Phone Business Data Essay - Answers To Questions Raised By Walter Waterson

o Clarity refers to how easy it is to read and understand the answer.
o Accuracy refers to the correctness of a calculation or response.
o Comprehensiveness refers to whether the whole question was answered and the required detail was provided.
o Appropriateness refers to how relevant and succinct a response is (long-winded and off-topic answers may not be appropriate).
o Professionalism refers to how well the responses are formatted, communicated factually, and related back to the business problem.
Instructions

• Provide your answers at the end of the “email chain” below. Ensure your reply is professional and follows all instructions in the email chain.
• Number your responses 1 to 20, corresponding to the questions in the email, and use full sentences.
• Do all calculations in Excel and save your workings (e.g. formulas or pivot tables etc.).
• Submit the assignment via Turnitin in MyPIA. The submission should be in two parts: This document (when completed) in .doc or .docx format, and the Excel file with your calculations in .xls or .xlsx format.
Introduction

Imagine you have just been hired as a part of the graduate recruitment program at Acme Accounting Services Pty Ltd. Once you pass a six-month probation period, the firm will sponsor you to complete your Chartered Professional Accounting (CPA).

The firm offers accounting and analytical services to other businesses. You’ve been given a client account to help with Polytechnic Plumbing Services, who provide emergency and regular plumbing services to consumers and small businesses.

Dataset and Analysis of Phone Calls Received

With reference to the email sent by Walter Waterson regarding his new implementation of the phone business, the following questions have been answered. The analysis conducted to answer his questions is mainly based on the dataset provided by him.

1. There has been implementation of a new phone system in their business which will automatically record the data on all the incoming calls on the telephone. The values 2 and 69 given in the image indicates the minimum and the maximum number of calls respectively received by the company in a day. 46 and 61 are respectively the first and the third quartiles of the data on the number of calls. 46 indicates that 25 percent of the calls received in a day are less than 46 and 25 percent of the received in a day are more than 61. 52 indicates than 50 percent of the calls received in a day are less than 52. This is known as the median.
2. It is correctly understood by Walter that the plot given above shows that the distribution of the number of phone calls in a day in actually left skewed (negatively skewed). This indicates that the number of phone calls above the median number of phone calls (48) are widely scattered whereas the number of phone calls less than 48 are close to 48.
3. It has been observed that the minimum number of calls in a day was 2 calls but mostly, on an average the minimum number of calls received in a day was 35. There can be several reasons for receiving 2 calls in a day. There can be network error which made people unable to reach to the company. Thus they received less number of calls. There can also be some system upgrades for which the company shut down their telephone lines. This can also result in the lesser number of calls received by the company.
4. From the box plot generated, it can be said that 2 is an outlier. There are no other existing outliers in the dataset. A data point is called an outlier if it lies 1.5 times of the interquartile range above the third quartile and 1.5 times of the interquartile range below the first quartile. These two ranges are known as the upper and the lower outlier range.

In this data, the first quartile is 44 and the third quartile is 60. Thus, the interquartile range is 16. Therefore, the upper outlier range is (60 + 1.5 * 16) = 84 and the lower outlier range is (44 – 1.5 * 16) = 20. There were two days when the number of phone calls received is less than 20. In one day, only two calls were received and on the other day, the number of calls received was 3. More than 84 calls were not received on any day.

5. From the ratings given by the customers on their satisfaction with the services of the company, it has been observed that the average rating given by the selected 100 customers is 3.88 which is considerably high on a scale of 1 – 5. The standard deviation of the ratings has been found to be 1.22.
6. The chart showing the average satisfaction rating for each of the job type is given below:

7. It can be observed from the analysis of the responses of the 100 callers that the number of people that have the emergency service from the company is 27, the number of people that called in for improvements are 37 and the number of people that called for maintenance is 36.

It is not possible to find the average type of calls as the type of phone calls is not numerical. But it can be said that people mostly call the company for services related to maintenance and improvements. Lesser emergency calls are received in comparison to the other two call types.

From the given chart, the approximate mean duration of calls has been found to be 4.99 minutes with a standard deviation of 2.32 minutes.

The data on call duration was started to be captured in the last few days. It has been observed from the data on call duration that the average time people talk on the phone with the company is 4.50 minutes with a standard deviation of 1.07 minutes.

It can be seen that there has not been much difference in the duration of calls from 2015, though the standard deviation of the call duration has decreased in 2017. Moreover, it can be said that the duration of calls has decreased in 2017. One reason behind this can be such that the company has been more efficient in solving the problems over the phone. Thus the call durations were all close to the average duration in 2017 than in 2015.

From the data summary given in the contingency table, it can be seen that the number of people that has given a five-star rating is 35 out of 100 chosen callers. Thus, the probability of receiving a five-star rating is 0.35.

From the contingency table, it can also be seen that the number of callers who have given ratings less than four-star on jobs classified as “improvements” is 21. Thus, the probability of getting a rating less than 4 star on the job classified as “improvement” is 0.21.

Average Duration of Calls

Walter wants to attach the average of the facebook ratings for advertisement. The average ratings for maintenance and emergency will be considered for it. Ratings for improvements will not be considered. Thus, the average facebook rating to be displayed for the advertisement is 4.38.

The customer ratings that has been collected are on the basis of the responses of the callers of only three days. Originally, there are a lot of more customers. The ratings might be a lot different from what has been received by the three day customer satisfaction. By conducting respective analyses, it can be said with 95 percent confidence that the average rating given by the callers on facebook is not less than 4.

ANOVA test has been conducted to test the existence of any reliable difference in the satisfaction scores between different job types. From the results of the ANOVA test, it can be said with 95% confidence that there is significant difference between the satisfaction scores of the callers of different job types as the p-value is less than 0.05 (the level of significance).

On the basis of the last question, independent sample t-test has been conducted considering two job types at one time and comparing their mean satisfaction scores. It can be seen from the results of the three different t-tests that there is significant difference between the satisfaction scores for each of the job types with the other two as the p-value is less than 0.05 (the level of significance) for all the tests. This can be claimed with 95 percent confidence.

The improvement job has been difficult for the company. Thus, they decided to stop the improvement plumbing work and focus on the emergency work and the improve the business in that sector. It has been observed from the results of the t test that the p-value (0.182) is greater than the level of significance (0.05). Thus, there is no significant difference in the return on investment for the two different job types. This can be claimed with 95 percent confidence.

The null hypothesis for answering the above problem was “No significant difference exists between the return on investments (ROI) on the two different job types improvements and emergencies”. The above analysis shows that the null hypothesis has been accepted.

If the statistical test shows that there is difference between the ROI of improvements and emergencies, there is an error in the analysis. The probability of that error is 0.05. This type of error when the null hypothesis is rejected when it is true is known as type one error.

It has been observed from all the analysis that there is no significant difference between the return on investments of the two different job types, emergencies and improvements but the overall satisfaction scores of the customers are much higher for the emergency type of business than the improvement type of business. Thus, the improvement business has to be made better in order to satisfy the customers. Otherwise, it will be better to switch to the emergency business and shut down the improvement business from the company side. This is more likely to increase the customer satisfaction.