For unlimited access to Study Guides, a Grade+ subscription is required.
Chapter 1: Research Strategies: How Psychologists
Ask and Answer Questions
Limits of Intuition and Common Sense
Intuition can lead us astray: Imagine folding a sheet of paper on itself 100 times.
How thick is it? How big should a group be to have a 50% change of a birthday
Hindsight bias - The tendency to think that you would have forseen an
outcome after it happened
E.g. Separate people in two groups and tell them: 1) Separation weakens romance
“Out of sight, out of mind” 2) Separation strengthens romance “Absence makes
the heart grow fonder”.
Both groups will justify a result.
In police lineups, people might be uncertain, but if told they were correct, after
they will say “There was no maybe about it”.
Sometimes intuition is wrong too.
We tend to think we know more than we do.
e.g. People are given three anagrams with solutions, and asked how long they
think it would have taken to solve. They vastly overestimate the actual time it
takes most people.
Even when our initial predictions our wrong, those who were wrong often still say
“I was almost right”. Overconfidence is hard to overcome.
The Scientific Attitude
Underlying all science is a passion to understand or explore. Scientific inquiry
sometimes proves crazy ideas. But more often it shows they are unbelievable.
Psychologists ask two questions over and over: What do you mean? How do you
Practicing scientific attitudes requires skepticism and humility because we might
have to reject our own ideas.
Critical thinking – examines assumptions, evaluates evidence, and gives
Psychology’s own critical inquiry has bee open to some surprising findings. E.g.
(massive losses of brain tissue early in life may have minimal long-term effects,
diverse groups have roughly comparable levels of happiness).
Psychologists use the scientific methods to observe, make theories, and then
A theory explains through a set of principles that predicts observable behavior.
A good theory of depression will help organize many observations into a shorter
list of principles.
A good theory needs testable predictions called hypotheses.
Operation definition – statement of the operations used to define research
A theory is useful if it: 1) effectively organizes a range of observations, and 2)
gives clear predictions that anyone can use to check
The Case Study
Psychologists study some individuals hoping to find broad general principles.
Individual cases can suggest fruitful ideas, but they can sometimes be misleading.
Commonly used in descriptive and correlational studies looks at many cases in
Wording: Asking questions is tricky; even small changes can have a major
impact. We have to phrase questions carefully
Sampling: We tend to overestimate others’ agreement with us (the false
consensus effect). The challenge is to pick a representative sample. Don’t
generalize from a few vivid but unrepresentative cases.
Random sampling works in national surveys. A larger sample, if not random,
merely gives better estimates of a misleading number.
Naturalistic observation is recording the behavior of organisms in their natural
environment. It doesn’t explain behavior but describes it.It can also be used with
correlational and experimental research.
Correlation effect: a statistical measure of how things are related, and how one
factor can predict the other. It tells us nothing about cause and effect, but it can
help us see things more clearly by telling us the extent two things are related.
Correlation and Causation
Correlation doesn’t prove causation!! If watching TV violence positively
correlates with aggressiveness, does that mean TV violence influences
aggression? Or the other way around? Or both?
Same thing with negative correlation between self-esteem and depression. Does
low self-esteem cause depression?
Correlation indicates a possibility of a cause-effect relationship, but it doesn’t
A perceived correlation that doesn’t exist is an illusory correlation. When we
believe something, we will more likely remember things that prove our belief.
E.g. we believe infertile couples are more likely to conceive after adopting. But
this is due to overreliance on the top left-hand corner in the table below:
Conceive Do not conceive
Do not adopt
We need data from all four cells. This is why people believe myths like sugar
makes children hyperactive, getting cold and wet catches a cold.
When we notice random coincidences, we may forget they are random and see
them as correlated.
Perceiving Order in Random Events
Illusory correlations come out of our eagerness to make sense of our world. We
look for patterns even among random patterns. E.g. people think HHHHHH is
less likely than HTTHTH in coin flips, when they are actually equally likely. Also
in lotteries, people are less likely to pick repeated numbers because they look less
By not recognizing random occurrences, people may seek extraordinary
explanations for ordinary events. “With a large enough sample, any outrageous
thing is likely to happen”. (e.g. people winning lotteries twice)
An event that happens to one in 1 billion people every day happens six times a
day, 2000 times a year.
The best way to isolate cause and effect is thorugh experiments. They manipulate
the factors of interest while holding constant other factors. They are different
than correlational studies, which try to uncover naturally occurring relationships.
Experiments manipulate a factor to find out its effect.
We have a tendency to seek new remedies when we’re sick, and it can produce
misleading effects. We’ll think that our healing came as a result of some new
thing we tried. We need to experiment to determine if something is really
working. One way to do this is through “blind” testing groups by giving them a
placebo. This helps researchers check a treatment’s actual effects.
Placebo effect – just thinking one is getting a treatment can cause relaxation
and symptom relief.
Double-blind procedure – An experiment where both the research
participants and research staff are blind about whether someone received a
The “double-blind” procedure creates an experimental condition of people
getting a treatment, and a control condition without it. After random
assignment, this can be pretty accurate. The experimental factor is the
independent variable, because we can vary it. Then we look at the effect of an
Confirming evidence Disconfirming evidence
Disconfirming evidence Confirming evidence