You are to analyse data from the file plALL2017.xlsx, using univariate statistics (i.e., one variable at a time). The file contains information I have assembled about nearly 2700 postcode districts in the UK (e.g., AB12, CF13, BS2, YO8, etc). There are enough variables (columns) for you to develop your own focus. Note that it is important to develop a perspective; doing so will inevitably mean you ignore many of the variables. Do not try to analyse all the data from all perspectives! It really does help if the assignment seems to be about something, or a connected set of issues. It also helps if the assignment seems to be going somewhere. There are a number of ways to get this sense of momentum going. For instance, you may present a finding, then try to rule out competing interpretations. Or you may show that an unexpected interpretation is really what is going on. Or you may present a series of linked findings that make the same kind of point even more forcefully. All of the data applies to England and Wales, but only some of the data is available for Scotland. You must therefore decide whether it is relevant to include Scotland or not. You may also may need to take a view whether to analyse postcode districts with very few habitations. There may be something strange about them. One such postcode I have already removed for you is PE35 (Sandringham), one of the places Her Majesty The Queen lives. Other oddities are postcodes consisting of out-of-town shopping malls, car sales showrooms (where licensed cars may greatly exceed people), universities, factories, GCHQ, DVLA, and so on.

Start with a brief paragraph setting the scene for what your perspective is. This can already start momentum by setting up expectations of what is to follow. In the main body, be sure to consider using (1) informed descriptive statistics and their interpretation; (2) graphical exploration of the data; (3) formal testing of hypotheses. Develop hypotheses on common sense grounds (no need to go to a background literature), and remember that unless a strong case can be made for a direction in what you expect, the default setting must be to use 2-tailed testing. In theory, this “strong case” should be made before you ever look at the data. Please lay out hypotheses as follows:
H1: The more computer lab classes a student attends, the better their assignment mark.
H0: Students who attend more computer lab classes do not get better assignment marks.
This may be tested via correlation, so that more algebraically:
H1: R(lab, mark) > 0.
H0: R(lab, mark) ≤ 0.
Or it may be tested another way. There are often several ways to test a hypothesis. But don’t just do all possible analyses to cover your back, hoping that one of them is the right one; or to show me you can do everything. It is better that you choose the right analysis, arguing why it is the appropriate analysis or others are not so appropriate (e.g., t-tests on data with outliers or large skew do not meet the assumptions upon which they are based). However, occasionally students want go technical, and test a (small) set of findings in a highly systematic way to establish the robustness of their findings. In so doing, they may hope to shed light on the nature of the tests undertaken. However, this should be done in an aware kind of way, noting why you are doing it and relating it to features of the distribution(s) under analysis. If you think this is the same as “doing all possible analyses to cover your back”, then my advice to you is to avoid it.

Please write in simple, grammatical English. Microsoft Word has tools that may help you achieve this . Figures and tables are different kinds of thing, and should be numbered as separate series. A simple sentence or two should describe what each is about. Even though this is a statistics / data analysis assignment, clarity of presentation is something I do care about in written reports, and you should too. Only present those figures and tables and analyses that really help your flow of argument and investigation. The rest can go in Appendices as a safety net. I should never need to read your Appendices, but having said that, you are to lay them out in a systematic way with numbering, should I ever need to go there. Record the R commands that you have used in footnotes where they arise. Finally, the main body of your text (i.e., not including Appendices) should be no more than 15 pages long. (Many fewer is better.) This constraint is to prevent the text being flooded by unnecessarily long tables and figures, making it difficult to follow.

Credit will be given to students who are able to successfully add information to this data. A whole lot more data exists about postcode districts. Also, ONS (Office of National Statistics) has its own parallel classification of districts in the UK. You may use their data instead of postcode data if you prefer, but I will not be supplying any such data. (Great) credit will be given to students who are able to link the ONS data into my postcode data as a preliminary to analyses. Often this is the way research proceeds, by linking in new data to old. However, at your stage of expertise, this may be tricky and you are certainly not expected to do so.