Transcript Sociological Research

Sociological Research
METHODOLOGY
From Reading: Short Answer
 Write down the first three steps of the scientific
method. Think of a broad topic that you are
interested in and which would make a good
sociological study—for example, ethnic diversity in a
college, homecoming rituals, athletic scholarships, or
teen driving. Now, take that topic through the first
steps of the process. For each step, write a few
sentences or a paragraph:



1) Ask a question about the topic.
2) Do some research and write down the titles of some articles
or books you’d want to read about the topic.
3) Formulate a hypothesis.
Quantitative vs. Qualitative
Qualitative
 Attempt to collect
information about
the social world that
cannot be converted
to numeric form
 Collected in a
number of ways:



Participant observation
Interviews
Reviewing archives
Quantitative
 Attempt to collect
information about
the social world that
can be converted
easily into numeric
form

Then uses statistical
analysis to determine
how changes in one
factor affect another
social outcome, while
factoring out other
simultaneous events
Purpose of Qualitative and Quantitative Research
 To establish a causal
relationship between social
elements
 Qualitative Researchers

Describes social processes in
such detail as to rule out
competing possibilities
 Quantitative Researchers
 By eliminating all other
possibilities through their
research design, hope to
state with some certainty
that one condition causes
another
Deductive vs. Inductive
 Deductive Approach:

A research approach that
starts with a theory, forms
a hypothesis, makes
empirical observations,
and then analyzes the data
to confirm, reject, or
modify the original theory
Deductive vs. Inductive
 Inductive Approach:

A research approach that
starts with empirical
observations and then
works to form a theory
Journal:
 A sociologist observes the work-seeking habits of
welfare recipients. After weeks of observation, trends
emerge and the researcher forms a theory about the
behaviors of this group.


Is the sociologist in this example using a deductive or inductive
approach?
How would the sociologist study this phenomenon using the
other approach?
Causality vs. Correlation
 CORRELATION
(association) DOES NOT
EQUAL CAUSATION!

Correlation is a
simultaneous variation in
two variables
Correlation: Example
 A correlation exists between poor health and low
income
 But to say two things are correlated is very different
from stating that one causes the other



For example: You might assert that bad health causes you to
have a lower income and health– you get sick and can’t work,
lose your job, etc.
OR: Higher income leads to better health– can afford better
doctors, access to fresh, healthful foods, gym in the office, etc.
OR: A third reason that causes both?
Possible Third Factor: Reckless Tendencies
 For example: you love spending your paycheck on
designer clothes, fast cars and partying.
 This short-sighted behavior could negatively affect
health. Could also affect income.
Correlation Continued
 If we simply observe the two correlated phenomena–
poorer health and lower income, they might not be at
all affected except through a third factor (in this case,
“reckless behavior”)
 In social science, it is MUCH easier to say there is a
correlation between one factor and another than to
say there is a causal relationship
Causality
 Causality: the notion that
a change in one factor
results in a
corresponding change in
another
 Three factors needed:
correlation, time order,
and ruling out alternative
explanations
Determining Causality Example:
 Observation– notice variations in nutritional food across
countries, also notice height differences which seem to
correspond to differences in nutrition

Now we need to establish a time order
Have people in country A always been taller than people in country B?
 Or did changes in nutrition occur before increases (or decreases) in
height?


Rule out alternative explanations for the variations observed in both
nutrition and height

Third factor possibilities: groundwater supply– could lead to better
nutrition through crop yields, but could also lead to cleaner drinking
water and less infection (which is turns out to matter for height)
 In this case, the causal relationship between nutrition and height
would be termed “spurious” or false, whereas the relationship
between infection and height might be described as a “true” causal
relationship
The Problem of Reverse Causality
 Reverse Causality: a situation in which the
researcher believes that A results in a change in B,
but B, in fact, is causing A

Does low income lead to poor health, or does poor health lead
to low income?
You might have heard…
 “Every time I eat chocolate, I get acne!”

Third factor which causes both? - perhaps this person suffers from anxiety and stress. The
stress causes him to eat chocolate, but at the same time causes acne to break out. This will
lead to the two occurring very close in time, even though one isn't actually causing the other.
 Recent studies have proven that watching too much violence on television
leads to people being violent in real life.

Other variables? Example: both increasing violence on television and increasing violence in
society might be caused by changing economic circumstances or something else entirely.
 Drinkers are more likely than non-drinkers to get lung cancer, suggesting
drinking causes lung cancer.

It turns out there is a strong correlation between consuming alcohol and developing lung
cancer. The problem would be asserting that alcohol consumption causes lung cancer; the
actual reason is that people who drink more also tend to smoke, or smoke more, than non
drinkers.
 Many claim that marijuana is a “gateway drug” because those who have
smoked marijuana are more likely than those who haven’t to go on to try
other drugs.

The problem would be asserting that marijuana use leads to increased use of other drugs;
other variables? Those who are willing to try one drug are also willing to try other drugs: the
cause – willingness to try or use drugs – must necessarily exist before one tries pot.
Journal: Think, Pair, Share
 What is the difference between causality and
correlation?

Provide one example to illustrate the difference between the
two.
Variables
 Dependent Variables: The
outcome the researcher is
trying to explain
 Independent Variables: a
measured factor that the
researcher believes has a
causal impact on the
dependent variable

Key independent variable:
the most important one
 Change in your dependent
variable DEPENDS on your
independent variable
Hypothesis
 Hypothesis: a proposed
relationship between two
variants, usually with a
stated direction

Direction of the
relationship refers to
whether your variables
move together (positive)
or in opposite directions
(negative)
Hypothesis Direction Example:
 We know that income is positively related to
education: As people’s education increases, usually
so too does their income
 Overt prejudice, on the other hand, is negatively
related to education: as people’s education level
increase, generally their levels of expressed prejudice
decrease
With a Partner:
 Create a 4 panel comic that describes and illustrates
(colored neatly, with narration) one of the bullet
points below




Correlation/Causation
Reverse Causality
Variables (dependent, independent)
Hypothesis/direction (positive, negative)
Hypothesis Testing
 Operationalization: the
process of assigning a
precise method for
measuring a term being
examined for use in a
particular study
 “Poverty
Operationalization”



One study focuses on
poverty who fall below the
official poverty
One study focuses on the
UN’s definition of poverty
($1 per day)
Very different ideas of
“poverty” to be studied
Poverty Study Continued
 Once you decide how to
define poverty, you must
begin to consider all the
variables related to the
concept




Education
Employment
Race
Gender
Poverty Study Continued
 Now, is poverty the
dependent or
independent variable?
 Interest: Want to
examine the factors that
cause poverty, and
especially interested in
the effect of parental
education on children’s
poverty levels
Poverty Study Continued
 Now ready to pose the research question: “What
effect do parents’ educational levels have on
children’s chances of living in poverty as adults?”
 Hypothesis: The lower the educational level of
parents, the greater the chance that their children
will be poor as adults.
Poverty Study Continued
 Other variables: race and family structure could be
moderating variables

Variables that affect the relationship between independent and
dependent variables
Poverty Study Continued
 Now we need to tell stories that explain why the
hypothesis might be true

Parents who are more educated have acquired more
confidence and skills for succeeding in our economy and are
more likely to pass on this knowledge and positive outlook to
their kids at home
Validity, Reliability, Generalizability
 Validity: the extent to
which an instrument
measures what it is
intended to measure
 Reliability: likelihood of
obtaining consistent results
using the same measure
 Generalizability: the extent
to which we can claim our
findings inform us about a
group larger than the one
we studied
Validity
 That the study measures what you intend it to
 If you step on a scale, and it measures your height, it’s not
valid.
 If I ask you how happy you are with your life in general, and
you tell me how happy you are with your school life in
particular, it is not a valid measure of your life satisfaction
Reliability
 How likely you are to obtain the same result using
the same measure the next time



A scale that’s off by ten pounds might not be totally valid– but
it is reliable if every time I step on it, it reads exactly 10 pounds
less than my true weight
A clock that is five minutes fast is reliable, but not valid (not
providing the actual time, which is what it is intended to
measure)
As for a measure in life satisfaction on a scale of 10, and you
give a 10 out of 10 one week, and a 7 out of 10 the next, it might
be a valid test, but not reliable.
Generalizability
 Can we generalize our
findings to a larger
population?
 How do we determine if
we can?
Journal: Think, Pair, Share
 Give an example of a measure that is valid, but not
reliable
 Then give an example of a measure that is reliable,
but not valid.
Journal:
 A sociologist analyzes the effects of social class and
gender on intelligence based on standardized test
scores. The test consistently gets similar results but
actually measures the ability to read quickly.


What are the independent and dependent variables in this
example?
Are the results of this study reliable, valid, both or neither?
Creating and Testing Theory
 The most important thing to determine is what you
want to learn and then contemplate the best possible
way to collect the empirical data that would answer
your question.
Data Collection
 Participant Observation: A
qualitative research method that
seeks to observe social
interactions in practice


What this usually entails is
“hanging out” and documenting
people’s practices in a given
society
Some focuses more heavily on the
participating, and some focuses
more heavily on the observing–
depending on the interests of the
researcher and the
appropriateness of participating
 The context in which the
observations occur are called
sites

Usually several different sites are
chosen, in order to compare
findings
Data Collection: Interview
 Interviews are another way to gather qualitative data
 Open-ended interviews: allow the subjects to go off
on tangents
 Semi-structured or structured interviews– the
researchers develop a specific set of questions to
address with all respondents in a relatively fixed
sequence
Data Collection: Survey Research
 Survey: an ordered series of questions intended to elicit
information from respondents
 A survey targets a specific population, people who are the
focus of a study, such as college athletes, international
students, or teenagers living with type 1 (juvenile-onset)
diabetes.


Most researchers choose to survey a small sector of the population,
or a sample: that is, a manageable number of subjects who represent
a larger population.
The success of a study depends on how well a population is
represented by the sample.
 In a random sample, every person in a population has the
same chance of being chosen for the study.
Quantitative and Qualitative Data in Surveys
 Quantitative Data: yes/no questions that would be
easy to tabulate
 Qualitative Data: Complex questions with complex
answers


Ex) How do plan to use your college education?
Harder to analyze and tabulate
Research Design: Create a survey
 Follow the directions on the handout provided
Field Research
 Field research refers to gathering primary data from
a natural environment without doing a lab
experiment or a survey.
 Researcher interacts with or observes a person or
people, gathering data along the way.

The key point in field research is that it takes place in the
subject’s natural environment

coffee shop or tribal village, a homeless shelter or the DMV, a
hospital, airport, mall, or beach resort.
Field Research
 Field research often begins
in a specific setting

The study’s purpose is to
observe specific behaviors in
that setting.
 Field work is optimal for
observing how people
behave. It is less useful,
however, for understanding
why they behave that way.

You can't really narrow down
cause and effect when there
are so many variables
floating around in a natural
environment.
Ethnography
 Ethnography is the
extended observation of the
social perspective and
cultural values of an entire
social setting.

Ethnographies involve
objective observation of an
entire community.
 The heart of an
ethnographic study focuses
on how subjects view their
own social standing and
how they understand
themselves in relation to a
community.
Case Study
 A case study is an in-depth
analysis of a single event,
situation, or individual.
 Uses: documents and
archival records, conducts
interviews, engages in
direct observation, and
even participant
observation, if possible.
 Criticism: Difficult to
general a case study, since
it is designed for a single
person or situation
Experiment
 One way researchers test
social theories is by
conducting an
experiment

Investigate relationships
to test a hypothesis.

Scientific approach
 Two types


Lab-based
Natural/Field-based
Experiment
 In a lab setting, the research can be controlled so that
perhaps more data can be recorded in a certain amount
of time.
 In a natural or field-based experiment, the generation of
data cannot be controlled but the information might be
considered more accurate since it was collected without
interference or intervention by the researcher.
 As a research method, either type of sociological
experiment is useful for testing if-then statements: if a
particular thing happens, then another particular thing
will result.
Experiment
 The Hawthorne effect—people changing their
behavior because they know they are being watched
as part of a study.
 The Hawthorne effect is unavoidable in some
research.
 In many cases, sociologists have to make the purpose
of the study known.

Subjects must be aware that they are being observed, and a
certain amount of artificiality may result (Sonnenfeld 1985).
Method
Implementation
Advantages
Challenges
Survey
•Questionnaires
•Interviews
•Yields many
responses
•Can survey a large
sample
•Quantitative data
are easy to chart
Field Work
•Observation
•Participant
observation
•Ethnography
•Case study
Yields detailed,
accurate real-life
information
•Can be time
consuming
•Can be difficult to
encourage
participant response
•Captures what
people think and
believe but not
necessarily how they
behave in real life
•Time consuming
•Data captures how
people behave but
not what they think
and believe
•Qualitative data is
difficult to organize
Experiment
Deliberate
manipulation of
social customs and
mores
Tests cause and
effect relationships
•Hawthorne Effect
•Ethical concerns
about people’s
wellbeing
Research Design Project
 Create a research design to collect empirical data to
either support or disprove one of the following
claims:



People on welfare are lazy and don’t want to work
Women are worse drivers than men
African Americans are naturally more athletic
 Remember to think about the different variables
involved, whether you’re aiming to establish
correlation or causality, and what method of
data collection is best suited for your research
question