Psychology as a Science. In this discussion we will explore: limits of common sense science - a method for understanding methods of science description correlation experimentation evaluating data with statistics sources of error and bias in research. How do we know things?.
Can we trust our intuition?
Why don’t we rely on intuition (common sense)?
Try these anagrams: WREAT
As a check on their bias, psychologists report their findings with –
a statement of the procedures used to define research variables
Operational Definitions allow others to replicate the study (repeating the essence of the study to get similar results).
In-depth study of one individual
Recorded observation of people or animals in their natural environment – focus on a specific behavior
e.g., more democratic parents have children who behave better
(slope of the points)
(amount of scatter)
Correlations make visible relationships that we might otherwise miss.
They also restrain our “seeing” relationships that actually do not exist – illusory correlations – perception of a relationship where none exists
(e.g., superstitious beliefs)
Believe – likely to notice and recall
As humans we want to make sense of our world, so we look for meaningful patterns.
Prob – Random sequences often don’t look random – leads to illusory corr.
If someone flipped a coin six times, which of the following sequences of H and T would be most likely?
HHHTTT HTTHTH HHHHHH
A little note about correlation –
DOES NOT PROVE CAUSATION!!
(e.g., gets the new drug)
(e.g., gets the placebo)
(created when using a double-blind procedure)
Validity – extent to which a test measures or predicts what it is supposed to
Internal Validity – controlled for confounding variables (if random assignment is violated, can destroy internal validity)
External Validity – generalizability (if random sample is violated, can destroy external validity)
Double Blind Procedures allow researchers to check for a treatment’s actual effects apart from the placebo effect –
- experimental results caused by expectations alone; such as a research participants’ enthusiasm for it or the healing power of belief
Confounding Variables -
(e.g., placebo takes a study skills seminar, non random assignment)