Psych 203 Definitions

Dependent variable

What the experimenter measures or records; value depends on the level of the independent variable.

Independent variable

What the experimenter manipulates; must have at least 2 levels; suspected cause of an effect.

Observed score

Quantitative label produced by measuring an event.

True score

Measure obtained by independent and different procedures free from error.

Measurement bias

Nonrandom error in one direction.

Systematic error

Every time this variable influences the response, it is in the same direction and by the same amount.

Random error

Every time this variable influences the response, it varies in amount and direction.

Accuracy

Degree to which observed value matches true value.

Validity

Measuring what (construct) you think you’re measuring.

Face validity

Content of the test is appropriate.

Criterion-related validity

Test score predicts future or current behavior.

Predictive validity

Test scores correlate with future behavior.

Concurrent validity

Test scores correlate with current behavior.

Convergent validity

Measures of the same concept are correlated.

Discriminant validity

Measures of different concepts are not correlated.

Construct validity

Successful measurement of the intended construct.

Research question

Hypothesis about relationship between IV and DV.

Descriptive statistics

Use data to describe population with graphs and summary stats.

Mean

Average of dataset affected by extreme values.

Median

Middle value when ordered.

Mode

Most frequent value.

Standard deviation

Measure of variation around mean.

Range

Difference between largest and smallest value.

Variance

Measure of spread around the mean.

Normal distribution

Symmetrical bell-shaped distribution.

Outlier

Data point that does not follow the pattern.

Nominal scale

Categories only.

Ordinal scale

Categories with rank order.

Interval scale

Categories with rank order and equal spacing.

Ratio scale

Categories with rank order, equal spacing, and true zero.

Statistical conclusion validity

Accuracy of conclusions from statistical analysis.

External validity

Generalizability of results.

Direct replication

Repeating study with same conditions.

Conceptual replication

Repeating study with differences to extend findings.

Subject pool

Group of students used for research participation.

Ecological validity

How well study reflects real-world settings.

Mundane realism

Similarity between lab and real world.

Experimental realism

How real the situation feels to participants.

Extraneous variable

Unwanted variable that may affect DV.

Confound

Variable that systematically affects results besides IV.

Internal validity

IV is responsible for changes in DV.

Treatment integrity

Accuracy and reliability of IV implementation.

Subject representativeness

Participants may not represent population.

Setting representativeness

How well setting reflects real world.

Inferential statistics

Use sample data to make predictions about population.

Chi-square test

Tests differences between observed and expected frequencies.

Linear regression

Predicts value of one variable from another.

Independent samples t-test

Compares means of two different groups.

Dependent samples t-test

Compares means of the same group under two conditions.

ANOVA

Tests differences between multiple group means.

Null hypothesis

No effect.

Alternative hypothesis

Effect exists.

Alpha level

Probability threshold for rejecting null hypothesis.

p-value

Probability result occurred by chance.

Statistical significance

p ≤ .05 indicates significant result.

Effect size

Magnitude of relationship between variables.

Eta squared

Percentage of variance explained.

Cohen’s d

Difference between means in standard deviation units.

Fail to reject null

p-value greater than alpha, no significant effect.

Reject null

p-value less than or equal to alpha, significant effect.

Level

Condition of independent variable.

Factor

Independent variable.

Between-subjects design

Different participants in each condition.

Matched-group design

Participants matched on variables before assignment.

Cross-sectional design

Compare different groups at one time.

Cohort effect

Effects due to shared experiences of a group.

Experimenter bias

Researcher influences results.

Subject reactivity

Participants change behavior because they are observed.

Single-blind

Participants unaware of condition.

Double-blind

Both participants and researchers unaware.

Placebo

Control group receives inactive treatment.

Placebo effect

Improvement due to expectation.

Demand characteristics

Cues that influence participant behavior.

Deception

Misleading participants about study purpose.

Ex post facto design

Study of existing differences without manipulation.

Interrupted time series

Repeated measures before and after treatment.

Control variable

Variable held constant.

Subject variable

Participant characteristic not manipulated.

Quasi-experiment

Includes non-manipulated variables.

Ceiling effect

No increase possible because scores are already high.

Selection bias

Groups differ before experiment.

Posttest only design

Measure after treatment only.

Pretest-posttest design

Measure before and after treatment.

Nonequivalent control group

No random assignment, control group used.

History

External events affect results.

Regression to the mean

Extreme scores become less extreme over time.

Random assignment

Assigning participants randomly to groups.