Learning Objects and Evaluation

Some searching machines in a learning object repository need more than one letter. To get a result you should us short, often used words like “and”, “the” or in German “und”.

But I am afraid that the result is not as good as using “e”. But it is enough to get an impression. 

After citing the data from 2004, I just found the actual number of objects in LON-CAPA:

148.483 resources (Fall 2005), including more than 50.000 problems!

More details: http://www.lon-capa.org/sharedpool.html

My last weblog entry shows my typical working weakness – I just publish results. Now I have to add my ideas and experiences ;-(

One indicator (others will follow) for the success of a learning object repository (LOR) is the number of content. The number is important because if somebody wants to use learning objects (LO) for producing a course s/he needs many LO. And the more content a LOR owns the higher is the probability of finding useful LO.

(I think specialised LOR like the Health Education Assets Library (HEAL) can increase the probability of finding the needed material.)

For finding existing LOR I started with (i) a search with Google. (ii) I used collections of links. (iii) I analysed publications containing several names of LOR.

The number of LO are sometimes shown in an exemplary way. Connexions offers on its homepage an approximate number and a precise number in the content area. Secretive is the “European Knowledge Pool” (ARIADNE). ARIADNE is a LOR based on a yearly fee. But on their webpage I did not found any number, I had to find a research paper (http://ariadne.cs.kuleuven.ac.be/silo2004/silo.jsp Search: ICETA).

Some repositories offer a search engine, but no number. In this case you can use the character e to get an approximate number. If the number of search results is not shown, sometimes is it useful to count the results.

The publications mentioned above about LOR sometimes offer some number for different LOR, which can be used as approximate number.

I wanted first to plan my weblog but I just have to write about this thoughts

In our project CampusContent we make research about learning objects and think about installing a playground for special features. I am evaluating learning object repositories (LOR) around the world to find out their special features and highlights.

And I would like to know how much content the repositories own. Here are some ways to count:

• Search on the web site for an actual number
• Search on the web site for an approximate number
• Sum up the numbers of the number shown in the search via browsing for each theme
• Search with a blank search field
• Put “e” into the search field
• Count each learning object (only possible in small LOR)
• Search in publications about this LOR
• Search in publications comparing different LOR
• Send an e-mail to the site manager of the LOR

It is sometimes difficult to find out a number. Here are some results from November 2005:

Content:

LON-CAPA (USA): > 60.000
Health Education Assets Library (HEAL): 20.786
Knowledge Agora (Canada): > 20.000
SMETE: > 12.000
MERLOT (USA): 12.368
NEEDS: 7.286
ARIADNE (EU): > 4.500
Connexions (USA): 2588
Wisc-Online (USA): 1.903
iLumina: > 1.500
The Learning Matrix: 1.355
Maricopa Learning eXchange (MLX): 1.232

CourseWare:

MIT (USA): >1000
China Quality OpenCourseWare (CNQOCW) (China): 451
Connexions (USA): 103
FETP OpenCourseWare (Vietnam): 66
OSAKA University Open Courseware (Japan): 28
Kyoto University (Japan): 22

Meta Search Engines (some kind of federated search):

OAIster (USA): 5.797.078
Education Network Australia (EdNA, Australia): 1.600.000
National Science Digital Library (NSDL, USA): > 100.000 (?)

Here you find more data and the links to the repositories for learning objects and courses.