1

Digital Transformation in Modern Business Success

Student's Name

Institutional Affiliation

Professor's Name

Course Name

Date
 2

Digital Transformation in Modern Business Success

Project Data Set Choice

I'll use a pre-existing data collection for simplicity's sake. On websites like Kaggle, the

UCI Machine Learning Repository, and official ones, many open datasets are accessible online.

A data set from a poll regarding customers' happiness with a well-liked item or service would be

an excellent match for this project.

Variables in the Data Set:

1. Age (Quantitative)

2. Gender (Qualitative: Male, Female, Prefer not to say)

3. Monthly Income (Quantitative)

4. Duration of daily phone usage (Quantitative: hours)

5. Brand loyalty (Qualitative: Brand Loyalist, Neutral, Brand Switcher)

6. Satisfaction with battery life (Qualitative: Satisfied, Neutral, Unsatisfied)

7. Camera quality rating (Quantitative: 1-10 scale)

8. Purchase reason (Qualitative: Brand, Features, Price, Recommendation)

9. Likelihood to recommend (Quantitative: 1-10 scale)

10. Previous phone brand (Qualitative)

1. Research Questions:

 Is there a gender disparity among mobile phone users?

 Does monthly income, daily phone usage, or camera quality rating vary depending on the

brand loyalty status (Loyalist, Neutral, or Switcher)?

 Does the current phone brand influence the reason for purchase (Features, Brand, Price,

or Recommendation)?
 3

 Is there a correlation between age, daily phone usage, and age and camera quality rating?

 Can monthly income, daily phone usage, and battery satisfaction predict brand loyalty?

2. Importance & Existing Literature:

Mobile phones have increased to the point that they are now commonplace in daily life.

Given the crowded market and fierce brand rivalry, manufacturers and marketers must

understand consumer preferences as smartphone technology develops quickly. Analyzing prior

literature shows subtle consumer preferences, actions, and expectations differences.

The impact of age is among the most often discussed subjects in the literature on

technological adoption and adaptation. According to several studies, young people seem to adapt

to technology changes more rapidly and are more eager to try new features. This may be

attributable to sociocultural effects, cognitive flexibility, and prior tech exposure.

Age-related differences exist in smartphone expectations and preferences. Being digital

natives, younger consumers often give priority to characteristics like camera quality, the app

ecosystem, and gaming performance. Older users, however, could place more importance on

dependability, battery life, and simplicity of use. For goods to be tailored to certain demographic

groupings, it is essential to comprehend these subtleties.

The question of brand loyalty in the smartphone sector has also received substantial

attention. Brand loyalty is a firmly held commitment to repeatedly purchase or patronize a

preferred product or service in the future, leading to repeat same-brand or same-brand-set

purchases. In the case of smartphones, brand loyalty affects the ecosystem (apps, accessories)

and influences repeat purchases.

The strength of recommendations is a crucial aspect that supports brand loyalty.

Customers devoted to a company are likelier to tell their friends about it, serving as brand
 4

ambassadors. In his study, Reichheld (2003) created the Net Promoter Score (NPS) as a metric to

gauge this potential for recommendations, indicating a substantial association between customer

loyalty and the inclination to suggest (particularly in tech sectors).

Knowing what motivates client loyalty is essential at a time when gaining new customers

is substantially more costly than keeping current ones. According to studies, brand loyalty and

the possibility for recommendations might provide useful information for developing client

retention tactics. For instance, giving loyalty benefits, using word-of-mouth marketing, or

customizing goods based on consumer input may boost brand equity and raise retention rates.

3. Sample and Data Description

The dataset under discussion was compiled by a reputable tech review publication

renowned for its in-depth analyses and insights on the newest tech goods. Mobile phone users

from various demographic backgrounds make up the survey's participants. Participants had to

have used their cell phone for at least one month as a requirement for participation. Ensuring the

input is based on sufficient usage makes it possible to portray user experiences and preferences

accurately. The dataset includes both quantitative and qualitative information:

Quantitative variables: Numbers or scales are used to represent these variables. For instance, a

responder may give their phone's camera a score between 1 and 10, depending on how good it is.

These measurements provide user input a quantifiable and comparative component. The

classification of the Qualitative variables are based on user input. Categories like "satisfied,"

"neutral," or "dissatisfied" may be included in qualitative data in place of numerical values. This

kind of information captures the respondents' complex experiences and arbitrary viewpoints.

4. Statistical Analyses

Univariate Analysis
 5

Descriptive statistics for all variables, including frequency distributions for qualitative variables

and means for quantitative data.

Gender
Cumulative
Frequency Percent Valid Percent Percent
Valid male 52 47.3 47.3 47.3
female 58 52.7 52.7 100.0
Total 110 100.0 100.0

Brand Loyalty
Cumulative
Frequency Percent Valid Percent Percent
Valid Loyalist 37 33.6 33.6 33.6
Neutral 38 34.5 34.5 68.2
Switcher 35 31.8 31.8 100.0
Total 110 100.0 100.0

Battery Satisfaction
Cumulative
Frequency Percent Valid Percent Percent
Valid Satisfied 41 37.3 37.3 37.3
Unsatisfied 25 22.7 22.7 60.0
Neutral 44 40.0 40.0 100.0
 6

Total 110 100.0 100.0

Purchase Reason
Cumulative
Frequency Percent Valid Percent Percent
Valid Features 32 29.1 29.1 29.1
Brand 17 15.5 15.5 44.5
Price 44 40.0 40.0 84.5
Recommedation 17 15.5 15.5 100.0
Total 110 100.0 100.0

Previous Phone Brand

Cumulative
Frequency Percent Valid Percent Percent
Valid Apple 27 24.5 24.5 24.5
Huawei 24 21.8 21.8 46.4
LG 15 13.6 13.6 60.0
OnePlus 25 22.7 22.7 82.7
Samsung 19 17.3 17.3 100.0
Total 110 100.0 100.0

Current Phone Brand

Cumulative
Frequency Percent Valid Percent Percent
Valid Apple 52 47.3 47.3 47.3
Huawei 10 9.1 9.1 56.4
LG 7 6.4 6.4 62.7
OnePlus 17 15.5 15.5 78.2
Samsung 24 21.8 21.8 100.0
Total 110 100.0 100.0
 7

The gathered data offers fascinating new information about mobile phone users. The

range of monthly salaries, which ranges from $2600 to $6200 with an average of $4306.36,

demonstrates a noticeable income variation. In terms of device use, participants use their phones

on any given day for an average of 4.5 hours, however this might vary from 2 to 7 hours. With

an average score of 7.26 out of 10, the general opinion on camera quality is good. Furthermore,

the mean score of 7.91 indicates a clear tendency to suggest these items. On a scale of 1 to 3,

battery satisfaction averages out at 2.03, indicating a neutral to neutral feeling. Age-wise,

participants are between the ages of 21 and 41, with a peak at age 28. The ratio of men to women

is about equal, however women have an advantage at 52.7%.

The breakdown of brand loyalty is complex, with "Loyalists" making up 33.6% of

respondents, Neutrals 34.5%, and Potential Switchers 31.8%. According to battery response, the

majority either indifferent (40%) or pleased (37.3%), with just a small percentage (22.7%) being

dissatisfied. Price (40%) is the main driving factor behind phone sales, followed by alluring

features (29.1%). In the past, participants tended to choose companies like Apple and Huawei,

but in the present, Apple wins out with a staggering 47.3% of the vote, followed by Samsung

with 21.8%.

Bivariate/Multivariate Analyses

ANOVA Analysis

We do a one-way ANOVA to cornfirm whether there is a statistically significant

difference in participants' average monthly salaries according to their brand loyalty status

(Loyalist, Neutral, or Switcher).

ANOVA
Sum of Squares df Mean Square F Sig.
 8

Monthly Income ($) Between Groups 1817887.888 2 908943.944 1.592 .208

Within Groups 61087657.567 107 570912.688
Total 62905545.455 109
Daily Usage (hrs) Between Groups .897 2 .448 .498 .609
Within Groups 96.333 107 .900
Total 97.230 109
Camera Quality Rating Between Groups 6.052 2 3.026 3.470 .035
Within Groups 93.302 107 .872
Total 99.355 109

The Neutral, Loyalist, and Switcher groups all report similar monthly income levels,

indicating no correlation between brand loyalty and revenue. This opinion also applies to daily

phone usage, as there is no discernible difference between any loyalty group's daily phone usage.
 9

On the other hand, camera quality rankings. At least one of these classes distinguishes

camera quality from the others. The Tukey HSD test was used to narrow the attention on these

subtleties. No group stands out above the others regarding monthly revenue and daily phone

loyalty. But the picture shows a different story. The Switchers, those considering switching

brands, are on average 0.647 points more complimentary of their phone's camera than the

Neutrals. However, the admiration for camera quality holds true whether comparing Neutrals to

Loyalists or Loyalists to Switchers.

Chi-Square Test for Categorical Variables

We performed chi test to Test the relationship between Purchase Reason and current

Phone Brand to check if a specific brand influences the purchase reason more than others.

Purchase Reason * Current Phone Brand Crosstabulation

Count
Current Phone Brand
Apple Huawei LG OnePlus Samsung Total
Purchase Reason Features 15 4 1 5 7 32
Brand 9 1 1 3 3 17
Price 21 3 3 6 11 44
Recommedation 7 2 2 3 3 17
Total 52 10 7 17 24 110

Chi-Square Tests
Asymptotic
Significance (2-
Value df sided)
Pearson Chi-Square 3.241a 12 .994
Likelihood Ratio 3.213 12 .994
N of Valid Cases 110
a. 13 cells (65.0%) have expected count less than 5. The minimum
expected count is 1.08.
 10

The information shown is a crosstabulation of "Purchase Reason" and "Current Phone

Brand" relationships. The observed frequencies for each combination of Purchase Reason and

Current Phone Brand are shown in the table. For instance, 15 individuals who bought a phone

because of a feature have an Apple phone, whereas 7 people who bought a phone because of a

feature have a Samsung phone.

The degree of freedom (df) is 12, and the Pearson Chi-Square value is 3.241. The p-value

(Asymptotic Significance) is.994 in this case. The Likelihood Ratio is 3.213, and the p-value

is.994 as well.

The p-value is close to 1 and much higher than the threshold for Significance, which is

0.05. This indicates that the relationship between Purchase Reason and Current Phone Brand is

not statistically significant. In other words, the phone brand they presently possess does not seem

to impact their decision to purchase.

In this situation, there is no significant difference between the anticipated and observed

frequencies for phone brand and purchase reason, consistent with a p-value closer to 1. This

implies that the observed data match the predicted data well.

Correlation Analysis

Correlations
Daily Usage Camera Quality
(hrs) Rating Age
Daily Usage (hrs) Pearson Correlation 1 .101 -.836**
Sig. (2-tailed) .295 .000
N 110 110 110
Camera Quality Rating Pearson Correlation .101 1 -.166
Sig. (2-tailed) .295 .083
N 110 110 110
**
Age Pearson Correlation -.836 -.166 1
 11

Sig. (2-tailed) .000 .083

N 110 110 110
**. Correlation is significant at the 0.01 level (2-tailed).

Age and daily phone usage seem to be at odds with one another. A high negative

correlation is shown by the Pearson Correlation value of -0.836. This is relevant statistically

because it suggests that people may use their phones less often as they age. Simply stated,

younger people are more absorbed in their phones than older people are.

The story becomes slightly cloudier when age and camera quality scores are compared.

Only a weakly negative association is shown by the Pearson Correlation of -0.166. This tenuous

link shows that there may be a little tendency for people to judge the camera quality somewhat

lower as they age. This isn't certain, however, since the p-value falls short of the typical threshold

for statistical significance. Although the pattern is fascinating, it is impossible to declare with

certainty that older people have a different perspective on camera quality than younger people.

Lastly, a marginally positive association of 0.101 is seen between daily phone usage and

camera quality evaluations. This suggests that those who use their phones more could have

somewhat greater expectations for the camera's quality. However, similar to the previous

association, this one lacks statistical significance, making it difficult to identify the link between

daily usage and camera ratings.

Regression

Model Summaryb
Adjusted R Std. Error of the
Model R R Square Square Estimate
1 .176a .031 .003 .749
a. Predictors: (Constant), Battery Satisfaction, Monthly Income ($),
Daily Usage (hrs)
b. Dependent Variable: Brand Loyalty
 12

Coefficientsa
Standardized
Unstandardized Coefficients Coefficients
Model B Std. Error Beta t Sig.
1 (Constant) 2.405 1.333 1.804 .074
Monthly Income ($) -8.122E-5 .000 -.082 -.464 .644
Daily Usage (hrs) -.003 .141 -.004 -.021 .984
Battery Satisfaction -.133 .081 -.157 -1.633 .105
a. Dependent Variable: Brand Loyalty

Equation:

Brand Loyalty=2.405−8.122×10 −5 (Monthly Income)−0.003(Daily Usage (hrs))

−0.133(Battery Satisfaction)

Where:

"2.405" is the y-intercept (constant).

"-8.122E-5" is the coefficient for Monthly Income.

"-0.003" is the coefficient for Daily Usage (hrs).

"-0.133" is the coefficient for Battery Satisfaction.

According to this model, none of the independent variables—monthly income, daily use,

and battery satisfaction—are statistically significant predictors of brand loyalty at the 0.05 level

of significance. This implies that either this model may not be the best fit for the data or that

there isn't a strong linear connection between these factors and brand loyalty.

It's also important to keep in mind that the sort of regression model used and the variables

chosen should be based on the theoretical knowledge of the phenomena being examined. It's
 13

conceivable that additional factors not taken into account in this model or an other kind of

regression model might provide more noteworthy findings.

4. Limitations

1. Non-random sampling can produce bias since people willing to participate in surveys

may have more strongly held views.

2. Recall bias may impact feedback since some users may not recall their experiences

properly.

3. User response may be impacted by outside variables like software changes during the

survey.

5. Relevance to Management

 Marketing Strategy: With price being the major purchase driver, businesses should focus

on cost-effective offerings and highlight the value proposition of their products.

 Product Development: As users are generally neutral about battery satisfaction, there's an

opportunity to innovate in this domain and offer longer battery life or faster charging

solutions.

 Customer Segmentation: The even distribution across brand loyalty segments suggests a

need for tailored marketing strategies for each element. For instance, loyalty programs for

Loyalists and feature-driven campaigns targeting Switchers.

 Age-Targeted Campaigns: Younger users are more engaged with their phones, suggesting

they might be more receptive to mobile advertisements or app-based campaigns.

 Feedback Mechanisms: Given the potential recall biases, companies could invest in real-

time feedback mechanisms to get more accurate user feedback..