As you know, adults are resistant to change. Teachers are asked to produce data but given minimal training outside of “compare average test scores”. And without a math background, this may even make sense to those educators and superintendents. Therefore, when it’s easier to compare a mean and it cleans up the mandatory paperwork faster, this is the way things are done.
Question: How will (only) comparing averaging actually help individual student learning?
Question: If teachers lack a background of statistics and, even more frustrating for the educator, lack the time to learn the basics, how will they begin to leverage their own student data to improve learning? Ultimately, it is what they WANT to do. But how?
Solution: Educators need to answer deeper questions about their students using data without additional statistical training all while using their time efficiently. It must also be priced for teachers: free. And it’s here. It’s called Tableau. It’s data visualization. Instead of looking at a sea of numbers, Tableau produces pictures. Without a math background, anyone can look for trends and draw conclusions. And it’s free to educators.
Tableau allows teachers to import student gradebook data (most gradebooks export as a .CSV). Once the educator is in the Tableau workbook, one can merely hold down the CTRL key, click on whatever variables they would like to compare/explore. A “show me” set of suggested graphs pops up (if it doesn’t automatically pop up, after taking fingers off the keyboard, CTRL+1 will do the trick). You can also just drag and drop into the workbook. Drag and drop students to color. Play with it. And sometimes an ID will need to be set to a string (so the software knows you’re talking people, not calculations) and sometimes you’ll need to switch columns and rows for a better visual. I recommend sorting students by whatever measure (assessment? assignment? overall grade?) you are asking your data to compare. Playing with the visualization is a fun way of learning how to use the software. It won’t take long.
My first visualization
This graphic sorts messy data from Unit 3 (The Linear Regression unit) into a clean, organized dashboard to help me compare my students’ formative and summative assessments (sorted on Unit Test score, ascending).
I was shocked to see the overall trend in the formative to summative scores: They went DOWN. And they shouldn’t. And that’s a validity problem from my end. But this was not so evident in looking at the aggregate data. A t-test would tell me there is “no significant difference” between quiz and test scores. But we’re talking individuals, my students. And my job is to GROW them. By looking within the data, I found trends about which types of students, for example, lost traction from quiz to test. And my ultimate conclusion was to take ownership on my end. (This could be another post for another day.)
After playing with Tableau some more, I realized rows worked better than columns for the above visualization.
And did you know that approximately 8 percent of men and 0.5 percent of women are red/green colorblind?
So my next 2 units looked more like this:
To support our school’s mission and vision, I began teaching other teachers how to leverage student data within their PLCs to draw meaningful conclusions about teacher methods and student learning with Tableau. And these teachers are excited to identify trends and answer deeper student needs questions – to ultimately help and grow each individual student.
It is time teachers stop looking only to aggregate data and averages. We need the tools to find trends within each student’s learning patterns in order to provide them with the best “differentiated” learning experience for them. Unfortunately, I have found there is a huge gap between what districts want and what teachers are asked to do.
Eventually, data dashboards that ultimately give teachers a visualization of their current student data, including growth and achievement data, is the future of education. Right now teachers who want this will have to figure out the software (thankfully, Tableau is easy to use for simple visualizations.) But ultimately, data visualization through dashboards are the next step in the journey.
Beloit College creates a “Mindset List” every year for the entering college class. For the college class of 2014 (or, those of you who graduated from Centennial in May) the list has just been posted. Here are a few touchstones of your life:
Wow. I used to have to go to a library, search through databases using Boolean operators just to find the location of a pile of microfiche in a basement. This Google thing has changed the face of research forever.
A student brought this to my attention on Monday — as heard on NPR, the placebo effect is apparently getting stronger.
Well if you think about it, there’s a pill for everything these days, right? The article suggests that people, when sick, go to the doctor, get a prescription, take a pill and get better (in general). So maybe people are programmed to think they will get better?
But what does this mean to medical research? What if a new medication significantly decreases, say, blood pressure — when tested against the placebo, will it fail?
“Suppose you’re on a game show, and you’re given the choice of three doors: Behind one door is a car; behind the others, goats. You pick a door, say No. 1, and the host, who knows what’s behind the doors, opens another door, say No. 3, which has a goat. He then says to you, “Do you want to pick door No. 2?” Is it to your advantage to switch your choice? (Parade Magazine, Whitaker 1990)
Here is a simulation and an explanation of the answer.
Here is an article from the NY Times entitled “For Today’s Graduate, Just One Word: Statistics” (dated August 5, 1009) highlighting the ever-growing field of statistics. Organizations such as Google and IBM are looking for good statisticians as the demand for data analysis increases:
“We’re rapidly entering a world where everything can be monitored and measured,” said Erik Brynjolfsson, an economist and director of the Massachusetts Institute of Technology’s Center for Digital Business. “But the big problem is going to be the ability of humans to use, analyze and make sense of the data.”
Another interesting quote in the article: “I keep saying that the sexy job in the next 10 years will be statisticians,” said Hal Varian, chief economist at Google. “And I’m not kidding.”
What is the first thing you should do when you encounter a mess of data?
Draw a….what? (It’s in your notes…)
Draw a PICTURE. A distribution. A graph.
LOOK AT IT.
But statistics doesn’t just revolve around histograms, boxplots and scatterplots. Statisticians have (marginally) grown personalities over the years and realize non-statisticians need something tangible to understand data trends. Enter: Nathan Yau of FlowingData.com, a PhD candidate in statistics who makes use of his background in computer science to explore and visualize data.
Since the word “data” sounds so dull…like “widgets” in economics…let’s look at a few examples Yau took from reality:
Evidence of “data-visualization” tools is very commonplace these days and you’ll find that many popular websites mix humor and/or pop-culture into their infographics (The Onion has been doing it for years).