Tuesday, 31 January 2012

Miners or analysts: What's in a Name?

For an educator using a web-based course delivery environment, it could be beneficial to track the activities happening in the course web site and extract patterns and behaviour prompting needs to change, improve or adapt the course... For a learner using a web-based course delivery environment, it could be beneficial to receive hints from the system on what subsequent activity to perform based on similar behaviour by other 'successful' learners.
This is from Zaïane (2001) – with 160 citations, a leading paper in the field of educational data mining (EDM), and absolutely on topic for learning analytics. So how many papers for LAK12 have cited it? None that I have seen (and, being on the programme committee, I've seen most of them).

I’m worried that we’re all so concerned with the new-minted term ‘learning analytics’ that we’re forgetting all the work in exactly the same field under the less fashionable heading ‘educational data mining’.
If pushed, we make reference to educational data mining as one of the roots of learning analytics – a root that can be handily referenced and then passed over with a perfunctory reference to Baker and Yacef’s 2009 overview of the state of EDM. We may even add another reference, or ask Ryan Baker to speak at the beginning of a MOOC. In either case, we frame EDM as the start and as the past, and go on to produce enthusiastic descriptions of a present and future that begin around the time of Educause’s discovery of learning analytics.

Are the theoretical differences so great, are the numbers working in the field so large, that the divisions between EDM and learning analytics are helpful? Or is it that we want to be academic analysts, not plebeian miners? Are we avoiding the established EDM journal, conference and society because they have little direct relevance to us, or is it that we’re too busy describing new projects to potential analytics funders to spend time investigating the years of work that has already been done?

Wednesday, 25 May 2011

Excel Ideas

I've just attended a course on Microsoft Excel for PC, 2003 version. Lots of useful ideas, but how many will I be able to implement without a lot of thought on Excel for Mac 2011?
1. Create your own toolbar
Handy if you have put together a set of macros that you want to use together.
View>Toolbar>Customise toolbars>New
2. Macros
I'm fairly sure this wasn't in the last version of Excel for Mac, but it's back now.
Tools>Macro>Record New Macro
Macros can be recorded and used with just one workbook, or set up globally. When recording them, it's useful to bear in mind the difference between absolute and relative cell references. It's the $ sign that makes a cell reference absolute - eg $G$4. ON the PC you can make a cell reference absolute by clicking F4. On my Mac that brings up the dashboard and, irritatingly, starts one of my apps making an interminable gurgling noise.
3. Macro Icons
Yes, you can assign your macros to various key combos, but who remembers those a few days later?
View>Toolbar>Customise toolbar>Command>Macros>Custom button
The default is a smiley face, but you can right-clidck to change the picture on a PC. I haven't yet worked out which combination of keys does this on a Mac. If you don't want the icon on your toolbar any more, you can drag it away and drop it for ever if you hold down ALT at the same time.
4. Ideas for macros
One that's fairly straightforward and could be saved for global use. Format page as Landscape, fit to page, add standard header and footer. That's all under File>Page Setup
5. Subtotals
Sort data on the relevant column, then go to Data>Subtotals. Not only do you get handy subtotals for each heading, but you get a set of numbers, plus and minus signs on the left, that allow you to focus in on certain categories.
6. Protecting sheets
I do have a tendency to mess up my data analysis by changing sections of my tables that need to stay the same. This one is rather counter-intuitve. Highlight the cells you want to remain open. Format>Cells>Protection>Unlock. Then head to Tools>Protection>Protect Sheet and limit users to selecting unlocked cells. No need to set a password. To make it all editable once again, just reverse what you've done.
7. Wrapping text
Format>Cells>Alignment>Wrap text. Who would have guessed that command would be hiding under the Alignment heading? You're supposed to be able to put soft returns into cell text as well, but I can't make that work on my Mac.
8. Splitting words into different columns
I haven't tried this out yet, but it involves using the SEARCH function to look for spaces, recording in a new column how many characters before (or after) the space and then using that information with a LEFT or RIGHT function to pick up the second word and put it in another column.
9. Show all the functions in your Excel spreadsheet
On a PC you do this by pressing CTRL together with a key with three little symbols on it at the top left of the PC keyboard. I haven't yet spotted how I could do this on a Mac.
10. Counting
I would have expected COUNT to count all cells, but it only counts all data items. COUNTBLANK counts the empty data cells and it's COUNTA that counts all the cells.

Monday, 10 January 2011

Learning Analytics and Knowledge course

Jumped right into this course with a look at Hunch.

When I arrived I realised I'd visited a couple of times before - it's a link I follow from blogs and Twitter on occasion. Answer a series of picture questions and Hunch learns enough about you to make some recommendations. Keep engaging with the questions and recommendations and you should, in theory, have some useful recommendations that could help you choose what to read, where to go, what to buy. In theory. As I say, I've been there before on occasion and I have neither remembered the name nor bookmarked it. But why not?

First of all, I didn't like the options. Multiple-choice always winds me up. I hate questioners who won't give me the freedom to choose my own words and interpretations. Often all their answers are wrong, either unequivocally wrong, or wrong from my perspective.

Q. Do you live in the suburbs, a major city or a rural area? I live in a good-sized town. Skip the question.
Q. Do you tend to support liberal or conservative politicians? I live in the UK. Ask me a sensible question. Skip.
Q. Which of these sorts of fries do you prefer? Fish and chips. Oh, but the options are American fast food. Skip.
Q. Is Barack Obama a Muslim? (!) Should the site be posing as a complete idiot? Skip and End.

So the style of it all wound me up. Assuming I live in the USA wound me up. And the summary of people's responses - which is supplied each time you answer - made me feel an outsider. Only thirty-nine percent of users had identified as female (or as aliens wearing dresses, it wasn't clear), 15% had identified as Europeans. Only 17% read more than three books a month. (And did 3% really say they had PhDs? Either Hunch attracts a crowd of intellectuals, or a lot of liars.) I didn't  feel at home.

And the recommendations. Well, it spotted that I like Apple products - but after I'd answered the 'Mac or PC' question that wasn't too tricky, was it? And I said I was European, but it  recommended me five New York city museums, and five US credit cards. I was recommended a dreary series of magazines, a series of books that you'd read only if really bored, and some blogs where I didn't get beyond the first post.

Did I bookmark it? No.
Is this what I would want from a set of learning analytics? Absolutely not.

Thursday, 25 November 2010

Assessing reliability

ReCal (”Reliability Calculator”) is an online utility that computes intercoder/interrater reliability coefficients for nominal ordinal, interval or ratio-level data. I haven't tried it out yet, but will need to test reliability on my analysis, so I'm bookmarking it here.


Wednesday, 10 November 2010

Counting words in an Excel cell


=IF(LEN(TRIM(A1))=0,0,LEN(TRIM(A1))-LEN(SUBSTITUTE(A1," ",""))+1)

I knew there must be a way of doing this - but hadn't tracked down the exact formula until now. This one piece of information made the whole Excel course worthwhile. Now that I have the link it looks like such an easy thing to Google that I can't think why I didn't manage to track it down by myself. Suppose I was just convinced that there would be a WORDCOUNT function buried somewhere in Excel.

Friday, 17 September 2010

Analysis of Retweets on Twitter

Noting a useful resource.
Blog post by Brian Solis on 'The Science of Retweets on Twitter', examining a report by Dan Zarrella.
This is an overall analysis of Twitter, which provides some possible benchmarks for comparison when focusing on a subset of Twitter.
He records the most retweeted words and the least retweeted words - noting that retweets require a slightly higher reading age than Tweets.
Punctuation is more common in Tweets than in retweets - except for semi-colons, which show up as the 'only unretweetable punctuation mark'.
And retweets vary by day of the week and by time of day - with Friday evening being a top time for retweeting.

Wednesday, 15 September 2010

Analysing Elluminate dialogue

Bit of a summer break between blog posts, during which I've turned my attention from the Twitter stream around the OU online conference in June to the Elluminate chat around the presentations.
I'm still looking for indicators of exploratory talk that can be used to identify where learning is likely to be taking place.
People at the conference used chat a lot. For example, to take the afternoon session of 22 June, there were 858 separate contributions. I can divide these roughly into four groupings:
  • chat about content (526 contributions)
  • chat about tools (101 contributions about eg the conference format, Elluminate and Twitter)
  • social chat (215 contributions including hi, bye and thanks)
  • and blank contributions (16). 
That means about 61% of the chat was focused on the content of the presentation. This seems pretty high - I've got a presentation from an old ALT conference running on my computer at the moment, and nobody has typed anything in the chat box during the first half hour.
I've picked out 94 words and phrases that could be indicators of exploratory dialogue. These include 'have you looked at', 'have you read', 'do you mean', 'my understanding' and 'next step'.
Here's an example of a chat contribution that my indicators flag as a possible example of exploratory dialogue.
An initial run-through suggests that this list is good for picking out the areas where learning dialogue seems to be taking place. As you'd expect, the exploratory dialogue is mainly in the sections of the chat related to content, and there are not only areas where these indicators are more common but also people who use these words and phrases more than others.