RCM exam 2

Nomothetic causal explanation

Explains general patterns or relationships across many people or cases; usually probabilistic rather than claiming a cause always produces an outcome.

Idiographic causal explanation

Explains why a particular event happened to a particular person or case by examining its specific circumstances and sequence of events.

Nomothetic vs. idiographic

Nomothetic explanations seek general patterns across cases; idiographic explanations seek detailed explanations of particular cases.

Empirical association

A criterion for causation: there must be an observable relationship between the independent variable and dependent variable.

Appropriate time order

A criterion for causation: the proposed cause must occur before the effect.

Nonspuriousness

A criterion for causation: the association must not be explained by a third variable that influences both the independent and dependent variables.

Causal mechanism

An explanation of how or why a cause produces an outcome.

Context

The circumstances or conditions under which a causal relationship occurs.

Five criteria for causation

Empirical association, appropriate time order, nonspuriousness, causal mechanism, and context.

Independent variable (IV)

The variable or treatment proposed to cause or influence a change.

Dependent variable (DV)

The outcome researchers measure to see whether it changes in relation to the independent variable.

True experiment

A design that tests a treatment and uses random assignment to create treatment and control/comparison conditions.

Random assignment

Allocating study participants to conditions by chance; helps reduce systematic preexisting differences and supports internal validity.

Random sampling

Selecting participants from a population; primarily helps with sample generalizability.

Random assignment vs. random sampling

Random assignment helps make experimental groups comparable; random sampling helps make a sample representative of a population.

Treatment/experimental group

The group that receives the intervention being tested.

Control group

The group that does not receive the treatment being tested, or receives a comparison condition; provides a baseline for evaluating the treatment.

Posttest

A measure of the outcome collected after the intervention.

Three components of a true experiment

Random assignment, at least one treatment group and one control group, and a posttest.

Quasi-experiment

A study that evaluates a treatment or intervention but lacks random assignment to treatment and comparison conditions.

True experiment vs. quasi-experiment

True experiments use random assignment; quasi-experiments do not. Quasi-experiments are often more vulnerable to preexisting group differences.

Nonequivalent control group design

A quasi-experiment with a treatment group and comparison group formed without random assignment, often before the treatment.

Main threat in a nonequivalent control group design

Selection bias, because the treatment and comparison groups may differ before the intervention.

Before-and-after design

A design that measures an outcome before and after a treatment, typically without a separate control group in its basic form.

Main weakness of a basic before-and-after design

The outcome may have changed because of maturation, testing, history, regression to the mean, or another factor rather than the treatment.

Ex post facto control group design

A quasi-experimental design in which the treatment has already occurred and researchers identify or construct a comparison group afterward.

Main concern in an ex post facto design

Self-selection and preexisting group differences can make it difficult to determine whether the treatment caused the outcome.

Internal validity

How confidently researchers can conclude that the treatment or independent variable caused the observed outcome rather than another explanation.

Selection bias

A threat to internal validity in which treatment and comparison groups differ in important ways before treatment.

Initial group differences

A selection-bias problem in which treatment and comparison groups already differ before the intervention.

Differential attrition / deselection

A threat in which participants drop out of groups at different rates, potentially making the remaining groups incomparable.

Endogenous change

Changes arising from processes within participants or from measurement itself, rather than from the treatment alone.

Testing effect

A threat to internal validity in which taking a pretest affects performance or responses on a later test.

Maturation

A threat to internal validity in which participants naturally change over time, independently of the treatment.

Regression to the mean

The tendency for unusually high or low scores to be less extreme when measured again, partly because of random variation.

Historical / external-event effect

A threat in which an outside event during the study affects the outcome and may explain the change instead of the treatment.

Contamination

A threat in which study conditions influence each other or participants alter behavior because they know about the study or their group status.

Compensatory rivalry / John Henry effect

A contamination threat in which the control group works harder because it knows it is not receiving the treatment.

Demoralization

A contamination threat in which the control group becomes discouraged or performs worse because it did not receive the treatment.

Treatment misidentification

A threat in which researchers cannot confidently attribute an outcome to the intended treatment because other aspects of the treatment situation may explain the result.

Researcher expectancy / self-fulfilling prophecy

A threat in which researchers' expectations influence how participants are treated or how outcomes are interpreted.

Placebo effect

A change in an outcome caused by participants' belief or expectation that they received an effective treatment, rather than by the treatment's specific active ingredient.

Hawthorne effect

A change in participants' behavior because they know they are being observed or studied.

Double-blind design

A design in which neither participants nor relevant researchers interacting with or assessing them know who receives the actual treatment, helping reduce expectancy effects.

Solomon four-group design

A design with treatment and control groups both with and without pretests; helps assess whether pretesting affects the treatment's apparent effect.

Internal validity vs. generalizability

Internal validity asks whether the treatment caused the outcome; generalizability asks whether findings apply beyond the studied participants and conditions.

Generalizability / external validity

The extent to which findings can apply to other people, populations, settings, or conditions.

Sample generalizability

The extent to which findings from a sample can be generalized to the population it represents.

Cross-population generalizability

The extent to which findings apply to populations or settings beyond the one studied.

Do quasi-experiments automatically generalize better?

No. Natural settings may improve real-world relevance, but generalizability depends on the sample, setting, treatment, and outcome.

Cross-sectional research

Research that collects data at one point in time or within one short period.

Longitudinal research

Research that collects data across multiple time periods to study change or relationships over time.

Repeated cross-sectional design

A longitudinal design that uses different samples from the same target population at different time points.

Fixed-sample panel design

A longitudinal design that repeatedly collects data from the same individuals over time.

Event-based / cohort design

A design that follows a defined group whose members share an event, characteristic, or starting point; data collection details may vary.

Repeated cross-sectional vs. panel

Repeated cross-sectional studies sample different people from the same population over time; panel studies follow the same individuals.

Attrition in panel studies

Participants dropping out over time, potentially biasing results if those who remain differ from those who leave.

Statistical control

A method for accounting for a third variable when evaluating the relationship between an independent and dependent variable.

Spurious relationship

An apparent association between two variables that is explained by a third variable rather than a direct causal relationship.

How can statistical controls reveal spuriousness?

Researchers account for a possible third variable and check whether the original IV-DV association remains within relevant groups or after adjustment.

Deception in experiments

Withholding or misrepresenting information to participants; may sometimes be methodologically useful but requires ethical justification and appropriate safeguards.

Debriefing

Explaining relevant study details to participants after participation, especially when deception has been used, when appropriate.

Selective distribution of benefits

An ethical issue that arises when some participants receive a potentially beneficial treatment while others do not.

Survey

A research method that asks people questions to collect information about attitudes, beliefs, experiences, behaviors, or characteristics.

Advantages of surveys

Efficient, standardized, versatile, often economical, and useful for collecting information from many people or estimating population characteristics with an appropriate sample.

Limitations of surveys

Responses may be inaccurate or socially desirable; low response rates can cause bias; surveys may lack detail and may not establish causal time order.

Leading question

A question worded to push respondents toward a particular answer.

Loaded question

A question using emotionally charged or biased wording that may influence responses.

Double-barreled question

A question that asks about two or more things at once, making it unclear what a single answer means.

Vague or ambiguous question

A question whose wording can be interpreted in different ways by different respondents.

Double negative / confusing wording

Wording that is difficult to understand because it uses multiple negatives or complex phrasing.

Overlapping response options

Answer choices that are not mutually exclusive, so a respondent may fit more than one option.

Incomplete response options

Answer choices that fail to include plausible answers for some respondents.

Question-order effect

A situation in which earlier questions influence how respondents answer later questions.

Split-ballot design

A design that gives different versions or question orders to different respondents to test question wording or order effects.

Mailed / self-administered survey: benefits

Often low-cost; respondents can answer privately and at their own pace; useful for geographically dispersed populations.

Mailed / self-administered survey: limitations

May have low response rates, and researchers cannot easily clarify confusing questions.

Group-administered survey: benefits

Many people can complete the survey at once, making it efficient and economical.

Group-administered survey: limitations

Only reaches people available in the group setting, and privacy may be a concern.

Telephone survey: benefits

Researchers can contact respondents directly, clarify questions, and make follow-up attempts.

Telephone survey: limitations

Questionnaire length and complexity are limited; people may not answer or may be difficult to reach.

Personal / in-person interview survey: benefits

Can handle longer or more complex questionnaires; interviewers can clarify questions and monitor completion.

Personal / in-person interview survey: limitations

Can be expensive and time-consuming; interviewer behavior may influence answers.

Electronic / web survey: benefits

Often fast and inexpensive; can reach many people and record responses electronically.

Electronic / web survey: limitations

May exclude people without reliable internet or digital skills; response rates and representativeness may be concerns.

Protection of survey respondents

Researchers should consider possible harm, explain relevant risks, support informed consent, and take precautions with sensitive topics.

Confidentiality

Researchers may know or be able to identify a respondent but protect the respondent's identity and information from improper disclosure.

Anonymity

Responses cannot be linked to a particular respondent by the researcher, given how the data are collected and stored.

Confidentiality vs. anonymity

Confidentiality means identity is known or potentially knowable but protected; anonymity means responses are not linked to identity.

Qualitative research

Research that emphasizes meaning, context, social processes, and participants' experiences, often using words, narratives, and observations.

Quantitative research

Research that emphasizes numerical measurement, comparisons, patterns, and statistical evaluation of relationships or hypotheses.

Qualitative vs. quantitative data

Qualitative data are primarily words, narratives, and observations; quantitative data are primarily numbers and measurable variables.

Inductive reasoning

Begins with observations and develops concepts or explanations from patterns in the data.

Deductive reasoning

Begins with a theory or hypothesis and tests it against evidence.

Inductive vs. deductive

Inductive reasoning moves from observations toward an explanation; deductive reasoning moves from a theory or hypothesis toward testing with evidence.

Participant observation

A qualitative method in which researchers observe people in a natural social setting and may participate in group activities.

Strength of participant observation

Captures behavior, social interactions, and context that participants may not fully describe in interviews.

Limitation of participant observation

The researcher's presence may change behavior, and findings from one setting may not generalize widely.

Intensive interviewing

A qualitative method using open-ended questions and follow-up probes to understand experiences, perspectives, and explanations.

Strength of intensive interviewing

Provides rich detail and allows exploration of unexpected responses.