Reading I NEED HELP ON MY HOMEWORK 2
Consumer Behavior in a Digital World, Winter 2024 Conjoint Analysis 1 Professor Hamilton
CONJOINT ANALYSIS ASSIGNMENT OVERVIEW Due: Thursday, February 15 by 11:59pm
This assignment is to be completed individually.
Deliverable:
1. A 3-4 page paper in APA format that includes a/an: a. explanation of your research methodology b. report of the major findings from your analysis (further described below)
2. An APA style appendix (in addition to the report) that includes a(n): a. matrix of features/levels with overall ratings (‘U’s), b. raw regression output (from Excel or other stats tool), c. graphical representation of the partworths you calculated, and d. table of your willingness to pay (WTP) calculations.
Overview: Conjoint analysis is a method for determining the relative importance of distinct features of a product or service for determining overall preferences.
In this assignment, you will create a conjoint choice task related to a product or service of your choosing, collect preference ratings from a friend or classmate, analyze these data using a regression, and interpret your results. Your results will yield insight into how much each feature of a product/service contributes to overall preferences, and you will use this insight to calculate your respondent’s trade-offs between levels of each feature in terms of willingness to pay (WTP). Most importantly, you will use the paper component of this assignment to communicate what your data mean and why they matter (see requirements and guidelines below). Just like the Laddering Assignment you just completed, your goal is to collect and analyze data and then interpret your data in a white paper report.
Consumer Behavior in a Digital World, Winter 2024 Conjoint Analysis 2 Professor Hamilton
Detailed instructions 1. Step One – Conduct a Survey
a. Choose a product (e.g., Garmin GPS). i. Specify four features of the product (price + three features of your choice;
e.g., price, accuracy, battery life, color), each with two levels (e.g., 10 feet vs. 32 feet).
ii. Deliverable: In your report, briefly explain why you picked your product. Then, provide a definition of the features of this product and the levels of each feature.
b. Survey someone else (not yourself). i. Ask your respondent to indicate their overall preference for a hypothetical
product with each combination of features. (Hint: Because there are 4 features with 2 levels each, there are 16 combinations. The 16 combinations are reflected in Step 2a).
ii. Here are some ways you can ask them: 1. “On a scale of 1-100, how likely are you to recommend the
[PRODUCT] featured to a friend or colleague? 2. “On a scale of 1-100, How likely or unlikely are you to purchase the
[PRODUCT]?” 3. (Hint: You can do this via survey or through an interview. But,
remember your goal is to ask them about 16 different products that each have their own unique attributes. The better you can help them visualize each product, the more reliable your results will be.)
iii. Record their preferences for all 16 combinations. iv. Transcribe their preferences to a Google sheet (or Excel) where preference
ratings are in Column A and the code for each level (e.g., 10 feet = 0 and 32 feet = 1 for Accuracy) of the four features are in Columns B-E
v. Deliverable: Include a screenshot of recorded data (similar to the image below) in Appendix A of your white paper.
Consumer Behavior in a Digital World, Winter 2024 Conjoint Analysis 3 Professor Hamilton
2. Step Two – Run a Conjoint Analysis
a. Conduct a regression predicting preferences (‘U’s) from features (‘x’s). Make sure to code the levels of your features as ‘0’s and ‘1’s. Then, select “Regression” where Y is the preference ratings column and X is the four feature columns. (I will include a demo in the lecture of the process of running your regression in Google Sheets in our lecture). i. Deliverable: Include a screenshot of your regression output (similar as below)
in Appendix B.
Consumer Behavior in a Digital World, Winter 2024 Conjoint Analysis 4 Professor Hamilton
b. Compute the partworths of each feature. i. The table below describes how to calculate partworths.
ii. Deliverable: In your report, describe the partworth of each feature and interpret the direction of the relationship between levels on preference ratings. For example, “This individual preferred a GPS with a 32 hour battery life over 12 hour battery life, which accounted for 32% of the variance in their preference ratings.”
iii. Deliverable: Show the relative importance of each feature using a bar graph or pie chart and include the illustration in Appendix C.
3. Step Three – Calculate and Interpret WTP a. Compute Willingness to Pay (WTP) for all the attributes.
i. Deliverable: In your report, explain what the trade-off for price you computed means using plain language and refer to these calculations while providing recommendations about the price trade-off of including different features.
ii. Deliverable: Include a table (like the one below) that explains your WTP calculations in Appendix D.
Feature Partworth (%) WTP Calculation WTP Price ($250 vs. $350) 43% $2.33*43 = $100.19 $100.19 Accuracy 10% $2.33*10 = $23.30 $23.30 Battery 32% $2.33*32 = $74.56 $74.56 Color 16% $2.33*16 = $37.28 $37.28
4. Step Three – Report your findings a. Report your findings in an APA style paper. Your paper must explain your research
methodology and your findings before providing an evidence-based recommendation about the design or promotion of your product. More specifically, your report must: i. Give detail about why you picked your product. Then, provide a definition of
the features of this product, and the levels of each feature (see step 1a.ii). ii. Give detail about your respondent; when presenting your results, make sure to
discuss the extent to which you think your respondent’s ‘w’s are representative of the general population, a particular segment, or perhaps just that individual. Your description of this person should inform your generalizable findings in the recommendation section.
Consumer Behavior in a Digital World, Winter 2024 Conjoint Analysis 5 Professor Hamilton
iii. Describe the partworth of each feature and interpret the direction of the relationship between levels on preference ratings (see step 2b.ii).
iv. Explain what the trade-off for price you computed means using plain language (see step 3a.i). A discussion of trade-offs can inform segmentation, new product creation, profit maximization, and so on.
v. Most importantly, talk about your results in depth. You must show that you’re making connections between the numbers produced by your analysis and what these numbers mean. If you have any odd results, discuss why they might have turned out this way.
vi. Discuss limitations of your data. Limitations could be related to the nature of the person you surveyed, the way you surveyed them, etc.
vii. Describe the ideal product to promote to the perceived “population” you sampled. For example, if you survey (i.e., sample from) a college student about their preferences of laptops, you should be careful not to generalize your insights to all people, and instead be clear to whom you believe you are generalizing your insights based on the details you provide about your respondent. Then, provide a more practical product that considers WTP tradeoffs. This product can have the same features as the ideal product, but you must rationalize why you believe including these features are worth the price they may cost. The insight you generate should be based on evidence you have aggregated about the importance of features, WTP trade-offs, or theoretical findings discussed in lecture.
Consumer Behavior in a Digital World, Winter 2024 Conjoint Analysis 6 Professor Hamilton
CONJOINT ASSIGNMENT EVALUATION CRITERIA
Your assignment will be evaluated out of 22 points, following the rubric below.
Description Level grade
Definition
1 Submission ● Was the assignment fully submitted and
on time?
0 No
1 Yes
1 Formatting and Required Elements ● Does the white paper follow the
prescribed format (i.e., 3-4 page APA format with an appendix)?
● Here is a reference: https://guides.rasmussen.edu/apa6th/abstra ct-appendix
0 No–not all formatting requirements are met.
1 Yes–all formatting requirements are met.
2 Methodology: Product details ● Does the report provide a detailed
discussion about 1) what the product/service is, 2) why the product/service was chosen, 3) definitions of the features of this product, and 4) levels of each feature of the product/service?
0 No-- definitions of the features and/or levels of each feature are not described in the report.
1 Somewhat-- product details are discussed in the report, but one (or more) crucial aspect is missing. This missing detail makes it harder to follow the conjoint analysis.
2 Yes-- All criteria are met.
3 Methodology: Respondent details ● Does the report provide details about the
respondent and whether this respondent may (or may not) be representative of a particular segment (or segments) of the population?
0 No-- details about who the respondent reflects in the population are missing.
1 Somewhat-- respondent details are discussed in the report, but one (or more) crucial aspect is missing about the product details. This makes it difficult to connect these details to the
Consumer Behavior in a Digital World, Winter 2024 Conjoint Analysis 7 Professor Hamilton
limitations they report.
2 Yes-- All criteria are met.
4 Conjoint: Product features + Regression ● Does the product/service reported in
Appendix A vary on four features (price + three others), with two levels for each feature?
● Does the report describe the regression analysis and include the raw regression output in Appendix B?
0 No-- one (or more) crucial aspect is missing from the conjoint analysis.
1 Somewhat-- it is clear a conjoint analysis was conducted, but one (or more) aspect is not reported or incorrect.
2 Yes-- All criteria are met.
5 Partworths: Computation + Interpretation ● Does the report include partworth
calculations for each feature, and explain what these partworths mean with respect to the relationship between levels of that feature?
0 No-- the partworth calculations are missing.
1 Somewhat-- it is clear partworths were calculated, but one (or more) aspect is not reported or the calculations are incorrect.
2 Yes-- All criteria are met.
Partworths: Computation + Interpretation ● Does the report include a graph (e.g., bar
graph, pie chart) of partworths in Appendix C? Give credit based on whether the illustration clarifies the weighted preference (i.e., importance) for a particular feature.
0 No– a graphical representation of partworths is missing or something about the figure is incorrect.
1 Yes-- All criteria are met.
6 Trade-offs: WTP + non-monetary features ● Does the report include WTP for each
non-monetary feature and explain what these WTP calculations mean?
0 No-- WTP calculations are not explained in the report
1 Somewhat-- it is clear WTP was calculated for each non-monetary feature, but one (or more) aspect is not reported or the calculations are incorrect.
2 Yes-- All criteria are met.
Consumer Behavior in a Digital World, Winter 2024 Conjoint Analysis 8 Professor Hamilton
Trade-offs: WTP + non-monetary features ● Does the report describe WTP
calculations in Appendix D?
0 No– WTP calculations are not reported in Appendix D or the calculations are incorrect.
1 Yes-- All criteria are met.
7 Recommendation ● Does the report provide details about an
ideal product with features that are based on partworth or WTP calculations, or are in consideration of theoretical findings discussed in lecture?
● Ultimately, does the recommendation illuminate how a company might design or promote this product based on their conjoint analysis data?
0 No-- the submission does not provide a recommendation for the product analyzed.
1 Somewhat-- It is clear the report fully attempts to provide a recommendation, but their recommendation is not clearly connected to their conjoint data or course material.
2 Yes– All criteria are met.
8 Insight ● Did the conjoint data yield insight that
was explained in a way that felt novel and interesting?
0 No-- The insights generated from the conjoint data were not interpreted in a way that yielded much insight (Hint: Usually in this case data are described, and then ignored).
1 Yes-- conjoint data were interpreted in a way that yielded insight someone could use to design or promote a better product.
2 Yes (+ special recognition)-- conjoint data were interpreted in a way that yielded insight someone could use to design or promote a better product AND the insight generated was probably more novel and interesting than 90% of class submissions.
9 Clarity ● Is it clear how data from the analysis
connects to the insight generated in the
0 No-- it is not clear whether insight was generated from data.
Consumer Behavior in a Digital World, Winter 2024 Conjoint Analysis 9 Professor Hamilton
report? Is the insight generated from the results confusing to understand?
1 Somewhat-- the insight generated was confusing to follow.
2 Yes-- the insight is clear and it is easy to see how this insight was informed by data.
10 Organization ● In general, was the paper well-organized
(i.e., did the formatting contribute to a better reading experience)?
0 No-- the formatting made the paper harder to read.
1 Somewhat-- the paper followed basic formatting standards.
2 Yes--the formatting structure or writing organization made the paper easy to read.