Data here. Data there. Data everywhere.

baseball
Data.  Big data.  People talk about data these days as if they know what they are talking about.  They have no idea.  What we do know is that the amount of data has grown exponentially with new systems and techniques, resulting in new types of data.  I find it fascinating that it is making possible to compare things that you couldn’t before or never thought of comparing.

telescopeThat is why you see a picture of a baseball and telescope.  You can capture what is happening in baseball as it moves, but it is more than the speed and distance of the baseball.  Baseball and astronomy are defining and capturing new datasets.  It’s a lot of data.  They know some of the information that they can get from the data.  Furthermore, they only have an vague idea of what is possible to do with that information.  Usually data is to support findings.  With today’s data it is not known what are the questions to ask.  That fascinates me.  So what about baseball and astronomy?

Major League Baseball has been measuring the velocity, movement, location and spin rate of every pitch since 2001 Bruce Schoenfeld wrote in Can New Technology Bring Baseball’s Data Revolution to Fielding?.  By triangulating views from two cameras  perpendicular to each other, you can calculate where that ball was at a given moment and where it went.  But how to track movements of something that it is not a given size, like people?

In 2015 STATCAST technology started to capture and record the movement of people from cameras placed on the field. The data about the movement of players is now layered with the data from the radar system to see where the ball went.  The technologies generate 3D snapshots of every movement on the baseball field.  It comes to about 40,000 frames per second converted into digital data. There’s a lot of data.

What can this data possibly show? One example is that it can show how well fielders play their position.  No longer it is is just “did you see that catch?”  Now looking at where the ball went and where the fielder started to catch it, a percentage of how often a fielder makes a catch can be calculated.

A statistic that could not be calculated before but clearly has huge significance.

Data here, data there, data everywhere.

On NPR’s Ted Talks, Andrew Connolly talked about data and the universe.  With 24 data points, the Hubble telescope in 1929 showed universe was expanding. 24 galaxies. Seventy years after Hubble, by looking at 42 data points over 3 years, it showed the universe was not just expanding but accelerating. 42 supernovas exploding stars.  Small changes can give rise to new ideas and theories, even with only 42 data points.

Yet there are 10s of 1000s of galaxies with 10 supernovas per second.  Galaxies merge and collide, stars born and die.

The digital camera in the Large Synoptic Survey Telescope (LSST), currently being built, will take a picture  5 1/2 feet across which is 3 billion pixels.  One image from LSST will be equivalent to 3000 images from Hubble. Instead of 42 supernova in 3 years, they expect to find 500-1000 supernova every night.

They can test and rule out theories with the data, first by asking the questions that they have been wanting to ask.  With better and different data collected, they can also change the questions or ask new ones.  Forty-two data points completely changed the way that they looked at the universe.  What can these new data points show? At the end of the survey in 2030, a new theory of physics could emerge about the universe.

Data here. Data there.  Data everywhere.

Daren Willman who analyzes STATCAST data said in the above-reference, “There will be a whole new baseball revolution based on information that we are just starting to get.”  Different types of data could bring about a whole new way to look at the world.  It will also be how we develop approaches to learning.

Experience API (xAPI) has the ability to create new types of data on learners’ experiences.  It gathers data from different sources and puts it in one place, a Learning Record Store, where data can be analyzed.

I have never used xAPI in a project before.  To start with, we are planning to take data from a project completed and apply xAPI to it (if we get the funding for it, of course).  We did a study looking at different changes in knowledge and skills of health providers and of health outcomes in the health facilities where the providers worked.  It was linear analysis.  Looking at each set of data in silos.

Using xAPI, we want to ask questions that we may have not asked or thought of before.
-If health outcomes are improving in one facility but not other facilities, why effect does the different mentors have?
-If a health provider performs well on the knowledge “test” of a skill but cannot perform the skill, what may be factors?
-If one health provider is doing well but others are not in the same health facility, why?
-What do health facilities that overall are improving health outcomes have in common?Hopefully, I’ll be writing about it in a few months.

With these new types of data, more than you could ever want or imagine, we can ask a different set of questions than we didn’t before.  We will find information that could be useful to answer those questions.  If we are lucky, we will figure out what to do with those answers.  If we are truly fortunate, after applying it, it changes the way we see or do things, for the better.