September 16 , 2013
CHAPTER TWO –SUDYING BEHAVIOUR SCIENTIFICALLY
Skepticism: “Show me your evidence”…“Might there be a better explanation”
Open-mindedness: Recognizing that the results and findings may go against your
EXAMPLE –KITTY GENOVESE MURDER
o Horrible crime witnessed by 38 bystanders, none whom called the police
o Bystander apathy?
o Darley and Latane
o Was there something more going on?
o Why would no one call the police?
DARLEY & LATANE EXAMPLE
Rather than apathy, perhaps it was “diffusion of responsibility”
Created an If-Then question:
o “In an emergency, IF multiple bystanders are present, THEN the likelihood
that any one bystander will intervene is reduced.”
Designed an experiment to test their hypothesis
How did they do it?
o Recording of someone yelling for help
o The one-on-one responded more quickly than the larger groups
What did they find?
Video clip of an individual laying on the ground calling for “help” in three
o Homeless man: Lasted at least twenty minutes without anyone stopping
o Middle class female: Lasted six minutes, then someone helped and others
o Business suit man: Lasted maybe six seconds, then every bystander was
by his side offering help Factors that can influence the bystanders –social status, age, gender, blood or
noticeable injury, being witness to the event (a fall) etc.
THE CORRELATIONAL METHOD
Study relationships among variable
Variable: anything that can be measured
Example: more smokes, more cancer
Example: higher IQ, more money
Relationship between variables is shown by a statistic, the correlation coefficient
r can be between -1.o and +1.0
Value tells us the strength of the relationship
Close to +1.0 or -1.0: very strong relationship
Sign tells us the direction of the relationship
+ means that both variables vary in the same direction
o If one goes up, so does the other
o Ex. More smokes, more cancer
- mean that as one variable goes up, the other goes own (like a see-saw)
o Ex. More smokes, fewer expected years to live
o Marijuana and grade point average?
o Marijuana and junk food consumption?
o Grade point average and height?
*Note: Cannot exceed past +1.0 or below -1.0.
*YOU CAN NEVR INFER CAUSATION… CORRELATION DOES NOT MEAN CAUSATION!!
Li (1975) tried to identify variables that best predicted contraceptive use for
people in Taiwan. Strongest positive correlation was the number of appliances in the house and use
of birth control.
September 18 , 2013
THE EXPERIMENTAL METHOD
1. Manipulate a variable
2. Determine whether the manipulation causes a change in a second variable
3. Control for extraneous factors that might cause the apparent change
- A prediction about a cause-effect relationship