Step 1: Import all necessary packages, convert the excel file into a csv file then use pandas to open the file.

import numpy as np import pandas as pd import matplotlib.pyplot as plt import seaborn as sns %matplotlib inline sns.set_style('ticks') import warnings warnings.filterwarnings('ignore') survey_data = pd.read_csv("task.csv")

survey_data.head()

Data Preparation/Cleaning

Checking the number of records(rows) and fields(columns)

survey_data.shape

survey_data['1. Which feature(s) on the SkyeXperience platform do you use regularly? (multiple answers allowed)'].value_counts(normalize=True)

Checking for null values

survey_data.isnull().sum()

Handling null values by dropping empty columns

survey_data = survey_data.fillna(0)

survey_data = survey_data.dropna(how = 'any')

Questions to be answered during analysis

1. What feature is mostly used by users in (categorized by location, gender and age group)? 2. What feature do users think is the most effective/ineffective and why (categorized by location, gender and age group)?? 3. What percentage of users think the platform is easy to use, has convenient banking, secure to use, consistent across digital platforms, reliable and convenient,  by gender and age group?5. what location do they have the highest users 6. what percentage of customers will recommend the platform 7. what new features or services will users want by gender and age group

survey_data.columns

Renaming columns and encoding ( relabeling) values for better identification for analysis

survey_data.rename(columns = {'1. Which feature(s) on the SkyeXperience platform do you use regularly? (multiple answers allowed)': 'Q1','2. Which particular feature(s) do you think is most effective? (multiple answers allowed)': 'Q2','What are your reason(s) for the answer(s) above?':'Q2a','3. Which particular feature(s) do you think is most ineffective? (multiple answers allowed)':'Q3','What are your reason(s) for the answer(s) above?.1':'Q3a','4. Please tell us to what extent you agree or disagree with the following statements; The SkyeXperience platform..... [is easy to use.]':'Q4a','4. Please tell us to what extent you agree or disagree with the following statements; The SkyeXperience platform..... [is convenient banking for me.]':'Q4b','4. Please tell us to what extent you agree or disagree with the following statements; The SkyeXperience platform..... [is consistent across all digital devices.]':'Q4c','4. Please tell us to what extent you agree or disagree with the following statements; The SkyeXperience platform..... [is secure to use. ]':'Q4d','4. Please tell us to what extent you agree or disagree with the following statements; The SkyeXperience platform..... [is reliable and convenient. ]':'Q4e','5. How likely are you to recommend the SkyeXperience product/platform to family, friends and colleagues?':'Q5','6. Kindly state any recommendation(s) on what you would like to experience or see added on the features of the SkyeXperience platform':'Q6','Demography: Age':'Age_Group'}, inplace = True)

print(survey_data.columns)

Index(['ID', 'Q1', 'Q2', 'Q2a', 'Q3', 'Q3a', 'Q4a', 'Q4b', 'Q4c', 'Q4d', 'Q4e',
       'Q5', 'Q6', 'Age_Group', 'Location', 'Gender', 'Unnamed: 16',
       'Unnamed: 17', 'Unnamed: 18', 'Unnamed: 19', 'Unnamed: 20',
       'Unnamed: 21', 'Unnamed: 22', 'Unnamed: 23'],
      dtype='object')

survey_data = survey_data[['ID', 'Q1', 'Q2', 'Q3a', 'Q3', 'Q3a', 'Q4a', 'Q4b', 'Q4c', 'Q4d', 'Q4e', 'Q5', 'Q6', 'Age_Group', 'Location', 'Gender']]

survey_data.Q1

survey_data.Q1.str.split(",").head(11)

from sklearn import preprocessing le = preprocessing.LabelEncoder() survey_data.Q1a = le.fit_transform(survey_data.Q1) survey_data.Q2a = le.fit_transform(survey_data.Q2) survey_data.Q3b = le.fit_transform(survey_data.Q3)

survey_data['Q1']

survey_data['Q4a'] = survey_data['Q4a'].replace({'Strongly Agree':'Agree','Strongly Disagree':'Disagree'}) survey_data['Q4b'] = survey_data['Q4b'].replace({'Strongly Agree':'Agree','Strongly Disagree':'Disagree'}) survey_data['Q4c'] = survey_data['Q4c'].replace({'Strongly Agree':'Agree','Strongly Disagree':'Disagree'}) survey_data['Q4d'] = survey_data['Q4d'].replace({'Strongly Agree':'Agree','Strongly Disagree':'Disagree'}) survey_data['Q4e'] = survey_data['Q4e'].replace({'Strongly Agree':'Agree','Strongly Disagree':'Disagree'})

survey_data['Q5'] = survey_data['Q5'].replace({'e. Most likely':'Likely','a. Most unlikely':'Unlikely','d. Likely':'Likely','b. Unlikely':'Unlikely'})

survey_data.head()

survey_data.Q5.unique()

Analysis and insights

Univariate analysis

survey_data.Location.value_counts(normalize=True)

Top 3 locations of SkyeXperience users are Lagos accounting for 28% (189 users), Abuja accounting for 8% (58 users) and Ibadan accounting for 4% (41 users)of total users across locations.

survey_data.Gender.value_counts(normalize=True)

71% of SkyeXperience users are male(471 users) and 28% are Female(186 users)

Q1- What features do users mostly use.

import seaborn as sns sns.countplot(x = survey_data.Q1a) #Insight: The top 3 uses of the platform by users are to transfer money, purchase airtime and do the both(transfer money and purchase airtime)

print(survey_data.Q1.str.split(",")[10]) print(survey_data.Q1.str.split(",")[23]) print(survey_data.Q1.str.split(",")[0])

['a. To transfer money']
['c. To purchase airtime']
['a. To transfer money', ' c. To purchase airtime']

Q1. insights

The most used feature on the platform is the Transfer money feature followed by the Purchase airtime feature and lastly a combination of both transfer money and purchase airtime

Q2. What features do users think are most effective and ineffective

import seaborn as sns sns.set(rc={'figure.figsize':(11.7,8.27)}) sns.countplot(x = survey_data.Q2a, )

print(survey_data.Q2.str.split(",")[10]) print(survey_data.Q2.str.split(",")[26]) print(survey_data.Q2.str.split(",")[0])

['a. To transfer money']
['a. To transfer money', ' c. To purchase airtime']
['a. To transfer money', ' d. To pay bills']

Q2. Insights

Users say the transfer money feature is the most effective and their top 3 reasons are to transfer money, to purchase airtime and to pay bills

Q3. -What features are ineffective and reasons

import seaborn as sns sns.set(rc={'figure.figsize':(11.7,8.27)}) sns.countplot(x = survey_data.Q3b)

print(survey_data.Q3.str.split(",")[36], "reason:" + survey_data['Q3a'].values[36]) print(survey_data.Q3.str.split(",")[19], 'reason:' + survey_data['Q3a'].values[19]) print(survey_data.Q3.str.split(",")[0], "reason:" + survey_data['Q3a'].values[0])

['c. To purchase airtime'] ['reason:It more easier' 'reason:It more easier']
['c. To purchase airtime'] ['reason:experience' 'reason:experience']
['a. To transfer money'] ['reason:I do transfer regularly' 'reason:I do transfer regularly']

Q3.Insights

Q4- What percentage of users think the platform is easy to use, has convenient banking, secure to use , consistent across digital platforms and reliable

97% of users agree that the platform is easy to use as they make use of the transfer money feature the most.

plt.figure(figsize=(10,7)) sns.set(style='darkgrid') sns.countplot(survey_data['Q4a']) plt.title('Distribution of users ease of use with the platform') plt.xlabel('Q4a') plt.ylabel('Frequency') plt.tight_layout()

survey_data.Q4a.value_counts(normalize=True)

92% of users agree that the platform has convenient banking system in place

plt.figure(figsize=(10,7)) sns.set(style='darkgrid') sns.countplot(survey_data['Q4b']) plt.title('Distribution of convenient banking with the platform') plt.xlabel('Q4b') plt.ylabel('Frequency') plt.tight_layout()

survey_data.Q4b.value_counts(normalize=True)

81% of users say that the platform is secure to use and 17% of users are indifferent about the security of the platform

plt.figure(figsize=(10,7)) sns.set(style='darkgrid') sns.countplot(survey_data['Q4c']) plt.title('Distribution of platform security') plt.xlabel('Q4c') plt.ylabel('Frequency') plt.tight_layout()

survey_data.Q4c.value_counts(normalize=True)

Majority of users 92% agree that the platform is consistent across digital platforms

plt.figure(figsize=(10,7)) sns.set(style='darkgrid') sns.countplot(survey_data['Q4d']) plt.title('Distribution of platform consistency acorss digital platforms') plt.xlabel('Q4d') plt.ylabel('Frequency') plt.tight_layout()

survey_data.Q4d.value_counts(normalize=True)

Majority of users 93% agree that the platform is reliable.

plt.figure(figsize=(10,7)) sns.set(style='darkgrid') sns.countplot(survey_data['Q4e']) plt.title('Distribution of reliabilty of the platform') plt.xlabel('Q4e') plt.ylabel('Frequency') plt.tight_layout()

survey_data.Q4e.value_counts(normalize=True)

Q5. Likelihood of users to recommend the platform to others

Majority of users (79%) would likely recommend the platform to others whereas 16% of users are of the opinion that they're unlikely to recommend.

plt.figure(figsize=(10,7)) sns.set(style='darkgrid') sns.countplot(survey_data['Q5']) plt.title('Distribution of platform recommendation likelihood') plt.xlabel('Q5') plt.ylabel('Frequency') plt.tight_layout()

survey_data.Q5.value_counts(normalize=True)

.css-9i38tn{letter-spacing:inherit;-webkit-text-decoration:none;text-decoration:none;}Step 1: Import all necessary packages, convert the excel file into a csv file then use pandas to open the file.