Methods

Report
NAEP-Howard Statistics
NAEP-Howard
Statisticsand
andEvaluation
EvaluationInstitute
Institute
This Workshop Still Available:
Qualitative Research Methods August 11-22 10am-2pm
School of Education Bld. Room 216
Attendees will learn to apply various methods of qualitative
inquiry, including ethnographic, structured and semi-structured
interviewing, focus groups, document and content analysis,
narrative inquiry, phenomenological studies, case study,
observation, historical research, and action research.
Instructor: Dr. Dawn Williams, Chair of Department of
Educational Administration and Policy.
You must apply separately for this workshop.
More information is available at www.HowardSEI.org.
NAEP-Howard Statistics
Statistics and
NAEP-Howard
andEvaluation
EvaluationInstitute
Institute
Quantitative Methods
for the Social and Behavioral Sciences
Dr. Jamie Barden
Department of Psychology
[email protected]
Social and Behavioral Sciences
:study systematic processes of human
behavior.
Level of Analysis
 within
individual: neuroscience, brain biology
 individual: psychology, behavioral genetics
 social structure: economics, anthropology,
sociology, political science, public health
“People like it when they understand
something that they previously thought they
couldn't understand. It's a sense of
empowerment.”
--Neil DeGrasse Tyson, 2008
“What is the principal of science?
The test of all knowledge is experiment.
Experiment is the sole judge of scientific 'truth'.”
“What business are you in as a scientist?
There is an expanding frontier of ignorance...”
-- Richard Feynman (1964)
Why use the scientific method?

To understand relationships between
variables in our social world.

Empirically test predictions.
(birds of a feather/opposites attract)

To allow others to independently verify
findings.
Hypothesis
Operationalize
Measure
Evaluate
Revise or Replicate
Hypothesis
: an explicit, testable prediction about the
conditions under which an event will
occur.
Useful hypotheses should be
1. a priori: before data collection
2. falsifiable: could be found false
Hypothesis
Where do hypotheses come from?
Segue to Inspiration
Has your hypothesis been explored already?
Segue to Literature Review
Operationalize

Conceptual variable: The general abstract
definition of a variable. (like a dictionary
definition)

Operational definition: The specific procedures
for manipulating or measuring a conceptual
variable. (concrete application)
Hypothesis (conceptual)
similar people will be more attracted to each other
Hypothesis (operational)
personality test
height, age
choice of interaction partner
attraction questionnaire
Construct Validity: How well measures and
manipulations reflect the variables they are intended
to measure and manipulate.
Variables
Conceptual
(dictionary)
Example
Feeling scared or a 1. distancing behavior
behavioral tendency 2. questionnaire items
to distance the self
3. facial expression
from a stimuli
4. skin conductance
Fear
Pick One of
Your
Variables
Operational (concrete
measures and manipulations)
Methodological Options:
Social and Behavioral Sciences
Data Collection Approaches
1. Life Record Data
2. Field Study
3. Survey Research
4. Laboratory Research
5. Case Study
6. Focus Group
7. Modeling
Types of Study
A. Descriptive
B. Correlational
C. Experimental
Which have you
used?
Measure
Three types of studies:
1. Descriptive: What is the level of 1 variable?
Ex: What is the president’s overall approval rating?
2. Correlational: How are 2 variables related?
Ex: How does survey respondent’s age relate to
approval rating? [Predictor is measured]
3. Experimental: Does one variable cause the other?
Ex: Does dark vs. light skin in Barack Obama’s
photos influence approval rating?
[The independent variable is manipulated]
Measure: Descriptive

Descriptive Research: describes people using
the level of a single variable (a thought, feeling
or behavior).

Types:
1. Observation
2. Historical records (archives)
3. Survey questionnaires
Examples?
Descriptive Research Example
Gallup Daily Poll
Measure: Descriptive

Random Sampling: Selecting participants to be in
a study so that everyone in the population has an
equal chance of being in the study.
Population
Sample
Population
A random sample (N=1000)Estimate
allowsofus
to generalize our
(mean +/- %)
findings back to THIS population.
Measure: Descriptive

Advantage: easy to do

Disadvantage: only involves 1 variable, so
no information about relationships
between variables.
Correlational Research: describes the
relationship between two or more naturally occurring
variables (predictor and criterion).
-Does having a resilient personality relate to mental
health outcomes following natural disaster?
-Does pre-existing STD infection increase
susceptibility to HIV infection?
-When the sun is out more, are people happier?
Which is the predictor variable? In correlational
research the predictor is measured not manipulated.
Measure: Correlational

Advantage: study naturally occurring
variables

Disadvantage: correlation is not causation
You cannot draw causal conclusions from
correlational results.
Measure: Experimental

Experimental Research: examines cause and
effect relationships between variables.

Independent Variable (IV)
 Variable
that is the CAUSE of the dependent variable
 Variable that is manipulated by the experimenter

Dependent Variable (DV)
 Variable
that is the EFFECT
 Variable that is measured
NOTE: The IV is manipulated, which helps make it
independent of other variables.
Measure: Experimental
Examples (name the IV & DV):
-Are children more likely to be aggressive after being
shown violent media content to children (or is there no
effect)?
-What impact does having a Black person (or not) in an
otherwise White group have on decision making?
-Is someone more likely to be attracted to you if you
emphasize your similarities or differences?
-How does alcohol consumption (or not) relate to male
decision-making regarding sexual encounters?
Measure: Experimental

Advantage: cause/effect relationships

Disadvantage: can’t manipulate all
variables (impossible or ethical reasons).
Demos
Name that method DEMO
 Name that method for your research.

The End
Measure: Experimental
random assignment—each participant in
the experiment has to have an equal
chance of being in any condition, so the
conditions start the same. [DEMO]

25 participants needed per condition for
a between-participants design.
½ are told about someone similar
½ told about someone different
Quasi-experiment

Lack of control over the assignment of
participants to conditions and/or does not
manipulate the causal variable of interest.

A quasi-independent variable is not a true
independent variable that is manipulated by the
researcher but rather is an event that occurred
for other reasons.
Examples






Does smoking cause cancer?
Did 9/11 cause an increase in prejudice against
people of middle-eastern decent?
Do Republican vs. Democratic presidents affect
the economy?
Do extreme events (i.e., winning the lottery or
being paralyzed) affect day-to-day happiness?
Does giving employees a raise or extra vacation
time boost productivity and job satisfaction?
Does campus crime affect applicants to a
university?
Measure: Experimental

Advantage: can investigate quasiindependent variables that are impossible
or unethical to manipulate

Disadvantage: internal validity threats
undermine causal conclusions

similar documents