unit 3 simplified notes

Comparison Groups/Conditions

Different groups of participants that experience different levels of the independent variable.

Concurrent-Measures Design

Participants see all conditions at the same time and choose their preference.

Construct Validity

Does the study measure what it claims to measure?

Control Group

The group that receives no treatment or a neutral condition.

Control Variable

Variables that are kept constant so they do not influence results.

Covariance

Two variables change together. When the independent variable changes, the dependent variable also changes.

Dependent Variable

The variable measured to see if it changes.

Design Confound

Another variable accidentally changes along with the independent variable and could explain the results.

External Validity

Can the results apply to other people, places, or times?

Independent Variable

The variable manipulated by the researcher.

Internal Validity

How confident we are that the independent variable actually caused the change in the dependent variable.

Order Effects

The order of conditions affects results.

Post-Test Only Design

Participants experience one level of the independent variable and measured once afterward.

Pre-Post Design

The dependent variable is measured before and after the treatment.

Repeated-Measures Design

Participants experience every level of the independent variable.

Selection Effects

Groups are different from the start, which may affect results.

Statistical Validity

Are the data analysis and statistical conclusions accurate?

Temporal Precedence

The cause must happen before the effect.

Treatment Group

The group that receives the independent variable being tested.

Unsystematic Variability

Random factors that influence participants, but are not part of the study.

Confound

an additional variable that could explain empirical findings. The presence of a confound threatens the internal validity of the study

how researchers can design studies to prevent internal validity threats

use random assignment to avoid selection bias.

include control groups to rule out alternative explanations; also useful for detecting history and maturation effects.

keep environment and interactions consistent across all participants

use single or double blind studies

measure participants before and after the intervention (pre-post test design)

control do confounding variables

keep studies engaging so minimize participant dropout

use alternate forms of testing to avoid testing effects

use reliable testing measurement to avoid instrumentation threats

do shorter time frames to reduce maturation effects

how researchers can design studies to minimize possible obscuring factors

maximize statistical power

use sensitive research designs

improve measurement quality

strengthen the independent variable

control extraneous variables

increase variability

minimize attrition (drop outs)

reduce random error

use multiple measures

maturation threat

changes in the variable that emerge spontaneously over time

history threat

when some external or real-world event affects members of one of the experimental groups

maturation vs history threats

maturation: personal development & people mature

history: contextual development & history happens around people

countermeasure: include a control group

regression

a statistical concept, when an extreme level in an observed variable is likely to return to the mean level in the future

attrition threat

reduction in participants from pre to post test

countermeasures for attrition threats

remove extreme data points from attritting participants

investigate why participants drop-out and adapt incentives

use statistical methods that can handle missing data

testing threat

a change in a participant’s response on the DV as a result of experiencing the DV more than once (fatigue effect)

countermeasures for testing threat

include a control group

use different tests/instruments

instrumentation threats

when an instrument or measurement tool changes over time

Ex: a rater changing their rating criteria over time

Ex: using non-equivalent instruments to measure the same construct

countermeasures for instrumentation threats

rigorous training of raters

use post-test only design

testing vs instrumentation threats

testing threat: participant changes; testing activity changes, not the test itself

instrumentation threat: measure change, instrument changes

null effects

either not enough between variance or too much within variance

not enough between variability

insensitive measures

ceiling and floor effects

weak manipulations

insufficient power

too much within variability

measurement error

individual differences

situation noise

insufficient power

insensitive measures

measure hasn’t been operationalized in a way to distinguish differences in the conceptual variable

Ex: Pass or Fail versus A+ to F grading scale

ceiling and floor effects

questions are “too easy” or “too hard”

“too agreeable” or “too disagreeable”

insufficient variability

insensitive measures vs ceiling/floor effects

both have some kind of scale limitation but insensitive measures has precision/discrimination problem while ceiling/floor effects has a boundary problem

weak manipulation

manipulation of the IV did not have an effect or change the thing that wanted to change

measurement error

measurement error of psychological instruments introduces noise to the dependent variable

use reliable and valid measures

individual differences

can restrict the sample

can account for individual differences with a within-groups or pre-post design

situation noise

any kind of external distraction that could cause variability within-groups that obscure between-groups differences

control the surroundings of the experiment to minimize situation noise

insufficient power

doesn’t have enough ability to detect a real effect

how to maximize power

have a large sample size

use strong experimental manipulations

study theoretically plausible solutions

reduce causes of variability that can cause error

Factorial design

A research design with at least two independent variables. They are either manipulated or specifically selected

Reasons to use factorial design

To test theories and test limits (external validity)

Interaction effect

Degree to which an independent variable effect depends on another independent variable.

Interaction effect with words

Ex: Caffeine helps those with low sleep and has little to no effect on people with high sleep.

Interaction effects with tables

Helps to see patterns across conditions

Interaction effects with graphs

Show lines that are not parallel. Non-parallel lines = interaction is present

Between groups factorial design

Each participant is in only one condition. Ex: a 2×2 design with 4 separate groups, each person only experiences one combinations.

Within groups factorial design

Each participant experiences all conditions

Mixed factorial design

Includes both between and within groups.

Example: sleep level (between groups) each person is assigned to either high or low sleep. Caffeine level (within groups) everyone completes both caffeine and no caffeine trials.

Increasing the number of IV levels

Adding more levels to a variable increases how detailed your data can be.

Increasing the number of IV

Adding another independent variable increases the design’s complexity