Exploring Tuva's Embed Feature

At Tuva, we've heard from users they want to use Tuva's tool for analyzing their own data. We also noticed that occasionally, people analyze a dataset that they want to share in an interactive format. So, we explored this feature, and I'm creating this blog post as an example of how Tuva's tool appears within a blog/article format.

I'm going to explore this feature in the context of data about dogs.

Let's say we would really like to know the average height of different dog breeds. Fortunately, I already found the data and uploaded it to my personal account. Then, via Tuva's tool, I can embed the dataset here and show the results.

From this graph, we can hover over the red horizontal lines to read the different mean values. We can conclude working dogs are taller on average.

The key to using any data analysis tool is to ask a guiding question. Then, once we have the right data, we can begin kneading the graph until we arrive at a visualization that answers our question.

Let's try asking a question that's a little more interesting: Based on this data, what seems to be the greatest influence on a dog's maximum life span?

After trying several different graphs to see if it would answer my question, I settled on two different graphs that seemed to answer it best:

The first one is maximum weight vs maximum life span.

The second one is maximum height vs maximum life span.

We will notice a downward trend in both of these graphs. At the same time, one of the variables, maximum height and maximum weight, seems to be more strongly correlated with maximum life span. Maximum weight is more strongly correlated to maximum life span than maximum height. So, from this data we can conclude that the maximum weight of a dog seems to have a greater influence on the maximum life span of a dog than maximum height.

What else do you think we can ask? What might be interesting to learn about dogs that we think this data can reveal? Play around with the graphs above and see what else you can discover.

No comments:

Post a Comment