CH 1-3 Flashcards
Give two reasons for studying measurement
- create flexibility 2. improve communication
Define “measurement”
Measurement consists of rules for assigning symbols to objects so as to scale or classify attributes
Describe the scaling aspect of measurement
representing quantities of attributes numerically
Describe the classification aspect of measurement
defining whether the objects fall in the same or different categories with respect to a given attribute
Ordinal Scale
variables whose categories have a meaningful, hierarchical order
Interval Scale
variables whose order has meaning and the numbers used to represent the attribute have meaning; no true zero point
Nominal Scale
used to classify variables that can be placed in categories , characterized by kind
Ratio Scale
variables whose order has meaning and the numbers used to represent the attribute have meaning; true zero point
Reliability
consistency; does the instrument produce scores that are internally consistent or stable across time?
Validity
legitimacy; does the instrument measure what it is intended to measure?
What are the four primary methods of measuring health behaviors?
- Self-report 2. Observation 3. Biobehavioral measure 4. Electronic monitors
List some examples of self-report.
- Interviews 2. Questionnaires 3. Journals/Diaries 4. Response Scales 5. Indexes
List some examples of scaling methods.
- Visual Analog Scale 2. Thurstone Scale 3. Likert Scale 4. Guttman Scale 5. Semantic Differential Rating Scale
Measurement Error
the degree to which a measure deviates from the ideal level of reliability and validity
Random Error
error that is due to chance factors that influence the measurement of a variable; affects the reliability of a measurement
Systematic Error
error that is due to factors that systematically increase or decrease true scores of an attribute; affects the validity of a measure; can be additive or correlational
Additive Systematic Error
a consistent deviation from the true score in the same direction (e.g. a scale that always adds 2 pounds to the actual weight)
Correlational Systematic Error
measures consistently inflate or deflate scores but do so in different ways for different responders (e.g. some subjects answers may be consistently more positive)
Response Sets
patterns of responses or bias in responses
What are the types of response sets?
- Social Desirability 2. Acquiescence and Nay-Saying 3. End Aversion/Central Tendency 4. Positive Skew 5. Halo
Social Desirability
the tendency of subjects to attribute socially desirable traits to themselves; e.g. faking-good or faking bad
Acquiescence and Nay-Saying
the tendency of subjects to choose true over false or agree over disagree (or the opposite)
End Aversion/Central Tendency
avoidance of extremes (e.g. rarely picking strongly agree/disagree)
Positive Skew
the tendency of people to hold extreme attitudes about a topic
Halo
the tendency for rating of specific traits to be influenced by a general attitude, or set, toward a person (e.g. course evaluations)
Recall
a major source of self-report error in which the ability and motivation of respondents to provide accurate information about behaviors or events that occurred in the past
A graphic rating scale that uses a straight line and gives a numeric score based on the distance from one end of the line to the subject’s response
Visual Analog Scale
A rating scale that pre-weights items to generate a score for endorsed items
Thurston Scale
A summated rating scale that uses several items to asses a single attribute through adding individual response ratings to obtain a single score
Likert Scale
A rating scale structured so that an affirmative response to one item in a set suggests affirmative responses to other items in that set. Total score is obtained by adding all the affirmative answers
Guttman Scale
A rating scale that assesses meaning. To score this scale, response numbers are reversed and added.
Semantic Differential Rating Scale
variables whose categories have a meaningful, hierarchical order
Ordinal Scale
variables whose order has meaning and the numbers used to represent the attribute have meaning; no true zero point
Interval Scale
used to classify variables that can be placed in categories , characterized by kind
Nominal Scale
variables whose order has meaning and the numbers used to represent the attribute have meaning; true zero point
Ratio Scale
the tendency of subjects to attribute socially desirable traits to themselves; e.g. faking-good or faking bad
Social Desirability
the tendency of subjects to choose true over false or agree over disagree (or the opposite)
Acquiescence and Nay-Saying
avoidance of extremes (e.g. rarely picking strongly agree/disagree)
End Aversion/Central Tendency
the tendency of people to hold extreme attitudes about a topic
Positive Skew
the tendency for rating of specific traits to be influenced by a general attitude, or set, toward a person (e.g. course evaluations)
Halo
a major source of self-report error in which the ability and motivation of respondents to provide accurate information about behaviors or events that occurred in the past
Recall
Visual Analog Scale
A graphic rating scale that uses a straight line and gives a numeric score based on the distance from one end of the line to the subject’s response
Thurston Scale
A rating scale that pre-weights items to generate a score for endorsed items
Likert Scale
A summated rating scale that uses several items to asses a single attribute through adding individual response ratings to obtain a single score
Guttman Scale
A rating scale structured so that an affirmative response to one item in a set suggests affirmative responses to other items in that set. Total score is obtained by adding all the affirmative answers
Semantic Differential Rating Scale
A rating scale that assesses meaning. To score this scale, response numbers are reversed and added.
What kinds of factors influence measurement error?
- Respondent Factors
- Instrument Factors
- Situational Factors
