Test for interval and nominal variables

#1
Hello
I am doing a study on error correction. I would like to know if years of teaching have any effect on the method of error correction. I have surveyed 51 teachers. I have coded their responses to different situations involving student errors. I believe my independent variable is their years teaching which is interval and the dependent variable is the method of error correction which is nominal or categorical. I would like to know which test I should run to see if their is a relationship between years of teaching and method of correction. I have all my data in SPSS ver 20.
 

Karabiner

TS Contributor
#2
What exactely do you mean by "relationship between years of teaching and method of correction."?
For example, for any method, you could compare years of teaching between those teachers who
use it at least in one situation, and those who do not use it at all (a series of t-tests). Or, for any
method, you could correlate number of situations where it is applied by a teacher,
and years of teaching experience. Or, if you put teachers into distinct categories
("usually prefers method A", "uses methods B and C" or something like this), then you
could compare teaching experience between categories using oneway analysis of variance.
Or, for each situation you compare years of experience between methods used
(a series of oneway analyses of variance).

But maybe you had something else in mind.

With kind regards

Karabiner
 
#3
Karabiner:

Thank you for the reply. My question is if years of experience have any effect on the use of correction methods in each of the situations. I would like to be able to say, for example, the less years of experience the more they would use recasts, or the more years of experience they had the more they would use metalinguistic correction. On the other hand, I need to know if years of experience have no impact on the method of correction used by the teachers in each situation. I know I may not be using the correct statistical terms.

Thank you again,

John Burrell