SOC*2120: Introductory Research Methods
Instructor: J. I. (Hans) Bakker
Office Hour: Wednesdays 4:00-4:50 in Johnston Hall
Room 115, Extension 53545
Fall, 2005 Tentative Syllabus
Wednesday evenings 7:00-9:50 in Axelrod 100
Goals:
This is an introduction to both qualitative and
quantitative research in sociology. It is merely an introduction. But
you will learn some important foundational concepts.
In the more advanced courses you will learn more
about the techniques of quantitative (Qt) and qualitative (Ql) methods.
But here you will learn some of the basics of: (1) foundations of doing
research, (2) planning and preparation, (3) data collection, (4) data
analysis, and (5) reporting on your findings.
No one will be a sociological researcher merely
as a result of taking this introductory course, but you will be on the
path to a more sophisticated understanding. At the very least you should
be able to read scholarly refereed journal articles and know what research
Methodology and method/technique has been employed in any specific research
study. For example, most Qt research in sociology is based on survey
research and the analysis of Qt data generated by such surveys. On the
other hand, most Ql research in sociology is generated by some form
of Afield research@ or ethnography, and the Ql Adata@ or information generated by such
ethnographies. Note that the mere presence of numbers (e.g. percentages,
measures of association) does not necessarily mean that the study is
fully Qt in terms of research design, etc. Much of what you learn here
will also be relevant in the other social sciences, especially anthropology.
Texts: There is one required texts. Neuman
(2003) will also be available as a used book in the fourth edition.
(A sixth edition is expected in January.) There is no research project
required for this course, but you may find the use of Research Navigator
useful for your other courses.
Suggested Future Reading (helpful for term
papers in other courses; will be referred to from time to time in lecture
as Jacoby): It is shrink wrapped with the text.
Recommended Texts (widely available)
-
Bakker, J. I. (Hans) 1993. Toward A Just
Civilization. Toronto: Canadian Scholars' Press. (This book
will be useful for the Critical Book Review option.)
-
Delaney, Tim. 2005. Contemporary Social
Theory: Investigation and Application. Upper Saddle River,
N.J.: Pearson/Prentice-Hall.
-
Macionis, John J. and Linda M. Gerber. 2005. Sociology: Fifth Canadian Edition. Toronto: Prentice-Hall.
[Widely used in Introductory Sociology. It is very helpful to review
Part I. The terminology is a bit different. You can also refer to
the fourth or earlier editions.]
Grades: Your final grade will be calculated
as a result of three exams. The Mid-Term will be in two parts (on two
separate days).
-
Exam I (23.33%) True-False & Multiple-Choice
(60 questions, in class)
-
Exam II Part I (33.33%) same, (This is the "Mid-Term")
-
Exam II Part II (10%) One Quantitative Problem
-
Exam III (33.33%) same, plus one Qt question
and one ESSAY (worth one third of the exam, Short Definitional Essay
[SDE] 500 words) = FINAL EXAM
There will not be any other essay requirement in
this course. There will not be a research study or research report.
It would be too early in the process of learning about research for
most students to do a reasonably good job at actually conducting a significant
research study, even if such a study were merely a mock study. Moreover,
the process of getting Ethics-Human Subjects (IRB) approval has become
very cumbersome.
However, if you miss Exam I or Exam II for any reason
you can make it up by writing a Critical Book Review of any book
(or set of four articles) referenced in the text. . (That is, your CBR
could also be of four very current research articles in refereed journals
not explicitly mentioned in the text.) One other option is to write
a CBR of Bakker's book on Gandhi's social theory and its relevanceThe
CBR is not just a book report but must analyze a conceptual problem
and have a Athesis statement@ that is defended ( 9- 10 pages, no more than 12 pages or 3500 words,
including all apparatus, e.g. bibliographical references).
T.A. Assistance: The TAs will be available
to assist you to interpret the text and lectures. There will not be
seminars to which you are assigned, but we may run a few seminars during
the day to help those who feel they need it. The TAs will help you to
understand where you may have gone wrong on Exam I or Exam II. He will
hold regular office hours and be available through email. Please direct
your questions to Sanchari Quader or Robert Pontsioen in the first instance.
See them during their office hours or send them an email. Only after having consulted Sanchari or Robert should you come to see the professor,
Hans, during office hours (Wed. at 4:00). Please do not send unnecessary
emails.
Office Hours Wednesday : 4:00-4:50 in Johnston
Hall Room 115. Please do not email me unless strictly necessary.
Call extension 53545 and leave a message if you were not able to communicate
after class. Also, if you drop notes in my mail box put them on full
size pieces of paper. (Scraps of paper get lost very easily.)
Web: The web has many, many resources available
on the classical and contemporary authors we will be briefly discussing
in lecture. Moreover, each of the topics has a great deal of material
that is worth looking at through refereed academic journals. But please
try to avoid the Aspam@ material that also appears in
the form of popular opinion, unless you are able to take a certain critical
distance from such writing. For example, popular press items are not
the best source of information. Jacoby and Barr (2004) may be useful
to you in other courses where you have writing assignments.
Tentative Schedule of Classes (Syllabus in
the narrow sense): This may vary due to all kinds of things. The key
to learning is to read the text. Skim the whole text first. Get a feel
for it. Then read a chapter that really grabs you right away. Try to
read the designated chapter BEFORE the lectures and then review it again
after the lectures.
Week One: Experimental Research
Sept. 14: Introduction to the Class; Syllabus handed
out and choice of text justified.
07: Introduction to Experimental Research & Logic of Experimental Design.
Introductory Comments on ADimensions
of Research@
Basic versus Applied
Exploratory, Descriptive and Explanatory
The Time Dimension (cross-section versus
longitudinal/time-series-panel, cohort/ case study/ historical
Read: Neuman 2003: Chapter 9 [Hereafter: N: C9:
237-262]
N: C2: 20-40
Skim N: C6: 139-145. Pay special attention to Table
6.1 p. 145
Note that Qt tends to be Positivistic and Ql tends to be Interpretive; but, the mere presence of numbers
does not make a study Positivistic. (It is an association and not a
perfect fit.)
All "science" (Latin scientia, knowledge)
is based on what is often called Athe
scientific method@.
However, not all "scientists" agree on what the scientific method is.
A physicist who has not studied the philosophy of science may not know
as much about the epistemology of science as a philosopher who HAS studied
the philosophy of science. A sociologist may not be a physical scientist
and yet may have a more sophisticated understanding of the history of
science than a chemist.
One version of the scientific method that is widely
accepted, however, is the logic of experimental design. The problem
with experimental design in the social, cultural, historical, human
sciences is not Aepistemological@ but practical. It is difficult to carry out an experiment on human beings
without violating basic ethical principles. In medicine it is done all
the time on the basis of the notion that only a true experiment will
provide a valid and reliable test of the effectiveness of a medication
( a stimulus ) on the health of patients (a response).
The logic of experimental design is a "method" that
is central to the Positivistic or "Positive" approach, which
is one that often stresses the use of quantitative variables.
However, ingenious ways have been invented to circumvent the problem
of not really being able to do an experiment. Hence, in Positivistic
(or Positive) survey research there is much attention paid to
how an Aindependent
variable@ ( similar to the stimulus or medication
) will affect a Adependent variable@ ( similar to the effectiveness
of a medication ). The IV is said to be correlated with the DV because, in the absence of a true experimental design, we cannot determine
any kind of causation. Correlation does not equal causation! (Also, "association" does not equal causation.) There is a Aco-relationship@.
But other variables ( Aintervening variables@ ) may have been the true cause.
(Think of storks, babies and season of the year.)
When we do not have good measurement of variables
at the ratio or interval level of measurement then it
is even harder to claim that the use of statistical techniques helps
us to really overcome the problem of not doing a real experiment. Yet,
many researchers try to make generalizations based on measurement at
the ordinal level of measurement. We will learn a little bit
about the Chi squared statistic that is used when Pearson=s
Product-Moment Correlation Coefficient is not justified by the lack
of a ratio or interval level of measurement.
Note that at the nominal or ordinal level of measurement
we can only do a measure of "association." It is a co-relationship,
but it is NOT a "correlation" in the strict, technical sense. (The term
correlation is often used in a less strict manner, even in textbooks;
but, the point is very important for understanding the logic of methods,
Methodology.) Many social and physical scientists are not particularly
sophisticated about Methodology, although they may be very well
versed in the use of their own specific "methods." They are, strictly
speaking, method-ists and not Methodologists. (The Methodist
Church should not be confused with the "methodist cult"!)
The extreme "cult" is to think that it is only "real"
if you can measure it. But, on the other hand, many social scientists
reject all forms of measuremente. Many social scientists believe that all quantitative measurement of sociological variables can be
misleading. A famous statement of the misleading nature of the use of
quantitative variables comes from Herbert Blumer, the sociologist who
coined the label ASymbolic Interactionism@.
Blumer=s SI is one example of a research approach that rejects the use
of quantitative variables and the statistical techniques used to overcome
the lack of experimental design in human social sciences. There are
many others. Together they are often discussed as AInterpretive@ rather than APositivistic@ approaches to research. Blumer=s SI is a very well known type
of Interpretive research and sometimes the label ASI@ is used for all such approaches.
(That is, some sociologists will speak of SI when they mean the broader
concept of Interpretive approaches.)
One radically different Interpretive approach to
the study of sociological problems is the Comparative-Historical Sociological
(CHS) approach. (It is radically different from SI, but it is
also very different from most quantitative studies that are based purely
on responses by individuals to sample surveys, etc.) Rather than study Adata@ that is the product of questionnaire
surveys, the CHS will study information about societies. Usually, those
societies are conceptualized as Anation-states@ (e.g. France, UK, US, Russia, PRC). A CHS may compare two societies,
three societies, or more than three societies. But the comparisons are
usually stated in qualitative terms. (There may be some use of numbers,
e.g. percentage figures; but the underlying logic is not quantitative
and probably not Positivistic.) The excellent work done by Charles Ragin
on the logic of method (Methodology) was initially inspired by his desire
to establish better ways of doing CHS.
The underlying logic of the way in which research
will be presented in this course is that: (1.) we will start with some
aspects of the analysis of quantitative data and then (2.) move to
the analysis of qualitative data. Exam II (the Mid-Term) will cover
analysis of data (Qt and Ql ) in a general way. Then we will get into
the variations and expand our awareness of the different implications
of choosing to use either Qt or Ql Adata@.
Ultimately the sociological research must also have a good grasp of
sociological research theory in order to do good research and produce
valuable results. (Moreover, to understand research theory it is also
necessary to be able to link that to more abstract levels of theorizing
at the paradigmatic level of theorizing.) But since this is a course
on research methods we will not emphasize the theoretical questions
that go into Methodology, the logic of method, any more than is absolutely
necessary. Some AMethodology@ is necessary, however, in order to have more than simply a mechanical
and Acook
book@ understanding of Amethods@ in the sense of specific techniques.
Again, no one can become a competent research as
a result of taking just one course. But you can use this course to lay
a foundation that will help you to read the research literature and
that will prepare you for the third year courses on Qt and Ql methods.
Do not hesitate to ask questions during lectures.
If you do not understand what is being said then chances are other students
also do not understand. Moreover, unlike many Asubstantive@ courses in sociology (e.g. sociology of the family, sociology of crime,
sociology of delinquency, etc.) this course is somewhat cumulative.
That means that to understand what is being said at the end of twelve
weeks will require learning what is taught in the first weeks of the
course.
Week Two: Analysis of Quantitative Data
21: Analysis of Quantitative Data: One Variable
Read N: C12: 331-340
Examine Table 4.1 items 1 - 8 for Positivism and
Interpretivism.
[Do not worry about Critical, Feminist and Postmodern
at this time.]
We will get into the question of Alevels
of measurement@ in more detail later but it is important for now to know that there
are four levels of measurement which are generally recognized: nominal,
ordinal, interval and ratio. Other words are sometimes used to convey
the same levels.
21: Analysis of Quantitative Data; Two Variables
Read N: C12: 340-350
Week Three: Analysis of Quantitative Data
28: Analysis of Quantitative Data: Three + Variables
Read N: C12: 351-357
28: Analysis of Quantitative Data: Chi Squared as
an Inferential Statistics
Read N: C12: 358-362
Know each term in the glossary (e.g.
Type I & Type II Error)
Week Four: Exam I
Oct. 5: Exam I (23.33% of total grade; T-F &
M-C; no Qt. Calculations )
Oct. 5: Further Clarification of Logic
of Experimental Research Design
Week Five: Field Research (Ethnography) &
Analysis of Qualitative Data
Oct.12: Logic of Field Research versus Logic of
Experimental Design
Read N:C13: 363-401
12: Analysis of Qualitative Data Read
N: C15: 438-448
Week Six: Analysis of Qualitative Data
19: Analytic Strategies for Qualitative Data Read
N: C15: 447-467
Know Each Term in the Glossary N: C15: 466
19: Comparative-Historical Sociological Research
Design and
Bakker=s Expanded View of the importance
of Ideal Types and
Ideal Type Models (ITMs) Read N:C14: 402-413 &
N:C15: 456-467.
Week Seven: Exam II (Mid-Term)
26: Exam I Part I (33.3%) (T-F, M-C)
26: Exam I Part II (10.0%)(one Qt. Problem)
Week Eight: Quantitative-Experimental versus
Qualitative-Ethnographic
Research Designs
Nov. 02: Comparing the Qt & Ql Research Designs
Read N:C6: 139-167
Know the Terms in the Glossary on p. 167
02: The AMeaning@ of Positivist versus Interpretivist Social Science
Read N: C4: 68-80
{We will not tackle the CSS, Feminist and
Postmodern, yet!}
Week Nine: Measurement
09: Reliability and Validity Read N:C7: 178-188
09: Levels of Measurement in Quantitative-Positivistic
Research
Using Variables Read N:C7: 188
Week Ten: Scales and Indices & Survey Research
16: Scales: Likert, Thurstone, Bogardus, Semantic
Differential
Read N: C7: 195-209
16: Survey Research Questionnaires N:C10: 267-288
Week Eleven: Sampling & Survey Research
23: Nonprobability & Probability Sampling N:
C8: 210-236
23: Survey Research N: C10: 263-267, 289-307
Week Twelve: Closing the Theory-Research Gap!
30: Theory and Research Methodology and Methods/Techniques
Read N: C1: 1 - 19 & N: C3: 41-67
30: The AMeanings@ of Methodology as opposed to Methods/Techniques
Read N: C4: 68-94
What did we skip? We did not look at the details
of Chapters 5, 11, & 16. But you should skim those chapters in preparation
for the Final Exam.
Week Thirteen: Exams
Dec 01: No class
Dec 5 - 19: Final Exam December 14 2005 8:30-10:30
probably in the Gym.
Format of Exams: T-F, M-C, two Qt problems,
one Short Definitional Essay (500 words).
All Exams will have a number of True or False questions.
Remember that a statement is only true if it is entirely true. Thus,
for example, it is common for those who wish to make a rhetorical point
to make a number of true statements before they make the statement which
they want to convince others is true.
True or False?
____1. Canada is a very large country geographically,
although there are regions where very few people live. Much of the urban
core of Canada is located between Windsor and Quebec City, although
there are also a number of other major cities such as Vancouver. The
Prime Minister of Canada was Jean Chretien but is now Paul Martin. The
capital of Canada is Toronto.
You will immediately recognize that Toronto is not
the capital of Canada. But that requires a certain amount of expertise.
As a person living in Canada you are well acquainted with the capital
region being Ottawa-Hull. But many Americans or Europeans would not
necessarily know that. Is this a fair question? It is if you know some
very basic facts about Canada.
Multiple-Choice
____2. The theory of stratification concerns inequality.
However, when we use the term Astratification@ it tends to imply that the theoretical approach ( paradigm ) is in the
Neo-Durkheimian Structural-Functionalist tradition ( research paradigm
). Such an approach would definitely not include:
a. Marxian and Neo-Marxian analysis of economic
class (e.g. Erik Olin Wright)
b. Weberian and Neo-Weberian analysis of class,
status and power ( e.g. A. Giddens)
c. Feminist analyses of the impact of class on gender/sex
d. All of the above are not, strictly speaking,
S-F approaches to Astratification@ in the technical sense ( stratification as a technical term, tt ).
e. a, b and c do not have to be referred to as contributions
to the study of inequality since they all clearly deal with the existence
of strata in society.
The correct answer is Ad@. But knowing that the term Astratification@ if often read to imply a Neo-Durkheimian Structural-Functional research
paradigm ( Pr ) is a matter of detailed knowledge, a bit like knowing
the capital city of Canada is not Toronto.
Appendix I: A Note on Methodology versus Methods/methods/techniques
We frequently use the word "methodology" in a more
limited sense to mean methods and techniques of research. This course
is not just about methods in the limited sense. We will also look at
the "logic of method" or Methodology (Mg). The Methodological decisions
we make directly influence the methods and techniques we can use. If
we decide on a Methodology that calls for other techniques then it would
be inappropriate to use techniques which are not designed for that Methodology.
Moreover, our Methodology (Mg) is often closely aligned with our Theory-in-general
(Tg). There are basically three Methodological-Theoretical (Mg-Tg) Approaches:
Positivistic, Interpretivistic and Criticalistic. (Some people would
add Postmodernistic as a fourth. Others might add Feministic as a fifth.)
The simplest example is the Positivist (Positivistic)
Methodology (Mg) and "variable analysis." If we have a Positivistic
Theory (Tg) and a Positivistic Methodology (Mg) then we can use the
analysis of variables and quantitative techniques of correlation or
association. But such work requires that we START with our hypotheses
(deduced from our Research Theory, Tr). If we do not already have hypotheses
formulated then we cannot test hypotheses and therefore we really cannot
apply quantitative techniques of analysis without violating fundamental
assumptions. (It is a bit like tossing a coin and THEN, when we have
seen it on the ground, calling heads or tails!) We may choose to consciously
violate the assumptions, but when we do so we also have to move away
from a Positivistic Approach (Tg, Mg) to a modified approach we might
call "Positive." Most Introductory textbooks do not make this very clear.
Neuman 2003: 68-94) is clearer than most. (See able 4.1 on p. 91.)
