Experimental Design Notes Ii

Front 1

Halo Effect

Back 1

the tendency for an impression created in one area to influence opinion in another area.

Front 2

Hawthorne Effect

Back 2

the alteration of behavior by the subjects of a study due to their awareness of being observed.

Front 3

John Hentry Effect

Back 3

In a controlled social experiment if a control is aware of their status as members of the control group and is able to compare their performance with that of the treatment group, members of the control group may actively work harder to overcome the "disadvantage" of being in the control group

Front 4

Reactive or interaction effect of testing

Back 4

a pretest might increase or decrease a subject's sensitivity or responsiveness to the experimental variable.

Front 5

Multiple treatment interference

Back 5

as multiple treatments are given to the same subjects, it is difficult to control for the effects of prior treatments.

Front 6

Reactive arrangements

Back 6

this refers to the artificiality of the experimental setting and the subject's knowledge that he is participating in an experiment.

Front 7

Unobtrusive measures

Back 7

measures that don't require the researcher to intrude in the research context.

Front 8

Static Group Comparison

Back 8

This is a two group design, where one group is exposed to a treatment and the results are tested while a control group is not exposed to the treatment and similarly tested in order to compare the effects of treatment.

Front 9

within-groups variability

Back 9

the variability of the observations within a group combined across groups

Front 10

between-groups variability

Back 10

variability of the observations between the groups

Front 11

F-Value

Back 11

F = MSbetw/MSwith

Front 12

Effect Size Measures

Back 12

indicate the strength of association between X and Y, that is, the relative strength of the group effect

Front 13

η2 (eta squared)

Back 13

ranging from 0 to +1.00, is known as the correlation ratio (generalization of R2 ) and represents the proportion of variation in Y explained by the group mean differences in X. Positively biased statistic (overestimates the association).

Front 14

ω2 (omega squared)

Back 14

interpreted similarly to eta squared (specifically proportion of variation in Y explained by the group mean differences in X) but which is less biased

Front 15

f developed by Cohen (1988)

Back 15

can take on values from 0 (when the means are equal) to an infinitely large positive value. This effect is interpreted as an approximate correlation index but can also be interpreted as the standard deviation of the standardized mean

Front 16

Confidence Intervals

Back 16

useful in providing an interval estimate of a population parameter (i.e., mean or mean difference); these allow us to determine the accuracy of the sample estimate.

Front 17

Power

Back 17

the probability of correctly rejecting a false null hypothesis. s primarily a function of α, sample size, and effect size.

Front 18

Analogue Experiment

Back 18

An experiment that manipulates a cause that is similar to another cause of interest in order to learn about the latter cause.

Front 19

Blocking

Back 19

The process of dividing units into groups with similar scores on a blocking variable, each group having the same number units as the nubmer of conditions

Front 20

Carryover Effects

Back 20

The effects of one treatment do not end prior to the administration of a second treatment, so that the effects observed in the second treatment include residual effects from the first

Front 21

Ceiling Effect

Back 21

Responses on a variable closely approach the maximum possible response so that further increases are difficult to obtain

Front 22

Counterbalancing

Back 22

In within-participants designs, arranging the order of conditions to vary over units so that some units are given Treatment A first but others are given Treatment B first.

Front 23

Double-Blind Study

Back 23

An experiment in which both the treatment provider and treatment recipient are unaware of which treatment or control condition is being administered, primarily used in medical clinical trials.

Front 24

Ethnography

Back 24

Unstructured exploratory investigation, usually of a small number of cases, of the meaning and functions of human action, reported primarily in narrative form.

Front 25

Floor Effect

Back 25

Responses on a variable approach the minimum possible score so that further decreases are difficult to obtain.

Front 26

Matching

Back 26

Sometimes synonymous with blocking, sometimes more specific to imply blocks in which units are exactly equal (rather than just similar) on a matching variable

Front 27

Meta-Analysis

Back 27

A set of quantitative methods for synthesizing research studies on the same topic

Front 28

Nested Designs

Back 28

Designs in which units are exposed to some but not all conditions

Front 29

Nesting

Back 29

When some units are grouped together into aggregate units, units are said to be nested within aggregates

Front 30

Odds Ratio

Back 30

An effect size measure for the difference between groups on a dichotomous outcome

Front 31

Omitted Variables

Back 31

Variables that are not in a model or an analysis that influence both the cause and the effect and so may cause bias.

Front 32

Order Effects

Back 32

The outcome of a study is affected by the order in which the treatments were presented

Front 33

Practice Effects

Back 33

Participants become better at something the more often they do it, a potential problem within-participants designs in which repeated tests aregiven to the same participants

Front 34

Propensity Score

Back 34

A predicted probability of group membership based on observed predictors, usually obtained form a logistic regression

Front 35

Purposive Sample

Back 35

A method by which units are selected to be in a sample by a deliberate method that is not random

Front 36

Standardized Mean Difference Statistic

Back 36

An effect size measure for continuous variables, computed as the difference between two means divided by the variability of that difference

Front 37

Testing Effects

Back 37

Effects due to repeated testing of participants over time

Front 38

Type I Error

Back 38

Incorreclty rejecting a true null hypothesis; in an experiment, this usually implies concluding that there is an effect when there really is no effect

Front 39

Type II Error

Back 39

Failing to rejct a false null hypothesis; in an experiment, this usually implies concluding that there is no effect when there really is an effect

Front 40

Within-Participants Design

Back 40

The same units are studied in different conditions