Lab 2



  1. scaleTime is used for domains expressed as an array of dates. We used scaleTime for our x axis in this lab because we were working with time series data. scaleLinear is used for continuous domains. We used scaleLinear for our y axis values because they were a range of numbers. scaleOrdinal is used for discrete domains, such as names or categories. We could use this with the webtree data when looking at classes or departments. We used scaleBand, which is similar, but it is better when working with bar charts.

  1. From this visualization, it is clear that California experienced the most confirmed cases of the West Nile Virus in 2016 and 2015. 2010 and 2011 had the least amount of cases. The majority of the counties represented in this graph follow a similar path. For most of the years, the county with the ‘peach’ color line has the most confirmed cases. Also, all of the peaks in the graph fall towards the end of each year around the fall season.
  2. The most obvious deficiency is not differentiating what each line represents. They are divided by color, but there is nothing to tell you what each color represents. We don’t even know that they represent different counties, let alone which county. Furthermore, there are many lines that follow a similar path, so it is hard to differentiate the path of one line from another. The graph is very crowded. To fix this problem, I would add a legend that specifies that each color line represents a different county, and then I would list each county’s color. I would also add a mouse over or click function to look at them individually. That way, we could get a better sense of the path of each individual county.

    Another deficiency that would be helpful to clarify is denoting what percentage of the county population these cases represent. We can only compare cases of a certain year to the others, so 35 seems really high and 5 seems low. But looking at the entire population is 35 actually a lot, or is it still a low number? This change would help clarify the severity of each year – if 2015 and 2016 were actually as bad as they seem in this visualization. A simple design improvement would be breaking the graph into divisions by horizontal lines. We would need additional information to determine where these lines would lie, but then you could divide the graph into a scale from less severe to more severe. However, there are a lot of lines already on the graph, so this addition might make the visualization harder to read. Therefore, I would make the y-axis a color scale of red from light red to dark red with a label noting that the scale goes from less severe to more severe.

Leave a Reply

Your email address will not be published. Required fields are marked *