Today we were tasked with looking at data from the Survey of Income and Program Participation (SIPP). My area of focus was Assets. My first thought was that it would be interesting to see the range of value for different assets such as cars/ properties/ bonds, among different demographics. Today was a reminder that you cannot pick what data is available, and the dream asset data that I had in mind did in fact not exist within SIPP datasets. Instead the data that I found looked at the Mean Value of Assets for Households, by Type of Asset Owned and Selected Characteristics for 2016. Assets were something that I thought I had a fundamental understanding of from finance classes, when I was reviewing the data however, I was slightly confused by what some of the terms referred to, and had to spend quite a bit of time researching.
The data came in an excel format, the main task for data preparation was working with merged cells and subtitles – which I did in Alteryx.
Having never Tableau-ed survey data before I had a look online for some advice on how to deal with it. Doing this highlighted the importance of showing the uncertainty of results – the data is from a survey of a sample of the population not the whole population. As such, it does not represent the entire population. To show the uncertainty it is important to use a margin of error and confidence intervals to allow the viewer to see an approximate range.
My aim was to replicate calculations that calculated asset correlation (to give a number between +1 and -1), after spending far too long figuring out how to do the calculations for this, it did not work. Due to the time limit, I could not spend an unlimited amount of time troubleshooting the calculations, and instead re-planned my dashboard. As the majority of my dashboards are explanatory, I wanted to make an exploratory one and still incorporate how different asset types correlate to one another. To do this, I made a table with set actions, so that users can drill down into demographic breakdowns within any category and see their average asset value (USD). After this, I was hoping to add slightly more interactivity, so the user can choose 2 variables (asset types) to compare. By selecting one of the categories from the table on the right, it populates the view with demographics from that category.
To interact with the visualisation, you can click the image below.