Sociology 2155: Outline 9

I. Examples of elaboration using SPSS

A. Theory: Glock--People who are denied status gratification in secular life are more likely to be strongly religious. Two groups he identifies to which he thinks this dynamic applies are women and minorities and it should also apply to lower social classes. Use GSS2000Selection 1

B. Secondary research (why that term?) Let's explore this using cross tabulation and then elaboration

Step 1: Recoding .

a. Let's use church attendance as a measure of religious involvement, but we should probably recode "attend" into a simpler version that just shows weekly/less than weekly (attendr2). Why?

b. Then for social class as a measure of status deprivation, , we can use the self-report variable, "class," but we might also want to use income(why?) , and again a recode is probably appropriate. Why? So let's recode income (income98) into lower/medium/higher (incomer3).

Just to confirm the importance of recoding, let's do a cross tabulation of "income98" and "attend".

Step 2. We now have four different potential measures of status deprivation: sex, race, and two measures of class (class, incomer3) Let's run bivariate tables with these as independent variables and attendr2 as the dependent variable.

Do you see support for Glock's hypothesis?

Step 3: Elaboration (partial tables)

1. Intervening variables.

a. Explore the relationship between race and religious attendance controlling for class; then income

b. Explore the relationship between sex and religious attendance, controlling for class; then income.

2. Possible spurious relationships. Control the relationship between class and religious attendance, controlling for race. Control the relationship between income and religious attendance, controlling for race. Do the same thing, controlling for sex. Notice that it would be nice to be able to control for both race and sex.

C. Class, race, and approval of police use of force: Open GSS2000Selection2 and explore the relationship between class and approval of police use of force, controlling for race. Is this a direct relationship or an interaction?

II. Entering the data from your surveys into SPSS and analyzing the results

III. Writing research papers:

A. Abstract: Purpose of the study and its findings. What you intend to show and why. This should take the form of an abstract, which immediately precedes most journal articles. (150-200 words is a typical limit).

1. You need to get the readers attention with your introduction of the research problem. Everyone has more than enough to do. Why should they care about your research and its findings?

2. You would typically write this as the last step in writing your research paper.

Here's a version for the Michael Kearl example I used in presenting about research proposals.

The issue of physician assisted suicide is one on which many Americans have strong feelings. Only one state,Oregon, allows such assistance, and their law was recently challenged by the Bush Administration and uphelpd in the Supreme Court by a 6-3 margin. National surveys of Americans over the last 50 years have shown consistently increasing support for active euthanasia. In this research, we will draw on the General Social Survey to show that a likely factor in the increased support for physician assisted suicide is a decline in the religiosity of Americans and that this dynamic is particularly prevalent among the middle and upper middle classes.

B. Review of literature and statement of the research problem/research question.

1. This is where you show, in relation to what other researchers have done, what makes your project worth doing, and what you expect to find.

2. Be sure to give credit to those whose work helped you understand the problem or whose ideas you borrow. DON'T PLAGIARIZE. It doesn't detract from the value of scientific work to be using other people's work and ideas. In fact, generally little of value is accomplished in any science without that kind of foundation.

Isaac Newton: "If I have seen further [than others] it is by standing on the shoulders of giants."

3. The literature review you did for your research proposal is a good beginning point, though you would typically expand it with material from more of the articles you found in your search of the relevant research literature.

C. Study design, including research hypothesis or hypotheses.

1. What research method (s) did you use and why?

2. How did you measure the key concepts?

3. Was sampling involved? If so, how was carried out? Are there issues with the representativeness of your sample?

4. If experimentation, how were subjects recruited

5. If field research, how was the site (s) determined, and what implication does that have for your ability to generalize?

6. You should present enough information here that other researchers could replicate your research.

7. The simplest research hypothesis identifies two concepts (or variables) and specifies the relationship between them, usually a causal relationship.

Example: Among high school students, social class is one of the major determinants of political ideologies.

Here we're clearly hypothesising a causal relationship.

8 . Your hypothesis can also be more specific about what you expect to find.

Example: People from higher social class backgrounds are more likely to choose college majors in the liberal arts.

Notice that your hypothesis doesn't have to be true; it's just your hypothesis--what you expect to find, based on some combination of theory, previous research (by yourself or others), and practical experience. Even if your results don't support your hypothesis, at this point in the research report you should be writing about what you expected to find.

9. With two independent variables and one dependent variable, you can express that in terms of 1 hypothesis or in terms of 2 (or more) related hypotheses. You can also incorporate control variables in your hypotheses.

a. Among UMD students, one of the major determinants of political ideology is social class background. Another is gender. We hypothesize that the relationship between social class and politics will hold primarily for whites and not for minorities.

b. Among UMD students, both social class background and gender affect people's political ideologies. We hypothesize that the relationship between social class and politics will hold primarily for whites and not for minorities.

10. Again, your research proposal is a good beginning here, but as you proceeded with your research, you may have made some additions or modifications.

Remember the research proposal example that I derived from Michael Kearl. It looked something like this:

RESEARCH HYPOTHESIS. It is predicted that more strongly religious one is the less likely one agrees that physicians should be allowed to end the lives of incurably ill persons. We also predict that this will be true primarily for the middle and upper classes, since the lower and working classes tend to be morally conservative and much less supportive of euthanasia.

D. Analysis and interpretation.

1. If your research is quantitative and your variables are nominal or ordinal, you should begin with frequency distributions for the relevant independent and dependent variables or else graphs of their distribution. The idea here is to establish that there was enough variation so that you can proceed with your analysis. Again, remember the Michael Kearl example. If he had found that everyone in the general social survey supported euthanasia, that would be the end of the project.

Remember to use the "valid percent" for the frequency distributions. Let's look at discaff and race, for an example.

2. Then proceed to cross tabulations and their interpretation.

a. Be sure to put independent variables in columns.

b. Be sure to have SPSS calculate percentages within the columns

Do cross tab of race and discaff; then another controlling for sex.

3. . You can copy/paste from SPSS output into your Wordperfect or Word document, though you will need to do some cleaning up within the table.

a. Do an example at this point of both a frequency distribution and a cross tabulation, to illustrate how to do the cleaning up process.

b. When you control for a third variable, the table generated by SPSS will be very large, and you may need to create partial tables within your word processor (again, an example)

Copy/paste the race frequency distribution, the discaff by race cross tab, and the discaff by race controlling for sex. Use page layout/orientation to switch to landscape. Then use view/zoom to make the table smaller.

Here's an example from Michael Kearl again.

______________________________________________________________________________
Responses to variable EUTHANASIA revealed that individuals were more than twice as likely to agree as opposed to disagree. Interestingly, although numerous professional books and journal articles on the subject detail considerable ambiguities and numerous moral shades of gray, only 5 percent of the public admitted not knowing where they stand on the issue.

Table 1: FREQUENCY DISTRIBUTION of EUTHANASIA

(Next you will want to look at the cross tabulation between your dependent variable and your primary independent variable(s). )

Example:

Table 2: Support for Euthanasia, by response to Religiosity I

Religiosity

In Table 2 it is evident that the more religious one is the less likely one supports euthanasia. Thus, for example, while only 50% of those with strong religiosity support euthanasia, 90% of those with no affiliation support euthanasia.

__________________________ _____________________________________________________________________

E. Summary and conclusions. Did you find what you expected, in terms of your research hypothesis? If so, why does that matter; what have you been able to show in this research? If you didn't get the results you hypothesized, do you have any ideas about why not? This is probably the place to note anything you would do differently if you repeated your research or anything that showed up serendipitously in your findings and might merit additional research.

F. References. Here you need to recreate the earlier list of references from your research proposal, supplemented by any you found Remember that references need to be listed in the American Sociological Association's format, as outlined in our Citations and References guide. This is not a complete bibliography of everything that's been written on the subject. You should reference only what you find that's relevant to your particular research problem.

____________________________________________________________________________
Kearl (continued)

References

Baird, Robert M., and Stuart E. Rosenbaum. 1989. Euthanasia: The Moral Issues . Buffalo, N.Y.: Prometheus Books.

Davis, James Allan and Tom W. Smith. 1994. General Social Surveys, 1972-1994. (Machine-readable data file). Principal Investigators, James A. Davis and Tom W. Smith. Produced by the National Opinion Research Center, Chicago. Tape distributed by the Roper Public Opinion Research Center, Storrs, CT. Micro diskette and codebook prepared and distributed by MicroCase Corporation, Bellevue, Washington.

Kearl, Michael. 1997. "The Research Paper." (WWW document).
URL http://www.trinity.edu/departments/soc_anthro/research.html.

Sherlock, Richard. 1986. "Selective Nontreatment of Newborns." Pp. 147-151 in Robert F. Weir (ed.), Ethical Issues in Death and Dying . New York: Columbia University Press.

Wood, John B. 1986-87. "The Birthday-Deathday Effect: Fact or Artifact?" Omega 17(4):321-326.

That's all, folks!