Transcript October 5th - Seattle Central College

Social Psychology as
Science
A couple of concepts…
…that make research especially important
Hindsight bias
 This
is the tendency to overestimate our
ability to predict events, once we know
the outcome
 Everyone is susceptible to this bias
“I just KNEW the Mariners would lose a
hundred games!”
Confirmation Bias
A tendency to search for, and
see, information that confirms
one’s preconceptions.
“I’ve been noticing that students
seem a lot more stressed out
these days because of the
economy.”
The Scientific Method
 Wonder
why
 Formulate hypothesis
 Test hypothesis
 Draw conclusions
 Report results
Primary Concerns with Research
Are
the data reliable?
Are the data valid?
Aronson uses “experimental
realism” and “mundane realism”
 Ethics of the experiment

• Effects on subject
• Use of results
Experimental Design:
What is involved?
 Independent
Variable
 Dependent Variable
 Control Group
 Experimental Group
 Random Selection
 Random Assignment
Experimental Design
 Strengths

Only way to determine cause and effect.
 Weaknesses



Validity –
Ethical considerations
Practicality
 Why


the only way to determine causation?
Random assignment
Control of variables
Most famous Social Psych
Experiments
Milgram
Zimbardo
An Example
 Your
observation:
that human service professionals care very
much about client well being, but tend not to
be very politically active on issues that affect
their clients.
Hypothesize
Hmmmm....

they're so busy, they don't have time to learn about the
issues or call/email legislators
 they don't know who to contact or how to contact them
 they don't think their input matters
 etc.
 "When people are given a concise and easy to
understand explanation of the issue and what they can
do, they will take action to influence politics."
Test your hypothesis


Two groups: Experimental group gets an email with
concise talking points, specific request and contact
info. The control group gets an email with just the
request.
How are subjects selected and assigned to
groups? What if people volunteer to participate and then
are randomly assigned? What if Sound Mental Health
provides the researchers with email of all staff and then
fifty of those are randomly selected and then randomly
assigned to one of the groups? What if all the staff at
the Tukwila clinic are assigned to the experimental group
while the Capitol Hill staff are assigned to the control
group?
Draw conclusions
 Statistical
analysis
 Probability of results due to chance


.01
.05
In other words, how likely is it that the results
are due to chance.
Report Results
 Scholarly/peer
reviewed journals
Design an experiment!
 Wonder
why – what do you want to
know?
 Formulate hypothesis – specific
question?
 Test hypothesis
 Draw conclusions
 Ethical considerations
Report Back
 What
did you wonder about?
 What is your hypothesis?
 How would you test it?
 What difficulties did you encounter in
planning the research?