University of Michigan School of Information Experimental Design with Web Analytics project
Over the next few weeks several students in SI422 will be working with the Drupal project to create and design a web analytics experiment.
The students are aware of the redesign effort and the research that has been done. The will propose a web analytics experiment that includes:
* Hypotheses: These are statements of predicted outcomes that you will be testing, including specific, operationalized variables that would allow you to know whether your test had been confirmed or disproved.
* Method: Describe how you would set up the experiment and how long you would run the experiment.
* Data and Analysis: Although you won't be running this experiment, in a real write-up, you would have a results section. Instead, describe what data you would collect and the analyses you would run to test your hypotheses.
* Discussion: A summary of expected findings that would tell you if your experiment had been successful.
If the experiment seems worthy, then members of the Drupal community can step forward to provide assistance in implementing some of these experiment on Drupal.org. The community should expect to see each team in this class submit an experiment in this group. The students will be given access to the Drupal.org web analytics reports so they can design their experiments.
After some discussion in this group, the students can choose to file a webmaster issue to have the experiment actually implemented, although it is not required for their class. The project starts today and will run over the next several weeks.
This is the second collaboration with the University of Michigan, the first was the pivots recommendation system.
I'm one of the said
I'm one of the said students. Drupal is one of the best ideas I've ever seen. I'm fully aware that this community had great efforts to improve the overall design of the project and I'll pay full respect to it. Over the next few weeks I'm going to get more familiar with the website, and try my best to learn from the community as well as experiment with our own ideas. My appreciations to the community and Kieran Lal for this great opportunity.
Maming: Welcome to the
Maming: Welcome to the community! I hope you can see some of the real challanges we face with web analytics and the value we can get out of it. As probally already mentioned we have a few analytics running, that give us insights mostly for when things are going wrong - but we have yet to get data out of that can fastly improve the way we approach the sites design.
I would suggest to take a low key, but important landing page and see how its current use is + could be improved - one example of this is the documentation landing page. But, either way - take something not to intense (like the homepage, but a page with clear goals) as it might be to fluctuating data, during the impending switch to the new design.
It will be a good step. If
It will be a good step. If the experiment seems worthy, then members of the Drupal community can step forward to provide assistance in implementing some of these experiment on Drupal.org - indeed.
Web design
interesting
What a different and interesting approach to take. Any news on how this went it of interest to me for my internet marketing consultant business. Many thanks.