Week 8 – Important Stat

To be honest, most of my research and work has been done in a more qualitative than quantitative fashion. The few statistics that I have seen are not really that detailed or nuanced. However, I was one of the many who followed Nate Silver’s blog during the American presidential election, and was blown away with the accuracy and specificity of his claims. Elections are tricky, as anyone who followed the recent Alberta provincial elections would know, and so the pure confidence of Nate Silver’s claims, with the perfect results he got, were astounding. Not only were the predictions correct, but they were done in a very attractive format, that encouraged interaction and probing with layers and layers of information.

I think it may have partly been from the hype of Nate Silver’s visualizations that we in the Association of Information Systems student chapter invited Breakeven Inc to do a workshop on visualizations. They showed us how they design their custom dashboards for clients, and talked about how many clients are looking for a visual way to demonstrate metrics to their employees. I couldn’t find anything on my laptop but I’ll dig around for more (that being said, we have wrangled some internships with them if anyone is interested in user experience and design).

However, visualizations do not enjoy full support. XKCD had a few negative words about popular methods of visualizing in this comic ( http://www.xkcd.com/1273/ ) although it also shows some great examples of visualizations as well (http://xkcd.com/980/huge/#x=-6400&y=-6272&z=3)

What do you all think? Are visualizations gimmicky or do they sometimes very useful at explaining something that numbers don’t quite put across?

 

Advertisements

2 Comments

Filed under Uncategorized

2 responses to “Week 8 – Important Stat

  1. I don’t think visualizations are necessarily gimmicky… but I do believe that in order for visualizations to be effective they need to be explained. I think a good mix of explanation, backed up by numbers (statistics), then given in a visualization to the reader, is the most effective (at least for me). It’s almost as though you introduce a topic with theory and observation, then you back up your observation with evidence (statistics), and then you outline it with a visualization for the reader, which helps bring the observation/theory and the statistics together.

    If you were reading an article and the author quickly ran across a topic, but through in a visualization, I definitely do not think it would be effective, or as effective, as a visualization is meant to be. A visualization should be something to support a claim (in my opinion).

    Victoria Grant

  2. Liz

    I agree. Victoria. I think visualizations can help users make sense of the data. However, it could be challenging to make sure that the readers are interpreting the visualizations the way intended without technical knowledge and background. I believe that integrating both statistics and visualizations can maximize user benefits. Statistics can illustrate relationships between variables, while visualizations can highlight the important patterns and trends.

    Thank you.

    Liz

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s