Memorandum Assignment 2020
Date: 10th May, 2018
To: Michelle Yeoh, Chief Data Analyst
From: Grace Park, Research and Analysis Department
Subject: Analysis of Mobile Phone Data
Dear Michelle,
This memorandum provides a better understanding of statistics to make decisions in the business. You are intended to publish an article on the current smart mobile phone usage in Australia.
It is identified that Australia has 2nd position in the world behind Singapore for smart phone usage. This article can be important for the Businesses, including telecom to improve their own operations, marketing strategies, etc., for the digital age.
This article will include information related to the user’s expenditure, usage patterns and demographics. Now, there is task to conduct the market research and find the valuable information regarding the article.
For this, a random sample of 150 smart phone users in Australia is taken to determine their usage and other associated data.
Q1. An Overall View of Mobile Phone spend per month
Can you provide an overall summary of how much consumers spend on their smart phones per
month?
From the data analysis, it is identified that the consumer spend maximum $216 per month while minimum $11 per month.
At the same time, the average monthly bill for smartphone usage is $67.64 that is less than the maximum monthly bill due to difference between the average value and the different data points.
In addition, total monthly bill for the given sample of 150 participants was $10147.
Additionally, some of the people use less electricity as they spend less on electricity while some of them use more amount of electricity paying higher monthly bills.
Q2. Monthly Bill vs Lifestyle Factors
In particular, does there appear to be any difference in how much consumers spend on their smart phone per month, across the three most common geoTribe categories Achievers, Independents and
Suburban Splendour.
In order to determine the impact of lifestyle factors, regression analysis is conducted. From this it is identified that p-value is 0.8 which is higher than significance value of 0.05.
It means there is no significance of the results or there is no significant difference in dataset (Wheeler et al., 2013).
It shows that the monthly smartphone bill does not relate to lifestyle factors. Any difference does not appear regarding how much consumers spend on their smart phone per month, across the three most common geoTribe categories – Achievers, Independents and Suburban Splendour.
The life style factor does not affect the consumption of electricity for people.
Q3. Mobile Phone Affordability
- Can you estimate the average monthly bill for all smart phone owners in Australia?
In order to estimate the average monthly bill for all smart phone owners in Australia, z-test is conducted that reflects confidence interval for mean.
From this, it can be stated that the average monthly bill for all smart phone owners in Australia is between $62.66 and $72.64. At the same time, a point estimates for the true average monthly bill for all smart phone owners in Australia is $67.65 and we are 95% confident that the true mean is between $62.66 and $72.64. However, the margin of error is large (4.99).
- Using your smart phone as a payment device is the next frontier in a cashless society. Can you estimate the proportion of all smart phone owners in Australia that use their phone as a payment device?
Yes, it is right to say that using your smart phone as a payment device is the next frontier in a cashless society. From the data analysis, it can be estimated that the proportion of all smart phone owners in Australia that use their phone as a payment device is between 69.17% and 82.83%.
At the same time, a point estimates for the true proportion of all smart phone owners in Australia to use their phone as a payment device is 76% and we are 95% confident that the true proportion is between 69.17% and 82.83%.
It means there are high numbers of smart phone users who use their phone as payment device.
It is because there is increasing online trade trend in Australia showing the requirement for users to pay online by using their devices especially smart phones.
- Additionally, can we conclude that there is a difference between the proportion of male and female owners when it comes to using their smart phone as a payment device?
In order to determine difference between the proportion of male and female owners when it comes to using their smart phone as a payment device, Hypothesis Test for π (Proportion) is conducted. From data analysis, it is determined that p-value (1.00) is greater than 0.05 (p>0.05) means null hypothesis cannot be rejected (Raghupathi and Raghupathi, 2014).
It indicates that there is no difference between the proportion of male and female owners in using their smart phone as a payment device. It is because half of smart phone owners using smart phone as a payment device are male and remaining half are female.
It means both male and female equally use their smart phone as a payment device. Both have 50% proportion in using the smart phone as payment device. So we cannot conclude that that there is a difference between the proportion of male and female owners in using their smart phone as a payment device.
Q4. Mobile Phone Usage
- A previous report, published by Business Insider, indicated that the proportion of smart phone owners in Australia who use their smart phone for work‐related activities is no more than 75%. A colleague of mine believes that the proportion is higher, given that the smart phone synchronises email, calendars and documents. Can you check my colleague’s claim?
To check the colleague’s claim, hypothesis test is conducted. From this, it is determined that p value is 0.0366 that is lower than the significance value (0.05). It shows that we can reject null hypothesis regarding lower proportion of using smart phone for work‐related activities.
It means a previous report published by Business Insider was not right that the proportion of smart phone owners in Australia who use their smart phone for work‐related activities is no more than 75%.
The claim made by the colleague is right as the proportion is higher or more than 75%, given that the smart phone synchronises email, calendars and documents.
The most of the smart phone owners use the smart phone at work related activities like email, calendars and documents.
- A business rival stated that the average number of phone calls made by Australian smart
phone owners last month was at least 27 calls.
I feel that this average may be overstated as there are other ways in which communication can occur, including popular Internet based alternatives.
Is there any evidence to suggest that the average calls last month is less than 27?
To determine the claim regarding average calls last month less than 27, hypothesis test is conducted. From this analysis, it is identified that p-value is greater than 0.05 significance value. It means the null hypothesis cannot be rejected.
Null hypothesis indicates that the average number of phone calls made by Australian smart phone owners last month was at least 27 calls.
This average is not overstated because people make more calls to communicate rather than other communication modes like email, SMS, or social media.
The calling through phone is the most favourable way of communication for people in Australia. There is no evidence to suggest that the average calls last month are less than 27.
Q5. Relationships
I would like to see whether factors listed below provide any explanation in the variation of monthly phone bills between consumers. If so, can you also indicate which factor is the most important?
Number of Calls
During analyse the relationship between monthly mobile bill and call made by the user, it is found that in the increase and decrease of the bill, there is not much effect of the call on the bill.
It is because in regression analysis, the value of Y is found 0.353. At the same time, the value of the R sequare is found by 0.112. It depicts that there is not much strong relationship between the mobile bill and call.
The result shows that the relation is possible because R square value is found positive (Bain, 2017). But, it is below than 0.5 that means call made by the customers do not increase bill much.
SMS’s
At the same time, it is also assumed that SMS is also a part of the overall cost of the mobile bill. Due to this, it is possible the total used SMS affect the amount of monthly mobile bill. In regard of this, the finding of the analysis depicts that there is not much relationship between the monthly mobile bill and used SMS.
It is because the value of the R square is found negative as -0.15. It means that there is negative relationship between monthly mobile bill and SMS. It is because a negative value represents there is no relationship (Busk and Marascuilo, 2015).
Data
In the context of this point, it is found that in the increase and decrease of the mobile bill, the role of the data that is used by the consumer is important.
During the analysis of relationship between the monthly bill and data, the value of R square is found 0.515 that represent that there is high relationship between these both variables.
On the basis of this, it can be said that when monthly mobile bill increases then the role of the used data by the customer is important (Little and Rubin, 2014).
Hence, it can be said that the consumption of the data as the internet determines the monthly mobile bill.
In the context of the correlation between the three different monthly bill and with the calls, SMS and Data, it is found that most strong relationship with is between the monthly data bill and data because significant value is found 0.71.
On the other hand, there is also positive relationship between the calls – monthly bill and SMS – Monthly bill. In this, the significant value is found 0.41 and 0.32 that depicts that value relationship is positive but it is not much strong compared to data – mobile bill (Crowder, 2017).
It is because in these significant value is less than 0.5.
Q6. Appropriate Sample Size Finally, I am concerned that the sample of 150 smart phone users in Australia is too small to provide accurate results as this seems hardly enough data. For a study we intend to undertake next year, we would like to be able to:
- estimate the proportion of Mobile Phone users that have purchased an item online to within 6%, and
from data analysis, it is determined that the proportion of Mobile Phone users that have purchased an item online to within 6% is between 2.20% and 9.80%. it is because the most of the smart phone owners like to purchase more online. So there are less number of the users purchasing an item online to within 6%.
(b) accurately estimate the average monthly bill to within $4.
From data analysis, it can be interpreted that accurate estimation for average monthly bill to within $4 will be between -1.03 and 9.03. It means average monthly bill will be between -1.03 and 9.03 rather than average monthly bill to within $4.
2.2% mobile Phone users would we need to include in next year’s survey to satisfy both of these requirements. It means 33 users would be required to meet these requirements.
References
Bain, L. (2017) Statistical analysis of reliability and life-testing models: theory and methods. UK: Routledge.
Busk, P.L. and Marascuilo, L.A. (2015) Statistical analysis in single-case research. Single-Case Research Design and Analysis (Psychology Revivals): New Directions for Psychology and Education, p.159.
Crowder, M.J. (2017) Statistical analysis of reliability data. UK: Routledge
Little, R.J. and Rubin, D.B. (2014) Statistical analysis with missing data. USA: John Wiley & Sons.
Raghupathi, W. and Raghupathi, V., 2014. Big data analytics in healthcare: promise and potential. Health information science and systems, 2(1), p.3.
Wheeler, D., Shaw, G. and Barr, S., 2013. Statistical techniques in geographical analysis. USA: Routledge.
