Non-Experimental Method: Correlational and Qualitative Studies

Correlational Study

• examines how variables are naturally related in the real world, without any attempt to alter or control them

• used if research is unethical to conduct

Population

the broader group of people you want to draw conclusions about

Representative Sample

A small amount of people with the same characteristics of broader group

Random Sampling

Every member of population has an equal chance of being selected

Sampling Bias

when you get too many/few of certain type of participant

Convenience Sampling

• choosing participants at random, but based on who happens to be available

• causes sampling bias

Generalizability

relates to everyone in the population

Random Assignment

when subjects have already agreed to participate in a study and are randomly assigned to a condition withing the study (experiment/control group)

• EXPERIEMENTS ONLY

Why can’t correlational studies not establish causation?

the environment can’t be controlled, directionality problem, third variable problem

Directionality Problem

• Correlation cannot establish the direction of causation

• cannot tell which variable is affecting

Third Variable Problem

Correlation cannot account for the presence of a third variable that may affect both variables

Why don’t correlations have independent and dependent variables?

Nothing is controlled

Positive Correlation

two variables vary in the same direction

Negative Correlation

when one variable decreases as the other increases

Strong Correlation

very close relationship between variables

• confident predictions

Weak Correlation

relationship between variables is looser

• difficult to predict

Zero Correlation

no relationship at all

The closer r is to -1/1…

the stronger the correlation

The closer r is to 0…

the weaker the correlation

Are direction and strength related?

No

What are surveys used for?

• to gather data from people quickly

• ask fixed questions and use responses

Likert Scales

measure attitudes by converting them to numbers

Self-Report Bias

people aren’t always accurate/honest about thoughts, feelings, behaviors

• could genuinely don’t know

What are the drawbacks of surveys?

Self report bias and social desirability bias

Social desirability bias

giving answers you believe are socially/politically acceptable

Quantitative research

numerical data

Qualitative Research

in-depth descriptions of people’s experiences and observed behavior

Naturalistic Observations

study behavior in natural environment, rather than labs

• less controlled

Where does naturalistic observations work best?

observing human behaviors and how people naturally act

Weaknesses of naturalistic observation

• less controlled

• cannot draw casual connections

• if subjects know they are watched, behavior alters (Hawthorne effect)

Case studies

focus in depth on only 1 or few individuals, usually with real problems/conditions

What do case studies work best for?

• intensive examination

• general conclusions about mental processes and behaviors

• develop theories for mental disorders

What of Case Studies

• small sample size

• not generalizable

Meta-analysis

synthesize the results of a bunch of different studies on the same topic to get a clearer overall picture