Measurement, Research Design, and Sampling Practice Flashcards

Comprehensive vocabulary flashcards covering measurement error, validity, survey techniques, experimental designs, and qualitative methodologies derived from Chapters 9 through 20.

Social desirability measurement error

An error that occurs when a respondent consistently denies negative behaviors, such as child abuse, to present a more favorable image.

Social desirability bias

The tendency of people to answer questions through a filter that will convey a favorable impression.

Reliability

The consistency of a measure, often assessed through test-retest methods or by asking multiple questions about the same fact, like age and date of birth.

Discriminant validity

A measurement property tested by comparing a new scale with measures of unrelated constructs.

Face validity

A property of a measure that is often considered insufficient because it relies on subjective assessments by researchers or experts.

Content validity

The degree to which an instrument's items appear to measure what they intend, often assessed by asking experts.

Known-groups validity

A form of validity demonstrated when an instrument can accurately distinguish between different groups, such as scoring untreated parents worse than model parents.

Closed-ended questions

Question formats that ensure uniformity of responses and ease of processing by requiring participants to select an answer from a provided list.

Open-ended questions

Questions that allow participants to provide their own answers without being limited by predetermined options.

Double-barreled question

A survey error where two separate issues, such as filing federal and state taxes, are combined into a single question.

Contingency question

A question that is intended only for some respondents, determined by their response to a previous filter question.

Likert scale

A measurement format illustrative of an index using categories such as Strongly Agree (SA), Agree (A), Undecided (U), Disagree (D), and Strongly Disagree (SD).

Response set

A potential issue in matrix-format Likert questions where respondents develop a pattern of answering regardless of the question content.

Time sequence

An essential requirement for causality indicating that the cause must precede the effect in time.

Statistical regression

The phenomenon where extreme scores on a first test, such as the 99th percentile, move toward the mean on a second test.

Instrumentation effect

A threat to internal validity that occurs when the measurement instrument is changed between the pretest and the posttest.

Selection bias

A threat to research where differences between groups can be attributed to the way participants were chosen or assigned rather than the intervention.

Blind raters

Raters who are unaware of participants' group status to mitigate bias in experimental research.

Cross-sectional designs

Research designs that collect data at a single point in time, which makes it difficult to establish causality.

Quasi-experiments

Research designs often used when agency constraints make true random experiments infeasible.

Propensity score matching

A technique that matches study participants on many attributes to reduce selectivity bias.

Switching replications

A design feature that helps determine if differences between groups at the first posttest are due to lack of comparability or the intervention.

Visual significance

In single-case research, the extent to which shifts in the target problem coincide with shifts in the independent variable.

Research reactivity

A situation where client improvement can be attributed to the impact of being part of a research study.

Obtrusive observation

A form of data collection where the subjects are aware they are being watched, such as observing a conversation in an office.

Operational definitions

Specific definitions that enable researchers to observe and measure target problems precisely.

Triangulation

The use of multiple indicators to measure the same target problem in single-case designs.

Sampling error

An error that typically decreases as the sample size increases.

Judgmental sample

A non-probability sampling method that allows researchers to use their prior knowledge to select a sample.

Systematic sampling

A sampling method involving the selection of every kth element, which can be biased if the list follows a specific pattern.

Stratified sampling

A probability sampling technique used to ensure specific personnel or population categories are represented by grouping data via related variables.

Multistage cluster sampling

A sampling method that involves selecting larger groups (like cities), then sub-groups (like churches), and finally individual members.

Random digit dialing

A technique for drawing a random sample by selecting telephone numbers at random.

Deviant cases

Specific cases used in qualitative research for comparison, such as comparing parents who refused to participate versus those who participated most.

Going native

A risk in qualitative research where a researcher loses objectivity due to prolonged engagement with the subjects.

Oral history

A qualitative interviewing method using open-ended questions to explore life events and personal meanings.

Empowerment

The central paradigm of feminist research methods.

Emic and etic perspectives

The dual perspectives in qualitative research where a researcher attempts to understand a culture from both an insider’s and an outsider’s view.

Interview guide approach

A qualitative interviewing method that uses an outline of topics and issues to be covered during the session.

Grounded theory method

An inductive qualitative approach that is similar to both case-oriented and variable-oriented analysis.

Semiotics

A field of study focused on how meaning is associated with particular signs based on social agreements.

Coding

The process of organizing and categorizing qualitative data into specific themes.

Case-oriented idiographic analysis

A deep-dive research approach focusing on a specific individual's story to understand unique life histories.