Asking Good Questions – Week 6

Marketers measure qualities that most do not think of in terms of measurement

• Attributes of a product
• Satisfaction with a service or product
• Likelihood of a behavior
• Predicting the next behavior

Rules for assigning numbers to objects to represent quantities of attributes

• Measure the ATTRIBUTES of an object – NOT measuring the object
• Definition is broad because
  different attributes
  have different qualities that
  dictate the rules of
  how numbers are assigned

Scales of Measurement

• Nominal Scale
• Measurement in which numbers are assigned to objects or classes of objects solely for the purpose of identification
• Gender: 1=Female, 2=Males

 

Ordinal Scale

• Measurement in which numbers are assigned to data on the basis of some order (e.g., more than, greater than) of objects
• Also take on the properties of nominal scale (identification)
• Brand Preference: Rank the following mascara brands (1=Most preferred, 4=Least preferred)

   __ Coke

   __ Pepsi

   __ Mountain Dew

   __ 7 Up

Interval Scale

• Measurement in which the assigned numbers legitimately allow the comparison of the size of the differences among and between numbers
• Also take on the properties of nominal (identification) and ordinal (rank order) scales
• Brand Attitude: What is your feeling toward each of the following mascara brands (1=Negative, 5=Positive)?

   Coke                 1  2  3  4  5

   Pepsi                 1  2  3  4  5

   Moutain Dew   1  2  3  4  5

   7 Up                   1  2  3  4  5

Ratio Scale

• Measurement that has a natural, or absolute, zero and therefore allows the comparison of absolute magnitudes of the numbers
• Also take on the properties of nominal (identification), ordinal (rank order), and interval (rating) scales
• Units Sold: How many of each of the following sodas did you consume in the last week?

   Coke                 _____ sodas

   Pepsi                 _____ sodas

   Moutain Dew  _____ sodas

   7 Up                  _____ sodas

How can attitudes, perceptions, and preferences be measured?

• Most of the time they are assessed via respondent self-reports
  • A communication method is employed and respondents are asked for their evaluations (i.e., attitudes) with regard to an attitude object
• Two most common scale types are both at the interval level of measurement
  • Summated-Rating (Likert) Scale
  • Semantic-Differential Scale

 

Other Considerations

• Number of Items in a Scale
• The four previous examples had four items (i.e., questions) each
• Could a global measure (e.g., an overall evaluation) have provided the same insight?
• Number of Scale Positions
• The Summated-Rating and Semantic-Differential examples had five scale positions each
• Would six have been better?  Seven?

 

Including a “Don’t Know” or “Not Applicable” Response Category

• If you are unfamiliar with First National Bank and its attributes, would the opportunity to state that these questions do not apply be helpful?

Reverse Scaling

• The Summated-Rating example had positively worded items
• Positive: courteous service, convenient location, convenient hours, low-interest rate loans
• What if we used negatively worded items?
• Negative: discourteous service, inconvenient location, inconvenient hours,  high-interest rate loans

Errors

• Systemic Error or constant error – error the impacts the measurement in a consistent manner
  • Responses concerning satisfaction are consistently higher when taken via phone interview compared to email.
• Random Error – error due to temporary aspects of a person’s situation
  • Based on mood, fatigue or most recent encounter.

Validity and Reliability

• Higher the errors – Lower Validity – Less Reliable
• Lower the errors – Higher Validity – More Reliable