Chapter 7 Flashcards
Constructs
An abstract idea or concept formed in a person’s mind
Concrete –> anything that can be seen and measured
Abstract –> feeling/not directly measurable
Likert scales - ORDINAL
Asks respondents to indicate the extent to which they agree/disagree with a series of mental belief/disbelief statements
Can use a 5 or 7 scale descriptor
Semantic differential scale - ORDINAL
A unique bipolar scale format that captures a person’s attitude or feelings on a given object (NOT a yes/no–dichotomous)
Use an even-point or forced choice scale point format and incorporate a non-applicable response
Constant-sum scale
Asks respondents to allocate a given number of points based on the importance of product features to the respondents
Single vs. multi-item scale
Single: collects data about only one attribute of the object (ex: how old are you?)
Multi-item: collects data on several attributes of an object
Nominal scale
Uses questions requiring respondents to provide only some type of descriptor as the raw response (NO #’s)
Ordinal scale
Allows a respondent to express relative magnitude between the answers to a question, ranking questions
Interval scale
Demonstrates absolute differences between each scale point
Ratio scale
Allows the researcher not only to identify the absolute differences between each scale point but also to make comparisons between the responses
Enables a “true natural zero” or “true state of nothing” response
Double-barreled questions
Includes two or more attributes in the same question, but responses allow comments on a single issue (How warm and cozy does that sweater make you feel?)
Leading/loading questions
Leading: influences the respondent’s answers (In general, how healthy are you?)
Loading: suggests a socially desirable answer or involves an emotionally charged issue (So have you always had a drinking problem?)–makes an assumption
Double negative questions
Contains two negative thoughts in the same questions (Does it seem impossible that the moon landing never took place?)
Scale responses
Should be mutually exclusive and not overlap
Scale measurement
Assign a set of scale descriptors to represent the range of possible responses to a particular question
Miselading scaling formats
Cannot measure the impact of non-sampling errors