Writing up your findings - Business statistics with Excel and Tableau (2015)

Business statistics with Excel and Tableau (2015)

3. Writing up your findings

In chapter 1, I mentioned some of the questions I wished I had been able to answer when I was operating a business. Now I want to suggest ways in which you might structure your response to such questions. Theactual answers come when you’ve done the statistical work (carried out in the following chapters), but it might help to have at least a structure on which you can begin building your work.

Before you do any work on the problem, get straight in your mind exactly what you are being asked to do. You need to be clear and so does the decision-maker to whom you are going to submit your findings. Here isone way to do this:

Copy down the key question that you are being asked in exactly the way that it has been given to you. For example, let us say that you have been asked to answer this admittedly tough question: ‘What factors areimportant for the success of a new supermarket?’

Now keep rewriting it in your own words so that it becomes as clear as possible. Make sure that each and every word is clear. What does ‘success’ actually mean? —in the business context probably ‘most profitable’.What does ‘new’ mean? Is this a completely new supermarket or a new outlet of an existing brand? You might end up with: ‘when we are planning a location for a new outlet for our brand, what factors contribute mostto profitability?’ Doing all this helps you to identify the statistical method most suitable for the task. You also can include here the hypotheses (if any) that you will be testing.

Here is a suggested section order for your report. However, this probably won’t be the order of the work. The order in which you do the statistical work is in Section 3, your plan of attack. So the order

15

of doing the actual work is that you carry out the plan of attack, and then write up your results following the section by section sequence I am giving you now.

Section 1. State the question to be answered clearly and succinctly, as result of the rewriting you did above. Doing this makes sure that you clearly understand the problem and what is being asked of you. Writing out thequestion and stating it clearly right at the beginning of your report also ensures that the decision-maker (the person who posed the question in the first place) and you understand each other. You can also write in thehypothesis which you are testing.

Section 2. Provide an executive summary of the results. This should be perhaps six lines long and contain the key findings from your work. Don’t put in any technical words, jargon or complicated results. Just enoughso that a busy person can skim through it and get the general idea of your findings before going more deeply into your hard work.

Section 3. Outline your plan of attack, describing how you have approached the problem. You also can include here the hypotheses (if any) that you will be testing. A typical plan of attack follows later on in thischapter.

Section 4. Data. Provide a brief description of the data that you have used. The source of the data, how you found it, whether you think it is reliable and whether it is sufficient for the task. It is a good idea to includesummary statistics at this point. This includes information such as the number of observations, and what the variables are.

Section 5. Statistical methodology. Describe the statistical method- ology which you are going to use, for example ANOVA to test whether or not the means are the same.

Section 6. Carry out the tests, making sure to include a description of whether or not the hypotheses can be rejected. This is the central part of the report. Just make sure to focus on answering the question.

Section 7. Write a conclusion which describes the results of the test and how the results answer the question that was asked. You can also include here any shortcomings in the data or in your methodology whichmight affect the validity of the results. The conclusion is very important because it ties together all the previous sections. Here you could include a link to a Tableau ‘story’ as an additional or alternative way ofpresenting results.

Section 8. References/end notes or further information, especially on sources of data, should end the document. If you want to make your document look really good, and you have lots of references, consider usingthe Zotero plug-in for Firefox. It is an excellent free way to manage your bibliography.

3.1 Plan of attack. Follow these steps.

a. Identification of the statistical method that you will use and why you chose those methods

b. What data is required, and where can it be found?

c. Conduct the statistical tests

d. Discuss whether the results are reliable/answer the question. (If not, start again!)

3.2 Presenting your work

While it is probably best to concentrate on written work because the decision-maker might want to read your work in detail, and discuss it with colleagues, a Tableau presentation is an excellent way of making yourpoint over again. See Chapter 2 for more on visualization and Tableau.

Here is a link to some excellent notes by Professor Andrew Gelman on giving research presentations¹

¹http://andrewgelman.com/2014/12/01/quick-tips-giving-research-presentations/