CLUSTER ANALYSIS FOR HAPPENSTANCE SANDALS

CLUSTER ANALYSIS FOR HAPPENSTANCE SANDALS

Author: Bhavesh Choudhary, Paramjeet Singh, Vaishnavi Salunke, Sachin Kumar

-ITM Business School

 

 

Introduction:

Happenstance footwear is a pioneering brand focused on delivering extreme comfort through innovative designs and materials. Established to address the gap in the Indian market for stylish yet comfortable shoes, Happenstance employs advanced technologies like Fluffium and Buoyance to create footwear that molds to the foot and absorbs shock effectively. The brand emphasizes affordability without compromising quality, making it accessible to a wide audience. With a commitment to local production and ergonomic design, Happenstance aims to redefine comfort in everyday footwear while appealing to modern consumers’ aesthetic preferences.

 

Characteristics:

  • Durability: High-quality materials ensure long-lasting wear.
  • Aesthetic: Modern and attractive designs complement various outfits.
  • Design: Ergonomic and innovative features enhance comfort and fit.
  • Cushioning: Advanced cushioning systems reduce foot fatigue.
  • Material: Durable EVA and PU provide resilience and comfort.
  • Fabric: Soft, quick-drying, and skin-friendly fabrics improve comfort.
  • Breathability: Perforated uppers and mesh inserts keep feet cool and dry.
  • Sustainability: Eco-friendly practices and materials minimize environmental impact.
  • Affordability: Priced to provide excellent value for money.
  • Slip-on Style: Convenient slip-on design for easy wear and removal.

 

Objective:

conducting cluster analysis for Happenstance sandals focus on enhancing customer understanding and improving marketing strategies. The analysis aims to identify distinct customer segments based on demographics and purchasing behaviour, enabling targeted marketing efforts. It seeks insights into valued features such as comfort, style, price, and material to inform product development. The analysis will optimize product offerings by identifying popular styles while tailoring marketing messages to resonate with specific segments. Improving customer experience is essential for boosting satisfaction and retention. Finally, monitoring brand perception will guide necessary branding adjustments to drive growth in the competitive footwear market.

 

Data Collection:

The data of 50 respondents was collected via a survey done on google form for 10 different characteristics of Happenstance sandals, using a liked scale: rating from 1 to 10. By focusing on key features we can gain a deeper understanding of what makes a sandal appealing and desirable.

 

 

 

 

Cluster Membership

Case Number

Cluster Number of Case

Cluster

Distance

1

1

1

5.544

2

2

2

8.691

3

2

2

7.446

4

1

1

9.864

5

1

1

8.546

6

1

1

8.226

7

2

2

8.733

8

1

1

8.302

9

2

2

6.689

10

1

1

7.784

11

1

1

5.308

12

2

2

10.793

13

1

1

8.474

14

2

2

8.098

15

2

2

7.625

16

2

2

8.971

17

1

1

8.446

18

2

2

9.194

19

1

1

10.978

20

2

2

9.596

21

1

1

8.840

22

1

1

6.812

23

1

1

9.229

24

2

2

7.998

25

1

1

9.576

26

1

1

8.459

27

2

2

7.446

28

1

1

7.654

29

2

2

6.673

30

1

1

8.308

31

1

1

9.475

32

2

2

9.666

33

1

1

7.741

34

2

2

7.676

35

2

2

8.595

36

1

1

6.801

37

1

1

6.936

38

2

2

9.241

39

2

2

5.909

40

2

2

8.793

41

2

2

8.765

42

2

2

8.450

43

1

1

9.773

44

1

1

8.960

45

1

1

7.874

46

1

1

9.120

47

2

2

8.683

48

1

1

9.173

49

2

2

7.885

50

1

1

5.950

 

Data Overview:

The data provided consists of 50 cases, each assigned to one of two clusters along with the respective distances from each cluster center. Cluster analysis is a tool to understand how closely cases within each cluster relate to one another and their respective cluster center. Each case’s distance from the center provides insights into cluster compactness and potential outliers.

 

Detailed Analysis:

  1. Cluster Distribution:
  • Cluster 1: Contains 27 cases.
  • Cluster 2: Contains 23 cases.

 

Cluster 1 has a slightly larger number of cases compared to Cluster 2. This distribution allows for comparison in terms of cluster compactness and variability.

 

  1. Distance Metrics:

Cluster 1:

Minimum Distance: 5.308

Maximum Distance: 10.978

Sum of Distances: 215.946

Average Distance: 7.998

 

Cluster 2:

Minimum Distance: 5.909

Maximum Distance: 10.793

Sum of Distances: 189.612

Average Distance: 8.244

 

Interpretation of Distance Metrics:

 

Compactness: Cluster 1 has a slightly lower average distance (7.998) compared to Cluster 2 (8.244), indicating that Cluster 1 is more compact. This suggests a higher degree of similarity among cases within Cluster 1, as they are more closely grouped around the center.

 

Variability: Cluster 2 has a broader range of distances, indicating greater variability and a less tight grouping. This could suggest that cases in Cluster 2 have more diverse characteristics.

 

Outliers: Both clusters have cases with higher distances from their centers, with Cluster 1 having a maximum distance of 10.978 and Cluster 2 having a maximum of 10.793. These cases could represent potential outliers within their respective clusters.

 

Conclusion:

  1. Cluster Compactness:

Cluster 1 is more compact, with a lower average distance to its centre. This suggests that the cases in Cluster 1 are more like each other, making this cluster potentially suitable for applications that require consistency or homogeneity, such as targeted marketing strategies for a similar customer group.

 

 

  1. Cluster Variability:

Cluster 2, with its higher average distance and broader range of distances, appears to be less compact. This cluster’s greater variability might indicate that it encompasses a more diverse set of cases, which could be useful in applications where a range of characteristics within a group is desired.

 

  1. Outliers:

Both clusters contain cases that are farther from the cluster center, possibly representing outliers. Further investigation of these cases could reveal unique attributes, helping to determine if they should be addressed separately or included within the overall analysis.

 

This interpretation and conclusion provide a clear understanding of the structure and quality of the clusters. The insights on cluster compactness, variability, and outliers can guide decisions in clustering applications, such as customer segmentation and market analysis, where different strategies may be employed for each group based on their internal consistency or diversity.

 

 

 

 

 

 

 

 

 

 

 

 

ANOVA

 

Cluster

Error

F

Sig.

Mean Square

df

Mean Square

df

design

13.316

1

7.460

48

1.785

.188

fabric

100.557

1

7.992

48

12.582

<.001

affordable

2.616

1

8.873

48

.295

.590

aesthetic

82.139

1

7.343

48

11.187

.002

sustainable

22.725

1

8.360

48

2.718

.106

durability

154.041

1

5.245

48

29.372

<.001

cushioning

8.181

1

7.411

48

1.104

.299

sliponstyle

64.393

1

6.732

48

9.565

.003

material

142.842

1

5.781

48

24.710

<.001

breathability

46.997

1

7.681

48

6.119

.017

The F tests should be used only for descriptive purposes because the clusters have been chosen to maximize the differences among cases in different clusters. The observed significance levels are not corrected for this and thus cannot be interpreted as tests of the hypothesis that the cluster means are equal.

 

 

Interpretation of ANOVA Results:

 

The table presents the results of an ANOVA analysis, examining whether the means of various factors differ significantly between two clusters. The factors analysed include “design,” “fabric,” “affordable,” “aesthetic,” “sustainable,” “durability,” “cushioning,” “lifestyle,” “material,” and “breathability.”

 

Key Terms:

 

  • F (F-value): The ratio of variance between the clusters to variance within the clusters. Higher values indicate more variation between clusters.
  • Sig. (Significance): The p-value indicating whether the difference between clusters for each factor is statistically significant. A value below 0.05 is generally considered significant.

 

Results:

  1. Design:
  • F-value: 1.785
  • Significance: 0.188
  • Interpretation: The difference in “design” between clusters is not statistically significant.
  1. Fabric:
  • F-value: 12.582
  • Significance: < 0.001
  • Interpretation: The difference in “fabric” between clusters is highly significant, indicating this factor varies notably between clusters.
  1. Affordable:
  • F-value: 0.295
  • Significance: 0.590
  • Interpretation: No significant difference in “affordability” between clusters.
  1. Aesthetic:
  • F-value: 11.187
  • Significance: 0.002
  • Interpretation: The aesthetic factor significantly differs between clusters, meaning aesthetics play a distinctive role in the clustering.
  1. Sustainable:
  • F-value: 2.718
  • Significance: 0.106
  • Interpretation: The “sustainable” factor does not significantly differ between clusters.
  1. Durability:
  • F-value: 29.372
  • Significance: < 0.001
  • Interpretation: Durability shows a highly significant difference between clusters, suggesting it is a key distinguishing factor.
  1. Cushioning:
  • F-value: 1.104
  • Significance: 0.299
  • Interpretation: Cushioning does not significantly differ between clusters.
  1. Lifestyle:
  • F-value: 9.565
  • Significance: 0.003
  • Interpretation: The lifestyle factor is significantly different between clusters, indicating it contributes to the clustering.
  1. Material:
  • F-value: 24.710
  • Significance: < 0.001
  • Interpretation: Material has a highly significant difference between clusters, making it an important factor for clustering.
  1. Breathability:
  • F-value: 6.119
  • Significance: 0.017
  • Interpretation: Breathability differs significantly between clusters, suggesting this feature affects the grouping.

 

Conclusion:

 

From the ANOVA analysis, we can conclude that several factors significantly differentiate the two clusters:

 

  • Highly Significant Factors: Fabric, durability, material
  • Significant Factors: Aesthetic, lifestyle, breathability

 

These factors may be critical in understanding consumer preferences or behaviour patterns associated with each cluster. For example, Cluster 1 might prioritize fabric and material quality, while Cluster 2 might emphasize lifestyle alignment and breathability. This insight can help in tailoring marketing strategies and product offerings to meet the specific needs of each segment.

 

Factors like “design,” “affordable,” “sustainable,” and “cushioning” show no significant difference, implying these aspects are similarly valued across both clusters.

Leave a comment