Scribd
Upload
461 views3 pages

How To Conduct Conjoint Analysis

This document provides instructions for conducting conjoint analysis, which includes identifying relevant attributes and levels, using orthogonal designs to reduce attribute combinations, inputting data into SPSS, generating output files, interpreting the results to determine the importance of attributes and utility values for designing prototypes.

Uploaded by

aashish guptha
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
461 views3 pages

How To Conduct Conjoint Analysis

This document provides instructions for conducting conjoint analysis, which includes identifying relevant attributes and levels, using orthogonal designs to reduce attribute combinations, inputting data into SPSS, generating output files, interpreting the results to determine the importance of attributes and utility values for designing prototypes.

Uploaded by

aashish guptha
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

How to conduct conjoint analysis

1. To develop your problem statement

2. To identify relevant attributes to measure

3. To create levels for each attribute

4. To determine number of combinations

5. To use orthogonal designs to reduce the number of combinations

6. Create multiple files in the SPSS (not at the same time, but during the technique)
a. One, for attribute combinations
b. Second, for consumer response
c. Third, for creating syntax
7. Input in the SPSS – for the first file: Attribute combinations
a. Data – Orthogonal - Generate orthogonal design
i. Write Factor name (starting with alphabet)
ii. Describe Factor label (for future reference)
iii. Add
iv. Select the factor
v. Define values for one attribute
1. Auto fill (based on the number of levels for that attribute)
2. Describe the label
3. Continue
vi. Repeat the process for remaining attributes
b. Check reset random number seed to
i. Write a random number (Remember this number for future reference)
c. Click on Options
i. Minimum number of cases
ii. Hold out cases
1. Click on number of hold out cases
2. Define numbers
3. Check randomly mix with other cases
iii. Continue
d. Create a new data file (Remember the location)
e. OK

8. Generate first SPSS output file.


a. Observe the number of cards created by orthogonal designs
b. Add number of holdout cases
c. You will see the number of cards in the above-created new data file.

9. Copy the cards in Excel or word


a. Ask the respondents to rank them
b. Rank 1 as the most preferred… and so on
10. Create consumer response in the excel file

11. Read/open consumer response in the SPSS


a. Save the SPSS file (Remember the location)

12. Create syntax


a. File – New – Syntax
i. Conjoint plan = ‘file location of orthogonal combination plan with file extension’
ii. /Data= ‘consumer preference - SPSS file location with file extension’
iii. /Sequence= Pref1 to Pref ….
iv. /Subject= User OR Respondent OR ID
v. /Factors = Factor name (Linear less/ Linear more/ discrete)
vi. /Print = Summaryonly
vii. Run - Click on triangle Green button

13. Output interpretation


a. Importance of each attribute
b. Utility / part-worth utility
c. Correlation for robustness of the model
d. Reversal summary

14. Create utility for each card


a. Go back to cards in the Excel
b. Use part-worth utility for the respective card
c. Utility of a card = Respective part-worth utility + Constant
d. Cards with the highest utility are considered as the first choice to create a prototype

You might also like