This post continues my review of the online course, Using Data in the Health and Social Care Environment. The second part of Section 3 is about presenting data in a report, and it concludes the course.
In December there will be an evaluation day, which I might also review here, so this might not be the final post in the series.
Presenting data appropriately
This final part of the course starts by returning to the subject of data presentation, with two more reminders in bold text about crediting sources. This advice isn’t really satisfactory. In addition to giving credit, it’s important for anyone who reads your report to be able to verify its accuracy but the course doesn’t tell you this.
There is some general advice, much of which is not very useful. For example:
“Use a standard font style and size for all text and labels in tables that you design.”
Why? There’s no clue in the text. Many tables use bold text for row and column headings, and indeed the example table shown later does exactly that, contradicting the advice.
It goes on to say:
“Obviously, this is more difficult with external graphics you have cut and pasted.”
You can’t really cut and paste external graphics. You can only copy and paste them. And if all you do is copy and paste then you can’t control the font style at all, so saying it’s more difficult is a laughable understatement.
In the examples of copied and pasted graphics, some of the text is so small it’s impossible to read. It can’t be enlarged, and the link provided doesn’t take you to the original graphic, just to the website where the original graphic was allegedly generated. This means it’s virtually impossible to verify the accuracy of the examples.
There’s a list of four mistakes to avoid, and they all seem reasonable. For example:
“Don’t over-claim with your data: don’t try to draw general or universal conclusions from small samples of context-specific data.”
There wasn’t any discussion of sampling earlier in the course, though. By the time you come to present the data, it’s too late to fix the sample.
There are three examples copied and pasted from websites. Despite the previous advice to “Make sure any tables or graphics are clearly labelled…” not one of the three is clearly labelled.
In the example table there are incomprehensible code numbers, like: “Burnley 007C (E01024858)”.
In the example bar chart the axes aren’t labelled. There’s a little square blob at the right-hand side that’s just labelled “Score”. The caption does explain what the chart means, however.
In the example map hardly any of the text is readable. At the right there seems to be some kind of colour key, but there are colours in the map that don’t seem to be in the key, and if there’s any text on the key to explain what it means, it’s not readable.
Structuring and writing a report
The next part of the course explains that writing a report is like following a recipe, and it includes a template for structuring a report. As before in the course, there’s an underlying assumption that the report will be a bid for funding:
“If you are looking to persuade someone to invest in your service, you need to be clear about what you do, what your strengths are and how you can make a difference.”
A peculiar statement in the template indicates that the whole thing might have been copied and pasted from elsewhere without acknowledging the source:
“Use Harvard System – guidelines in pack”
There is no “pack” accompanying this course, but perhaps there was an information pack along with the original report template that this was taken from. There’s no link to any information about the Harvard system, and the course itself doesn’t use the Harvard system.
The report structure suggested by the template is conventional. No one needs a course to find out how reports are structured — it’s simple enough to copy the headings from some other report.
The template doesn’t include a glossary or a list of figures.
It does introduce a new term, core data:
“(e.g. age, gender, ethnicity, time in UK, citizenship, languages, religion, sexuality, disability)”
This might be because the template was copied from a context where these data items really were core.
The final course assignment is to create a report structure, not a complete report. However, it has to present actual data, which doesn’t entirely make sense because there was no data collection exercise following the design of the survey earlier in the course.
My attempt at the assignment will present some published background data about NHS complaints within the outline of a conventional report structure, although it won’t be exactly the same as the template in the course. Some of the data about complaints is quite interesting and I might publish it here, too.
Overall, the title of this course gave it grandiose pretensions: Using Data in the Health and Social Care Environment. The actual course seems to have been based on a guide to writing a certain type of funding application, with some parts omitted.
The wider context of health and social care, and the role that data plays in health and social care policy, strategy and implementation, were all ignored, as were indicators, targets, the various sources of bias, the concept of open data, and so on.
It would have been useful to explore all the broad categories of data relating to health and social care: paywalled research data, public research data, data published by public bodies, unpublished data held by public bodies (accessible using the Freedom of Information Act), unpublished data held out of reach of the FoIA, and things that no one knows (so that you have to do your own research).
The audience for the course was poorly defined, causing much weirdness. For example, the term average was explained in great detail while the term spreadsheet apparently needed no further explanation, and any possible connection between averages and spreadsheets wasn’t explored.
The course content was poorly structured. Instead of a focus on using data, a big part of the course was about collecting your own data. Most people who use health and social care data don’t need to collect their own. Indeed, it’s important for transparency in health and social care that anyone should be able to find stuff out without starting from scratch and doing basic research themselves.
In the detail of the content there were so many errors and omissions that I gave up trying to document them all here. One glaring omission throughout was the lack of further reading. It would have been a great improvement if there had been no factual course content at all, but only links to authoritative sources.
The assignments were not well designed. For example, one of the outcomes was supposed to be:
“Be able to plan a data collection exercise to support a case for change in health and social care.”
But there was never any planning for a data collection exercise, just a survey design.
The online presentation was badly designed, although the technology only failed in minor ways and the worst of these were addressed quickly.
Some of the attempts at webifying course content to make it more colourful and clicky only made it harder to read. The course could have been delivered more cheaply and effectively as a single PDF or e-book. Using a mixture of PDF, Microsoft Word forms and videos was pointless, as they were only used for things that could easily have been done on web pages.
Tinkering won’t fix this dreadful course. Someone needs to start again from scratch and get the fundamentals right.