Unit 3 - Beautiful Evidence in Full Color - Granata


What affordances and/or limitations did you experience with each of the tools?

In order to begin this project, I thought it would be useful to explore the two tools to determine what type of data they would need to create the visualization. I noticed that both required a strict convention of syntax, in that strings and numbers had to be coded properly before uploading to allow users to manipulate the data into useful visualization. I explored the various available chart types, which were standard. I tried to create a map with fancy colored dots, but found that there is no easy way to upload a JSON formatted map of school districts in Westchester. New York Counties were included as an option, but the steps to download just the school district boundaries proved to be beyond my current capabilities. Due to the limiting factor of my abilities, I changed the focus of my exploration.

I began the journey of storying the NYS 2018 Assessment data by first evaluating the entire set of data to determine what information was included in the report. I found the following information to be relevant to my inquiry:

  • District Name

  • Subject

  • Total Tested

  • Level 1 Count

  • Level 2 Count

  • Level 3 Count

  • Level 4 Count

  • Mean Scale Score

Once I decided to explore the results of the Grade 7 Math test, I began to filter and sort the data according to the information I was looking for. First, I hid the unneeded columns to remove them from view. Next, I selected filter view to be able to select just the subject I was looking to analyze, which in this case was Grade 7 Math. Next, I filtered the schools to include only those from Westchester County. I then highlighted the districts I wanted to analyze more deeply, in this case New Rochelle and Mount Vernon. A screenshot of this filtering and color coding is below.


You may notice two extra columns. I had to apply a formula to change the “Subject” data to only a number, and did so by using the [=Value(“value”)] formula to convert the test from “Grade 7 Math” to “7”. This is required to be able to use the DataWrapper app to create a visualization.


Next, in order to make sense of the data I had filtered out, I did some research with the United States Census Bureau to obtain information on the demographics and socioeconomic information for both of these small cities.



Mount Vernon

New Rochelle




Median Age



Median Income






Median Property Value



Student Teacher Ratio



# of Students



African American












Free/Discounted Lunch

751 (75.8%)

1173 (48%)



Finally, I used the spreadsheet data I had filtered to create a few visualizations.



Although the two tools we were instructed to use were both full of options and creative refinements to color code, sort and stack the data, I found both to be similar to the use of charting in Microsoft Excel. The same types of charts I can create in DataWrapper and RawData are available through Microsoft Excel 2010 and Excel 2016. The advantage of pre-loaded maps in the DataWrapper app sets it apart. I also feel that these tools are better suited to visualize the large data set available from the NYSED website for ALL of the districts in NYS, but when used for a more intimate view of results and demographics they pale in comparison to a simple Excel chart.


What kinds of stories did they allow or not allow you to tell with these data?


These tools were able to help me visualize the number of students who showed proficiency on the Grade 7 Math test from New Rochelle and Mount Vernon. They showed that of the total tested students, a greater percentage fell in the Levels 1 & 2 proficiency set than the 3 and 4. They also showed me that the mean scale score of both districts were almost identical.


The visualization of the demographics of the city was useful to clearly emphasize the disparity in poverty rates between Mount Vernon and New Rochelle and the number of students who are eligible for Free and Reduced Lunch. This number is frequently cited by administrators as a predictor of academic performance, and as such should be included in any broad examination of student achievement.


From an Education Technology standpoint, I feel that the use of these applications to visualize data of student achievement and performance would be overkill. The available visualizations from Google Apps for Education, such as Pie Charts, Scatter Point Graphs, Bar Graphs, etc. are more than sufficient to visualize the data needed to inform instruction and easier to utilize. In addition, when instructing students on the use of charts to visualize data in spreadsheets, it is much easier to manipulate the available data in Google Sheets to see the instant change in the visualization than it is in either RawData and DataWrapper. If I were to explore the larger datasets required to analyze overall student performance for the county or state, however, these graphs seem more capable of creating relevant charts with larger amounts of data. The data, however, must be filtered and coded appropriately before uploading to create meaningful results.

    • Gerald Ardito
      Gerald Ardito


      Your analysis really surprised me. I was stunned to discover that New Rochelle had a higher student to teacher ratio than did Mt. Vernon. It would be great to further investigate these gaps and similiarities.

      I have come to agree with you about the tools. The charting tools in Google Sheets and Excel seem more than sufficient to create these visualizations. The real skill, I have come to believe, is deciding which data to investigate.

    Computer Science for Teachers Spring 2019

    Computer Science for Teachers Spring 2019

    Here is the online home for CS for Teachers at Pace University for Spring 2019.

    Latest comments