Thursday, May 16, 2013

Without Data, All You Have Is An Opinion

A few years ago our math department created a new math class, sort of an "Algebra 1.5".  In the traditional Algebra 1-Geometry-Algebra 2 track, too many students just weren't ready after a 10-day intensive review of Algebra 1.  We created a "bridge" course just for them.

Our current principal doesn't like the course.  He doesn't think kids need it.  As a result, the course isn't scheduled for next year.  So instead of meeting the students where they are and offering a course than can help them progress, they'll now either sink or swim.

I provided much background information in the introduction to this post a couple weeks ago.

I received the data today.  Analysis begins next week!

Why next week, you ask?  Well, for reasons I won't go into in this post, our seniors graduate Monday, but our underclassmen attend school until the first week of June.  My statistics classes will be empty except for 2 juniors.  I'm going to give them the data and have them do a chi-square analysis to see if that Algebra 1.5 course adequately prepared students for Algebra 2.

You want real-world application?  I gotcher real-world application right here!

It may turn out that our principal is right.  I don't know what he's basing his opinion on, but it's not data.  I intend to find out.


Ellen K said...

Data driven educrats are killing us. I just ranted about it on my site. I don't know how much longer I can do this.

Darren said...

Is making decisions based on data worse than *not* making decisions based on data?

Dam Lies said...

It is not rare that data is used to support pre-conceived opinions.

You may very well do some fancy numerical analysis and then say, "Look: this proves I was right all along."

Be more impressive and tell us what "Algebra 1.5 was a success" numbers look like and what "Algebra 1.5 was a failure" numbers look like.

But tell us BEFORE the analysis is done. Not after. Otherwise, retrofitting is much too easy and tempting.

allen (in Michigan) said...

Sorry Ellen but it's data-driven educrats who only have to answer to rules laid down by legislators that are killing us.

The democratic process is so inherently inefficient at maintaining control over a dynamic process like education that the real wonder is that kids get educated, to the extent they do, at all.

Also, data isn't enough. Those in a position to need data to make decisions also have to have to proper motivations to make the best decisions based on that data. But the motivations that typically drive those decision-makers derive from the legislative and the judicial process neither of which has the education of children as its central focus. Until that situation's resolved, as it is being due to the expansion of parental authority, things can't improve generally.

That's why the education reform movement's so gratifying in that it proves my view that we, the people, no matter how wrong we get things in the short term will eventually see the error of our ways and right those wrongs in the longer term. In this case the inherently stupid idea that the state can do a good job of educating kids.

Darren said...

I haven't stated what I think it "right", but it's nice to know that someone thinks I would lie to my readers. I question why that person would read this blog in the first place.

I'm going to define "success" as getting an A or B in the first semester of Algebra 2. We will do a chi-square analysis to determine if the proportion of students who achieve success in Algebra 2 without the transition class is any different than the proportion who achieve success in Algebra 2 after having taken the transition class.

There might be better ways to do the analysis--if, for example, we could know what grades people would have gotten in Algebra 2 without having taken the transition course--but I think this is the best we can do with the data that's available.