| Course | BUSI 614 Marketing Research (BUSI614) |
|---|---|
| Module/Week | 8 |
| Assignment | SPSS Quantitative Assignment 7 |
| Paper type | SPSS quantitative analysis report |
| Length | about 904 words, 3 double-spaced pages plus title page and references. Course guide: an SPSS analysis of the data set provided in the course, running the appropriate statistical procedures and reporting the results with interpretation |
| Format | APA 7 student paper |
| Points | 100 |
| School | Liberty University Online |
| Program | MBA |
| Where it sits | Module/Week 8 in Liberty's published BUSI 614 course guide; current sections can differ |
| Updated | October 2026 |
Free sample paper for BUSI 614 Week 8
Segmenting Fitness App Users: Factor Analysis and Cluster Analysis in SPSS
[Student Name]
School of Business, Liberty University
BUSI 614: Marketing Research
SPSS Quantitative Assignment 7
[Instructor Name]
[Date]
The data set in this sample is illustrative, generated to show how the SPSS procedures are run, checked and reported; your assignment uses the data set provided in the course, and the research cited is real.
Segmenting Fitness App Users: Factor Analysis and Cluster Analysis in SPSS
Purpose and Data
A composite fitness app company wants to segment its users so that it can decide which features to develop and which messages to send to which users. The illustrative file contains 300 active users who rated nine statements about what they value in a fitness app on seven-point scales, from strongly disagree to strongly agree. Because the nine items probably reflect a smaller number of underlying motivations, a factor analysis is used first to identify those dimensions, and the factor scores are then used in a cluster analysis to group users with similar motivations. Clustering on a few uncorrelated factor scores rather than nine overlapping items keeps any single motivation from being counted several times.
Factor Analysis
The analysis was run through Analyze, Dimension Reduction, Factor, with all nine items entered. Under Descriptives, KMO and Bartlett's test of sphericity were requested; under Extraction, principal components with eigenvalues greater than 1 and a scree plot; under Rotation, varimax; and under Scores, regression factor scores were saved to the data file.
The data were suitable for factoring. The Kaiser-Meyer-Olkin measure of sampling adequacy was .71, and Bartlett's test was significant, χ²(36) = 819.7, p < .001. Three components had eigenvalues above 1 (2.51, 2.33 and 1.50), the fourth fell to 0.54, and the scree plot leveled off after the third, so three factors were retained. Together they explained 70.4% of the variance.
| Item | Social | Convenience | Tracking | Communality |
|---|---|---|---|---|
| Workouts with friends | .85 | -.12 | -.11 | .75 |
| Sharing progress | .85 | .05 | -.05 | .72 |
| Group challenges | .82 | -.13 | -.02 | .69 |
| Fits my schedule | -.11 | .84 | -.04 | .72 |
| Quick to start a workout | -.11 | .84 | -.06 | .73 |
| Works without equipment | .02 | .82 | -.06 | .67 |
| Detailed progress data | -.09 | -.06 | .82 | .69 |
| Personalized plans | -.13 | .05 | .82 | .69 |
| Syncs with my watch | .02 | -.17 | .80 | .67 |
Each item loaded at least .80 on one factor and no more than .17 on the others, a clean simple structure. After rotation, the social factor accounted for 24.0% of the variance, convenience for 23.9% and tracking for 22.4%. Scales formed from each set of three items were reliable, with Cronbach's alpha of .80 for social motivation, .79 for convenience and .76 for tracking, above the .70 often used as a minimum when purifying measures (Churchill, 1979).
Two choices deserve comment. Fabrigar et al. (1999) found that researchers often use principal components and orthogonal rotation by default, and they recommended common factor methods when the goal is to identify latent constructs, along with oblique rotation when factors may be correlated. Here the scales built from the three factors correlated only weakly, between -.13 and -.16, so the orthogonal solution is a reasonable simplification, but a principal axis analysis with oblique rotation would be a sensible check.
Cluster Analysis
Punj and Stewart (1983), reviewing cluster analysis in marketing research, recommended a two-stage approach: a hierarchical method such as Ward's to determine the number of clusters and starting points, followed by K-means to refine cluster membership, and some validation of the solution. Following that approach, a hierarchical analysis using Ward's method on the three saved factor scores was run through Analyze, Classify, Hierarchical Cluster. The agglomeration schedule and dendrogram were examined for solutions of two to five clusters. The two-cluster solution combined convenience-minded users with tracking-minded users, which hides the difference that matters most for feature decisions. The four- and five-cluster solutions split off groups of about 40 users defined mainly by low interest in both social features and tracking, groups too small to justify separate marketing programs. Three clusters were therefore chosen as the most usable structure. K-means clustering, run through Analyze, Classify, K-Means Cluster with the Ward cluster centers as starting values, converged within ten iterations.
| Segment | Size | Social | Convenience | Tracking |
|---|---|---|---|---|
| Social exercisers | 110 (37%) | 0.98 | -0.13 | -0.41 |
| Time-pressed convenience seekers | 103 (34%) | -0.71 | 0.85 | -0.20 |
| Data-driven trainers | 87 (29%) | -0.39 | -0.84 | 0.76 |
Values are mean factor scores, which have a mean of 0 and a standard deviation of about 1 across all users. The clusters differed sharply on all three factors: social, F(2, 297) = 197.5; convenience, F(2, 297) = 126.3; tracking, F(2, 297) = 47.7; all p < .001. These F tests describe how well the factors separate the clusters but are not formal hypothesis tests, since the clusters were formed to maximize those differences. To check stability, the file was split randomly into two halves and K-means was run on each; both halves produced three clusters with the same pattern of high and low scores, supporting the solution.
Segment Profiles and Implications
Each segment wants a different app, even though all three use the same one. Social exercisers, the largest group, rated workouts with friends and group challenges near 5 on the seven-point scale and rated the tracking items just below the scale midpoint of 4. Features such as team challenges, leaderboards and easy sharing should be marketed to them, and referral incentives may work well. Convenience seekers rated fit with their schedule and quick starts near 5 and showed low interest in social features; short workouts that need no equipment, reminders timed to their routines and messaging about saving time suit them. Data-driven trainers rated detailed progress data, watch syncing and personalized plans near 5; they are the natural audience for a premium tier with advanced analytics and wearable integration.
For product development, the results suggest that no single feature appeals to most users, so the company should tailor onboarding and in-app messages by segment. Users could be assigned to segments with a short set of questions at sign-up, scored using the factor results.
Limitations
The segments are based on stated attitudes and should be validated against behavior, such as how often each group uses social features or opens progress reports. Cluster solutions depend on the analyst's choices of variables, method and number of clusters, so the three-segment solution is one useful view rather than the only correct one. The sample includes only active users, so it says nothing about people who downloaded the app and stopped using it, a group that further research should reach.
References
Churchill, G. A., Jr. (1979). A paradigm for developing better measures of marketing constructs. Journal of Marketing Research, 16(1), 64-73. https://doi.org/10.1177/002224377901600110
Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 4(3), 272-299. https://doi.org/10.1037/1082-989X.4.3.272
Punj, G., & Stewart, D. W. (1983). Cluster analysis in marketing research: Review and suggestions for application. Journal of Marketing Research, 20(2), 134-148. https://doi.org/10.1177/002224378302000204
What the BUSI 614 Week 8 instructions ask
The last SPSS assignment in BUSI 614 comes due in Module/Week 8, alongside chapters 30 and 31 of Malhotra, one presentation and a website, and it is the course's final graded item. Like Assignments 5 and 6, it is worth 100 points. Across all seven SPSS assignments, the guide asks students to use the software to analyze data sets provided in the course, run the appropriate statistical procedures and report the results. The posted instructions determine which procedures this assignment requires. Whatever they are, the final assignment is a chance to show everything the series has built: matching methods to questions, checking assumptions, presenting output selectively and turning statistics into decisions about how to serve customers better.
How this BUSI 614 Week 8 example is built
The sample opens with the business purpose, segmenting app users so that product features and messages can be targeted. The factor analysis section reports the suitability tests, the number of factors retained and why, a rotated loading table and Cronbach's alpha for each factor, and it notes the choices Fabrigar and colleagues caution about. The cluster analysis section follows Punj and Stewart's recommendation to combine hierarchical and K-means methods, reports final cluster centers and sizes and checks stability across split halves. A segment profile table and a section on managerial implications translate the results, and a final section sets out the limits of the analysis.
BUSI 614 Week 8 grading rubric: where the points go
Grading looks at whether the procedures suit the question, whether they are run with appropriate options, whether the output is reported accurately in APA style and whether the interpretation supports a business decision. For factor analysis, graders expect correct use of the KMO measure, Bartlett's test of sphericity, eigenvalues, explained variance, the scree plot, rotation, loadings and communalities. For cluster analysis, they expect hierarchical and K-means methods, the agglomeration schedule, cluster centers and some check of validity. Because both techniques involve judgment, reports that explain each choice, such as the number of factors or clusters, and acknowledge alternatives earn more credit than reports that accept SPSS defaults without comment.
BUSI 614 Week 8 help: mistakes that cost points
Run the factor analysis first and check that the data are suitable: KMO above .60 and a significant Bartlett test. Decide on the number of factors using more than one criterion, such as eigenvalues, the scree plot and interpretability. Save factor scores to the data file and use them as the clustering variables. Run a hierarchical cluster analysis with Ward's method to judge the number of clusters, then K-means with that number. Name the segments only after examining their profiles, and check that the solution holds up in a random half of the data. Profile the segments on variables not used to form them, such as age or plan type, to show that they differ in ways a marketer can reach.
Related BUSI 614 sample papers
Other BUSI 614 weeks
- BUSI 614 Week 1: Forum 1: When Research Gets Used
- BUSI 614 Week 2: Qualitative Assignment 1: Interviews
- BUSI 614 Week 3: Qualitative Assignment 2: Surveys
- BUSI 614 Week 4: Qualitative Assignment 3: Fieldwork
- BUSI 614 Week 5: Forum 2: Big Data and Surveys
- BUSI 614 Week 6: SPSS Assignment 5: One-Way ANOVA
- BUSI 614 Week 7: SPSS Assignment 6: Regression
More MBA sample papers
- BUSI 606 Week 8: Literature Review: Climate Risk
- BUSI 607 Week 8: Marketing Plan: Cold Brew in Vietnam
- BUSI 610 Week 8: Forum 4 Part 2: Culture and Change
- BUSI 613 Week 8: Faith and Learning: Supply Chains
BUSI 614 Week 8 questions, answered
What is BUSI 614 SPSS Quantitative Assignment 7?
BUSI 614's last SPSS task, a 100-point report closing the course in the eighth week, analyzing a data set supplied in the course with the appropriate procedures and reporting and interpreting the output.
Where can I find a free BUSI 614 Week 8 sample paper?
This page has a complete BUSI 614 Week 8 SPSS report, using factor analysis and cluster analysis to segment the users of a fitness app.
What KMO value is acceptable for factor analysis?
Values above .60 are commonly treated as adequate and values above .80 as very good; the KMO measure shows whether the correlations among items are strong enough to be summarized by factors.
Why combine hierarchical and K-means clustering?
Hierarchical clustering with Ward's method helps choose the number of clusters and starting centers, and K-means then refines the assignment of cases, an approach recommended in reviews of cluster analysis in marketing.
What else is due in BUSI 614 Module/Week 8?
Nothing else is graded; the final week assigns chapters 30 and 31 of Malhotra, one presentation and one website along with the last SPSS assignment.
Use it as a model, or ask for your own
This paper is an original model written by our desk for study. It is not a submitted student paper and not an official Liberty University document. Read it for structure and reasoning, then write your own to the instructions in your Canvas course. If you want a version built to your exact prompt and rubric, the first custom sample is free and arrives in 24 to 48 hours.
Ask for a custom BUSI 614 Week 8 sample · All BUSI 614 weeks · All courses