Chapter 10 and 12: Experimental Psychology

Simple Experiments

One dependent and one independent variable

Within Subjects (repeated measures)

Every participant is exposed to every condition of the independent variable

advantages of within subjects design

1. allows the use of fewer subjects to obtain the same number of observations

2. allows for greater statistical power than a between subjects design

disadvantages of within subjects design

fatigue, attrition, carryover effects, order effects, practice effects

What are the two types of order effects?

Progressive and carry over

Progressive

changes in participant responses due their cumulative participation in the experiment (boredom, fatigue, etc.)

Carry over effects

Some form of contamination carries over from one condition to the next

Between Subjects (Independent groups)

Two separate groups are exposed to different levels of the independent variable

Disadvantages of between-subjects?

You begin with more error variance due to individual differences, you need more participants

Advantages of between-subjects?

you don't have to worry about order effects, if experiencing one condition makes experiencing the second condition irreversible you cannot do it with a repeated measure

Random groups design

Participants are randomly assigned to a level

Blocked Randomization

Participants are tested in "blocks" of treatment conditions and blocks are repeated until the researcher obtains the number of participants needed

Matched group design

Participants are paired according to some characteristic or matching variable then randomly assigned

(between groups)

- When the sample size is small

Natural groups design

when the groups are based on a naturally occurring characteristic (qualitative and discrete)

Participant variable

something that the participant brings with them (sex, hair color) ALWAYS A BETWEEN SUBJECTS VARIABLE

Distal Variables

those occurring in the distant past relative to the event

Proximal Variables

represented by the things that occur within the current or immediate context that influence the phenomenon of interest (i.e. stress, happiness).

Counterbalancing

involves different techniques to distribute order or sequence effects equally across the levels of the independent variable

Complete counterbalancing

The number of presentation orders equals the number of conditions factorally

Only use this when there are four or less levels to your IV

All of the possible presentation orders

The number of participants= K!

Incomplete counterbalancing

Include techniques a SAMPLE of the possible presentation orders where each level is produced equally as often (takes care of MOST order or sequence effects)

Number of participants= K

Complex Experimental design (Factorial Design)

More than one independent variable (factors)

The simplest factorial design is a 2x2 design (two numbers= factors, multiplied product= levels)

Every letter represents

a factor, the more letters there are it means it is a complex experimental

We use factorial designs to?

understand the complexity of human behavior

To find the total amount of levels,

multiply the levels (IE: factor one has 2 levels, F2 has 3, F3 has 2= 12)

Why do we perform factorial designs?

Better approximate what happens in the real world

It is more efficient (better to conduct one study with two variables than 2 studies with two variables)

Allows us to examine main effects and interactions

Main effect

The effect of ONE factor on the dependent variable, averaging across the levels of the other factor

The number of potential main effects =

the number of factors

In a 2x2 there are

4 factors

2 possible main effects

1 interaction

Interactions

When the effect of one factor depends on the level of another factor

Combination of factors

changes the outcome (alcohol and barbiturates= overdose)

The number of potential interactions depends on the

number of factors (IE: an AxB design has ONE potential interaction because A mixes with B)

AxBxC

three possible main effects, 3 two-way interactions, 1 three way interaction

2x4

two possible main effect, one possible interaction (DOES NOT CHANGE)

If there are no interactions, what do we not do?

report the main effects.

Marginal means

We calculate the MM to see if there are any main effects

If the lines are NOT perfectly parallel

there is an interaction

Mediator variable

is the explanatory variable between the IV and the DV

The MV between phone usage and mistakes while driving is distraction

Moderator variables

are variables that change the strength or direction of the relationship between two variables

Snow would be a moderator variable in the phone usage vs mistakes experiment

This happens during a complex experimental design (always a risk)

What section of an empirical report goes over the procedure of the study?

Method section