L11 - Conjoint measurement analysis

Question 1

Goal of the conjoint measurement analysis

Answer 1

Identify and measure the attributes and attribute levels that people care most about ( that have the strongest impact on preferences)
–> decompositional method

Question 2

What is the principle of the conjoint analysis?

Answer 2

Create options presenting combinations of different attribute levels and measure the preferences for different options; then infer the influence of the different attributes and attribute levels.

Question 3

Example conjoint analysis

Answer 3

Estimating the influence of individual product features on preferences.

Question 4

What are the steps in the conjoint measurement analysis?

Answer 4

Question 5

3 key assumptions of conjoint analysis

Answer 5

• preference is based on a combination of all attributes (they are CONsidered JOINTly)
• attributes can compensate each other
• the attributes are independent

Question 6

Is relevance important when selecting the attributes?

Answer 6

Yes. The attribute must be relevant for people’s preferences (-> identification of attributes using focus groups)

Question 7

Is control important when selecting the attributes?

Answer 7

Only choose attributes and levels that can be modified and realized during product design.

Question 8

Is independence important when selecting attributes?

Answer 8

Yes the attributes should be independent.

Question 9

Is compensatory important when selecting the attribute?

Answer 9

Yes. The attributes can compensate each other (e.g. a reduction of the calorie content can be compensated by an improvement in taste)

Question 10

Is an exclusion criteria allowed when selecting attributes?

Answer 10

No. The attribute levels should not be exclusion criteria (“knock-off criteria”).

Question 11

Do the number of attributes and attribute levels have to be limited?

Answer 11

Yes.

Question 12

How to calculate the number of profiles?

Answer 12

Question 13

What is a full factorial design?

Answer 13

When you list all the profiles (combinations of attributes and attribute levels). This can result in too many options for the customer.

Question 14

How can you reduce the number of combinations of the profiles?

Answer 14

Fractional design with latin square. -> You can maximize information without considering every option.

Question 15

What is the criteria you have to fulfill so that you can apply the fractional design?

Answer 15

Only if no interaction between the attributes is to be expected.

Question 16

What is the latin square restricted to?

Answer 16

Situations with 3 attributes and 3 levels.

Question 17

What is the principle of the latin square?

Answer 17

Each attribute level is combined once with each other attribute levels of the other attributes.

Question 18

What is another fractional design method?

Answer 18

Orthogonal design?

Question 19

When do you use the orthogonal design?

Answer 19

When you have more than 3 attributes and levels.

Question 20

What is the assumption of the orthogonal design?

Answer 20

That no interactions between attributes are to be expected.

Question 21

What is the goal of the orthogonal design?

Answer 21

That the attributes are uncorrelated. It is based on a simulation methods.

Question 22

Ordinal methods for preference assessment

Answer 22

Question 23

Metric methods for preference assessment

Answer 23

Question 24

What is the ranking method?

Answer 24

Please rank the 9 options according to how attractive you find them.

Question 25

What is pair comparison?

Answer 25

Please indicate which option you find more attractive

Question 26

What is the rating scale?

Answer 26

Please indicate how attractive you find option A on a scale from 1 to 10

Question 27

What is the dollar metric?

Answer 27

How much more would you pay for option A relative to option B
or
How much would you pay for option A

Question 28

Constant sum method

Answer 28

Distribute 100 points across the options such that a more strongly preferred option receives proportionally more points.

Question 29

What is the coding matrix about?

Answer 29

The presence of absence of attribute levels

Question 30

What is the part-worth matrix?

Answer 30

How important attribute levels are for a person.

Question 31

How can you use the results of a conjoint analysis?

Answer 31

• It can have implications for existing products
• simulation of preferences for novel products

Question 32

Predicting utility of other profiles

Answer 32

4.75-1(-1)+1.7-1-.70+-21-.51-.25-1

Question 33

What do we estimate with the logit model in the choice-based conjont analysis?

Answer 33

probability that a particular option is chosen.