Assignment 1: Chapter 1, Conjoint Analysis
Information
Earlier this semester, you answered a brief survey in which you ranked your preferences for 15 different potential music streaming services. These 15 options were:
brand
fee
quality
catalog
download
tracks
Apple Music
$9.99
low
small
no
no
Apple Music
$9.99
medium
small
no
yes
Apple Music
$14.99
high
large
yes
yes
Apple Music
$14.99
medium
small
yes
yes
Pandora
$4.99
low
small
no
no
Pandora
$4.99
medium
small
yes
no
Pandora
$9.99
high
large
yes
yes
Pandora
$9.99
medium
small
no
no
Pandora
$9.99
medium
small
yes
no
Spotify
$4.99
low
small
yes
no
Spotify
$4.99
low
small
no
no
Spotify
$9.99
medium
large
yes
no
Spotify
$9.99
medium
large
no
no
Spotify
$14.99
high
large
yes
yes
Spotify
$14.99
high
large
no
yes
In fact, these 15 options were unique permutations of the following attributes and levels:
brand
monthly fee
sound quality
music catalog
free
download
downloadable
tracks
Pandora
$4.99
Low (128 kbps)
Small
(<100M tracks)
yes
yes
Apple Music
$9.99
Medium (256 kbps)
Large
(>100M tracks)
no
no
Spotify
$14.99
High (320 kbps)
105 of you and your classmates filled out this survey. Your rankings are anonymized and combined in the “music_rankings.csv” dataset. In this homework, you will use this dataset to conduct a conjoint analysis and related calculations, create a spine chart, conduct a trade-off analysis, and determine market shares and attribute importance. Furthermore, you will use these analyses and tools to draw important marketing conclusions from the data.
Helpful hints – Step 0. Open R, set your working directory to your BUAD476 folder, load the spine chart code, and ‘read’ the csv. I recommend opening your Chapter 1 R code from class and copying/pasting that code into a new script/document, or ‘saving as’ a new file, and then editing the script from there. You will need to understand what parts of the existing code to change to apply to this new data set… you are not required to write new code entirely.
Questions
• How many unique product option combinations are possible?
• Write the formula for the conjoint model below. Run it in R and summarize the output using the summary function.
• Why are there 3 attributes for the monthly fee variable, but only 2 coefficients?
• What must be true about the sum of the coefficients for a given attribute?
• Write below the values of ALL 15 attribute level coefficients (hint: R serves as a calculator!) in the table below.
Attribute / Level
Part-worths
• Calculate the importance values of all 6 attributes and report them in the table below. What does an attribute importance value indicate? Explain which attributes have the highest and lowest values and what that means.
Attribute
Importance Value
• Based on your model, what is the most preferred combination possible? What would be the least preferred combination possible?
• Based on your model, which of the 15 product options do you and your classmates prefer the MOST? The least? Give the total utility for these two products.
• Create the spine chart of the coefficients for this model and save this image as a screenshot or image below. (HINT: use the CH1 code and only change the arguments/variables that are specific to the data, like ‘brand’, ‘startup’ etc.)
• Suppose you work for Pandora, and you have an actual product with the following attributes:
brand
monthly fee
sound quality
music catalog
free
download
downloadable
tracks
Pandora
$4.99
Low (128 kbps)
Large (>100M tracks)
yes
no
Imagine you wanted to increase the monthly fee from $4.99 to $9.99. Assuming all consumers are like you and your classmates, how much HIGHER should the sound quality for the product be in kbps to hold constant the valuation of the product (i.e., interpolate the trade-off value)? Assume that sound quality can be set to the nearest kbps (that is, a value besides 128, 256, and 320). Write out your steps to maximize your ability to receive partial credit.
• Imagine that the actual market for music streaming services contained the following products:
Product
brand
monthly fee
sound quality
music catalog
free
download
downloadable
tracks
Market Share (%)
1
Apple Music
$9.99
low
small
no
no
2
Apple Music
$14.99
high
large
yes
yes
3
Pandora
$4.99
medium
small
yes
no
4
Pandora
$9.99
high
large
yes
yes
5
Spotify
$14.99
high
large
no
yes
6
Spotify
$9.99
medium
large
no
no
• Calculate the market shares for all 6 products and write them in in the table above.
• Give one recommendation to a company that is NOT dominating in market share. What could they change about their available products to get a larger market share?
• Explain two ISSUES or DRAWBACKS of conjoint analysis. You can think about this from the perspective of the consumers who answer these surveys and/or the companies who are developing these or related products.
• Provide one potential solution or strategy that could help alleviate one of the “cons” you highlighted in the previous question.