Research Methods Terms

Cognitive dissonance

(unpleasant state of arousal when beliefs are inconsistent)

Research methods

(process of steps rooted in the scientific method to systematically collect and analyze information)

Testability principle

theories should be testable and falsifiable

Parsimony principle

when two theories are equally good at predicting behavior, we as scientists should prefer the simpler one

Law

universal statement that allow for reliable predictions

Theory

general statement about the relationship between two variables

Hypothesis

testable prediction derived from theories

Operationalization

defining theoretical constructs in terms of observable and testable measures

Inductive reasoning

reasoning from specific hypotheses to general theories

Deductive reasoning:

reasoning from general theories to specific hypotheses

Construct validity:

the degree to which your measured variables reflect the hypothetical variables (constructs) of interest

Nominal scales:

categorical scales with non numerical names

Ordinal scales:

scales that involve order/ranking but no distinct distance between categories

Interval scales:

have order/ranking and relative size/distance between units

Ratio scales:

interval scales but with a true and meaningful zero

Test retest reliability:

degree to which participants respond the same way to measurement at different times

Interrater reliability:

the degree to which different observers agree on a measurement/judgement

P value:

acceptable level of probability of falsely rejecting the null hypothesis

File drawer problem:

many studies that are conducted are never reported and results of those may differ from those that are

Publication bias:

tendency to report positive, not null findings

Internal validity:

the extent to which a set of research findings provides compelling evidence about causality

Covariation:

changes in one variable correspond with changes in another

Temporal sequence:

changes in causal variable must come before changes in outcome variable

Manipulation:

systematically altering the levels of the IV to see if it causes changes in the DV

Manipulation check:

survey items meant to insure that the IV has been changed across conditions

Noise:

extraneous variables that are evenly distributed across conditions but influence the DV and make it more difficult to detect an effect of the IV

External validity:

degree to which research findings provide an accurate description of what happens in the real world (generalizability)

Applied science:

applying findings from experiment to real world

Basic science:

using findings to understand processes and develop/support theory

Non experimental:

observational, correlational, no comparing across groups

Quasi experimental:

comparing groups but no random assignment

Experimental:

comparing groups with random assignment

Person confounds:

third variables associated with individual differences that account for the relationship between IV and DV

Environmental confounds:

third variables associated with context/environment that account for the relationship between IV and DV

Pseudoexperiment:

exposing people to a variable interest and then measuring the outcome variable of interest

Pre-post test:

tests participants before and after a treatment

Regression to the mean:

tendency for people who score high/low on a particular measure to score closer to the mean on the next testing

Experimenter bias:

experimenters’ expectations about their hypotheses influence their observations

Selection bias:

sampling from an unrepresentative sample

Nonresponse bias/self selection bias:

those who choose to respond to a survey might be systematically different from those who chose not to

Determinism principle:

all events have meaningful and systematic natural causes

Empiricism principle:

basing knowledge of causal relationships on thorough observation

Equifinality:

same phenomenon often has multiple different causes

Convergent validity:

degree to which different measures of same construct correspond

Discriminant validity:

measures of constructs that are not theoretically related should not correspond

Concurrent validity:

measures of constructs should correspond with measures of theoretically related constructs

Predictive validity:

measures of constructs should predict measures of theoretically related constructs

Content validity:

the degree to which a measure represents all dimensions of a given construct

Face validity:

the degree to which a measure appears to be relevant to the construct that it intends to measure

Reliability:

the degree to which a measurement/set of research findings produce consistent results

Validity:

how accurately your measures and design test what they are meant to test

Internal consistency:

the degree to which participants respond in the same way to multiple items in a single scale intended to measure one construct

Cronbach’s alpha:

measure similar to the correlation between items on a scale, high alpha means high internal consistency

Operational confounds:

manipulation does not adequately change construct of interest

Procedural confounds:

experimenter unknowingly manipulates more than one thing

Participant expectancies:

behaving in ways that are consistent with the experimenter’s hypotheses

Demand characteristics:

characteristics of a measurement that reveal the experimenter’s hypotheses

Participant reactance:

behaving in ways that are inconsistent with the experimenter’s hypothesis

Evaluation apprehension:

behaving in ways that make one look socially desirable

Artifacts:

variable that is held constant in a study but might define the context in which an observed relationship occurs

Heterogeneous attrition:

systematic difference between participants who are left in the study after withdrawal threatens internal validity

Homogenous attrition:

no systematic difference between participants who are left in the study after withdrawal threatens external validity