Affective Learning and Teacher Evaluation

Author of Tool: 

McCroskey, J. C.

Key references: 

McCroskey, J. C. (1994). Assessment of affect toward communication and affect toward instruction in communication. In S Morreale & M. Brooks (Eds.), 1994 SCA summer conference proceedings and prepared remarks: Assessing college student competence in speech communication. Annandale, VA: Speech Communication Association. 

Background: 

Experts in Educational Psychology have determined that there are three general categories of learning: Cognitive, Affective, and Psychomotor. The results of research in instructional communication suggests that instructor communication behavior may have its strongest impact on affective learning, although certainly impacting the other categories as well. Hence, measurement of affective learning has received considerable attention in this field. The first measures of affective learning in instructional communication research were developed in the 1970s. These evolved over a period of time. Early studies employed scales similar to those noted below but also included measures of probable use of the behaviors recommended in the classes in students= future lives.

Two problems were identified in this research. The first was that many courses do not have a focus on future behaviors, hence classes which do, and classes which do not, cannot be fairly compared using such measures. The second problem came from inappropriate use of factor analysis. All the items on all of the measures were factored as a single measure rather than being factored separately. In this early research it was determined that they could all be included as a single measure of affect. This was confused with affective learning. As a result, affect for content and affect for instructor were not considered separately. This was a case of the legendary adding of apples with oranges. The general score really could not be interpreted since affect for the course (and the likelihood of taking another course in the content) clearly related to the true affective learning construct but affect toward the instructor (and taking another course with the instructor) was measuring a construct of importance, but not one that was included in the affective learning construct advanced by educational psychologists.

More recent research has been sensitive to this distinction. Below is the set of measures (each with four bipolar scales) which is most commonly employed in this research. Each of the four measures can be collected separately or together. The first two measures (items 1-8) can also be used together as a measure of Affective Learning. Similarly, the third and fourth measures (items 9-16) can be used together as a measure of Instructor Evaluation. However, the items on the four measures should never be used together and referenced as Affective Learning. Instructor evaluation is not affective learning. It is a separate construct of importance in its own right.  

Psychometrics: 

Reliability:

The reliability of these scales in a large number of studies has been very good. The reliabilities for the Aaffect for content@ measure has ranged from .85 to well above .90. The other three measures consistently have yielded alpha reliability estimates above .90. Similarly, estimates for the 8-item measures of Affective Learning and Instructor Evaluation have consistently been estimated at well above .90.

Validity:

Face validity of the instruments is excellent. More importantly, predictive validity is also very strong. Many studies using these instruments have produced results in line with theoretical relationships of communication behaviors with affective outcomes.  

Keywords: 

Files: 

Digital Object Identifier (DOI):