Lectures 5 and 6 Flashcards Preview

Marketing Research > Lectures 5 and 6 > Flashcards

Flashcards in Lectures 5 and 6 Deck (23)
Loading flashcards...

Proprietary Info

Brand Name & Manufacturer


nominal scales

Mutually exclusive and collectively exhaustive categories


ordinal scales

Implies statement of greater than and less than (ranking)


interval scales

Equality of interval (Equal distance exists between


ratio scales

Absolute zero point; flat numeric data


characteristics of good measurement scales

1. Reliability: The degree to which repeated measurements yield consistent results
2. Validity: Are we measuring what we think we are
3. Sensitivity: Ability to discriminate between meaningful
differences in attitudes


scaling tips

Keep the scales relatively simple
• State instructions clearly
• A “don’t know” option, if necessary
• Have enough “white” space between scales &
• Limit open ended questions
• Pilot test the scales with groups similar to the
planned respondent group


questionnaire development

Step 1: Specify what information will be sought
Step 2: Determine type of questionnaire and method of administration
Step 3: Determine content of individual questions
Step 4: Determine form of response to each question
Step 5: Determine wording to each question
Step 6: Determine question sequence
Step 7: Determine physical characteristics of questionnaire
Step 8: Reexamine step 1-7 and revise if necessary
Step 9: Pretest questionnaire and revise if necessary


step 1

Precision at earlier stage
Dummy Tables
Interesting vs. relevant information


step 2

Questionnaire type:
Structure, disguise
Method of administration:
-personal, mail, telephone, e-mail, web…
Initial Questions: screeners/warm ups



• Helps to winnow the respondents to the most appropriate people
- Have you shopped at the Gap in the past month?
• Establish respondent “buy-in” for the survey



• Gets the respondent thinking about the topic at hand
- How often do you go shopping?


step 3

Is the question necessary?
Are several questions needed instead of one?
Do respondents have the necessary information and can
they remember it?
Will respondents give the information?


Branching question

1. Do you have a personal computer (PC) at home?
___ Yes ___ No—please go to Q.3
2. What brand of PC do you own? _________


step 4

Open-ended questions
-What do you like and dislike about brand A?

Fixed-alternative questions


Issues with multichotomous items

 Check one item vs. multiple items
 Exhaustive
 Mutually exclusive
 “Other” category
 “Don’t know” or “no opinion” option
 Order bias


step 5

Use simple and unambiguous words
Avoid leading questions
Avoid generalization and estimates
Avoid double barreled questions
Avoid implicit alternatives and assumptions


step 6

Use simple, interesting opening questions
Use the funnel approach, asking broad questions first
Carefully design branching questions
Ask for classification (e.g., demographics) information last
Place difficult or sensitive questions near the end


Transitions - First Third of Questions:

• Questions that set the tone for the more difficult questions to come
• Ordinal scale - sorting or ranking type questions
• Example: “Rank order the following products based on your preference”


Complicated - Second Third of Questions:

• Use of rating scales for attributes, attitudes, beliefs, opinion, etc
• Expands on warm-up questions with more quantifiable scales
• Example: “How likely are you to purchase the following products?
(scale 1 to 10, with 1 being least likely to buy, and 10 being most
likely to buy)?”


Classification - Last Third of Questions:

• Personal & demographic type questions
• Tackling controversial issues
• Example: “What is your household income?”


Advantages of the internet and questionnaires

 The questionnaire appearance
 The survey can be created quickly
 Skip patterns can be efficiently established
 The survey can be distributed quickly for expert review & input


Disadvantages of the internet and questionnaires

 Over reliance on electronic survey construction can lead
to the researcher’s getting sloppy as he might think “the
software will do the work and correct any errors