Research Methods: Measurement, Survey Design, Sampling, and Correlation

Vocabulary practice flashcards covering measurement scales, reliability, validity, survey construction, sampling techniques, correlation coefficients, and linear regression principles.

Measurement

The assignment of scores to individuals so that the scores represent some characteristics of the individuals.

Psychimetrics

The field and procedures used for the measurement of psychological constructs.

Self-Reports

Questionnaires or interviews that measure participants' thoughts, feelings, and behaviors, which can be vulnerable to social desirability concerns, distorted memory, and wording effects.

Behavioral Measures

Measures collected by carefully observing and recording participants' behavior, which can be limited by reactivity and observer error.

Physical Measures

Recordings of bodily activity (such as pulse, sweat, or glandular activity) taken with a piece of equipment.

Nominal Scale

A measurement scale using qualitative categorical labels without quantitative value or ranking (e.g., gender, political party).

Ordinal Scale

A measurement scale where items or scores are ranked in order, allowing determination of the difference and direction of the difference, but not the exact magnitude of difference.

Interval Scale

A measurement scale where scale points are ordered and equidistant, but lack an absolute or true zero point (e.g., temperature).

Ratio Scale

A measurement scale characterized by ordered, equidistant points and a true zero point, allowing values to be multiplied or divided (e.g., height, weight).

Discrete Variables

Variables representing counts that take on distinct, separate integer values with clear numerical gaps between them.

Continuous Variables

Variables representing measurements that can take on any numeric value along a continuum.

Reliability

The consistency or stability of a measurement of behavior.

Test-Retest Reliability

A form of reliability assessed by administering the same measure to the same individuals at two separate points in time.

Internal Reliability

The assessment of measurement reliability using participant responses collected at only one point in time.

Inter-Rater Reliability

The degree of agreement or correlation between the observations and ratings made by two or more independent raters.

Validity

The accuracy of a measurement reflecting whether an assessment actually measures the construct it is intended to measure.

Relationship Between Reliability and Validity

A measure may be reliable without being valid (consistent but off-target), but it cannot be valid unless it is first reliable.

Face Validity

The extent to which an assessment item or scale superficially appears, on the surface, to measure what it claims to measure.

Predictive Validity

The extent to which a measure accurately forecasts or predicts a participant's future behavior or outcomes.

Discriminant Validity

Evidence indicating that an assessment scale does not correlate with measures of other conceptually dissimilar constructs.

Response Set

A systematic tendency for survey participants to answer questions in a predetermined manner rather than reflecting their honest feelings or opinions.

Social Desirability

A response bias where participants choose answers they think will present them favorably or appear socially acceptable rather than reporting the truth.

Yea-Saying / Nay-Saying

A response tendency where participants systematically agree (yea-saying) or disagree (nay-saying) with survey items regardless of content.

Context Effects

Influences on survey answers produced by the surrounding environment or format of the questionnaire rather than the specific item content.

Item-Order Effect

A context effect where the sequence in which questions appear alters how respondents interpret and answer subsequent items.

Response-Option Effects

A context effect occurring when the provided response categories or numerical ranges steer how respondents evaluate and choose their answers.

BRUSO Model

A set of guidelines for writing survey questions: Brief (concise), Relevant (pertinent to research), Unambiguous (single clear meaning), Specific (clearly defined scope), and Objective (neutral wording).

Double-Barreled Question

A poorly constructed question that asks about two separate issues at once, making it impossible to answer each issue independently.

Loaded Question

A question written with biased language that pushes or nudges the respondent toward choosing a specific answer.

Forced-Choice Question

A question that compels participants to choose between only two options when additional alternatives or shades of opinion actually exist.

Response Sensitivity

The degree to which a survey item and its response options allow participants to express meaningful differences in feelings, behaviors, or opinions.

Ceiling Effect

A measurement limitation that occurs when the task or measure is too easy or generous, resulting in most participants clustering near the highest possible score.

Floor Effect

A measurement limitation that occurs when the task or measure is too difficult or restrictive, resulting in most participants clustering near the lowest possible score.

Restricted Range

A condition in which data shows very narrow variability because available response choices or measurement instruments are overly limited.

Likert Scale

A rating scale that assesses participants' attitudes or opinions using numbered response alternatives, commonly recommended with $5$ to $10$ options.

Scale Anchors

Descriptive words or adjectives attached to the endpoints (and sometimes intermediate points) of a numerical rating scale to define their meaning.

Bipolar Scale

A rating scale designed to measure a construct that ranges in two opposite directions away from a neutral zero midpoint (e.g., $-3$ to $+3$).

Probability Sampling

Any sampling procedure in which each member of the population has a known, specifiable probability of being selected.

Nonprobability Sampling

Any sampling procedure in which the likelihood of any specific member of the population being selected is unknown.

Simple Random Sampling

A probability sampling technique where every single individual in the target population has an equal likelihood of selection.

Stratified Random Sampling

A probability sampling technique where the population is divided into meaningful subgroups (strata) and individuals are randomly sampled from within each stratum.

Cluster Sampling

A probability sampling technique where the population is grouped into pre-existing clusters, several clusters are randomly sampled, and all individuals in those clusters are surveyed.

Convenience Sampling

A nonprobability sampling approach (also termed haphazard sampling) based on selecting cases that are easiest for the researcher to access.

Quota Sampling

A nonprobability sampling procedure where individuals are recruited non-randomly until a targeted number of participants in specific subgroups is filled.

Snowball Sampling

A nonprobability sampling technique in which an initial set of participants recruits or refers other eligible candidates to participate in the study.

Self-Selection Sampling

A nonprobability sampling procedure that occurs when people decide on their own to volunteer for a study upon viewing an announcement or advertisement.

Sampling Frame

The operational list or roster representing the target population from which sample units are drawn.

Response Rate

The percentage of individuals invited to take part in a study who ultimately complete and return the survey.

Correlation

A statistical procedure used to determine, measure, and describe the relationship between two variables regarding direction, strength, and form.

Correlation Coefficient ($r$)

A statistical index ranging between $-1.00$ and $+1.00$ where the algebraic sign dictates relationship direction and distance from zero reflects strength.

Coefficient of Determination ($R^2$)

A statistic obtained by squaring $r$, representing the proportion of variability in one variable ($Y$) accounted for or explained by another variable ($X$).

Regression Line

A straight best-fit line described by $Y = a + bX$ used to visualize linear patterns and predict values of $Y$ from known values of $X$.

Intercept ($a$)

The point where the linear regression line crosses the vertical $Y$-axis, representing the predicted value of $Y$ when $X = 0$.

Slope ($b$)

The rate of change in a regression model indicating how much the predicted value of $Y$ increases or decreases for every single-unit increase in $X$.