Chapter 12: Understanding Health Measurement Scales Flashcards
What is a scale?
An ordered system based on a series of questions or items, resulting in a score that represents the degree to which a respondent possesses a particular latent trait (attitude, value, or characteristic).
Should be unidimensional, representing a single construct.
Generic or related to specific patient groups.
What is a summative scale?
Total score is obtained by adding values across a set of items.
Items on the scale are generally on an ordinal scale.
All items contribute equally to the total score.
Types of formats may include:
- Likert scales
- Visual analog scales
What is a likert scale?
Ordinal scale (e.g.,disagree to agree) Range from low to
high
- Odd number (1-5), with neutral point
- Even number (1-6) forces choice
A commonly used format for summative scales is the Likert scale, in which individuals choose among answer options that range from one extreme to another to reflect attitudes, beliefs, perceptions, or values. Likert scales typically include five categories, such as: strongly disagree (SD), disagree (D), neutral (N), agree (A), and strongly agree (SA)
Describe the visual analog scale.
Fixed at 100 mm (10 cm) in length
- Measured in mm but underlying measure is ordinal
Anchors at extremes (best to worst)
Frequently used to measure the intensity of
pain from a patient’s perspective
Also used to measure other constructs such as
stress, anxiety, disability, fatigue, quality of life
What are the variations of the visual analog scale?
Numeric Rating Scale
- Example: rating pain from 0 to 10
- An ordinal scale
Faces Rating Scale
- Example: use of different faces to represent aspects of pain
- Developed for use in children but may be used with adults
What is a cumulative scale?
Items reflect increasing severity of the characteristic being measured
Each item is dichotomous
Total cumulative score based on the number of items the individual agrees with
Maximum score is the number of items
Different individuals with the same score should have the same level of ability
Also known as Guttman scales
What is a Rasch Analysis?
The scales described thus far are based on classical measurement theory, whereby a trait is assessed by getting a total numerical score that represents a person’s overall performance. Most scales include dichotomous or ordinal values that are summed, and this “summary” score is treated as a continuous measure for purposes of assessing the degree to which an individual exhibits a certain trait.
Based on Item Response Theory
What is the Person-Item Map fro Rasch Analysis?
Person ability and item
difficulty should match
- Person ability on left
- Item difficulty on right
Measured in logits
Can be converted back to
original scale scores
Allows comparison across
instruments for same
trait
What is computer adaptive testing?
Based on Item Response Theory
Designed to adjust the difficulty of items to
match the ability of the test taker
Used for standardized testing
Can be used to reduce the number of items
administered with a health status questionnaire
What is the Item Response Theory
Item response theory (IRT) is a different measurement approach that addresses these shortcomings based on two important premises.
First, scores should represent the amount of a unidimensional trait so that all persons with the same level of ability should get the same score.
Second, the probability that an individual will “pass” an item should increase as an individual’s ability increases
What is differential item functioning?
Differential item functioning (DIF) refers to potential item bias in the fit of data to the Rasch model. It occurs when there are differences across subgroups within a sample who respond differently to individual items despite their having similar levels of the latent trait being measured.
