You are on page 1of 5

Conjoint Analysis Exercise:

A tyre retailer wanted to study customer preference for various attributes of


tyres. The stimuli for the experiment are 22 hypothetical tyres. The stimuli
represent different brands (Goodstone, Pirogi, Machismo)*, prices ($69.99,
$74.99, $79.99), expected tread life (50,000, 60,000, 70,000), and road hazard
insurance plans (Yes, No). There are 3 ×3 ×3 ×2 = 54 possible combinations.
From these, 18+ 4(holdout) combinations are selected that form an efficient
experimental design for a main effects model. The combinations are then ranked
from 1 (most preferred) to 22 (least preferred).
Coding for Brand
1 = 'Goodstone'
2 = 'Pirogi '
3 = 'Machismo '
Coding for Price
1 = '$69.99'
2 = '$74.99'
3 = '$79.99'
Coding for life
1 = '50,000'
2 = '60,000'
3 = '70,000'
Coding for Hazard insurance
1 = 'Yes'
2 = 'No '

Prepare a conjoint plan and perform conjoint analysis. Report your results.

Answer:

It is often that the marketing analyst is asked to determine the attributes of a


product that are most and least important in determining a consumer’s product
choice.

After a product portfolio is shown to the customers and as the customers rank
the product profiles, the analyst can use the conjoint analysis to determine the
relative importance of various attributes.

The first task is to create the stimuli. Based on the stimuli, we can set up the
ranking given by the customer for the combination.
For Tyre companies, a conjoint analysis is done.
Going through, Data -> Orthogonal Design -> generate.
We have generated a file that contains 22 cases, 18 minimum number of cases
and 4 number of holdout cases.

Now the design is determined.

We have a customer preference data for the 22 cases.


Now we create the Conjoint plan.
We have received the following results.
As we can see from the
Brand, Price and Life has 3 levels and Hazard_Insurance has 2 levels.
Brand is discrete, and Price is Linear(less). Life and Hazard_Insurance has got
Linear(more). All the factors are orthogonal.

Fig.1 (Conjoint Analysis 1.0)


For the above figure (Conjoint Analysis 1.0), we can see for the table, Utilities,
the most preferred brand is Machismo (0.283) utility, highest among the three.
Similarly, the most preferred price is $69.99 with -0.525 as utility. The most
preferred life is 70,000 with 3.075 as utility and preferred Hazard_Insurance is
yes with -0.475 as utility.
Fig 2.0 (Conjoint_analysis) Fig 3.0 (Conjoint_analysis)

For Fig 2.0(Conjoint_analysis), we can see the component Important Values.


Important Values mean among the attributes, which has received the most
importance in determining the ranking.
So here, brand has received the highest with 29.006 and Hazard_Insurance the
lowest importance with 19.911 as values.
When it comes to Coefficients, Price and Hazard_Insurance is decreasing the
utility by every 1 unit. However, Life increases the utility per unit by 1.025.
Now in Fig 3.0 (Conjoint_analysis) in the table, Number of reversals, the
greatest number of reversals, we can see is for Hazard_Insurance. The people
who might have preferred the tyre to be insured are not coming as per the data.
So, it is a violation where, Hazard_Insurance (Linear(more)), should have
shown data with preferred cases is showing data that people do not prefer it to
be insured as per the rankings. Similarly, brand has zero reversals as it is a
discrete variable. In the table Reversal summary, we can see, 1 reversal has
been done by 3 subjects and 2 reversals have been done by 4 subjects.
Submitted by-

Group - 2

Abijith Puravankara
Anamitra Mandal
Varun Belgamwar
Sourav Pattnaik

You might also like