Research Methods Exam 3

social norms

widely held beliefs about what most people in a group believe or value

experiment

a study in which at least one variable is manipulated and another is measured

manipulated variable

a variable whose levels are manipulated by the researcher

measured variable

a variable whose levels are observed and recorded

frequency claim

describe rate or degree of single variable, often given in percentages

association claim

one variable correlates with another

causal claim

one variable causes change in another

independent variable

the variable that’s manipulated

condition

each level of the independent variable

dependent variable

variable that’s measured

control variable

a variable that a researcher holds constant on purpose

random assignment

each participant has an equal chance of ending up in each condition, individual differences was out across conditions

covariance

are different levels of the IV associated with different scores on the DV

temporal precedence

the IV is manipulated before the DV is measured

internal validity

rule out confounds

systematic variance

another variable fluctuates with levels of IV

unsystematic variance

another variable fluctuates across all levels of the IV

confound

when another variable varies systematically with an IV, threaten internal validity because they are alternative explanations

design confound

an experiment is designed so that another variable varies systematically with the IV

selection effect

participant characteristics vary systematically with IV

correction for design confounds

control variables

correction for selection effects

random assignment or matched groups

independent groups designs

each participant is assigned to one level of the IV

within-groups designs

each participant completes all levels of the IV

types of independent group designs

posttest-only and pretest/posttest

types of within groups designs

concurrent measures and repeated measures

order effects

exposure to one condition changes responses to a later condition

correction for order effects

counterbalancing with some participants completing levels of the IV in different sequences, order of completion is randomly assigned

maturation

when change occurs simply because time has passed

history threat

an external factor systematically affecting most participants systematically

regression to the mean

mean starts at an extreme, making it likely that it will become less extreme overtime

correction to maturation, history, and regression to the mean

include a comparison group

attrition

when there are more participants for pretests than posttest because participants drop out

correction for attrition

compare pretest scores of people who dropped out with those who stayed in; excluded anyone with incomplete data

testing threat

when there’s a change in the participant’s score because they’ve taken the test more than once

instrumentation threat

changes because a measuring instrument changes over time

corrections for testing threats and instrumentation

consider using post-test only design

demand characteristics

participants guess study’s purpose and change behavior

observer bias

researcher’s expectations influence interpretation of results

correction to observer bias

conduct a double-blind study

placebo effects

change due to believing they are receiving treatment

correction to placebo effects

double-blind, include placebo comparison group (double blind placebo control study)

simple experiments

have only one independent variable that may have different levels within that variable

factorial designs

two or more independent variables, allows us to ask more complex questions

format for describing factorial designs

number of levels in factor 1 x number of levels in factor 2 (etc.)

finding total conditions

multiply the numbers in the format for factorial designs

interaction effect

the effect of one independent variable depends on the level of the other independent variable

there is a “difference of differences”

the mathematical way to describe an interaction of two independent variables

cells

a condition in an experiment

quasi-experiements

has an “independent” and dependent variable, but lacks full experimental control

types of quasi-experiments

non-equiv control group posttest only design

non-equiv control group pretest/ posttest design

interrupted time series design

nonequivalent control group interrupted time series design

interrupted time series design

dependent variable is measured before, during, and after the independent variable, there is no control group

how quasi experiments differ from experiments

no random assignment, subjects are selected on the values of the IV, they have less internal validity

how quasi experiments differ from correlations

only select a certain range of values of an IV, more control

direct replication

when the original study is repeated as closely as possible to see if the same result is reached; same conceptual variables, same operationalization

conceptual replication

same conceptual variables but there are different operationalizations

replication-plus-extension

same conceptual variables, plus new variable(s)

meta-analysis

a way of mathematically averaging effect sizes of studies to see what conclusion the weight of the evidence supports