EDTE 250 - Education Research
Forum - Validity in Action Research
Due Midnight, April 11, 2009
Preparation and Notes
Please read this page carefully. Your assignment is described at the bottom of this page.
Introduction
There are many environments where an experimental design is clearly the methodology that should be used. There are also other environments that might call for another way of thinking about research. There are times where being limited to experimental requirements such as: detachment, large numbers of participants and proving generalizability may limit or preclude practitioners from conducting research. It can be argued that learning environments such as classrooms are environments that could be served well by having a wider variety of research tools. Practitioner Research or Action Research has the potential to provide practitioners with a methodology that goes beyond experimental design, but does not exclude many of its useful features.
Validity is a key concern in any research project. There are many definitions of validity. Ultimately, for action research purposes, validity refers to the degree to which the conclusions reached as the result of the study can be trusted (considered trustworthy).
There are different ways validity or trustworthiness can be established in an action research study. Two key methods for this are to use triangulation and to focus upon collecting data in ways that "show" the reader what happened, as opposed to "telling" them what happened. For example, a teacher might "tell" us that student motivation was increased when students were allowed to create electronic portfolios, but how are we, as readers, supposed to know that the reporting is trustworthy? A valid action research study in this case might "show" us through data collection that students had better attendance, completed more work and made positive comments about making portfolios when interviewed. Showing the reader these details, combined with a report of the researcher's own perceptions (such as reporting on notes of observations recorded in a journal) would allow the researcher to both describe what they mean by "motivation" and show us how they arrived at the conclusion that students were in fact more motivated when they were allowed to create electronic portfolios. In this case the researcher both triangulated their data (looked at the same data a number of different ways) and "showed" us a great deal of detail. This type of study design and reporting helps the reader to see that the researcher has gone to some lengths to make their study conclusions valid and trustworthy.
In response to changing definitions of validity, different models for establishing validity have emerged. These models consist of a series of conditions which, when met, lead to more trustworthy studies. In our last exercise each of you selected various types of data you might collect to yield trustworthy results. For our exercise this week, we will explore basic data analysis strategies and three models for establishing validity in action research. Once you have reviewed the resources, you will design a draft of a validity strategy for a research project you might pursue.
To complete this assignment:
1) Read about Models of Validity in Action/Practitioner Research.
2) Using Word, submit a plan to me (cowanj@csus.edu) that contains the following:
Each of the 3 models for establishing validity offers a series of strategies that help to assure trustworthy research results. Based upon your reading of the models of validity, discuss which of the validity models might fit for you and for the reasearch project that you have been condiseridng and why you chose this model. You are also free to create a hybrid plan by using parts of all three models if you like. If you choose this method, please explain why your hybrid model would work for your situation.
Instructor(s)
John Cowan
|
