, or courses with a single section, assess all students. See what internationally-recognized higher education assessment expert Linda Suskie has to say on her blog.
  • Design assessments based on the Bloom’s level of your SLOs and a “research question.”