research methods test 2

Double-barreled question

A poorly written question that asks about two things at once, making it unclear which part the respondent is answering (e.g., "Do you like your job and your coworkers?").

Double negative

A question containing two negatives, which confuses respondents (e.g., "Do you disagree that people shouldn't recycle?").

Question wording

The phrasing of questions; should be simple, neutral, and specific to avoid bias or confusion.

Response scale

The range of possible answers respondents can choose from; includes the number of options, labeling, and anchors.

Number of response options

Typically 5-7 choices on a scale to balance sensitivity and simplicity.

Closed-ended question

Provides a fixed set of responses (e.g., multiple choice, rating scales).

Open-ended question

Allows respondents to answer in their own words, providing richer but harder-to-analyze data.

Labeling alternatives

Assigning clear descriptors to scale points (e.g., "Strongly Agree," "Neutral," "Strongly Disagree").

Reference group effect

Respondents answer relative to how they perceive others (e.g., "I exercise a lot" compared to who?).

Question sequence

The order in which questions appear; can influence responses. Start with general items, put sensitive or demographic items last.

Yay-saying (Acquiescence bias)

The tendency to agree with all statements, regardless of content.

Nay-saying

The tendency to disagree with all statements, regardless of content.

Fence-sitting

The tendency to select the neutral or middle option to avoid committing to a side.

Social desirability bias

The tendency to give responses that make oneself look good or acceptable to others rather than truthful answers.

Likert scale

A psychometric scale used to measure attitudes or opinions, typically using 5 or 7 response options ranging from "Strongly Agree" to "Strongly Disagree." Assumes equal intervals between points.

Representative sample

A sample that accurately reflects the characteristics of the population being studied.

Probability sampling

Sampling method in which every individual in the population has an equal or known chance of being selected.

Simple random sampling

Every individual has an equal chance of selection (e.g., names drawn from a hat).

Stratified random sampling

Population divided into subgroups (strata), and random samples are taken from each.

Cluster sampling

Population divided into clusters (e.g., schools, neighborhoods); random clusters are selected, and all individuals within selected clusters are surveyed.

Systematic sampling

Selecting every nth person from a list after a random start.

Advantages of probability sampling

Minimizes bias, allows for generalization to the population.

Disadvantages of probability sampling

Expensive and time-consuming.

Nonprobability sampling

Sampling method where not everyone has an equal chance of selection; often used for convenience.

Convenience sampling

Using participants who are easy to reach (e.g., students in a class).

Purposive sampling

Selecting participants with specific characteristics.

Snowball sampling

Existing participants recruit future participants (useful for hidden populations).

Quota sampling

Setting quotas to ensure certain categories are represented.

Advantages of nonprobability sampling

Quick, inexpensive, easy.

Disadvantages of nonprobability sampling

May not represent the population; biased results.

Polls

Surveys measuring public opinion on an issue or candidate.

Margin of error

The range of expected difference between the sample result and the true population value, typically expressed as ± percentage.

Correlational research

Examines relationships between variables without manipulation.

Advantage of correlational research

Identifies relationships and predictions between variables.

Disadvantage of correlational research

Cannot determine causation.

Third variable problem

A hidden variable influences both variables, creating a spurious correlation.

Directionality problem

It's unclear which variable causes the other.

Reverse causation

The presumed effect actually causes the presumed cause.

Misleading correlations

Correlations that appear strong but are due to chance or bias.

Small sample size

Leads to unstable or misleading correlations.

Low reliability

Unreliable measures weaken observed relationships.

Outliers

Extreme values that distort correlation strength or direction.

Restriction of range

Limited variation in one or both variables weakens correlation.

Curvilinear relationship

Relationship where variables are related, but not in a straight line (e.g., stress vs. performance).

Correlation coefficient (r)

A statistic measuring the strength and direction of a linear relationship between two variables; ranges from -1.0 (perfect negative) to +1.0 (perfect positive).

Longitudinal study

Research that measures the same variables in the same participants over time.

Cross-lag panel correlation

Examines the relationship between two variables measured at different time points to assess directionality of influence.

Experimental research

Study design in which an independent variable is manipulated and participants are assigned to conditions to test causal effects.

Random assignment

Randomly assigning participants to conditions to eliminate preexisting differences.

Matched random assignment

Matching participants on key variables, then randomly assigning them to groups to ensure balance.

Independent variable (IV)

The variable manipulated by the researcher.

Dependent variable (DV)

The outcome measured to assess the effect of the IV.

Internal validity

The degree to which the results are attributable to the manipulation and not other factors.

Confound

A variable that systematically varies with the IV, making it unclear what caused the effect.

Systematic variance

Variation associated with the independent variable (can threaten internal validity if due to a confound).

Unsystematic variance

Random variation not linked to the IV (increases noise but not bias).

Posttest-only design

Participants are randomly assigned to conditions, experience the IV, and are tested once after the manipulation.

Within-groups design

Each participant experiences all levels of the independent variable.

Repeated-measures design

Type of within-groups design where participants are measured multiple times after different conditions.

Concurrent-measures design

Participants experience all conditions simultaneously, and one dependent variable is measured once.

Order effects

Changes in participants' responses due to the sequence of conditions (e.g., fatigue, practice, carryover).

Counterbalancing

Varying the order of conditions across participants to control for order effects.

Cover story

A plausible explanation for the study's purpose used to prevent demand characteristics.

Manipulation check

A measure to confirm that the IV was successfully manipulated.

Pilot study

A small preliminary study to test the design and procedures.

Ceiling effect

When scores are too high, making it hard to detect differences.

Floor effect

When scores are too low, making it hard to detect differences.

Maturation

Natural changes over time influence results (e.g., participants get tired or smarter).

History

External events occur during the study that influence outcomes.

Regression to the mean

Extreme scores tend to move toward the average on retesting.

Attrition (mortality)

Participants drop out, potentially biasing results.

Instrumentation

Measurement tools change over time, altering results.

Testing effects

Taking a test once influences performance on a later test.

Demand characteristics

Participants guess the study's purpose and alter behavior.

Experimenter bias

Researcher's expectations influence participant behavior or interpretation.

Placebo effect

Participants improve simply because they believe they're receiving treatment.

Proper comparison group

A control or baseline group used to rule out alternative explanations.

Selection-history interaction

Groups experience different external events that affect results.

Selection-attrition interaction

Groups lose participants at different rates, biasing outcomes.

t-test

Statistical test comparing means between two groups to see if they differ significantly.

p-value

Probability that results occurred by chance; if p < .05, results are statistically significant.