Skip to content
22 August 2013 / Erik Duval

Waarom ik niet teken…

(In Dutch, as it relates to a public discussion about ‘publish or perish’ that has hit the press here recently.)

Neen, ik ga de petitie tegen de publicatiedruk niet tekenen…

Ik heb veel sympathie voor de druk die jonge academici ervaren. Hard werken. Geen zekerheid dat je je doctoraat behaalt. Of dat je een post-doc positie kan vinden. Of dat je daarna een loopbaan kan uitbouwen. Of waar je dat dan zou kunnen.

Ik ben natuurlijk erg oud nu, en heb geen benoeming of bevordering meer nodig. Maar om een of andere reden heb ik daar ook vroeger nooit van wakker gelegen. Niet omdat ik zeker was dat het wel ging lukken. Maar omdat ik het werk leuk vond en er van uit ging dat gaandeweg wel zou duidelijk worden wat kon. Of niet. Misschien sta ik wat lichter in het leven dan veel van mijn jongere collega’s…

Maar ik heb dus wel sympathie voor de druk die velen wel ervaren.

Dat lijkt me echter geen reden om te tekenen: die druk ontstaat niet omdat ik verplicht word om een aantal publicaties te halen. Die druk heeft alles te maken met het feit dat we allemaal met elkaar concurreren. Ik ervaar dat meer als een soort sport: je wil sneller, hoger, verder. Of misschien meer nog als een ploegsport: je wil samen met anderen sneller, hoger, verder. En neen, het is nooit snel genoeg, hoog genoeg, of ver genoeg. Alles kan beter. Het is vooral competitie met jezelf. Ik geniet van dat spel. Zoals, denk ik, topsporters genieten van hun sport.

(En neen, niet iedereen moet aan topsport te doen. Dat is waarschijnlijk niet eens gezond. Maar we begrijpen wel allemaal dat alleen de besten meedoen aan de Olympische Spelen. En dat daar ook geluk mee gemoeid is. En dat zelfs diegenen die super snel lopen niet eens mogen deelnemen als er een paar zijn die nog sneller lopen.)

Het gebruik van die gegevens over publicaties wordt in de discussie ook vaak verkeerd voorgesteld. Ik zit geregeld in benoemingscommissies om professoren te benoemen. Overigens is de concurrentie vaak veel groter dan in de pers wordt gesuggereerd: 40 tot 70 kandidaten voor één positie is niet ongebruikelijk. Maar er wordt niet louter geteld hoeveel publicaties of citaties een kandidaat heeft. Er wordt naar veel meer gekeken, inclusief het soort van onderzoek dat iemand verricht heeft, waar, met wie, wat hij of zij wil doen als ze benoemd wordt, of kinderen de uitbouw van de loopbaan hebben beïnvloed, enz. Kandidaten kunnen omstandig uitleggen waarom ze zichzelf geschikt achten voor die job. Enz. Het gaat echt wel om meer dan louter publicaties tellen – anders hoeven die commissies ook geen lange beraadslagingen meer te doen, natuurlijk…

De petitie wil meer aandacht voor kwaliteit: daar kan niemand tegen zijn natuurlijk. Maar dan moeten we ook afspreken hoe we die kwaliteit willen meten. De basisidee van het huidige systeem is dat we kwaliteit valideren door het gepresteerde werk door onze collega’s te laten beoordelen. Dat is de idee van ‘peer review’. En door op te volgen hoe vaak onze collega’s verwijzen naar ons werk wanneer ze hun eigen resultaten publiceren. Daarom dat we naast publicaties ook citaties bekijken. Mij lijkt dat redelijk. En sterk te verkiezen boven het vroegere systeem waarin alles binnenskamers werd bedisseld. Het zou gerust veel transparanter mogen – de idee van open science, die ik zeer genegen ben. Maar ik zie niet direct een reëel alternatief?

Overigens ben ik het er wel mee eens dat de administratieve druk te groot is. Dat is blijkbaar niet alleen in de academische wereld zo…

Zo, nu ga ik terug aan mijn publicaties werken ;-)

4 July 2013 / Erik Duval

Learning Analytics and Quantified Self at LASI13

LASI13 is in full swing: it’s a nice group here in Stanford and there are 10 local events throughout the world.

Image

(Networked conferences should really be explored further. Face to face meetings are nice, but sometimes difficult to fit in and I wonder whether we wouldn’t be able to make on-line events work better than face-to-face ones with all the technology we have. Hey, I sent a twitter message to the person sitting next to me in a session yesterday, despite the fact that we had both travelled across an ocean to be in Stanford…)

Above are the slides I used for my session on ‘learning analytics and quantified self’. I would write a bit more about what happened at the workshop, but you can get all the details from Doug’s wonderful #lasi13 lifeblog

Basically, Abelardo and I asked participants to identify data that would be useful to track in order to have a good view on how the learning process is proceeding. We then asked people to design a scenario that demonstrates the added value of a quantified self approach. We ended the afternoon with a discussion on the research challenges in this area. Again, Doug provides many more details… A big thank you to all who participated – sure got me thinking about new things to try.

And maybe you have some suggestions for data to track in a learning context? Or maybe you know of some other work in this area? If so, I’d love to hear more…!

7 June 2013 / Erik Duval

Something cannot be bad when it’s open – or not for a very long time…

Openness is a recurring theme in much of what I do.

But let me first make clear that openness is not the only thing that matters: I do not believe that open bad courses are better than closed good courses, or that open source software or content is always better than closed source content or software. In fact, I worry sometimes that open advocates use the openness of their content or software as an excuse for not making it better: ‘this tool may be difficult to use, but you should use it nevertheless, rather than the tool you are currently using, and which works better for you, because it is open’ – as if there were some sort of moral imperative to suffer for using Open Stuff.

Yet, when something is open, it becomes more difficult to have it be bad for a long time. If I write a new wikipedia article, and there is sufficient interest in the topic, then, even if the original article is not very good, it will get better over time. If I teach in an open way, and my course sucks, then this will be apparent to all who care about my course, and I will get questions about what I do, which provides me with an opportunity to improve it. Or word will get out and students will take another course. If I write nonsense about someone or something on an open social network, then someone will react. Etc. Etc.

Compare this with closed systems, where low quality and abuse can continue for a long time. In the old days (i.e. when I was a student), a professor could do whatever he wanted in his courses. (Yes, ‘he’ and ‘his’ – academia was a rather masculine environment…) Some did Great Stuff. Some not so. And some did awful. But there was such absolute trust in professors, that those in the latter category could continue to “teach” until they retired.

Similarly, if I spread lies to you about a common friend, late at night in a pub, then you may never check whatever I said with anyone else, and my lies may influence your impression about our common friend for a very long time… Another obvious example is the multitude of scandals that have come to light related to the catholic church: as long as authority enabled the organisation to impose a closed culture, much of this abuse could take place without challenge. Another, even more uncomfortable example: most child abuse takes place at home, because parents can create a closed environment where that kind of behaviour does not ‘leak out’ and thus remains unchallenged.

This is why I favour an open society: an open arrangement makes it much more difficult to conceal abuse. Thus, abuse will be challenged – a first step to stopping it. And that is one of the reasons why I favour open learning…

How about you?

2 June 2013 / Erik Duval

Open Learning Analytics at TedX@UHOwest – and a confession

Yesterday, at tedxuhowest, I presented our work on open learning and learning analytics.

I started my talk with a confession: I actually don’t like TED talks all that much… Many can be summarised as “hey, if we all do Good Stuff, then the world will become a better place”. That is probably true, but, paraphrasing one of my PhD students, “that’s just hippie talk”…

What I do like about TED talks is that they are short ;-). And I liked the idea of trying to summarise the essence of what we do in 18 minutes. The video will be posted soon, I think. My slides are below.

The other thing I like about TED events is that you meet Interesting People: I was quite impressed by what Tom Christiaens does with quindo, where students from ‘difficult’ backgrounds make radio: very different setting from mine, but very similar inspiration, views and goals… Roel Berger presented the jini quantified self framework – wish their app would finally be public! And Vero Vanden Abbeele showed a nice example of the tension between apps that try to help us improve our lives and … just living our lives.

And I liked the music too: Goedele Taveirne and Ynoji – quite different, but then I am a man of many tastes…

25 April 2013 / Erik Duval

Activating students

We have a bit of a discussion going on how to ‘activate’ students, i.e. how to make them not passively sit and listen in sessions but make them participate in an active way…

Thought I’d share what I mailed earlier today and ask for your feedback…

I think it’s important to realise that activating the students is often a challenge, for all of us. Sometimes, my students can’t get enough and don’t want to stop working. Sometimes, it feels like I’m talking to bags filled with sand. The important thing is to keep trying: falling back in old, safe habits is not a solution – it just hides the problems.

A talk with the students on why they don’t engage can help, but can also turn into a pseudo-psycho-analysis that gets you nowhere.

We train students from age 6 to sit still, be quiet and listen. Those who do well at that in schools end up at the university. And then we wonder why they are not active ;-)

My main point: don’t think this is easy or always works for anyone. Keep trying. That’s much better than not even trying…

What do you do to make your students participate actively? Or, if you’re a student, what makes you engage?

 

21 April 2013 / Erik Duval

Come join us!

Come join us in Leuven: we have an opening for a full-time tenured academic position in Computer Science for Digital Humanities at the Department of Computer Science!

Deadline for applications: 30 June 2013

From the official text:

We are looking for a candidate with expertise in applications of computer science in digital humanities. A successful candidate will already have obtained excellent research results in the area of applying computer science techniques in research applications in the digital humanities. For the applied computer science techniques, examples could be data mining, or archiving or disclosure of information from large e-archives, or human-computer interaction, or visualisation techniques, or e-learning. Concerning the applications domains, examples could be the Arts, or the Social or Instructional Sciences. Research experience with the application of multiple techniques from computer Science and/or with multiple application domains in the Humanities or Behavioral Sciences are considered are a strong added value of the application.

The official announcement gives more details…

Feel free to contact me for additional details!

17 March 2013 / Erik Duval

Learning Analytics: what to measure?

Last Tuesday, I had the pleasure and privilege to do an afternoon session on learning analytics with a group of teachers at the Alberdingk Thijm College, Hilversum.

One of the reasons why I liked the session is because it was more a working session than a presentation. After a brief re-cap, participants brainstormed in small groups on things one might want to measure in a learning analytics context. Slide 18 below summarises the results.

In translation:

  • environment, light, carbon dioxide, oxygen,temperature, colour, weather, air quality, space, familiarity, background noise
  • experience, involvement, stress, boredom, well being, fear, emotions, motivation, feeling of being competent, experiencing success, compliments from teachers and other students, security, bullying, acceptance
  • duration of instruction, attention, learning effects, learning time, measuring every 10 mins whether you get through, best moments for instruction, before or after break, visual or auditory, result
  • physical condition, sleep, eating, brain activity, heart rhytm, movement, teacher and student, breathing, voice, conductance, palm humidity
  • web sites (related to learning or not), search activity, how long, relationship with domain being learned, facebook, alertness, thinking steps, brain activity
  • social, pairing good with less good students or friends, gender, age, talking, keyword extraction, English or Dutch, space, quantity, interaction, between teacher and student,
  • teacher behavior, walking around, interaction, patience, voice, compliments, humour, choice of words, volume, duration, clothes

OK, so this is isn’t perfectly structure, but, obviously, a group of 20 teachers can identify more relevant characteristics in 20 minutes than we will know how to measure, analyse and visualise in the next 20 months ;-)

Maybe you have some additional suggestions for things we could also measure?

Follow

Get every new post delivered to your Inbox.

Join 3,756 other followers