Linear Regression and Hierarchical Linear Models

Authored by: Delena M. Harrison , Stephen W. Raudenbush

Handbook of Complementary Methods in Education Research

Print publication date:  April  2006
Online publication date:  January  2012

Print ISBN: 9780805859324
eBook ISBN: 9780203874769
Adobe ISBN: 9781135283315

10.4324/9780203874769.ch24

 Download Chapter

 

Abstract

Educational researchers often want to know whether specific characteristics of student background or experience predict educational outcomes. Answering such questions might help us identify particular subsets of children who are not faring well or children who are thriving in their educational settings. This might have implications for improving practice. Moreover, different educational theories often produce different predictions for which children will do well. Testing predictions made on the basis of theory can then yield tests of the theories. For example, under one theory, a highly specified, phonetically oriented reading curriculum in Grades 1 and 2 will produce high levels of word recognition and reading comprehension in Grade 3. Under an alternative theory, such a method would produce high levels of word recognition but not comprehension. One might generate and test these predictions by collecting data on children’s instructional experience in Grades 1 and 2 and on student outcomes in Grade 3 and then assessing the accuracy of these predictions through statistical analysis. Thus, there are good practical and theoretical reasons to be interested in generating and testing predictions of student outcomes.

 Cite
Search for more...
Back to top

Use of cookies on this website

We are using cookies to provide statistics that help us give you the best experience of our site. You can find out more in our Privacy Policy. By continuing to use the site you are agreeing to our use of cookies.