Friday, September 8, 2017

The Secret about "Good Data"

Everyone wants "good data", but so often it's rarely found.  The minute you receive low scores, you have "bad data", right?  If you receive high scores, skepticism creeps in, and you doubt if they really knew it at high levels.  Does your data leave you with doubt?

Here's the reality.  Good Data is Fool's Gold.  There's no such thing as "good data".


What we need is VALID DATA

The secret to getting valid data is to create valid assessments.  Think about it.  If you fail to spend time creating assessments that you believe in, you will discount the data it yields and therefore blame the assessment for giving you "bad data" or invalid data.  If you have little involvement creating the assessment and the data is "good", then you will wonder if the assessment truly met the depth and complexity of the standards you wished to assess.  In short the quality of the assessment leads to the validity of the data.  There's no other way around it.


The Secret to Valid Data
Bottom line is there are a few things to consider if you want to move from getting good data to generating valid and actionable data.

  1. Know your standards.
  2. Create assessment items that assess those standards at the basic, intermediate and advanced level.
  3. Provide students a variety of assessment items that require students to transfer their learning in predictable as well as unpredictable situations.
  4. Analyze the student data by student by standard by level of proficiency.

At the end of the day, you will never get data that will drive your work with laser-like precision until you create assessments that are intentional, meaningful, and most importantly reliable in driving your response to student learning.  Good data tells you how you did.  Valid data drives you to know what to do next.

No comments:

Post a Comment