Lesson 8 will consist of the following topics
For lesson 8, read pages 101-157 in Educational Research: Competencies for Analysis and Application, Seventh Edition by L. R. Gay and Peter Airasian (2003, Merrill/Prentice Hall, ISBN 0-13-099463-4)
or read pages 98-150 in Educational Research: Competencies for Analysis and Application, Eighth Edition by Lorrie Gay, Geoffrey Mills, and Peter Airasian (2006, Prentice-Hall, ISBN 0-13-118534-9)
or read
pages 150-181 in Educational Research: Planning, Conducting, and Evaluating
Quantitative and Qualitative Research, 3rd Edition by
John W. Creswell (2008, Merrill/Prentice Hall, ISBN-13: 978-0-13-613550-0)
The way in which we select subjects for a study determines how we are able to generalize the results of our study. That is to say, to what group of subjects does our study refer?
1. Population and sample
Population - an entire group of persons or elements that have at least one thing in common (Minnesota fourth graders, MSUM summer school students).
Sample - a small group of persons or elements (observations) selected from the total population
We want the sample to be representative of the population, so we need to use a sampling method that will achieve this.
2. Sampling methods - methods of selecting a sample from the population
§ simple random sampling - equal chance for each member of the group to be selected
§ systematic sampling - selecting every 10th member (for example) from an alphabetized list
§ stratified sample - dividing the population into subgroups and then randomly selecting from among these subgroups. Number selected from each subgroup can be proportional or equal-sized.
§ cluster sampling - groups are selected rather than individuals. For example randomly select 5 elementary schools from among the 25 elementary schools in the district.
§ incidental or convenience sampling - with this type of sampling, just take any group that you have access to. This, of course, will cause problems in generalizing the results of your study.
3. Determination of Sample Size
§ Samples should be as large as possible; in general, the larger the sample, the more representative it is likely to be, and the more generalizable the results of the study are likely to be.
§ Minimum, acceptable sizes depend on the type of research. Generally you would need 30 subjects in each group for correlational and descriptive research but may be able to get by with 15 subjects per group in experimental or quasi-experimental designs.
§ Even large samples can lead to erroneous conclusions if they are not well selected.
§ The authors of your text provide a table (Table 4.2, page 135 in Gay & Airasian, 2000 or Table 4.1, page 125 in Gay, 1996) based on work of Krejcie and Morgan which shows recommended sample sizes based on various population sizes.
4. Sampling Bias - sampling bias is caused by systematic errors (rather than random errors), in the sampling process.
Data collection must be acccurate, where tests are used they must be scored correctly, and observations must be made systematically.
In some cases data may be coded - for example males coded as 1 and females coded as 2
An electronic spreadsheet is an excellent place to keep the data for your study, both raw data and coded data. In most cases you will also be able to perform the desired statistical calculations from within the spreadsheet. The Excel spreadsheet program, for example, has an Analysis Took Pak you can install in your Excel spreadsheeet program. This will allow you to calculate such statistics as chi-square, correlation coefficient, t-test, z-test, and analysis of variance.
According to Gay, the author of the text for this course, there are three major ways to collect data.
Standardized tests of one sort or another are used in most educational research studies. You will frequently use standardized tests to measure one or more of the variables in your study. It is important that you get as much information as you can about the tests that you are going to use in your study. In some cases no suitable instrument exists to measure the variables of your study. In that case you will have to design your own instrument. Look at the instruments (such as questionaire forms) which have been used in studies similar to yours and then modify the form to use in your own study. Many of the parameters on which standardized tests are rated, will not be available for an instrument you design.
Characteristics of standardized tests
9. Reliability - reliability is a measure of how consistent the results from a test are. If you administer a test to a subject twice do you get the same score on the second administration as you did on the first? The reliability of the test will answer this question.
There are many different types of tests that you might consider using in your study. Some of the most commonly used types of tests for educational research are:
11. personality tests
12. aptitude tests, including tests of academic aptitude (intelligence tests)
There are many sources available, in most university libraries to assist you in identifying an appropriate test to use in your study and also to provide evaluative information about a test you would like to use. You should avail yourself of one or more of these sources to assist you as you design your research study.
14. Tests in Print - Tests is Print is frequently updated and is a good source to initiallly look at for information on a test. It provides references to the appropriate Mental Measurement Yearbook for more information on the test.
15. Test Critiques - Test Critiques are another source of information about tests you may wish to look at.
Click on Lesson 8 Assignment to go to the assignment for Lesson 8.
Click on Lesson 8 Quiz to go to the Quiz for Lesson 8.
Please send electronic mail to the course instructor if you have any questions about this lesson or other concerns.