Mirror Sites:
CSUDH - Habermas - UWP - Archives

California State University, Dominguez Hills
University of Wisconsin, Parkside
Soka University Japan - Transcend Art and Peace
Created: June 30, 2003
Latest Update: July 2, 2003
jeannecurran@habermas.org
takata@uwp.edu

Chapter 1: Getting Started: Fundamentals of Research Design

Site Copyright: Jeanne Curran and Susan R. Takata and Individual Authors, June 2003.
"Fair use" encouraged.

Following our principle of community life-time education, Susan and Pat and I want this material to be of use (free access) to anyone who needs the information. Because our universities use Statistical Product and Service Solutions (SPSS), the new name for Statistical Package for the Social Sciences, you will need to have access to the SPSS program. If that's a problem and you really want to learn this introductory statistics, you can purchase a student version at most university bookstores. If you have access to a different program, such as MicroCase, SAS, or Minitab, you'll find that most of the procedures are similar enough that you should be able to translate them with no problem. There's also free access to an SPSS manual, sponsored by a California State faculty group. Nag me to find that URL for you.

We are using the two texts listed below to structure this introductory course. That's a lot for one semester, but if you're doing this on your own, that's probably a lot saner, because you can go at your own pace. Both texts are peculiarly adapted to a realistic understanding of the use of statistics in everyday lived experience. And we have tried to see that there is enough information on this Web site that you can substitute any traditional book you happen to have available. Any questions? E-mail jeanne at jeannecurran@habermas.org.

The Texts:

• J. Richard Kendrick, Jr. Social Statistics: An Introduction Usng SPSS for Windows Mayfield Publishing Co., Mountain View, Calif. 2000. ISBN 0-7674-1001-7
• Timothy J. Lawson. Everyday Statistical Reasoning. Wadsworth and Thomson Learning. 2002. ISBN: 0-534-59094-2 (paper).

If you compare Kendrick's to Lawson's text, you'll see that Kendrick does not mention ANalysis of COVAriance (ANCOVA), and he does not go into Bayesian probability. No matter. Getting through these two books is going to take some planning. Kendrick's text offers you the SPSS windows, so that you can review even when you're not in the lab. I think you'll need that because these are new processing skills and they require practice. Kendrick's text is set up to permit that. And Lawson's text is set up to make you consciously aware of how all this fits into your lived experience.

Jeanne's Notes from Kendrick's First Chapter: Elements of a Research Study

This title is an "advance organizer." (Ausubel) That means that it tells you what you're about to learn. It identifies the material so that you will know where that fits in your cognitive map. It fits with research study. So maybe it will be related to methods and research as those come up in your courses. And they will come up whenever your text reviews studies and /or you read research articles. The advance organizer tells you that it consists of the elements of any research study. So it's a list. Lists are good. They give you a structure, like a skeleton, on which you can hang what you are learning and find connections with other things you already know. That helps you remember.

Now, I could have written that you will learn here the elements of a research study, and then in the summary write that you can now explain what they are. That kind of language is almost as patronizing as "This was an interesting field trip." you were taught to write in middle school, and now find your teachers saying that's not college English. It isn't.

Instead I expect you to realize that the titles I set off, the key phrases or words I underline in another color, are advance organizers, flags, put up to alert you to what we're going to study, and how you could measure your learning of that concept. I will not patronize you and waste my time and energy saying "You will learn to spell ten words." "You have learned to spell ten words." Yuk. Both Kendrick and Lawson tend to do this. But they are not retired, and the school likes that kind of junk. And once they've done it for twenty years, they'll probably go on doing it out of habit, and we will have dulled down the academic world again. Yuk. Ausubel meant advance organizers to be shorthand flags you would recognize that you could use like paper clips in your apperceptive mass (Herbart), to pull up other ideas you attached them to from your memory. He didn't mean them to reflect patronization that disrespects your intellectual maturity.

That said, consult Advance Organizers and Note Taking for a complete explanation of advance organizers and their use in learning and teaching.

So these are my notes from reading Kendrick. I've tried to pull out what I'd like you to understand. I'll outline it, leave white space so you'll notice what I want you to, and use colors and form and visual aids to give you my reactions to the chapter. Then I'll give you a choice of ways in which you could let me know what you've learned. No traditional testing. So you should be able to focus on learning. These are job skills. You need them.

Elements of a Research Study: (Kendrick, at p. 7.)

First, I was very uncomfortable with the Elements of Research Design Kendrick gives. That's because he has assumed lived experience I don't think you've had. Here I've rewritten how you should really approach social research for the first time:

• Read with curiosity some articles of interest to you in a well-recognized newspaper. Example, (Ibid., at p. 4), on condom use.

• Ask questions that you think they should have answered. Or think about how what they say might apply to you or your friends or your parents or your kids.

  The Associated Press article (Ibid., at p. 7) says that "[s]afe-sex lessons for children are more effective if condom use instead of abstinence is emphasized, researchers found in a study of inner-city Blacks." One of the people who taught abstinence for the eight-hour health educatiion thought abstinence would have been more effective if they'd had more time. They had eight weeks. They measured success by the students' self-report of whether they had engaged in sex within six and twelve months of the study."By 12 months, 20% of the abstinence group had recent sex, compared to 16.5% of the condom group and 23% of the control group. The students in the study were in the sixth and seventh grades.

  What questions might you ask?

  • Would the results be different in a suburban school? Is race a factor?

  • Did they really need "more time" or did they need to disperse the messages throughout the whole curriculum for a few minutes here, a few minutes there, so that it sinks in more effectively than a lecture. Notice that they didn't tell us how the eight hours was used.

  • How would you measure successful sex education with some group that matters to you? Could we find a measure more sensitve than just did you or did you not have sex in the last 12 months? What about agreement with an attitude that says unprotected sex is very dangerous, somewhat dangerous, dangerous, not very dangerous?

  • Is this a health issue or a moral issue? The results reported in the AP article suggest that one of the researchers who taught abstinence needed and felt she should have more time to teach abstinence. The article itself says that:

  • Would church groups be a way to replicate or compare the study? What would the church group say about teaching condom use? Does that depend on the church group? Could you get permission to ask such questions of a church group?

    CAUTION Past indiscretions have led the government to enact protections for research subjects. Thes protections are designed to prevent your needlessly embarrassing or hurting someone by the questions you ask. All educational and research institutions abide by a process for determining the appropriateness of questions to be asked of human subjects. So don't forget to check with any institution in whose name you are planning to do any research in the real world.

    For example, IT IS NOT OK to walk up to a strange woman and ask: "Have you ever had an abortion?"

• Use the articles and your questions to find academic articles on the topic.

  In the library look in Sociology Abstracts of Psych Abstracts or Education Abstracts for keywords like teen sex, condoms, family planning, family research, teen pregnancy, etc. From the article you might glean: sexually transmitted diseases, abstinence, safe-sex. You might also learn that they measured sexual activity by self-report data, so search for self-report data.

• Notice the delimitations of the study: inner-city Blacks, sixth and seventh graders, schedule that allowed eight hours of instruction. Delimitations are the limits that the researcher places on the study to make it feasible and cost effective. For example, not a good idea to suggest that your first research study will draw a random sample of sixth and seventh graders in California. You don't have either the time or the money for such a study at your disposal. The delimitations you recognize should tell you that.

  Now figure out how you could either replicate the study or look for a comparison group. Could you just walk into a school and test sixth and seventy graders on self-report of sexual activity? We hope you said no. Schools wouldn't like that. Could you look at a different age group, and ask some of your college friends questions about their attitude towards safe-sex, where they learned that attitude, and their moral position on safe-sex? CLUE: Don't ask your friends if they've engaged in sex and whether it was safe-sex over the last 12 months, if you want to still have friends. Common sense still prevails.

All these questions ought to come before you ever begin to consider your research goals.(Ibid., at p. 7.) I recommend a replication or a slight variation of a previous study for your first venture into research. We don't do enough replications in sociology, anyway. And this way, you can use earlier research as a model.

Now for Kendrick's list that starts as mine ends. Kendrick's list of steps in the research process (Ibid., at p. 7):

1. Specify research goals.

   After you've found some models and asked lots of questions. Yes, then specify your research goals. Having a real study in front of you should prevent your deciding to investigate how effective teaching safe sex, supplying home care, and supplying drugs are in reducing AIDS in South Africa . Why would I want to prevent that? You're not in South Africa. You don't have access to samples. Previous studies have shown a relative lack of success on all counts. The topic is huge, mammoth. Start with something that intrigues you that is small enough for you to find a sample, collect data, and analyze it within a semester. That's not very long.

   If you really must take on a big question like reducing AIDS infection in Africa, then do a secondary analysis of published data. Find articles, read them, ask questions, see if you can describe the situation so that your friends and classmates can understand it. Then do the analysis and interpretation without collecting your own data.

   Goals in an introductory stat class need to be small and doable. In other words, quick and dirty. Share data. Reanalyze data. Use any of the massive data available on the Internet.

   And remember that the goal of your work in this class is a grade and certification of having passed statistics, not a Nobel Prize.

2. Review the literature.

   Choose an area, a carefully delimited area, that you've been introduced to in a text, or in class, or in your general reading. Be absolutely certain that you conceptually link your specific research goals to the theoretical and methodological slant of the course. You're doing this to get a grade, remember? So it should relate to what you're going to get a grade in.

   For samples of conceptual linking consult Concept linking - Process Texts: Working Together . . . Getting the Concept Straight

   then tomorrow. list of key words and definitions

   list of sample word problems they could do orally or verbally

   list of sample SPSS actions they could use to measure their learning.

   Discussion questions for each lecture.

   Enter induction / deduction wherever it fits with Lawson and Kendrick.

   Update statpreps and writing.