Chapter 9-12

Experimental Design

The general structure of the experiment

Experimental Design

Made up of things such as the number of treatment conditions and whether the subjects in all conditions are the same or different individuals

Experimental Design

The design of an experiment details an experimenter’s plan for testing a hypothesis

Experimental Design

The design is the experiment’s structure or floor plan, not the experiment’s specific content

Between-Subjects Designs

Different subjects take part in each condition of the experiment

Between-Subjects Designs

A subject participants in only one condition of the experiment

Subjects

The representativeness of our sample determines whether we can generalize our results to the entire population from which the sample was drawn

Effect Size

A statistical estimate of the size or magnitude of the treatment effect

Subjects Effect Size

Determines the number of subjects required to detect a treatment effect

Two-Group Design

When only two treatment conditions are needed, the experimenter may choose to form two seperate groups of subjects

Two-Independent Groups Design

Subjects are placed in each of two treatment conditions trough random assignment

Two-Independent Groups Design

We assign subjects to one of two levels of the independent variable

Two-Independent Groups Design

A design where there is one IV with two levels and subjects are randomly assigned to one of the two conditions

Random Assignment

Means that every subject has an equal chance of being placed in an of the treatment condition

Control Group

The subjects in a control condition

Control Group

Tells us how subjects ordinarily perform on the dependent measure

Control Group

A point of comparison

Two-Experimental-Groups Design

Used to look at behavior differences that occur when subjects are exposed to two different values or levels of the IV

Matching

Used to create groups that are equivalent on potentially confounding subject variables

10 to 20

You should have at least ______ subjects in each treatment condition to detect a strong treatment effect

Random Assignment

equally distribute subject variables between the treatment groups to prevent them from confounding an experiment

Experimental Conditions

We apply a particular value of our independent variable to the subjects and measure the dependent variable

Experimental Group

Subjects in an experimental condition

Control Condition

Used to determine the value of the dependent variable without an experimental manipulation of the independent variable

Two-Matched-Groups

There are also two groups of subjects, but the researcher assigns them to groups by matching or equating them on a characteristic that will probably affect the dependent variable

Precision Matching

We insist that the members of the matched pairs have identical scores

Range Matching

We require that the members of a pair fall within a previously specified range of scores

Rank-Ordered Matching

The subjects are simply rank ordered by their scores on the matching variable, and subjects with adjacent scores then become a matched pair

Rank-Ordered Matching

We do not specify an acceptable range between members of each pair

Multiple Groups Design

There are more than two groups of subjects and each group is run through a different treatment condition

Multiple Groups Design

A between-subjects design with more than two levels of an Independent variable

Multiple-Independent Groups

The most commonly used multiple groups design

Block Randomization

The experimenter creates random sequences of each experimental condition, and subjects are randomly assigned to fill each treatment block

Pilot Study

To pretest selected levels of an independent variable before conducting the actual experiment

Pilot Study

Like a mini-experiment in which treatments are tested on a few subjects to see whether the levels seem to be appropriate or not

Pilot Study

Allows you to make changes before you invest the time and resources in a large-scale experiment

Pilot Study

A trial run of the experiment that uses a few subjects

Factorial Designs

Designs in which we study two or more independent variables at the same time

Factorial Designs

Can provide information about both treatment and interaction effects

Factors

The independent variable in the factorial designs

Two-Factor Experiment

The simplest factorial design that only has two factors

Main Effects

The action of a single independent variable in an experiment

Main Effects

The action of a single IV on the DV

Interaction

Present if the effects of one independent variable changes across the levels of another independent variable

Interaction

When the effects of one factor depend on another factor

Interaction

The joint effect of two or more IVs on the DV

Higher-Order Interactions

An interaction among three or more IVs

Design Matrix

Quick and easy way o create a visual image of your experimental design

Factorial Design

Combines several one-factor experiments and allows us to study interactions

Crossover Interaction

The effects of each factor completely reverse at each level of the other factor

Within-Subjects Design

A design in which each subject serves in more than one condition of the experiment

Repeated-Measure Design

Also known as _________ because subjects serve in more than one condition of the experiment and are measured on the dependent variable after each treatment

Power

An experiment’s ability to detect variable’s effect on the dependent variable

Within-Subjects Factorial Design

A factorial design in which subjects receive all conditions in the experiment

Within-Subjects Factorial Design

Assigns subject to all levels of two or more independent variables

Mixed Design

A design that combines within and between subjects variables in a single experiment

Order Effects

Subjects’ responses might differ from one treatment to another just because of the position, or order, of the series of treatments

Order Effects

Positive (practice) and negative (fatigue) performance due to a

condition’s position in a series of treatments

Fatigue Effects

Subjects get tired which can cause their performance to decline as the experiment goes on

Fatigue Effects

Performance declines on the DV due to tiredness, boredom, or irritation

Practice Effects

As subjects become familiar with the experiment, they could relax and do a little better

Practice Effects

Subjects on the DV may improve across the conditions of a within-subjects experiment

Practice Effects

May be due to relaxation, increased familiarity with the equipment or task, development of problem-solving strategies, or discovery of the purpose of the experiment

Progressive Error

All of the changes, both positive and negative

Progressive Error

As the experiment progresses, results are distorted

Progressive Error

Includes any changes in the subjects’ responses that are caused by testing in multiple treatment conditions

Linear Progressive Error

Effects can be plotted as a straight line

Nonlinear Progressive Error

Effects can be plotted as a curve

Counterbalancing

Controls order effects

Counterbalancing

Controls order effects by distributing progressive error across different treatment conditions of the experiment

Subject-By-Subject Counterbalancing

Controls progressive error for each subject

Across-Subjects Counterbalancing

Distributes progressive error across all subjects

Subject-By-Subject Counterbalancing

A technique for controlling progressive error for each individual subject by presenting all treatment conditions more than once

Reverse Counterbalancing

A technique for controlling progressive error for each individual subject by presenting all treatment conditions twice, first in one order, then in the reverse order

Reverse Counterbalancing

We administer treatments twice in a mirror-image sequence

Block Randomization

Present each treatment several times, resulting in a sequence containing a number of randomized blocks

Block Randomization

Commonly used in cognition, perception, and psychophysics experiments in which treatment conditions are relatively short

Block Randomization

Researchers assign each subject to several complete blocks of treatments

Block

Consists of a random sequence of all treatments

Across-Subjects Counterbalancing

Used to distribute the effects of progressive error so that if we average across subjects, the effects will be the same for all conditions of the experiment

Across-Subjects Counterbalancing

A method for controlling order effects in research by assigning different participants to different orderings of conditions

Complete Counterbalancing

Using all possible sequences of the conditions and using every sequence the same number of times

Complete Counterbalancing

Different subjects are assigned to the sequences at random, and we give each sequence to an equal number of subjects

Partial Counterbalancing

We use this procedure when we cannot do complete counterbalancing but still want to have some control over progressive error across subjects

Partial Counterbalancing

We present only some of the possible (N!) orders

Partial Counterbalancing

Controls progressive error by using some subset of the available order sequences

Randomized Partial Counterbalancing

Simplest partial balancing procedure

Randomized Partial Counterbalancing

When there are many possible order sequences, we can randomly select out as many sequences as we can have subjects for the experiments

Latin Square Counterbalancing

A matrix, or square, of sequences is constructed that satisfies the following condition: each treatment appears only once in any order position in the sequence

Latin Square Counterbalancing

Controls adequately for progressive error caused by order effects because each treatment condition occurs equally often in each position

Carryover Effects

The effects of some treatments will persist, or carry over, after the treatments are removed

Order Effects

Emerge as a result of the position of a treatment in a sequence

Carryover Effect

A function of the treatment itself

Balanced Latin Square

Each treatment condition (1) appears only once in each position in the order sequence and (2) precedes and follows every other condition an equal number of times

Large N Design

Compares the performance of groups of subjects

Small N Design

Studies one or two subjects, often using variations of the ABA reversal design

Small N Design

Test only one or a very few subjects

Large N Designs

Lack precision because they pool, or combine, the data from many different subjects to reach conclusions about the effects of independent variables

Large N Designs

The results of data aggregated over groups of subjects might not really be a good reflection of the reactions of individual subjects

Aggregate Effects

The pooled findings from many subjects