Having just read an article on the differences in the readability of texts in the Sciences, Social Sciences, and Arts and Humanities, I decided to distract myself from my massive pile of work with a quick and dirty investigation into the readability of my own writing in an attempt to answer the question:
Does my writing differ according the audience?
Methods adopted
Microsoft Word was used to analyse the readability of 15 texts: five academic papers where I was the primary author; five feature articles I have written for Research Information; and five blog posts I have written for www.davidstuart.co.uk. All tables, pictures, and bullet points were extracted from the documents before carrying out the analysis. The were analysed in terms of:
- Number of words per sentence - generally speaking longer sentances may be considered more difficult to understand.
- Characters per word – longer words are considered more difficult.
- Number of passive sentence – passive sentaces are thought to inhibit the flow of reading and make it more difficult.
- Flesch reading ease – a readability test where the higher the number the simpler it is thought to be. [0.0 - 30.0: best understood by university graduates].
If my writing does indeed vary according to the audience, it would be expected that the results would look like this:
| Academic Articles | Professional Magazines | Blog Posts | |
| Words per sentence | Highest | Middle | Lowest |
| Characters per word | Highest | Middle | Lowest |
| Passive sentences (%) | Highest | Middle | Lowest |
| Flesch Reading Ease | Lowest | Middle | Highest |
With the academic articles being more difficult to read than the professional articles, and the blog posts being the easiest to read of all.
Results
Academic Papers:
| 1 | 2 | 3 | 4 | 5 | Average | |
| Words per sentence | 26.8 | 26.3 | 33.0 | 29.6 | 31.6 | 29.46 |
| Characters per word | 5.1 | 5.5 | 5.2 | 5.1 | 5.4 | 5.26 |
| Passive sentences (%) | 26 | 56 | 52 | 42 | 48 | 44.8 |
| Flesch Reading Ease | 35.2 | 22.7 | 27.6 | 30.3 | 17.2 | 26.6 |
Professional Magazines:
| 1 | 2 | 3 | 4 | 5 | Average | |
| Words per sentence | 24.3 | 25.6 | 24.5 | 26.2 | 24.6 | 25.04 |
| Characters per word | 5.2 | 5.2 | 4.9 | 5.1 | 5.2 | 5.12 |
| Passive sentences (%) | 16 | 5 | 18 | 33 | 14 | 17.2 |
| Flesch Reading Ease | 35.0 | 34.6 | 40.8 | 32.1 | 30.2 | 34.54 |
Blog Posts:
| 1 | 2 | 3 | 4 | 5 | Average | |
| Words per sentence | 30.0 | 28.5 | 22.0 | 26.1 | 30.3 | 27.38 |
| Characters per word | 4.7 | 4.8 | 4.8 | 4.6 | 4.9 | 4.76 |
| Passive sentences (%) | 10 | 15 | 30 | 9 | 11 | 15 |
| Flesch Reading Ease | 48.5 | 39.2 | 47.1 | 46.0 | 26.9 | 41.54 |
Whilst I may not think that there are vast differences in the writing style or vocabulary I use in different forums, reducing it to a few crude numbers would certainly suggest that there are. The only unexpected result was that there were more words per sentance within my blog posts than within my professional magazine articles.
It is also comforting to see that the numbers go in the right direction. If I had discovered that my blog posts were more turgid than my academic articles, it would have probably been time to hang up my keyboard.
Conclusions
The main conclusion is that I am far too easily distracted from work, and if I continue I will probably end up in the poorhouse….although it might be interesting to do a webometric study combining readability with impact factor….