Scoring & Interpreting

Question 1

Rasch Analysis

Answer 1

Each items has a degree of complexity/difficulty

Reduces bias of observation

Stabilizes reliability and validity

Rasch score derived from
various studies
consistency ratings

Question 2

Berg Balance Assessment

Answer 2

14 Items; 56 max points

1. Sitting to Standing
2. Standing Unsupported
3. Sitting Unsupported
4. Standing to Sitting
5. Transfers
6. Standing with eyes closed
7. Standing with feet together
8. Reaching forward with outstretched arm
9. Retrieve object from floor
10. Turning to look behind
11. Placing alternate foot on stool
12. Placing alternate foot on stool
13. Standing with one foot in front
14. Standing on one foot (hold >10 sec)

Question 3

Likert Scale

Answer 3

ex: Short Sensory Profile

Question 4

Item Discrimination

Answer 4

Differentiates the items intended to be measured
Mastered versus not mastered

Appropriate distractors

Discrimination power

Extreme group validation
High scorers versus low scorers

Cross-validation
Outside group compared to original group

Question 5

Item Analysis

Answer 5

Difficulty of items:

Appropriate difficulty
Avoidance of clues or defects
Intended function
Effective distractors

Information gained from wrong answers

Spread of difficulty
50%
80-90% mastery

Question 6

3 Types of Scores

Answer 6

Raw score:
Derived: ex: IQ is converting by inputting raw data into formula
Standard:

Question 7

Why can’t raw data simply be used?

Answer 7

Standard score formula: (raw score - x=MEAN)/SD (std.dev)

need to be compared to the norm

Too many variables may be involved: each task included in an assessment

Question 8

T-scores

Answer 8

Avoid the negative numbers

Can calculate from the z-scores

10(z-score)+50 = T-Score

Become the same distribution as a z-score with large sample size

Question 9

Errors in test

Answer 9

Intra-individual error: physiological factors (ex: no sleep, distraction, etc.)

Standardization: tool must be used on the reference group

Administration errors: ex: wrong setting (too loud)

Scoring errors: calculations

Question 10

Scoring Errors

Answer 10

Generosity or Severity (scoring too generously or severly); Central tendencies: midpoint scores (average score)
Halo Effect: Examiner’s impression
Logical errors: evaluators prior knowledge may hold bias
Proximity errors: unexpected events; already accounted for during standardization
Ambiguity: interpretation differs from typical evaluators’ interpretations
Contrast errors: evaluator’s subjective response