Methods In Psychology

Review Questions

• What big problem did the structuralists

encounter that largely ended this perspective as
a major force within psychology?

• Which psychological perspective focuses on

thinking about the way the world was many
years ago to understand the way humans act
today?
Problem
• Facilitated communication
Overview
• Scientific Method
• Handling Data
• Correlational Studies vs
Experiments
• Samples
• Critical Thinking
• Ethics
The Art of Looking

• Empirical
evidence:
Information from
observation (and
the other physical
senses) Eadweard Muybridge/Corbis

• Scientific
Method: Set of
rules/procedures
for using empirical Frames 2 and 3 of this historic photo
by Eadweard Muybridge (1830–
evidence 1904) show that horses can indeed
fly, albeit briefly.
 Earliest elements of scientific method

• 1) Explain observations
without mentioning God or
miracles (even though many
of these scientists were very
religious)
• 2) Form a community
– A) That agrees on what counts
as empirical evidence
– B) In which people are allowed
to present evidence opposing
cherished ideas
Scientific method terminology

• Theory: Hypothetical explanation of a natural

phenomena
– E.g., some might theorize that sugar causes
hyperactivity
– Rule of parsimony: Simplest theory that explains all
the evidence is the best one
• Hypothesis: Falsifiable prediction made by a
theory
– E.g., If I feed sugar to a class of children, they will act
more hyperactive than usual
Operational Definition

• Definition: Description of a property in concrete,

measurable terms
• Purpose: Gives guidance for steps for
measuring or manipulating property

• E.g., sugar consumption and hyperactivity

– Time out of seat,
– Times not on task,
– Incidences of shouting out
Examples: Measuring helpfulness. Measuring
giving.
• Helpfulness in City: Confederate on crutches
dropped magazines or pretended to be blind or
injured while crossing the street. Timed how
long for help to arrive.
• Latin American cities were most helpful; watch
out if you go to Kuala Lumpur, Malaysia.
• Giving: Donation receipts reported on tax return
– $531 average yearly giving per Canadian
– $863 average Albertan giving (2013, Stats Canada;
report89-652-X2015001)
Measurement

• Measure: Device that can detect the condition to

which an operational definition refers
– Validity: Extent to which a measurement and a
property are conceptually related
– Reliability: Tendency for a measure to produce the
same measurement whenever it is used to measure
the same thing
– Power: Ability of a measure to detect small
differences in the property
• (Concept overlaps with operational definition)
Problem: Demand characteristics

Demand characteristics: Those aspects of

an observational setting that cause people
to behave as they think they should

Rex Features via AP Photo

 Problem:
Observer bias
• When people’s
expectations make them
see what is not there
• E.g., Sugar and
distraction
• E.g., Teachers with
subset of class
identified as ready to
bloom
• E.g., Whole language
reading
Overcoming Demand Characteristics and
Observer Bias
• Single-blind: Observer doesn’t know purpose of
study or doesn’t know which group
(experimental or control) is being observed OR
Participant doesn’t know which group they are in
or doesn’t know purpose of study.
• Double-Blind: Both participant AND observer
lack this knowledge
• Naturalistic observation: Technique for
gathering scientific information by unobtrusively
observing people in their natural environments
Overview

• Scientific Method
• Handling Data
• Correlational Studies vs
Experiments
• Samples
• Critical Thinking
• Ethics
• Variable: Property whose value can vary across
individuals or over time
– E.g., hours of sleep
– E.g., number of symptoms of anxiety
– E.g., score on test of how many Canadian prime-
ministers you can name
 Frequency Distribution

• Flip coin 6 times and record result

– E.g., HTTHHT,
– Repeat two more times: e.g., HTHHTT, HTHHHT
– 3 heads, 3 heads, 4 heads

# of times it
x
happened
x x

1 2 3 4 5 6

Number of heads in six coin-tosses

 Frequency Distribution

• Flip coin 6 times and record result

– Imagine that it’s done 24 times
– What shape is this forming?
– Text calls this a frequency distribution
x x
x x
x x x x
x x x x # of times it
x x x x x x happened
x x x x x x

1 2 3 4 5 6
Number of heads in six coin-tosses
Frequency distribution with smoothed lines

= hypothetical group of men and women scored on a test of fine motor

skills.

Test scores are listed along the horizontal axis, and the frequency with
which each score was obtained is represented along the vertical axis.
Frequency distribution for 15 people on test
with maximum score of 7
Some Descriptive
Statistics
This frequency
distribution shows
the scores of 15
individuals on a 7-
point test.
Descriptive statistics
include measures
of central tendency
(such as the mean,
median, and mode)
and measures of
variability (such as
the range and the
standard deviation).
• Normal distribution: Mathematically defined
frequency distribution in which most measurements
are concentrated around the middle
Descriptive Statistics

• Descriptive statistics: brief summary

statements about data.
– Central tendency (center or midpoint)
• Mode: Value of the most frequently observed measurement
• Mean: Average value of all the measurements
• Median: Value that is in the middle in set of ordered scores
– Variability (extent measurements differ)
• Range: Value of the largest measurement in a frequency
distribution minus the value of the smallest measurement
• Standard deviation: Statistic that describes the average
difference between the measurements in a frequency
distribution and the mean of that distribution
Skewed Distributions
When a frequency distribution is normal (a) the mean, median, and
mode are all the same, but when it is positively skewed (b) or
negatively skewed (c), these three measures of central tendency are
quite different.
Drawing Conclusions

• Statistical significance
– Finding is big enough that the odds of finding it by
chance were less than 5% (p < .05)
– E.g., Group receiving therapy scores lower on anxiety
test than control group. E.g., People who exercise live
longer on average.
– Part of inferential statistics
Overview

• Scientific Method
• Handling Data
• Correlational Studies vs
Experiments
• Samples
• Critical Thinking
• Ethics
Correlational Study Procedure
• 1) Measure two variables in a sample of
participants.
• 2) Calculate whether those two variables tend to
be associated (i.e., correlated)

• Usually cannot provide strong evidence that

the first variable is causing changes in the
second variable. But if you know the score on
one variable, you can make a guess about the
second variable (whether it’s higher or lower
than average) and be right more often than
wrong.
Can you see
whether
people who
sleep more,
score higher
on this test?
Correlation Coefficient: A measure of relation
between variables
• Positive correlation: When one variable is
higher, the other tends to be higher: shoe
size and height
• Negative correlation: When one variable is
higher, the other tends to be lower (and vice
versa): Brushing frequency and number of
cavities
• Think of correlations that are
pos, neg, zero
Correlation

• Correlation coefficient: Measure of the

direction and strength of a correlation (r)
• r ranges
❖ -1.0 (perfect negative correlation)
❖ +1.0 (perfect positive correlation)
❖ 0 (has no correlation)

Let’s take a closer look.

Three Kinds of Correlations
Positive Correlations of Different Strengths
Correlations
• Does a correlation mean the first variable is
causing the second?
– self-esteem and school grades?
– flossing and heart disease?
– Ice cream consumption and violent crime?
 Correlation: What could it mean?

If X (exposure to media violence) and Y (aggressiveness) are

correlated, then there are at least three possible explanations:
X causes Y, Y causes X, or Z (some other factor, such as lack of
adult supervision) causes both Y and X, neither of which causes
the other.
Causation

• Sometimes we perceive causal relationships

that do not actually exist (e.g., feeling cold and
getting a cold).
• Third-variable problem: Fact that a correlation
may result when the relationship between two
variables (e.g., feeling cold and getting a cold)
may be caused by something else (e.g., virus)
 Correlation coefficient

• Interpreting a correlational finding:

– People say “correlation does not mean causation.” They
correctly mean that if two variables are correlated, you
can’t conclude that one of the two variables is causing
the other.
– But, correlation usually suggests something is causing
something, but often can’t know which of the following is
true:
1) A->B or 2) A<-B or 3) C-> A, C-> B

• E.g., self-esteem and marks in grade 2

Experimental study procedure (also called
randomized control trial or RCT)
• 1) Select sample of population (may be randomly
selected or not)
• 2) Randomly assign participants to groups.
• 3) Manipulate one variable (Independent variable)
so groups differ.
• 4) Measure a second variable (Dependent variable)
in the whole sample to see whether the groups now
differ on that variable.

• Can provide strong evidence that the first

variable is causing changes in the second
variable.
Experimentation

• Random assignment: Procedure that uses a

random event to assign people to the
experimental or control group
– Experimental group: Group of people who are treated
in a particular way in an experiment
– Control group: Group of people who are not treated in
the particular way that the experimental group is
treated in an experiment
Manipulation
The independent variable is exposure to media violence and
the dependent variable is aggression. Manipulation of the
independent variable results in an experimental group and a
control group.
Random Assignment
Children with adult supervision are shown in orange and those without
adult supervision are shown in blue. The independent variable is
exposure to media violence and the dependent variable is aggression.
Random assignment tends to make the group averages very similar on
thousands of variables. Thus, ruling out other explanations for group
differences (Correlational studies don’t do this).
• Colleen asks some people to sleep eight hours and
asks others to sleep 6. Then she assesses anxiety in
them. Correlational study or experiment?
• Jagmeet asks people how many hours they slept last
night and asks how happy the feel. Correlational study
or experiment?
• Phia randomly selects 200 people to be in her study.
She asks them to report their number of friends and
their average happiness. Correlational study or
experiment?
• Name an advantage of an experiment as opposed to a
correlational study.
• Name an advantage of a correlational study as
opposed to an experiment.
• Why is random assignment so useful and important?
How does it tend to make groups equal?
• Which type of study could provides stronger
evidence that a particular drug causes symptom
reduction: Correlational study or experiment?
• Researcher gathers sample of 50,000 people, then
uses coin toss to divide into two groups. Group A is
allowed to sleep 9 hours the next night. Group B is
asked to sleep only 6. Then she assesses anxiety
symptoms in each person: Correlational study or
experiment? Could this provide strong evidence
that lack of sleep tends to cause anxiety?
• What’s the difference between causal and casual?
Overview

• Scientific Method
• Handling Data
• Correlational Studies vs
Experiments
• Samples
• Critical Thinking
• Ethics
Samples

• In an experiment, a researcher cannot

observe an entire population, so instead
draws a sample.
– Population: Complete collection of participants who
might possibly be measured
– Sample: Partial collection of people drawn from a
population
• Case method:
Gathering scientific
data by studying one
or a few people
• Show 1) what is

AP Photo/Sony,Bill Phelps
possible, and 2) what
might or might not be
typical
• E.g., video Cases like the study of a child
prodigy are interesting in their own
documentaries right, but they also provide
important insights into how the rest
of us work.
• Ideal Sample
– Representative sample: Very similar to population
about whom you want to make conclusions
• One strategy for getting ideal sample
– Random selection: Ensuring every member of a
population has an equal chance of being included in
the sample
• Make sample more representative
• Usually unethical
Non-random
sampling can
lead to errors.

Ullstein Bild / The Granger Collection

REUTERS/Jim Young
• In the presidential election of 1948, the Chicago Tribune
mistakenly reported that Thomas Dewey would beat
Harry Truman based on a telephone survey of voters.
Why did their study fail? (Hint: Think about life in 1948.
Study done by telephone.)
• What is the difference between a population
and a sample?
• What’s the difference between random
selection and random assignment?
• Why is the failure to sample randomly not
always a problem?
Overview

• Scientific Method
• Handling Data
• Correlational Studies vs
Experiments
• Samples
• Critical Thinking
• Ethics
 Thinking Critically of
Evidence
• Critical thinking:
Involves carefully
assessing ideas. Must
consider ideas of smart

Stock Montage/Getty Images

people who disagree with
you.

Sir Francis Bacon (1561–1626)

“The human understanding, once it
has adopted opinions... draws
everything else to support and
agree with them…”
Process of Maturation:
Dichotomous to Probabilistic

Younger
teens

Children

Mature

0%
100%
Confidence
Confidence
= Definitely
= Definitely
false
true
 How to think like a Bayesian scientist

• Place self on continuum from 0% to 100% confidence

– Frequently move in response to new evidence
– Spontaneously mention that self may be wrong
– Rarely use 0% or 100%
– E.g., Consider the following claims: Netflix use increases life satisfaction.
Believing you should be happy makes you sad.
– Children can’t imagine arguing for idea when lacking 100% confidence

0%
100%
Confidence
Confidence
= Definitely
= Definitely
false
true
Overview

• Scientific Method
• Handling Data
• Correlational Studies vs
Experiments
• Samples
• Critical Thinking
• Ethics
The Ethics of Science: First, Do No Harm

• Ethics board evaluates most research before

data collection and can disallow it.
• Informed consent
• Freedom from coercion
• Protection from harm
• Risk-benefit analysis
• Deception
• Debriefing
• Confidentiality