…living up to its name

six bloody years

I’ve been planning to put something new on to this blog.

This, finally, is the attempt that succeeded.

It turns out writing very little, and saying nothing, is the key.

Who knew?

Top UK Universities : Combined Rankings

There are various  league tables out there for UK Universities. I’ve collected the results from a number of them, a league table based on league tables. This should hopefully help to remove any biases or weaknesses in particular methodologies. The results are further down this post.

I collected results for just 53 Universities, not the full 120 odd that exist in the UK. This was due to laziness, and to be honest I’m more interested in the higher end of the numbers. However I’m fairly sure no university I’ve excluded would come higher that those I’ve included In fact it was originally going to be 50, but as I collected from the various sources I added a few more around the cut off point.

For each ranking, I’ve recorded the position (e.g. 5th) , and then converted it to a score. To a create a score I simply subtracted the ranking position from ‘101’, which ensures that the University ranked first will get 100 points. A good University (according to the rankings!) will have a low number ranking and a high score, e.g. a University ranked 5th will get a score of 96 (101-5=96).

Let’s just be clear at this point, I’m not a statistician, this isn’t remotely scientific, or fair, or well thought out, or thought out at all in fact. Did you get that? Perhaps read it again to be safe. These numbers are crap, and any conclusions drawn on them are without foundation! I’m also no Higher Education expert.


Comment on Sources:
I’m not going to go in to detail about each source, you can follow the links, and if that seems like too much effort, then this Wikipedia page provides an overview for some.
I’ve provided two totals, one for UK only based rankings, and the other includes the international rankings.

The UK only rankings –  and it is my impression that the Guardian in particular – focuses on Teaching. They are, after all, aimed at prospective students. Though there is a danger in focusing two much on teaching resources, as ultimately one University may have fantastic teachers, amazing classrooms and great support, but ultimately is seen as a bad University by employers and the public at large (and to be ‘highly respected’ normally requires a good research record, not to mention being very old). You see, that could be rubbish, I don’t really know, you’re taking this with a pinch of salt right?

The ‘Performance Ranking of Scientific Papers’ from ‘Higher Education Evaluation and Accreditation Council of Taiwan’ is perhaps the most controversial. A major ingredient is citation/impact factors from SCI and SSCI, so those stronger on the Humanities will suffer due to these disciplines being excluded . Interestingly those who focus on the Social Sciences also seem to suffer as well, notably the LSE and Warwick. As I added these numbers in last, it was very notable that some Universities moved several positions due to its inclusion.

The Result:

Click on one of the two following links:

Combined UK University Rankings (excel) (recommended)

Combined UK University Rankings (via Google docs) (as a spreadsheet)

You can order the list by any field. There are two totals: the first using the three UK only rankings, and the second, one of the middle columns, is a total which takes in to account both UK and worldwide rankings.

The rest of the columns are either raw league table data – in black text, or scores – in red.

A score is: 101 minus the ranking. The scores just make it easier to add up and order the totals by highest score, though working in this was does make things a little messy.

The worldwide rankings have an extra column, they include the world ranking as well as the UK only ranking (A University may be the 4th UK university in the list but the 28th University overall). You could potentially do something with the world ranking, e.g. if one comes 10th in the world results, the next comes 11th and the third comes 98th, then clearly it suggests that the first two are broadly similar while the third is not at the same level, though my method simply treats them as first, second, third, and does not take this in to account.

Some Universities did not appear in all the world rankings. Simply giving them a zero score seemed a little harsh, so I hacked it a bit. If, say, the lowest score was 60, then any University without a score may get 40. I know just about everyone will be pulling out their hair out at such random stupidity, though it seems to avoid those not appearing on certain tables being heavily penalised. Especially as some Universities do seem to be randomly missing from certain worldwide tables.

As mentioned above, the Performance Ranking of Scientific Papers is perhaps the most controversial here, and perhaps should not be excluded (comments welcome)? They do explain on their website the pros and cons of their method: The Humanities are more or less ignored, while the Social Sciences are treated like the Sciences, however, as they note, the datasets they use include far fewer Social Science journals, which means these subjects will score relatively lower than the sciences.

This seems true, the LSE amazingly does not appear at all (it normally appears in the top 5), and Warwick appears very low in the list, even though it has a medical school, something they say helps pull Universities up the list. In fact, before this data was added, the LSE was fourth over all, now I’ve added this data they are twelve! I’ve created a column which shows totals ignoring this ranking.

Top 20

UK-only rankings

  1. Oxford
  2. Cambridge
  3. LSE
  4. Imperial
  5. St Andrews
  6. Warwick
  7. UCL
  8. York
  9. Durham
  10. Loughborough
  11. Bath
  12. Exeter
  13. Edinburgh
  14. Leicester
  15. Nottingham
  16. Kings college
  17. Lancaster
  18. Southampton
  19. Bristol
  20. SOAS
All Rankings
1= Oxford/Cambridge
3 Imperial
5 Edinburgh
6 Warwick
7 Kings college
8= St Andrews/Bristol
10 Nottingham
11 LSE
12 York
13 Manchester
14 Durham
15 Southampton
16= Leicester/Sheffield
18 Birmingham
19 Glasgow
20 Bath


My thoughts:

  • First, a week a go, I asked on this blog for people to provide their top 20 lists, you can see them here. My question was badly phased, but the replies are interesting. It includes my results in the first comment (I wrote this without looking at any of these rankings first).
  • Looking at my guesses, I clearly have an aversion to Universities starting with L. Completely missed out Leicester, Loughborough, and Lancaster. The Scots also faired badly from my off-the-top-of-my-head list: St Andrews was no where to be seen, and yet is near the top of both lists. Aberdeen and Dundee both are close to the top 20, yet I would have probably failed to include them in a ‘top 30’. Oh, and somehow forgot Durham.
  • I think I’ve always put UCL as the ‘one after oxbridge’, yet according to these results Imperial, LSE, St Andrews and Warwick are more or less on an equal pegging.
  • I’ve also thought of the groupings a bit like the football league tables: Russell Group, then the 94 group and then the rest. With people joining/leaving these groups as they progress or stagnate. These results show this to be wrong. Looking at the UK-only top 20, 9 of them are 1994 group (and so coming out better than many Russell group Universities). In fact the LSE and Warwick were both in the 94 group until recently, which would have lead to the majority of Universities in the top 20 being in the 1994 group! There are Universities in neither of these groups who are easily ahead of some of those in the Russell group.
  • As you can see from my guesses, I put Manchester, Birmingham and Southampton higher than their actual results, so why were MY expectations high for these organisations? The first two being grand old Universities and Southampton perhaps being accounted for because the one department I know something about – Electronics and Computer Science – is very highly regarded.
  • If these results really do reflect the Research (and teaching) ability of Universities, and if the Russell group is, as it is often portrayed, the leading research Universities, and the 1994 group being smaller research Universities, then there is argument that their should be some movement in group membership (I shall leave it to the reader to look at the excel file and decide who should move up and down!).
  • Having said this, the Russell Group website reports that the group accounts for 68% of all research income, so not doing that badly.
  • Oxford and Cambridge were equal in the international results, Oxford just one point ahead in the UK-only results. So no conclusions there.
  • The Times notes in its own assessment how there is almost a clear split between pre and post 1992 Universities, the list starts with the ‘old’ Universities, and then the ‘new’ universities, with only a couple of exceptions.

And Finally…

I have tried to provide some comment, but this is just my personal view based on near total ignorance. By all means laugh, but don’t get upset.

Link to the results excel file again Combined UK University Rankings.

(this post was slightly updated in November 2008 to improve readability)

Brighton New England Quarter

A few years a go I created some new pages on the ‘New England Quater’, on the site of some old railway works next to Brighton station and near me. Originally the pages just had some photos of how the site looked before work started, and at various stages since (stupidly I normally took this late in the day as the sun was setting which produced poor photos), however at the time I couldn’t find any website providing general information on the project so provided some background and useful links. Since then there has been much more information put on to the web (and lot of information removed as well).

Anyway, I write this as I’ve noticed that the council have put out a report this year reviewing the current state of play, see the link ‘briefing note 2008’.

News Quiz

On a Saturday, I’m prone to remembering the fantastic News Quiz (or equally fantastic Now Show), which starts at 12.30 (why am I saying that, you know that right?), at exactly 1pm. At this point I dash to listen again (now the radio iplayer) to listen to it there. At this point I’m often frustrated as it is still playing the show from the previous week. Do the BBC not understand by problem!

Never really thought about what goes on behind the scenes but it all seems quite impressive. Never really thought about the fact that *shock* they might actually encode the recording before i can listen to it (lots more technical stuff can be found in the many posts here).

Welcome back

In August I signed up with Dreamhost.

Until now I have used Freedom2surf’s web hosting service. I originally signed up with them as they seemed a good company (and still do, though now owned by pipex) and were cheap, £25 a year. When I first started using them they transferred my domain to them (it was all a bit new to me so I just presumed this was a required step) and their domain name costs were not cheap (£10). A few years later I wanted email as well (another £25), and the cost started going up. The service was stable, but I was started to need more. They offered just one database, not much when you are playing around with different software, such as Joomla, running WordPress and trying to experiment with your own silly web apps which require a DB, plus the quote for the database was tiny (10mb). Plus their web stats were crappy, and the log files were fiddly (kept for a few days on a secure website, no ftp and wget didn’t seem able to get them which made automating difficult, plus each file would download with the same filename by default).

During the summer I ended up at the GoDaddy website. I had heard bad things about them, but they did seem to offer an awful lot for a very cheap price. But I wouldn’t get that warm cosy feeling having my website hosted by them, so I waited. Dreamhost were recommended but not the cheapest. However after seeing just how much you do get (shell account, cron, etc), and that it would still be cheaper that my current setup, I went for it.

So far impressed, a lot of nice things and a lot of stuff (databases, email addresses, easy to get logs, easy to upgrade WordPress, etc). There does seem to be an awful lot of login requests when using their backend, each service and area seems to have it’s own username and password system, but this isn’t a huge problem. Plus I seemed to join them at a bad time, and there has been some downtime of late. has been down for the last couple of weeks, and this is just down to me. I tried to transfer the actually domain across, though the move was rejected (I may not have set the new DNS servers up right before putting in the transfer request). I then basically did nothing for a few weeks. I finally got around to it yesterday, the transfer this time went fine and now everything seems good again.

So, hopefully this is still on some of your RSS feeds and you haven’t all deserted me! Welcome back.

Analog web statistics

I use the wonderful Analog for web stats. Unfortunately it hasn’t been updated for a while (2004 in fact).

However I noticed that some kind soles on the mailing list have updated analog so that it includes Vista in the OS report.

I’ve put a copy here, just download the zip (Windows), extract and replace the original analog.exe.
See this mailing list message (and the thread that follows) for more information and credit to those who produced this new version.

Analog exe (as a zip) support for vista

Look! I can post links straight from B3ta

The Credit Card Prank

Never understood those people who thought a signature is better than a pin.

Couple of useful tools for web authors

Found both of these on Phil Bradley’s blog.

first is which can produce a button like this, which will link to a page with every kind of “add to [rss reader]” button you will ever need.

Add to any service

Secondly, allows you to “add your favourite social networking badges to your blog or web site”. I can’t seem to paste the code in to a post, wordpress seems to remove the {script} code (probably with good reason).
Will have to add these to my template at some point

UK has second highest level of debt in the world