Transcript Chapter 3

Experimental Psychology
PSY 433
Chapter 3
Experiments -- Designs
Two Kinds of Influence on DV
 Systematic – affects the DV in the same way
each time.


Can introduce bias into results.
When it occurs due to the manipulation of the
IV we call it an “effect.”
 Random – varies and thus affects the DV
differently from observation to observation.


Can introduce “noise” into results.
Typically makes it more difficult to observe
systematic influences.
Confounding Variable
 Any variable, besides the IV, that can
influence the DV.
 A potential cause for the experimental effect,
other than the IV.

An alternative explanation for observed
findings in a study.
 Any variable whose values change
systematically across levels of the IV.
Control Variable
 A variable whose values remain the same
across levels of the IV (e.g., room temp, light
levels, time-of-day, etc).
 A goal in experimentation is to control as
many variables as possible, to eliminate their
potential effect on the DV.

Eliminates both confounds and noise.
 Except for the IV, each subject should have
as closely similar an experience in the
experiment as possible.
Random Variable
 Variable whose values vary randomly in an
unbiased way across levels of the IV.
 Random variables are usually created by the
process of random assignment to levels in
the experiment.
Subject Variable
 A personal characteristic

Sex, height, weight, age, education, ethnicity,
socio-economic status, etc.
 Sometimes called a “demographic” variable.
 Data describing subject characteristics should
be collected in every study.
 Can be controlled or left to vary
unsystematically (through random
assignment to levels).
Control Group
 The group that receives no intervention,
“zero” or “the absence of” the IV.
 Eg, the placebo group in a drug experiment.
 The group that serves as a baseline to
compare with the behavior of the
experimental groups.
Experimental Groups (Levels)
 The groups that receive non-zero values of
the IV, the groups receiving some
intervention.


The drug groups in a drug study.
Also called “levels” of the IV.
 The performances of these groups are
compared with the performance of the
control group.
Conceptual Definition
 Definition of a variable at the conceptual or
idea level.
 Tends not to be very precise, more general.
 May be expressed in terms of constructs
(intervening variables).
Operational Definition
 Specifies the operations or procedures
necessary to measure the variables
relevant to the conceptual definition.
 Very precise.
 Not general or vague at all.
 Tells how the dependent variable was
measured.
 There may be many operational definitions
(and DVs) for a single conceptual definition.
ODs and CDs - Example 1
 Conceptual - Amount of alcohol
consumed
 Operational - # of beers in 1 hour (0,1,2,3)
 Operational - grams of alcohol/kg body
weight
 Operational – blood alcohol content, BAC
(mg alcohol/deciliter blood).
ODs and CDs - Example 2
 Conceptual - Helping behavior
 Operational - # of people who help a “victim”
in a defined context.
 Operational - duration of helping behavior.
 Operational - # seconds before helping
occurs (latency).
Complex Designs
 More than one IV

Left/Right and 1, 5, or 10 spaces from
center
 More efficient than single IV experiments
 Gives more information
 Allows analysis of main effects and
interactions.
Complex Designs - Terminology
 An IV is called a factor
 number of numbers = how many IVs there are
 values of numbers = how many levels each IV has
 Examples:
 “2 X 2 design” (two IVs, each with 2 levels)
 “2 X 3 design” (first IV has 2 levels, second IV has 3
levels)
 “2 X 8 design” (first IV has 2 levels, second IV has 8
levels)
 “2 X 2 X 4 design” (first IV has 2 levels, second IV
has 2 levels, third IV has 4 levels).
Main Effects
 There is one potential main effect for
each IV


A 2 X 8 design has two possible main
effects
A 2 X 2 X 4 has three possible main
effects
 A main effect is present if an IV had a
significant effect on the experiment’s
outcome (regardless of the effects of the
other IVs).
Interactions
 Please memorize: “An interaction occurs if
the effect of one IV varies depending on the
level of the other IV”
Designing Experiments
 Two general types of designs
 Between-subjects (independent groups) =
each group gets one level of the IV
 Within-subjects (repeated measures) =
each subject gets all levels of the IV
 Equivalency of groups at each level is built-
in for within-subjects and achieved by
random assignment for between-subjects
 Within designs require fewer subjects
because there is less variability between
levels.
Order Effects
 Order effects (practice effects) =
experiencing one level affects behavior in
another level

Effects of practice, boredom, fatigue
 Example: Does content (biology text vs.
novel) affect proofreading speed? Order is
Biology-Novel
 Order effects are controlled in withinsubjects designs by randomizing or
counterbalancing the presentation orders.
Differential Carryover Effects
 Carryover effects, differential/asymmetrical
transfer effects occur when experiencing
one level affects performance on the next.
 The effect of the first level on the second
level differs depending on which comes
first.

Effect of B following A ≠ effect of A following
B
 Confound occurs when one level
consistently precedes the other.
Order Effects in Proofreading
Group 1 (no practice)
Biology
1
(practice)
Novel
2
(no practice)
Novel
1
(practice)
Biology
2
Group 2
Differential Carryover Effects in
Problem Solving
Group 1
Group 2
(no practice)
Neutral
instructions
1
(practice)
Special
instructions
2
(no practice)
Special
instructions
1
(practice)
Neutral
instructions
2
Other Considerations
 Mixed designs (some between, some
within)
 Small-n designs
 Matched groups designs
 Demand characteristics = cues that tell
subjects how they should behave (eg, drug
studies)
 Blind and double-blind procedures
 Internal and external validity
 Quasi experiments.