Psychological Research Methods

Purpose of conducting an experiment

To establish a causal relationship between an independent variable (IV) and a dependent variable (DV).

Posttest-only design

Measure DV after the experimental manipulation.

Pretest-posttest design

Measure DV before and after manipulation; helps check group equivalence but may sensitize participants.

Independent groups (between-subjects) design

Different participants are assigned to each experimental condition.

Repeated measures (within-subjects) design

The same participants are used across all experimental conditions.

Order effects

Effects due to the order of conditions (practice, fatigue, contrast); controlled by counterbalancing.

Matched-pairs design

To match participants on key variables to ensure equivalent groups.

Probability sampling

When accurate representation of a population is important (e.g., political polling).

Convenience sampling

Testing behavioral hypotheses when full representativeness isn't critical.

Straightforward manipulations

Directly altering variables in an obvious way (e.g., changing lighting).

Staged manipulations

Setting up situations, often using confederates, to create specific psychological states.

Types of DV measurements

Self-report, behavioral, and physiological.

Ceiling and floor effects

When a measure is too easy (ceiling) or too hard (floor) to detect differences.

Demand characteristics

Participants guessing study purpose; reduced with filler items.

Balanced placebo design

Manipulates both expectations and actual substance received to study expectancy effects.

Single-blind and double-blind studies

Single-blind: Participant doesn't know condition; Double-blind: Participant and experimenter both don't know.

Benefit of increasing levels of an IV

Allows detection of curvilinear relationships.

Factorial design

A design with two or more IVs studied simultaneously.

2x2 factorial design conditions

Four.

Main effect

The direct effect of an IV, ignoring other variables.

Interaction effect

When the effect of one IV depends on the level of another IV.

Mixed factorial design

Combines between-subjects and within-subjects variables in one study.

Quasi-experiment

No random assignment.

ABA/Reversal design

Baseline → Treatment → Baseline to demonstrate treatment effect.

Multiple baseline design

Measure across multiple settings, behaviors, or participants.

One-group posttest-only design

Only one group, measured after treatment; very weak design.

Nonequivalent control group design

Groups compared without random assignment.

Interrupted time series design

Measure DV over time, implement treatment, and measure again.

Control series design

Interrupted time series with a comparison group.

Cross-sectional method

Study different age groups at one point in time.

Longitudinal method

Study the same group of participants over time.

Sequential method

Combination of cross-sectional and longitudinal designs.

Four scales of measurement

Nominal, ordinal, interval, and ratio.

Pearson correlation coefficient (r)

The strength and direction of a linear relationship between two variables.

Measures of central tendency

Mean, median, and mode.

Measures of variability

Standard deviation, variance, and range.

Effect size

The strength of a relationship (e.g., r, r², d).

Null hypothesis (Ho)

The assumption that there is no effect or difference between groups.

Rejecting the null hypothesis

There is a statistically significant effect.

t-test

To compare the means of two groups.

F-test (ANOVA)

To compare means when you have more than two groups or IVs.

Type I error

Rejecting Ho when Ho is actually true (false positive).

Type II error

Failing to reject Ho when Ho is actually false (false negative).

"Power" in statistics

The probability of correctly rejecting a false null hypothesis.

"File drawer" problem

Tendency for only significant results to be published.

Limitation of using only college students

They are not representative of the general population.

Experimenter characteristics

Their gender, personality, or experience can influence participant behavior.

Mundane realism

How much an experiment resembles real-world situations.

Experimental realism

How engaging and involving the experiment is for participants.

Conceptual replication

Testing the same hypothesis using different methods.