[Assessment 1481] Re: Using Data: Where to Start?

Archived Content Disclaimer

This page contains archived content from a LINCS email discussion list that closed in 2012. This content is not updated as part of LINCS’ ongoing website maintenance, and hyperlinks may be broken.

Cathay Reta cathayreta at sbcglobal.net
Mon Dec 8 16:07:09 EST 2008


Hello All,
I’m happy to jump in here with this question.  Where do you start?  I'd like to back into this a bit.  Let me first respond by posing some questions to you (and all interested in deciding where to start).  1)  What do you want to know?  What are you trying to improve?  2)  What data will give you that information?  And if you aren't sure, then as Tina suggested, conducting assessments with your students and staff may be the place to start.
For your consideration, here are a few places other programs have started.  One is with student attendance.   How many hours a week/month are students attending?  As you may have looked at in the Hawthorne Family Literacy Program case study, there is a definite correlation between student attendance and student learning gains.   So perhaps you want to get a handle on your student attendance rate and go from there to decide how you can increase it.  Another program looked at their student attendance data, analyzed it and saw that attendance dropped drastically when the weather was bad.  So they had their instructors review inclement weather policy with their classes.  Another program had a similar finding related to holidays.  So they changed their schedule to not have classes during those low attendance times.
Another area may have to do with increasing the number of students you pre and post test.  That's also a good starting place, which leads me to my next point -- probably the most important (and least glamorous) point. 
You need to make sure your data is valid and accurate.  Is information entered into your database in a consistent manner?  One program in our project found that student information was not always entered the same way.  For instance, sometimes a nickname was used (Joe) and sometimes the full name was used (Joseph).   So perhaps Joe had a pretest score in the database, and Joseph had a post-test score, but the two were not recognized as the same person.  After they cleaned up their data, and implemented a new procedure in which fewer people were involved in entering data, they increased the number of students with pre and post test scores by 28%.  We found one of the model program promising practices is that they have one person in charge of data.  Others may help to input it, but it is the responsibility of one person.
So all of that to say, start with clean and accurate data.  Once you decide what you want to look at, make sure the data you will base your decisions on is valid. 
Does that help at all?   
Cathay Reta

Cathay O. Reta
Cornerstone Concepts
6670 Southside Drive
Los Angeles, CA 90022
Ph: 323) 728-4302
cathayreta at sbcglobal.net


 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lincs.ed.gov/pipermail/assessment/attachments/20081208/4aa1d87b/attachment.html