Tableau Tips & Other Useful Info from Dr. Lance
Making Data Make Sense
One of the goals of the Research and Evaluation department is to be able to provide summarized and visualized assessment data to all educators in a helpful and timely fashion. To that end, it occurred to me that there are many factors influencing exactly how to present such data. Here is just a brief list of such factors:
1. Which student to include (i.e. current only, full academic year, all student who tested, cohorts)
2. How to summarize the data (i.e. mean or median, percent proficient, distributional)
3. Filters to include (i.e. subgroups, test windows)
4. The purpose (i.e. to provide information about current students, summarize performance)
5. How to display the data (i.e. bar chart, box plot, heat map, table)
Why Box Plot is Best
After considering these and many other factors with respect to Scantron data, it occurred to me that the box plot is the best approach for the following reasons:
1. The data is based on a norm-referenced assessment. Such assessments do not have proficiency cut scores. Their purpose is to rank and compare performance to a national norm sample to provide context and insight into relative strengths and weaknesses.
2. Simply providing the median (the mean would not be appropriate due to the possibility of outliers due to the large range of possible scores) does not communicate the distribution of scores per aggregation (school, grade, class).
3. Using the box plot allows one to see the median score in relation to the distribution of score, as well as to identify any group of students along that distribution.
Lessons from November 4 PD Day
Based on the needs of teachers and administrators using the data, it occurred to me that the views should be:
1. Based on currently-enrolled students only
2. Printable to fit one standard 8.5 by 11 inch paper
3. Formatted for use in a data dig
Scantron Data Dig:
On November 4th (“PD Day”), teams of teachers with Scantron data met at SIA and UA to reflect on their data based on the assessment typology. This enabled them to make comparisons across and within grades and classes. The typology is as follows:
Upon hovering over one of the “cells” above, the user is presented with links to data.
Here is Mr. Naim reflecting on Scantron “Across Classes” results.
The assessment typology is ever-improving based on your feedback. If you ever have any specific views that you would find helpful, please feel free to contact me and I will gladly work with you to see if, how, and when I can produce them for you.
· Contains many helpful training and reference resources
· New and need help with data and/or assessments? Look here.
We Want to Hear From You!
We would like to know how these resources worked for you. Take a moment to participate in this 5 question survey. Let us know about the following:
Usefulness; Recommendations; Needs - CLICK HERE
If you have specific questions, feel free to contact Ms. Bruzewski and/or Dr. Lance. Contact information is on the front page.
Thank you for being an engaged partner!