Research Methods in Psychology
The Scientific Method
© 2009 by The McGraw-Hill Companies, Inc.
Psychology Questions
 Do we always know how people will
behave or what they will think?
 Answer “True” or “False” to the following
questions:
© 2009 by The McGraw-Hill Companies, Inc.
Psychology Questions, continued
 Mothers talk to their younger children differently
than they talk to their older children. True or
false?
• False. Haden (1998) found that mothers use the
same conversation styles (“elaborative” or “repetitive”)
with their different-age children.
Haden (1998). Reminiscing with different children. Relating
material stylistic consistency and sibling similarity in talk about
the past. Developmental Psychology, 34, 99–114.
© 2009 by The McGraw-Hill Companies, Inc.
Psychology Questions, continued
 Few students will confess to ruining a computer
program if they didn’t do it. True or false?
• False. Kassin and Kiechel (1996) found that 69% of
students in their study falsely confessed to ruining a
computer program and signed a written confession.
Kassin, S. M., & Kiechel, K. L. (1996). The social psychology of
false confessions: Compliance, internalization, and
confabulation. Psychological Science, 7, 125–128.
© 2009 by The McGraw-Hill Companies, Inc.
Psychology Questions, continued
 Most individuals will notice if a person they are
talking to is replaced by another person. True or
false?
• False. Simons and Levin (1998) found that only 47%
of participants in one study and 33% of participants in
a second study noticed that the person changed to a
different person midway through their conversation.
Simons, D. J., & Levin, D. T. (1998). Failure to detect changes to
people during a real-world interaction. Psychonomic Bulletin
and Review, 5, 644–649.
© 2009 by The McGraw-Hill Companies, Inc.
Psychology Questions, continued
 Writing about adjusting to college improves
students’ grades. True or false?
• True. Pennebaker and Francis (1996) found that
students who wrote about adjusting to college had a
higher GPA (M = 3.08) the following semester than
students who wrote about superficial topics (M =
2.86).
Pennebaker, J. W., & Francis, M. E. (1996). Cognitive, emotional,
and language processes in disclosure. Cognition and Emotion,
10, 601–626.
© 2009 by The McGraw-Hill Companies, Inc.
The Scientific Method
 Way to gain knowledge about behavior
and mental processes
• not a particular technique or tool
• a general approach to gaining knowledge
© 2009 by The McGraw-Hill Companies, Inc.
The Scientific Method, continued
 Compare scientific method to “everyday” ways of gaining
knowledge:
•
•
•
•
•
•
•
•
general approach
observation
reporting
concepts
instruments
measurement
hypotheses
attitude
© 2009 by The McGraw-Hill Companies, Inc.
General Approach
 Nonscientific
• intuitive
• judgments based on “what feels right”
 Scientific
• empirical
• judgments based on direct observation and
experimentation
© 2009 by The McGraw-Hill Companies, Inc.
Observation
 Nonscientific
• casual, uncontrolled
• personal biases and other factors influence
observation
 Scientific
• systematic, controlled
• control = essential ingredient of science
• greatest control in an experiment
© 2009 by The McGraw-Hill Companies, Inc.
Observation, continued
 Control
• investigate factors one at a time in experiment
• an experiment has at least
 one independent variable
 one dependent variable
© 2009 by The McGraw-Hill Companies, Inc.
Observation, continued
 Independent Variable (IV)
Factor researchers control or manipulate in
order to determine the effect on behavior
• minimum of two levels
 treatment (experimental) condition
 control condition
• Example IV: write about adjusting to college (experimental
condition) or about superficial topics (control condition)
© 2009 by The McGraw-Hill Companies, Inc.
Observation, continued
 Dependent Variable (DV)
Measure of behavior used to assess the
effect of the independent variable
Example DV: measure students’ GPA
• Most studies involve several dependent
variables.
Example DV: measure characteristics of students’ health and
well-being
© 2009 by The McGraw-Hill Companies, Inc.
Reporting
 Nonscientific
• biased, subjective
• personal impressions
 Scientific
• unbiased, objective
• separate observations from inferences
• interobserver agreement
© 2009 by The McGraw-Hill Companies, Inc.
Concepts
 Nonscientific
• ambiguous
• Are we clear in the meaning of words we use?
 Scientific
• clear, specific definitions
• construct = concept
© 2009 by The McGraw-Hill Companies, Inc.
Constructs
 Many psychological constructs
• examples: aggression, depression, emotion,
intelligence, memory, personality, stress, wellbeing
 Operational definition
Specific procedure used to produce and
measure a construct
© 2009 by The McGraw-Hill Companies, Inc.
Constructs, continued
 Advantages of operational definitions
• define constructs specifically
• allow clear communication
 Disadvantages
• potentially limitless number of operational
definitions for any construct
• some operational definitions may be
meaningless
© 2009 by The McGraw-Hill Companies, Inc.
Constructs, continued
 Match each construct with an operational definition:
Construct
Operational definition
Aggression
A. score on Minnesota Multiphasic Personality
Depression
Inventory
Intelligence
B. score on final exam for this course
Memory
C. number of times person hits another person
Knowledge of
D. number of depression symptoms
Research Methods
from Diagnostic and Statistical Manual
Personality
E. score on the Wechsler Adult
Intelligence
Scale
F. score on the Digit-Span Test of
memory
© 2009 by The McGraw-Hill Companies, Inc.
Instruments
 Nonscientific
• inaccurate, imprecise
• examples: clocks, gas gauges, measuring cup
 Scientific
• accuracy: difference between what an
instrument says and what is actually true
• precision: instruments have different levels
© 2009 by The McGraw-Hill Companies, Inc.
Measurements
 Nonscientific
• not valid or reliable
• measure of concepts are inaccurate or
inconsistent
 Scientific
• valid and reliable
 valid measures are truthful
 reliable measures are consistent
© 2009 by The McGraw-Hill Companies, Inc.
Measurements, continued
 Physical measurement
• dimensions have agreed-upon standards and
instruments
 examples: length, weight, time
 Psychological measurement
• constructs have no agreed-upon standard nor
instrument
 examples: beauty, intelligence, aggression
• Researchers develop measures to assess
psychological constructs
© 2009 by The McGraw-Hill Companies, Inc.
Measurements, continued
 Measures must be valid and reliable
• Validity refers to truthfulness
 example: course exams?
• Reliability refers to consistency
 interobserver reliability
• A measure may be reliable but not valid
© 2009 by The McGraw-Hill Companies, Inc.
Hypotheses
 Nonscientific
• untestable
 Scientific
• testable
 concepts are clearly defined and measured
© 2009 by The McGraw-Hill Companies, Inc.
Hypotheses, continued
 Hypotheses are not testable if:
• constructs are not adequately defined
• circular: the event itself is used as an
explanation for the event
• appeals to ideas or forces not recognized by
science
© 2009 by The McGraw-Hill Companies, Inc.
Attitude
 Nonscientific
• uncritical, accepting
• accept claims without evidence, ignore
contradictory evidence
 Scientific
• critical, skeptical
 behavior and mental processes are complex
 human mistakes are made (even in science)
© 2009 by The McGraw-Hill Companies, Inc.
Goals of the Scientific Method
 Four research goals
•
•
•
•
description
prediction
explanation
application
© 2009 by The McGraw-Hill Companies, Inc.
Description
 define, classify, catalogue, or categorize
events and their relationships
• example: psychologists describe symptoms of
depression
• one operational definition of depression:
the list of symptoms in the DSM
© 2009 by The McGraw-Hill Companies, Inc.
Description, continued
 Most psychology research is nomothetic,
not idiographic.
• nomothetic: large sample sizes, “average”
performance of a group
• idiographic: individual case studies
• Nomothetic researchers emphasize
similarities among individuals.
© 2009 by The McGraw-Hill Companies, Inc.
Description, continued
 Most psychology research is quantitative,
not qualitative
• quantitative: statistical summaries of behavior
• qualitative: verbal summaries of research
findings
© 2009 by The McGraw-Hill Companies, Inc.
Prediction
 Correlations (relationships) among
variables allow researchers to predict
mental processes and behavior
• example: as level of depression increases,
individuals exhibit more helplessness (failure
to initiate activities and pessimism about the
future)
© 2009 by The McGraw-Hill Companies, Inc.
Prediction, continued
 variable =
dimension on which people differ, or vary
• examples: childhood loss of parent (yes/no),
symptoms of depression, aggressiveness,
age, emotional problems, stressful life events,
physical illness
© 2009 by The McGraw-Hill Companies, Inc.
Prediction, continued
 correlation =
two measures of the same people, events,
or things vary together or go together
• example: the more stressful life events
persons experience (one variable), the more
likely they are to experience physical illness
(a second variable)
© 2009 by The McGraw-Hill Companies, Inc.
Prediction, continued
 Based on a correlation:
• if we know people’s score for one variable, we
can predict their score for a second variable
• example: if we know how many stressful life
events a person has experienced, we can
compute number of physical symptoms (and
vice versa)
© 2009 by The McGraw-Hill Companies, Inc.
Prediction, continued
 Correlation does not imply causation
• example: amount of hair on one’s ears and
heart disease are correlated (true)
• Does hair in one’s ears cause heart disease?
• Does heart disease cause hair to grow in
ears?
• Could some other variable account for the
relationship?
© 2009 by The McGraw-Hill Companies, Inc.
Explanation
 Researchers understand and can explain
a phenomenon when they can identify its
cause(s)
• example: research participants exposed to
unsolvable problems become more
pessimistic and less willing to do new tasks
(i.e., they become helpless) than participants
who complete solvable problems
© 2009 by The McGraw-Hill Companies, Inc.
Explanation, continued
 Conduct controlled experiments to identify
causes
 Control
• manipulate factors one at a time to determine
their effect (= independent variables)
• measure dependent variables
© 2009 by The McGraw-Hill Companies, Inc.
Explanation, continued
 Remember
• the word “experiment” is often used in
everyday language to mean the same thing
as “research”
• “experiment” refers to a specific type of
research study
© 2009 by The McGraw-Hill Companies, Inc.
Explanation, continued
 Causal inference
= statement about the cause of an event or
behavior
 Three conditions
• covariation of events
• time-order relationship
• elimination of plausible, alternative causes
© 2009 by The McGraw-Hill Companies, Inc.
Explanation, continued
 Example: causal inference
Exposure to media violence causes an increase
in the likelihood of aggressive thoughts,
emotions, and behaviors immediately after the
exposure
• Based on this, we know:
 exposure to media violence and aggression vary
together
 aggression follows after the exposure (not before)
 other explanations for the relationship have been
ruled out
© 2009 by The McGraw-Hill Companies, Inc.
Explanation, continued
 Causal inferences and confounding
• Confounding = two independent variables
covary together
→ cannot determine which IV caused effect
on DV
• For causal inference, experiment must be free
of confoundings
© 2009 by The McGraw-Hill Companies, Inc.
Explanation, continued
 Describe the confounding:
A psychologist seeks to demonstrate the effectiveness of a new
therapy for helping students to cope with stress. One group of
students receives the new treatment during the fall term; a
second group of students is placed on a waiting list to receive
the treatment during the next term (control group). To make sure
the students in the control group maintain their interest in the
research project, an assistant calls them every week to “check in
and see how they’re doing.” The psychologist measures the
coping of students in both the treatment and control groups at
the end of the fall term and discovers no difference in coping for
the two groups and both are coping well. The researcher decides
to abandon the new therapy.
© 2009 by The McGraw-Hill Companies, Inc.
Generalization
 Researchers are not interested in just the
one sample of people or one set of
circumstances tested in a research study
 They wish to generalize a study’s findings
to
• other people
• other settings
• other conditions
© 2009 by The McGraw-Hill Companies, Inc.
Generalization, continued
 Can we generalize findings from
psychology studies involving
• college students to people of other age
groups and circumstances?
• highly controlled laboratory settings to realworld settings?
• conditions of aggression in a lab with college
students to real-life aggression?
© 2009 by The McGraw-Hill Companies, Inc.
Application
 Apply knowledge and research methods to
improve people’s lives
• example: treatment that encourages
depressed individuals to attempt tasks that
can be easily achieved decreases
helplessness and pessimism
© 2009 by The McGraw-Hill Companies, Inc.
Application, continued
 Basic and Applied Research
• Applied: research to change people’s lives for
the better
 often “real-world’ or natural settings
• Basic: research to understand behavior and
mental processes
 “seeking knowledge for its own sake”
 often in laboratory settings
 goal of testing theories
• Both are needed
© 2009 by The McGraw-Hill Companies, Inc.
Scientific Theory Construction and
Testing
 Theories = proposed explanations for the
causes of phenomena
• explain who, what, when, where, how, and
why of behavior and mental processes
• logically organized set of statements
 define events (concepts)
 describe relationships among events
 explain the occurrence of events
© 2009 by The McGraw-Hill Companies, Inc.
Psychological Theories
 Theories vary in scope and complexity.
 Successful theories
•
•
•
•
•
organize empirical knowledge
suggest testable hypotheses
guide research
survive rigorous testing (e.g., falsification)
logical, internally consistent, precise,
parsimonious
© 2009 by The McGraw-Hill Companies, Inc.
Psychological Theories, continued
 Intervening Variables
• processes or mechanisms used to explain
relationship between IVs and DVs
 example: emotional writing about adjusting to
college, compared to superficial writing, caused
students to have higher GPAs.
 Intervening variable: emotional writing causes
cognitive change, which causes higher GPAs.
© 2009 by The McGraw-Hill Companies, Inc.
Psychological Theories, continued
 Intervening variables
• “hidden” processes represented by constructs
• What intervening variable may explain these
relationships?
Independent Variable Intervening Variable
insult (present/absent)
?
amount of time spent studying ?
time without liquid
?
amount of positive feedback
?
Dependent Variable
aggressive response
score on a test
water consumed
improved performance
© 2009 by The McGraw-Hill Companies, Inc.