Research, Experimentation and Ethics

Overconfidence


Thinking we know more than we do

Hindsight bias


“I-knew-it-all-along” phenomenon

Experiment


Observing factors’ affect on behavior/mental processes

Population


All cases in a group, sample drawn for study

random sample

Fairest representation of the population

Convenience sample


More convenient for scientist

Random assignment


Experimental/control conditions by chance; reduce experimenter bias

Correlational methods


Study behaviors as they are (case studies)

Scientific method:


Hypothesis, experiment, conclusion

Descriptive research


Observation/record behavior

Correlational research


Natural relationships

experimental research


cause and effect

Case study


Deep dive into one person/situation

Survey


Self-reported attitudes/opinions/behaviors

Naturalistic observation


No control/interaction just observation in natural environment

Independent variable


The factor that is manipulated

Dependent variable


The factor that is measured

Extraneous variable


Any variable that can influence the dependent variable and create bias

Confounding variable


Has an unwanted effect on the dependent variable

Experimental group


group exposed to the independent variable

Control group


Group not exposed to the independent variable

Placebo


Inert substance that is in place of the independent variable in the control group

Stratified sampling


Sample represents the population on some criteria

Experimenter bias

Unconscious bias to confirm hypothesis

Double-blind procedure

eliminates experimenter bias and placebo

Hawthorne effect


people behave differently when being watched

Central tendency 


Mean, median, mode

Variation


Range, standard deviation

Inferential statistics


Estimating what’s happening in a sample population for the purpose of making decisions about that population’s characteristics.

Statistical significance


Difference between two groups is probably not due to chance; indicates real difference

“Null” Hypothesis


Independent variable had no effect on dependent variable

Generalizability


create a generalization from sample to make a conclusion about the population

Correlation


how well does A predict B?

Correlation coefficient


Statistical measure of the extent two factors vary together, thus how well either factor predicts the other

Golden rule

Correlation does NOT imply causation!

Illusory correlations


Perceived nonexistent, correlation (superstition)

Ethics


Moral principles that govern a persons behavior or the conducting of an activity

Stanley Milgrim

Experiment on obedience

Approving an experiment

Must be approved by ethics board (American Psychology Association) and protect rights/welfare of humans and animals