SOC 200--Fall, 2009
M. Senter
Reading: Social Statistics for a Diverse Society: Chapter 8, Chapter 9, Chapter 14 (pages 477-89 and 498-500), Chapter 10
Homework (on the web page): Practice on Sampling (Part B), Practice on Bivariate Analysis, Practice on Multivariate Analysis
Consult Web Page: Tables; Introduction to SPSS
The format of this exam will be the same as Exam #1 and #2--multiple-choice and short-answer. This exam focuses a great deal on "doing" analyses either by hand or from SPSS output that I will give you.
1. Be able to calculate the standard error of a sample proportion and to compute the margin of error (sampling error) once you have the standard error (at the 95% confidence level AND at the 99% confidence level).
2. Be able to construct a confidence interval for a proportion once you have a margin of error. Be able to interpret the confidence interval. What does it tell you about the population?
3. Be able to calculate the standard error of a mean from a sample and to compute the margin of error once you have the standard error (at the 95% confidence level AND at the 99% confidence level).
4. Be able to construct a confidence interval for a mean once you have a margin of error. Be able to interpret the confidence interval. What does it tell you about the population?
5. What are the three factors that affect the size (width) of the confidence interval? What can you do realistically do to reduce the margin of error (sampling error) in your probability sample?
6. Be able to distinguish between independent and dependent variables. Be able to create research hypotheses and null hypotheses from independent and dependent variables.
7. Be able to calculate the t-statistic when you have two mean scores and the standard error. Be able to calculate the degrees of freedom and to use the Distribution of t table (Appendix C in your textbook) to determine whether your research hypothesis or your null hypothesis is supported.
8. Be able to make sense of SPSS output (which I will give you) or to determine whether variances around two means are equal and to determine whether two means differ, given the significance levels of the t-statistics that are presented in SPSS.
9. Be able to calculate the confidence interval for a difference between two means once you have the means, the standard error of the difference in means, and the confidence level (95 percent or 99 percent). What does the confidence interval tell you?
10. Be able to make sense of Analysis of Variance SPSS output (which I will give you) to determine whether three or more means differ. Be able to interpret the F test and its significance level to determine whether our research hypothesis or null hypothesis is supported.
11. Be able to use the Distribution of F table (Appendix E) to determine whether a F-statistic is statistically significant.
12. Be able to create, percentagize, and make sense of a cross-tabulation to test a hypothesis. Be able to do this by hand and to make sense of SPSS output. (Make sure that you can distinguish between an independent and a dependent variable.)
13. Make sure you know when to RECODE variables as appropriate to the variable you are analyzing. Review the material on levels of measurement--nominal, ordinal, and interval-ratio. Recognize that we will be viewing dichotomous variables as nominal variables.
14. Be able to calculate and make sense of a percentage-point difference (epsilon) from a cross-tabulation.
15. Be able to find the cell N's (count) when you have the percentages and the column totals.
16. Be able to read and interpret a multivariate crosstabulation. Know how to figure out whether an independent variable "matters" in a three-variable multivariate table. Know how to figure out whether a control variable "matters" in a three-variable table.
17. Be able to create a multivariate crosstabulation to determine whether a bivariate relationship is spurious. Be able to interpret your findings (in English).
18. Be able to create a multivariate hypothesis that includes an intervening control variable. Be able to create the crosstabulation that introduces the intervening control variable. Be able to interpret your findings (in English).
Other Concepts and Terms You Want to Understand:
level of measurement,
nominal
variables, ordinal variables, interval-ratio variables,
dichotomous variables,
population,
parameter, mu, pi, sigma, survey sample,
sample statistic, Y-bar, p, standard deviation, generalizing from a sample to the population,
inferential statistics, point estimation, interval estimation, population distribution, distribution of sample
observations, sampling distribution of the mean, sampling distribution of the
proportion, proportion, percentage, mean, standard error of a proportion, standard error of the mean,
appropriate z to find sampling error or margin of error, sampling
error or margin of error of a proportion, sampling error or margin of error of a
mean, confidence interval for a proportion, confidence interval for a mean, 95
percent confidence interval for a proportion and a mean, 99 percent confidence
interval for a proportion and a mean, 95 percent confidence level, 99%
confidence level, bivariate analysis,
independent variable, dependent variable, research hypothesis, null hypothesis,
alpha, know what sig. or
p < .05 means,
confidence level (1-sig. x 100), a "statistically significant" relationship,
t-statistic, degrees of freedom (df), significance level of t-statistic,
independent samples (sub-groups), equality of variances, F- statistic to test
equality of variances, significance level of F-statistic, one-tailed test,
two-tailed test, critical value in the Distribution of t table (Appendix C),
confidence interval for the difference in means, t to find sampling error or
margin of error for the difference in means, between groups sum of
squares, within groups sum of squares, total sum of squares, mean square
between, mean square within, degrees of freedom between (dfb), degrees of
freedom within (dfw), F-statistic associated with analysis of variance,
critical value in the Distribution of F table (Appendix E), bivariate tables,
crosstabulation, column variable, row variable, cell of a table, percentagizing a cross-tabulation to test a hypothesis,
epsilon, percentage-point difference, direction of
relationship, control variable,
intervening control variable, the elaboration model, spurious correlation,
partial tables, conditional relationship