Ja ary 6, 2014
What is a Theory?
A system of ideas that are used to explain the causes and/or consequences of social phenomena eg.
Mead’s symbolic interactionism (we understand ourselves based on how we believe others perceive us),
Weber’s rationalization theory (why we saw the rise of the bureaucracy and why it got spread across
society), Durkheim’s suicide theory.
3 Main Components of a Theory:
1. Definitions: what do the key terms in a theory mean?
Eg. Durkheim’s suicide
Suicide: purposefully ending one’s own life.
Also define social integration, anomie etc.
2. Descriptions of the Phenomena of Interest:
Ex. The suicide rate in Canada is about .15/1000 population, which is about average for
OECD countries. This rate varies by gender, age, and visible minority status, with men,
1519 years old and persons of Aboriginal descent having the highest rates etc.
3. Relational Statements
Theories suggest ways that concepts are related to one another
If you know something about one, you know something about the other.
As social integration increases, the suicide rate decreases
As age increases, social integration increases.
2 Basic Types:
Deterministic: two concepts or variables that always go together
As social integration increases, the suicide rate decreases.
This is an “every time” kind of thing.
Probabilistic: they regularly, but don’t always go together
Use terms like more likely, less likely…make reference to the odds of one thing
happening, given the other.
Suicide becomes less likely as social integration increases.
* Most common in social science research. \
‘Size’ of a Theory
Grand theories vs. Middle range theories
All encompassing with regards to time and space
Ex. Structural functionalism, symbolic interactionalism, critical theory, post
structuralism, post modernism.
Tends to be difficult to link with the real world in a testable way.
Very useful as a way of seeing the world, but not very useful for directly
Middle Range theories (Merton): More limited in scope and less abstract – tend to refer to a more specific
Ex. Durkheim’s Suicide
Ex. Merton’s Anomie theory
When people are denied legitimate means to achieve a status, we tend to use
illegitimate means to achieve that same status.
What is Research?
a mixture of observations and interpretations that either –
Sheds light on an existing theory, or
Helps us to build new ones
How do we move between theory and research?
Theory > observations/ findings
Observations/ findings > theory.
Relationship between Theory and Research
Steps to deductive research:
2. Hypothesis – testable statements about your theory “if this theory is true, which of
these situations is true?”
3. Data collection
5. Hypotheses confirmed or rejected
6. Possible revision of theory
Steps to inductive research:
1. Gather data – often ‘loose’ in subject and very detailed
2. Make statements or generalizations about data
3. Derive explanatory theory from these statements
Also called grounded theory – starts ‘on the ground’ with observations.
Based on how the world looks we develop a theory.
Deductive more common than inductive
Almost never just one or the other
Going back and forth called iterative research process
Very commonly used.
Lecture 2 January 13, 2014
What is knowledge?
How do we come to ‘know’ something?
The philosophy of knowledge How do we ‘know’ that we ‘know’ something?
In the social sciences:
What kind of knowledge is appropriate to seek?
How may we best seek it?
Positivism and Interpretivism (both classical positions)
Knowledge in the social sciences should be gathered in the same way as in the natural sciences;
science is the way that we should find out about people.
Empirical (evidence based on the senses)
There are social laws and principles just like natural ones.
Social laws eg. laws like gravity except for humans it’s thoughts and beh’rs.
We can use deduction to find support for these laws.
Less often, we can use induction to discover new laws.
Eg. Comte’s Law of 3 Stages
Came up with the word positivism. All human societies, no
matter where they are located, go through 3 stages:1 theological,
2 metaphysical, 3 positivist stage
Eg. Marx’s Historical Materialism.
You can predict that the way that society will change based on
the relationships with resources. Communism is an inevitable
product of material relationships.
Big difference betw’n theory and research
It’s possible to collect observations without any reference to preexisting theories and to
develop new theories purely based on those observations.
Social science can (and should be) valuefree
Terms objectivity, intersubjectivity (between us, we agree)
Normative Statements are not scientific
Are certain acts or social conditions morally acceptable?
Place of religion or philosophy to say
Can’t be empirically tested
Studying people and social life is fundamentally different than subject matter in the natural
People act based on their own interpretations of the symbolic meaning of a situation.
You as a human are not reacting to the situation; you are acting to how you feel about the
Thomas’ theorem: situations perceived as real become real in their consequences. It
doesn’t matter what the situation is, it matters how we react to it.
Mead, Weber, and ‘symbolic interactionism’.
Mead: We interact based on the symbols that people give to us
Weber: Empathetic understanding.
Social scientists should be attempting to:
Understand the meaning people attach to their environment and their actions.
See things from the point of view of the people involved, this is the opposite from the
Goffman and ‘dramaturgy’ People believe that they have different roles (public and private) and
studied on how these roles affect the way that they interpret different
What is knowledge?
Positivism: explanation of social beh’r
Interpretivism: empathetic understanding (Verstehen) of social beh’r.
Critique of Positivism
Can social science really discover social laws?
Can it really be valuefree?
Is positivism still dominant in the natural sciences?
Critique of Interpretivism
Social scientists’ interpretation of the actor’s interpretation.
The actors involved are subject to forces they don’t identify or know about.
What is the nature of being? Of reality? Of existence? What is social reality made out of? This is
In social sciences:
What is social reality? What is its nature?
2 major debates
Do social phenomena have an objective reality independent of our perceptions?
Is social reality merely a set of mental constructions?
3 Major Positions:
Yes, there is an objective social reality
Social scientists should be discovering and explaining it.
There is no objective reality against which our conceptions and
views of the world may be tested
There are no facts, only interpretations.
There is an objective social reality, but it may not reflect our
ideas about it at all
What we see is different than what’s really out there) we can’t
even access anything outside of our perceptions.
Structure vs. Agency debate (comes up a lot in feminism research & debate)
The social world is ‘out there’ something that we act within, but have
little control over
Durkheim’s social facts
The social world is created and recreated out of our everyday actions.
We act within our gender roles; they are choices that we make for
ourselves. The sums of the choices that we make are the sum of the
reality that we live by.
So Far: Deductive and Inductive Research
Positivist and Interpretivist approaches to knowledge
Objectivist and Constructionist views of social reality
Can group them to distinguish between Quantitative and Qualitative methodologies.
Qualitative vs. Quantitative
Mostly use words as description, you don’t need to use the senses
Try to determine how people interpret their social world
See social reality as an ‘emergent’ (always just coming up) property of peoples’
Mostly use numbers as descriptions
Theory testing and observation important
Natural science/positivist model
Social reality is external to the people in it (objective) we are all experiencing social
reality but we are not in social reality.
Values: Personal beliefs and morals, feelings, preconceptions.
Can we be valuefree in social research? Should we be?
1) Classical positivists
Valuefree, neutral, unbiased research is both wanted and possible.
Increasingly rare attitude.
2) Social research cannot be completely valuefree
But we should try.
We should practice reflexivity
Understanding how you as a researcher are affected by your values, and how
they impact your social research.
3) Valueladen research can be good
Sometimes certain values are necessary to an understanding of the research process, and
especially those under study.
Much less common way of thinking.
Researchers sometimes ‘take sides’
Funding: Who gets it? What strings are attached?
Research subjects: Who gets access? Under what conditions?
Research findings: What is ‘acceptable’ for publication/dissemination?
Your Research Question:
States the purpose of a study, in the form of a question
E.g. To what extend does coital frequency impact how intensely traditional gender roles
are played out in the home?
Guides your choice of research design
Often start’s with a general area of interest and a very general question
If deductive method Question becomes steadily narrower
Becomes a hypothesis linked to the theory
Characteristics of Good Research Questions:
Should be researchable
There should be a way to answer your research question
Should have a clear focus and set realistic boundaries
Should be possible in terms of time and money
Should relate to existing theory and research in the area (and add something too it)
What is a research design?
A framework for the collection and analysis of data
Both quantitative and/or qualitative research designs
Nature of the research question is usually important to choice of design
Research ethics are important
Choice of design depends first, on kind of explanation you want
Two basic traditions
Nomothetic Design > quantitative
Attributions of cause and effect, expressed in terms of broad
Eg. The rate of suicide in a society is a function of the average
level of social integration in it.
Applied to people who were not actually in the study as well as those
who were (representative sample).
Cause and effect two terms in Nomothetic research
Independent variable the proposed cause
Dependent variable proposed effect
Ideographhic Design > qualitative
A rich description of a person or group
Cause and effect relations are focused on interpretation and
Usually meant to apply only to the person/people/group actually under
Criteria for Evaluating Social Research
In choosing a research design, 3 main types of criteria need to be considered:
Anyone should be able to get the results you got, using the same methods and samples.
Should also work on different but similar samples
If not, may be evidence of bias, inappropriate design, poor question, poor
Very important principle in science more generally
Is the measure/method consistent?
Classic example ‘The bathroom scale’
Say Professor Maynard steps on the bathroom scale 5 times, and I get: 115lbs,
100lbs, 125lbs, 110lbs, 130lbs all close to my real weight
Not reliable because it’s different each time.
So I get another scale and step on it 5 times: 325, 325, 325, 325, 325lbs. Reliable (even though it’s clearly wrong!)
Consistently measured. Even if it’s not valid, if it’s consistent
Same criteria used to evaluate existing social research
These mostly apply to Quantitative methods.
3 Main Types:
Measurement (construct) validity:
Are you actually measuring what you think that you are measuring?
In the bathroom scale example, neither is ‘valid’
One is unstable, and the other is not measuring her
Classic example here is IQ A reliable measure but does it
really measure intelligence?
Has causality actually been established?
Is it really in the direction you’ve suggested? Are the
independent variables you’ve suggested the right ones?
1) Are the findings applicable to situations outside the research
Naturalistic studies are best at satisfying this.
2) Can the findings be generalized beyond the people or cases studied?
Studies that use representative samples from a population can
generalize to that population?
For Qualitative Methods only:
Criterion of Trustworthiness:
1) Credibility: how believable are the findings?
Parallels measurement and internal validity
2) Transferability: do the findings apply to other people in other situations?
Parallels external validity.
3) Dependability: Are the findings consistent over time (where we expect them to be)?
4) Confirmability: Would another researcher reach the same conclusions?
The traditional means of ‘doing’ science
Are in sociology (more common in psych) because:
Many variables of interest are not subject to experimental manipulation
E.g. Gender, age, ethnicity, income etc.
Ethical concerns preclude performing experiments
Many phenomena of interest have longterm, complex causes that cannot be simulated
Lecture 3 MISSED January 20, 2014 Lecture 4 January 27, 2014
50 MC’s (no multiple multiples, only a > d) and true & false (about 15)
3 short answer (choose from 5)
Less than a page each
No point form but keep it short and simple
1.5 hrs to complete
Emphasis will be on what was covered in class today
Continuation of Chapter 3
Nature of Quantitative Research:
4 main goals of quantitative research:
Systematic measurement (very specific, exact process), establishing causality (causes
and consequences), generalization (doing the research to learn more about people in
11 basic steps in doing quantitative research
Finish 11 Basic Steps:
Step 9 Analysis
Big Step! Covered in some detail later in this course and in stats 2205
Choose statistical techniques and test relationships between variables
Check for validity and reliability issues
Always concerned with the consistency of measure of a concept, yet several definitions:
(1) Stability over time:
Consistency in a single measure over time (assuming no change in what is being
E.g. bathroom scale example
(2) Internal Reliability
Consistency in measurement when using multiple indicators to measure the same
concept at a single point in time
How to test?
Cronbach’s alpha coefficient > .80 (NOT USED IN THIS CLASS)
Test each thing to see if they are all related to an underlying concept.
Randomly split them in half and do some similar sort of test on each
(3) Interobserver Reliability
Consistency in measurement across researchers
E.g. coding open ended questions in a questionnaire or coding in content
Everyone has to be doing the same things when interviewing/observing and have the
same ideas Measurement validity
(1) Face validity: does this measure appear ‘on the face of it’ to be valid?
Soft of about common sense
Do other researchers agree?
Much easier to establish with more straight forward concepts
(2) Concurrent validity: does the measure correlate to another measure that is also relevant to the
Ex. Research on ‘job satisfaction’
Ask people to rate how satisfied they are with their jobs (our measure)
Find a measure related to job satisfaction – maybe absenteeism
Do workers who score low on absenteeism score high on job satisfaction?
If yes, concurrent validity!
(3) Construct validity: concepts relate to each other in a way that is consistent with the
Ex. Theory states that ‘poor parenting’ causes child ‘hyperactivity’
We see that ‘hyperactivity’ increases, as parenting ‘gets worse’
We thus suggest that the measures used to gauge ‘poor parenting’ and
‘hyperactivity’ are valid
Problem with this is that we are checking our measurements with our theory
and establishing our theory with our measurements
Be careful with this!!!!
It might not be nonvalid measures; it might be real difference between theory
(4) Convergent validity: a measure of a concept correlates with a second measure of the same
concept that uses a different measurement technique
Two completely different measures for the exact same technique
Survey based measure of income (asks respondents how much they make)
Information from Revenue Canada on income
If averages are close, measures are considered valid
Steps 10 & 11
Step 10: Findings/ conclusions
Is your research question answered or your hypothesis supported?
Are there implications for theory, for social life or social policy?
Step 11: Writeup findings/conclusions
Way easier to do with quantitative rather than qualitative research
Qualitative: you’re never really sure when to stop
Publish it and let others judge the quality/usefulness of your work, potentially replicate it etc
Critique of Quantitative Methods (Generally)
Will be posted online ** Reality and Practice
The ideals of quantitative research and how it is actually conducted may be very different.
Less attention may be paid to matter of reliability and validity than what one may think.
Some of that comes about through practical limitations of time, cost and feasibility.
Chapter 4 Survey Research
By far the most common data collection technique
Want to know? Ask people!
With representative samples
Contain many closedended questions
Types of Survey Research
Two Basic Types:
Structured interviews (face to face or phone)
Face to face
Researcher is present during the interview
Hard copy of questions
Created well in advance
Concept of Standardization
Reducing Interviewer Error & NonResponse
Everything done the same way each time
In both the asking of questions and the recording of answers
Interview schedule: a formal list of questions that the interviewer must follow in detail.
The questions must be asked in exactly the same way each time
Must also be asked in the order given—the same order each time.
In structured interviews, standardization can address error sources 2 & 5 by:
(1) Reducing interviewer variability
Intrainterviewer variability: interviewer is not consistent in asking questions
or recording answers
Can happen during 1 interview or across many
Interinterviewer variability: multiple interviewers are inconsistent compared
to each other
Can be stopped before it’s started
(2) Enhancing accuracy and ease of data processing
By using more closedended questions, and taking care with question order. Question Wording and Order
The question should be asked exactly as stated
Small differences in wording can make a large difference in meaning
Answers should be recorded as exactly as possible
To avoid ‘interpreting’ by the interviewer
Questions should be asked in the order they are given on the interview schedule
Previous questions make us think differently about the current one
Eg. Question wording *Posted on OWL*
Eg. Question order *Posted on OWL*
Part of source #7
The characteristics of the interviewer may influence the responses given
Sex, race/ethnicity, class, size, age, apparent sexual orientation, immigrant status etc.
Two Basic Types:
(1) Item nonresponse: respondents won’t or can’t answer specific questions
(2) Survey nonresponse: potential respondents refuse to participate at all
Reducing item nonresponse
Interviewers should try to develop a rapport with respondents
Trustworthiness, friendliness, professionalism
Must be consistent within and across interviews (interviewer
Must not overdue it
Ex. Being too friendly may bias responses (social desirability)
If respondent does not answer a question at all, or as fully as hoped, the interviewer may
Tend to be standardized
For closedended questions
May simply repeat the question or answer choices
For openended questions
Gently push for more info
Ex. ‘Can you give me an example?’ ‘What do you
think are the reasons for that?’
Probing is NOT prompting
Occurs if the interviewer suggests a specific answer for the respondent
Considered very inappropriate
Better to record ‘no response’
Easier to avoid with closedended questions
Show cards help even more
For closedended questions, possible answers are printed and given to
Face to face interviews only
There is a long list of possible answers
The question is of personal nature
The same set of answer categories applies to several questions
(eg. scale questions—happiness, depression etc) Cati/Capi
* Slides will be posted on OWL *
Other Respondent Problems
Response Sets: the respondent is not motivated to provide a genuine response
(1) Acquiescence: the respondent is just trying to please the research
Tend to either agree or disagree with all questions
Researchers should create lists of questions that force
respondents to choose from both sides of the spectrum to
maintain a consistent attitude.
(2) Social Desirability: respondents give what they believe is the
politically/socially correct answer
Do so out of a desire to be liked, considered respectable etc.
Related to bias due to innate characteristics (error #7)
(3) Laziness or Boredom: may prevent a respondent from giving a genuine
May just want to get it over with
Can use same strategy as for #1
Sources of Error
Page 63 & 64 in the textbook
7 most important errors
Mistakes are always made; therefore this is used to deduct the potential error
Test #1 Review
*NOT ON THE TEST from CH. 4*
No cover letters
Advantages vs. disadvantages with phone interviews
Different kinds of questionnaires
Researcher driven diaries
Chapter 4 MISSED LECTURE Cont’d Survey Research February 10, 2014
Face to face phone interviews
Questionnaires (in person, mail out, email, web, online pools)
Researcher Driven Diaries
Lecture 5 MISSED LECTURE February 10,
2014 Types of questions in surveys
Attitude and belief questions *
Compared people by their ascribed status’
Factual questions about the world * worst question to ask
People usually don’t answer this
Rules for writing good questions
Lecture 6 Structured Observation February 24, 2014
Finish Ch. 4
Finish Ch. 5
Review specific rules for writing good questions
Open or closed ended questions
Few other considerations
Some feminist theories are opposed to structured interviews and questionnaires
These methods are exploitative
Involved asymmetrical power relationship between the researcher and the respondent
Researcher is in a position of power
Some opposition has declined in recent years
Why decline in opposition?
Increased attention given to the rights of research respondents (ethics and rights)
Privacy rights, the right to end the interview at any time etc
Rights to help formulate and have access to research
Some findings using these methods are consistent with feminist ideals (feminist as in theory, not
male and female)
Findings on sexual assault, sexual harassment, etc.
Various forms of domination
All posted online
Prestige Bias: namedropping a famous or well known person in the question
Respondent answers in their own words
Get to meaning and interpretation of question from their point of view
Allows rare responses
Responses researchers have not thought of or hear before
Potentially greater detail
Good for exploring new or changing areas of research
Ex. Gender norms
Can use the answers to come up with fixedchoice responses (closed questions)
Timeconsuming Inaccuracies/mistakes easily made
If used in questionnaires:
High item nonresponse or incomplete answers
Answers have to be coded
Open questions have an extra step to coding
Coding Open Questions
Must first move from detailed answers to categories that we can later quantify
Categories are decided upon before the data are gathered
Data is gathered then themes or categories are discerned
Often a mixture of both
Easy to process answers
Standardization of Q’s and A’s
Reduced bias in recording answers
No chance to ‘interpret’ responses
Fixed responses help clarify what the question means
Easier and quicker to complete
But categories must be mutually exclusive and exhaustive
Mutually exclusive: no overlap between categories
Ex. posted online “What is your annual income?”
Exhaustive: everyone can answer the question
Spontaneity and authenticity in answers is lost
Difficult to make answers exhaustive
Lose out on rapport in interviews
Potential for Response Sets:
The respondent is not motivated to provide a genuine response
Being untruthful in order to look socially desirable
Examples posted online
A slight tendency to choose the last category listed
Can try to make last category unappealing