How Sociologists Do Research
2.1 Science and Experience
Puzzle: One, Two, Three, Four, Five, Six, Seven, Eight, Nine, Ten.
Levels of Experience
Concrete experience – obtained by seeing, touching, tasting, smelling or hearing.
o Percepts – the smallest bits of concrete experience.
o Patterns – collections of related precepts.
o Eg. A single dot – precept, selection of dots – pattern.
o Two characteristics of concrete experience
Experience you share with all other living creatures
Concrete level of experience is meaningless by its self
Full of sensations but no meaning.
Abstract experience – the imaginary world of the mind
o Concepts – abstract terms used to organize concrete experience.
o Propositions – ideas that result from finding the relationship between
o Eg. You see a pen, and develop different experiences for each one: colour,
size, and flawed/unflawed. Each of them is a “pen”, naming process known
as conceptualization, organize concrete experience by placing the objects
into a single, meaningful category.
Scientific VS. Unscientific Thinking
In everyday life, believing is seeing; our bias influences what we observe
o Chicken soup gets rid of colds- its works for my family.
Knowledge based on tradition
Although some are valid, science determines validity
o Read about it in newspaper, magnets heal illness.
Knowledge based on authority
Authoritative sources can be wrong
o I was walking when I saw the accident
Knowledge based on casual observation
Observations could be uncertain, better to be fully conscious
o People, who are poor and work hard, get ahead, just like my family.
Knowledge based on overgeneralization
Scientists do sample (part of the population of research interest that
is selected for analysis) cases that represent the population (entire
group about which the researcher wants to generalize).
o I’m right because I can’t think of any contrary cases
Knowledge based on selective observation
We tend to ignore evidence that challenges us, scientists however
use random sampling to avoid this so that the population is equally
o Smith is poor even though he works hard, but he has a disability. So
disabilities are the only exception to that rule. Knowledge based on qualification
Easily accepted as valid, but scientists treat them as statements that
must be further researched.
o Because it happened before, they’ll win next Thursday.
Knowledge based on illogical reasoning
We expect reoccurrence without reasonable cause, ignoring chance.
Scientists use statistics to avoid this.
o I just can’t be wrong
Knowledge based on ego-defence
Passionate scientists, but scientific community puts strict limits.
o The matter is settled once and for all
Knowledge based on premature closure of inquiry
We believe all relevant evidence has been gathered. However,
science believes things are only temporary true
o There must be supernatural forces at work here
Knowledge based on mystification
When something can’t be explained by attribute it to forces that
aren’t observed nor understood. However, science is skeptical
2.2 Research Preliminaries
The Research Cycle
1. Formulate a research question
o Must be answered by systematically collecting and analyzing data
o Require evidence more than faith
2. Review the existing research literature
o Must elaborate the question seeing what others have already done
o Stimulates imagination
3. Selecting a research method
o Each data method has strengths and weaknesses
4. Collecting data
o Observing subjects, interviewing, reading documents by/about them.
5. Analyzing the data
o Confirm expectations, confound others
6. Publish the results
o Allows scrutiny and criticizing so that errors can be corrected and new
questions can be made
Respect the subjects rights
o Do no harm. Right to decide whether they can be studied and in what way –
right to safety
o Right to decide whether their attitude and behaviors may be revealed to the
public and in what way – right to privacy
o Cannot use data that allows them to be traced to a particular subject – right
o Subjects must be told how the information will be used They must be
allowed to judge the degree of personal risk involved in answering questions
– right to informed consent. Plagiarism is a no no.
Operationalization – process of translating concepts into variables and propositions
Variable – a measure of a concept that has more than one value or score
Hypothesis – a testable form of a proposition (educated guess)
After operationalization, the original idea (proposition), which expressed a
relationship between concepts, can be translated into a relationship between
2.3 The Main Methods of Sociological Research
1960s teens watched TV, and the rate of violent crime began to rise
Experiment – a carefully controlled artificial situation that allows researchers to
isolate hypothesized causes and measure their effects precisely.
o Eg. Social scientists investigated the connection between media and real
world violence using experimental methods
Randomization – an experiment that involves assigning each individual by chance
processes to the group that will be exposed to the presumed cause or to the group
that will not be exposed to the presumed cause.
o Eg. Drawing numbers, assigning group eliminates any inequality
Dependent variable – the presumed effect in a cause-and-effect relationship
o Eg. Aggressive behaviour towards toys
Independent variable – the presumed cause in a cause-and-effect relationship
o Eg. Violent TV show
Watching violent TV increases violent behavior
Experiments allow researchers to isolate the cause of theoretical interest and
measure its effect with high reliability – the degree to which a measurement
procedure yields consistent results. Removing people from natural environment
lowers validity – degree to which a measure actually measures what is intended to
Why do experiments on the effects of media violence lack validity?
o Violent acts towards toys enable children to relieve anger, lowering chance
of acting violent in public
o Aggressive behaviour is not controlled in a lab as it is in the real world
Survey – sociologists ask respondents questions about their knowledge, attitudes, or
behavior, either in a face-to-face or telephone interview or in a paper-and-pencil
Close-ended questions (options, numbered) – easy to input data into computer
Open-ended questions (uses own words) – useful when researchers don’t have
enough knowledge to create a list of answers
They can make sure the survey questions elicit valid responses by guarding against
o 1. The exclusion of part of the population from the sampling frame
o 2. The refusal of some people to participate in the survey
o 3. The unwillingness of some respondents to answer questions frankly o 4. The asking of confusing, leading, or inflammatory questions or questions
referring to several, unimportant, or non-current events
Contingency table – a cross-classification of cases by at least two variables that
allows you to see how, if at all, the variables are associated.
Independent variable – x axis (presumed cause)
o Eg. TV viewing
Dependent variable – y axis (presumed effect)
o Eg. Attitude towards a man every punching an adult male
Relationship – between two variables exists if the value of one variable changes with
the values of the others
o Positive relationship – Eg. Higher approval of punching of those who watch
o Negative relationship – Eg. Lower approval of punching of those who watch
o Stronger the correlation – stronger the relationship – greater percentage
difference between the 2 relationships
Researchers are interested in casual relationships
Three casual criterion:
o The researcher must demonstrate that the variables systematically change
together – relationship test.
o Researchers must establish sequencing - showing that the independent
variable changes before changes in the dependent variable occur.
Control variable – identify the context for the relationship between
independent and dependent variables.
Existence of control variables ad an important implication for the
interpretation of relationship:
The appearance of relationship between two variables does
not always signify that the relationship is real.
Appearances and reality do not always coincide.
Relationship authentic or fake?
Spurious relationship (phony)– exists between an independent and a
dependent variable when a control variable causes change in both
the independent and dependent variables.
Variation in the third variable causes this
In a graph no line would connect the independent and
dependent variable showing that they're not really
In comparison of the two variables you with see independent
and dependent variables changing together systematically
o Eg. Relationship between the