ANCOVA Validation and Discriminant Analysis

I’ve finally finished my project at CSU.  I’ve used multiple methods of discriminant analysis to predict student proficiency level performance on the Alabama Reading and Mathematics Test using the Thinklink predictive assessment.  Before using these techniques, I used an ANCOVA model between different schools containing different grade levels to show validity.  My error rate was about 28% misclassification compared to the predictive assessments misclassification rate of 50%.  I will have my paper published later on the site.  I’m scheduling a presentation date at CSU this week.  I’d love for anyone who reads the blog to attend.  Here’s an excerpt from my paper:

Figure 9:  General Linear Models For Three Schools In ACS School District From      Independent Testing Data


Model Validations
 Elementary School Middle School Junior High School
y = .38x + 81.0 y = .34x + 134.8 y = .22x + 355.8
Slope p-value < .001Intercept p-value = .36Pearson’s R = .83 Slope p-value < .001Intercept p-value = .21Pearson’s R = .68 Slope p-value <.001Intercept p-value = <.001Pearson’s R = .83

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s