Course
Summary
This
course introduces students to various ways that sociologists study society.
Studying several methodologies, students should discover some common
concepts and the logic underlying all empirical social research.
The focus of the course, however, is survey methodology (as the most
common way sociologists gather data) and quantitative research methods
(analyzing data by measuring it with numbers).
Such a focus requires, among other things, that students acquire some
basic understanding of probability sampling and some techniques for statistical
analysis.
I
will encourage you to learn by doing. I
prefer for you to view this as a "how to" course -- more a workshop
than a series of lectures. Note
that a primary objective is to afford students the opportunity to gain
first-hand knowledge of how sociologists actually go about doing quantitative research. In
fact, throughout the course we will do quite a bit of quantitative research
ourselves, taking each step together as a class.
By the end of the semester, we will have worked through the major steps
involved in carrying forth a professional research project by doing our own
survey and analyzing the results.
Students
sometimes harbor certain myths about courses that require statistical analysis:
that the course assignments will be impossibly hard, that they require
sophisticated mathematical skills, that the software will be difficult to
master. By design, this is not a
particularly difficult course, but it is necessarily a demanding one.
A lot of material must be absorbed in a relatively short time and there
are a considerable number of homework assignments.
The material each week builds on that from the previous weeks.
Therefore, it is crucial that you keep pace with your weekly assignments,
come prepared for class, and, above all, ask questions -- no student
learns well without asking questions. While it is true that we grapple with some difficult concepts
in this course, the math required is not difficult. Any necessary calculations are fairly uncomplicated, and you
are encouraged to use a calculator for homework assignments and tests.
We leave any complicated calculations to the computer.
Finally, current software programs make using computers for statistical
analysis easy and even fun.
Let
me emphasize immediately that I am committed to helping you succeed.
Contact me when questions or problems inevitably arise and, of course,
I’m always available during my office hours, or by appointment.
Course
Objectives
1.
To help students identify and understand some concepts and underlying
logic common to different research methodologies.
2.
To provide students with opportunities for gaining "hands-on"
experience in social research practices and a working knowledge of how
sociologists and other social scientists typically go about doing research,
especially quantitative research.
3.
To make students critical consumers of social research findings as
reported in professional journals and by mass media.
This involves, at least, understanding the presentation and
interpretation of tabular data and univariate statistics, and includes the
ability to identify utter nonsense (of which there is plenty out there).
4.
To cultivate the ability of students to "think methodologically,”
including the ability to translate simple observations into testable hypotheses.
5.
To provide students with basic skills for doing computer analyses.
6.
To give students a firm foundation for more advanced course work in
research methods and statistics in the social sciences.
Course
Requirements
Students
are responsible for weekly reading and homework assignments.
In addition to the homework there will be two semester exams and a final
exam. Students are also required to
provide their own diskettes for assignments involving the computer; these
personal diskettes may be filed in the computer lab, Leggett 606.
Texts
Required
text:
Moore, David, Statistics: Concepts
and Controversies (5th
ed.)
Suggested
text, on reserve:
Schroeder, Larry D. et al., Understanding
Regression Analysis
Additional
texts on reserve:
Clegg, Frances, Simple Statistics
Huff, Darrell, How to Lie with Statistics
Johnson, Allan G., Social Statistics Without Tears
Randolph-Macon
Social Survey (RMSS)
The RMSS is
a questionnaire designed to give students hands-on experience with social
research. Conducted by this class
every other fall, the survey collects information on Randolph-Macon students.
Rather than merely studying survey instruments and techniques, you will
actually participate directly by conducting a real survey.
Rather than writing hypothetical and arbitrary survey questions, the
class will produce some survey items on a topic of their choosing for inclusion
on the RMSS. Rather than merely
analyze "canned" or existing data about groups holding little interest
for the class, students will learn how to analyze data using information about
their peers. Since the RMSS asks
some basic questions every year, data collected each fall may be compared with
data from previous years. This
aspect makes RMSS data particularly useful for discovering trends in the
opinions or views of Randolph-Macon students as they change over time. Finally, the RMSS data sets are readily available to
sociology and anthropology majors for senior papers and independent study
projects.
Homework
Policy
Unless
otherwise indicated, all homework assignments are due at the start of the
following class period and should be completed by each student individually
without the aid of others. Late
homework assignments will be accepted for half-credit until one week after they
are due; after one week, late homework assignments receive a zero.
Grading
The
combined homework scores, two semester exams, and the final exam will each count
equally (25%) toward your course grade. The
grading scale follows:
93
- 100 = A
90
- 92 = A-
87
- 89 = B+
83
- 86 = B
80
- 82 = B-
77
- 79 = C+
73
- 76 = C
70
- 72 = C-
67
- 69 = D+
60
- 66 = D
0
- 59 = F
In
borderline cases, your class participation will determine the course grade.
Course
Outline
I.
Studying the Social World
A.
Learning to study society: why
bother?
B.
Approaches to social research
C.
The field of statistics
II.
Stocking Your Toolbox: Statistical
Jargon on Parade
III.
The Logic of Social Science Research
A.
Discovery: premises and assumptions
B.
Traditional model of scientific inquiry
1.
theory
2.
hypotheses
3.
operationalization: how do
you measure that?
4.
observation
C.
Two systems of logic
1.
deductive: theory then research
2.
inductive: research then theory
IV.
Methods for Collecting Data
A.
Experiments
B.
Field research
C.
Available data
D.
Surveys
E.
Some ethical issues
V.
Survey Research Methods
A.
Collecting and coding data: getting our feet wet
1.
survey instruments
a.
writing survey questions: do's and don'ts
b.
some tricks of the trade
2.
samples and sampling
a.
overview
b.
the SRS - a special sample
c.
common sampling techniques
3.
survey formats
a.
personal interviews
b.
telephone surveys
c.
mail questionnaires
d. e-mail questionnaires: wave
of the future?
4.
from survey to coding
a.
coding decisions
B.
Creating and processing computer files: up to our necks
1.
file building and editing
a.
storing and cleaning data
b.
analyzing data
2.
tailoring output
Exam
#1
VI.
Organizing Data
A.
Tables
B.
Graphs
1.
line graphs
2.
bar graphs
3.
frequency histograms
4.
scatterplots
C.
Lying with statistics, tables, and graphs
D.
Distribution curves: "frequency histograms with
fancy dresses
on"
VII.
Univariate Statistics and Normal Curves
A.
Univariate statistics: describing characteristics of
samples
1.
central tendency: statistical
measures of center
a. mean: an "average" statistic
b. median
c. mode
d. comparing the averages
2.
variability: statistical measures of spread
a. percentiles and the interquartile range
b. variance
c. standard deviation
B.
Normal curves and distributions: toward making
inferences
1.
the properties of normal curves
2.
sampling distributions: a pivotal concept
3.
the standard normal curve and standardized scores
Exam
#2
VIII. Analysis of
Bivariate Relationships: Discrete and Continuous Data
A.
Contingency tables (discrete data)
1.
2 x 2 tables
a. one-way vs. two-way tabulation
("cross-tabulation")
b. percentage differences and Yule's Q
(dichotomous)
c. direct vs. inverse relationships
2.
r x c tables
a. nominal: measures of association based on
chi-square (c2)
b. ordinal: gamma
B.
Correlation (continuous data)
1.
"positive" vs. "negative" associations
2.
graphing associations: scatterplots and r
3.
The correlation coefficient (r)
4.
r2 as "explained
variance"
5.
computing r and r2
C.
Regression (continuous data)
1.
The regression coefficient (b)
2.
predicting Y from X: the linear formula (Y = a + bX)
3.
which is the regression line?
4.
computing b
5.
interpreting the regression coefficients, "a" and "b"
IX.
Analysis of Multivariate Relationships: Introducing Control Variables
A.
The concept of "control": experimental vs. statistical
B.
Contingency table analysis (discrete)
1.
elaborating the bivariate case: adding control variables
to
"test" statistical relationships
2.
"real" vs. "spurious" relationships
C.
Multiple regression analysis (continuous)
1.
elaborating the bivariate case:
adding control variables
to the equation
2. multiple
R2
3. standardized
regression coefficients
4. examples
of multiple regression analysis from professional
journals
X.
Statistical Inference and Significance
A.
The underlying logic: probability
1.
creating hypotheses for the population
2.
using sample results to test the plausibility
of hypotheses
B.
Significance tests
1.
the null hypothesis
2.
distributions for tests of statistical significance
a. normal distributions
b. c2 distributions
c.
t distributions
3.
the critical region
4.
type
I and type II errors
Final
examination
