• Assessment (esp. standardized tests) is both a political and professional issue
• Externally-mandated tests are used as both barometers and levers of educational reform
• Test results:
• Are not self-interpreting
• Come in great variety
• Serve a variety of purposes
• Fairness is always a big issue
• There are different definitions, some of which conflict
• Different definitions serve different social goals (e.g., improvement vs. equalization of school achievement)

• It is an evidence-gathering procedure for making a decisions
• The procedure can be formal or informal
• Tests are formal assessments and therefore can provide more systematic and more objective evidence. They supplement informal assessment.

1. How realistic are my plans for these students?
2. How should I group students?
3. Are my students ready for the next learning experience?
4. How well are my students meeting my learning goals for them?
5. How far are my students progressing beyond the minimum?
6. When would a review be most useful?
7. What learning problems are students having?
8. Which students need to be referred for help?
9. Which students have poor insight into their performance?
10. What grades should I assign?
11. What do I tell parents about their children’s progress?
12. How effective was my teaching??

Assessment = Any procedure (formal or informal) used to obtain information about student performance. Used to make a value judgment about learning, etc.

Test = A particular type of assessment. It generally fits this description:

1. Consists of a set of questions
2. Administered during a fixed time period (e.g., 1-2 hours)
3. Give under reasonably comparable conditions for all students

Measurement = A process of assigning numbers to individuals or their characteristics according to specified rules. This quantifies the value judgment.

NOTE: Each term is successively more limited in scope.

1. Set clear learning objectives
2. Assessments should be appropriate (relevant) for those objectives
3. A variety of objectives therefore generally requires a variety of assessment techniques
4. Know the limitations of each form of assessment (e.g., sampling error, measurement error)
5. Don’t forget that assessments are a means, not an end in themselves. That is, don’t forget their purpose.

• Signals to students what they should learn
• Provides feedback to students about their learning
• Provides feedback to teachers about student learning, appropriateness of instruction, and appropriateness of goals
• Allows both to modify their behavior to be more effective

• Maximal
  • test of the "can do" (e.g., aptitude, ability, and achievement tests)
  • right/wrong answers

• Typical
  • assessment of the "will do" (e.g., vocational interest or personality inventories)
  • no right or wrong answers

Fixed-choice vs. complex performance (format)

• Fixed-choice
  • "selection" item
  • objective
  • reliable
  • efficient
  • cost effective
  • cannot measure complex performance
  • requires reproduction only, not production

• Complex performance
  • "supply" item
  • subjective
  • not statistically reliable
  • costly in time
  • measures complex performance
  • requires production, not just reproduction

• These two are actually ends of a continuum ranging from questions requiring simple, fixed responses to those requiring free, complex answers.

Location in instructional process (purpose)

• Placement (have entry skills needed for instruction?)
• Formative (how is learning going? What snags?)
• Diagnostic (what are the causes of persistent learning problems?)
• Summative (how much was learned? What grade should be assigned?)

• Norm-referenced
  • Scores are compared to other students (e.g., rank, percentile) in a clearly defined group (e.g., same classroom, all American students of the same race and age)

• Criterion-referenced
  • Scores are compared to an external standard (e.g, basic, proficient, advanced) in a clearly defined domain of performance (e.g., reading in 8^th grade)

Other

• Informal vs. formal
• Individual vs. group
• Mastery vs. survey
• Supply vs. selection
• Speed vs. power
• Objective vs. subjective
• Formal vs. informal

1. Require specification of achievement domain to be measured
2. Require relevant, representative sample of test items
3. Use same types of test items
4. Use same rules for item writing (except item difficulty)
5. Judged by same technical standards (validity, reliability)
6. Useful in educational assessment

• Domain coverage
• Norm: Covers larger domain, meaning fewer items for each specific task
• Criterion: Covers narrower domain, with more items per task
• Focus on discrimination vs. description
• Norm: emphasizes discriminating among individuals’ learning levels
• Criterion: emphasizes describing an individual’s level of learning
• Concern with item difficulty levels
• Norm: favors items that discriminate (that is, favors items of average difficulty and eliminates very easy or very hard items)
• Criterion: matches item difficulty to learning tasks, so may include many easy items
• Standard for comparison
• Norm: Requires clearly defined group of students
• Criterion: Requires clearly defined achievement domain

Instruction and assessment are interacting partners in promoting student learning!