Experimental vs. Non-Experimental Methods, Reliability, and Validity

Flashcards covering experimental vs. non-experimental methods, types of reliability, and types of validity based on lecture notes.

When investigating relationships between variables, the two main approaches are experimental and __ methods.

non-experimental

Observational methods involve solely __ behavior in its natural environment without any manipulation.

observing

Correlational studies examine the strength of the __ between two or more variables without manipulation.

relationship

__ are used to collect a lot of data from many participants by asking questions about a concept or topic.

Surveys

__ are in-depth studies of individuals or small groups, often investigating rare phenomena, such as the case of Phineas Gage.

Case studies

A limitation of non-experimental methods is that they cannot establish __, only associations.

causation

Non-experimental methods are vulnerable to __ variables, which can reduce internal validity.

confounding

The 'gold standard' for experiments is the __, which involves manipulating the independent variable to see its effect on the dependent variable.

Randomized Controlled Trial (RCT)

__ are similar to experiments but lack random assignment, such as comparing reading scores across existing classes.

Quasi-experiments

__ are conducted in actual community or natural settings rather than in a laboratory.

Field experiments

A key advantage of experimental methods is their ability to establish __ by controlling confounding variables and using random assignment.

causality

A limitation of experimental methods can be a problem with __ validity, meaning results may not generalize to other settings or populations.

external

__ refers to the consistency and stability of a measure, meaning it should give similar results if measured multiple times.

Reliability

__ reliability assesses how stable a measure is over time, expecting similar scores upon retesting.

Test-retest

__ consistency means that all items on a test should measure the same underlying construct, like all questions on a depression scale correlating with depression.

Internal

__ reliability refers to the agreement between different observers or raters using the same measure.

Interrater

__ ensures that a study is measuring what it is supposed to measure.

Validity

__ validity tells us if a test is truly measuring the intended psychological construct, like happiness, and not something else.

Construct

__ validity refers to the generalizability of a study's results to other settings, populations, or times.

External

__ validity focuses on whether the independent variable is actually causing the observed change in the dependent variable, without influence from confounding factors.

Internal

__ validity is about whether a test, at first glance, looks like it measures what it claims to measure.

Face