Chapter 7 class version(1)
description
Transcript of Chapter 7 class version(1)
Chapter 7: Control Techniques
October 23, 2012
Roadmap
• Discuss Reflection Assignment #2– Due Nov. 1 via BlackBoard
• Coming Up: Exam #3 next Tuesday, Oct. 30• Quick review• Chapter 7
Factorial Designs
• When to use• Main and interaction effects• Effect patterns in data displays
OverviewControl Techniques
• Control at the beginning of experiment– Random assignment– Matching
• Control during the experiment– Counterbalancing– Controlling for participant effects– Controlling for experimenter effects
Create equivalent experimental groups
Treat groups the same during the experiment
Methods for Matching Participants
• Holding variables constant• Building the extraneous variable into the
design• Yoked control• Equating participants
Matching by Holding Variables Constant
• Hold extraneous variable constant for all groups in the experiment
• All participants in each treatment group will have same degree or type of extraneous variable
• Requires selection criteria for participant sample
Build Extraneous Variable into the Research Design
• Especially useful if you are interested in:– Differences produced by the levels of the
extraneous variable– Interaction between levels of IV and levels of
extraneous variable
• Sound familiar?– What kind of research design would this be?
Example: Effect of a study skills intervention on grades in a Quantitative Methods course
Intensive tutoring program Study packets (usual)
But the literature suggests that learning style may affect how students respond to different study skills training methods.
Learning style is a potential confounding extraneous variable….but we can build it in to the design!
Learning Style
Visual Auditory Kinesthetic
Intensive tutoring program
Study packets
Inte
rven
tion
A
B
Matching by Equating Participants
Precision control• Match each participant in experimental group
with a participant in control group on variable(s) of concern
• Example: Scholtz (1973) compared defense styles in suicide attempt vs. no attempt
Matching by Yoked Control
• Match participants on the basis of the sequence of administering an event
• Each control participant is “yoked” to an experimental participant
• Controls for the possible influence of participant-controlled events
• Example: Sklar & Anisan (1979)– stress and immune response
CONTROL DURING THE EXPERIMENTwww.xkcd.com
Control During the Experiment
• Must treat the different groups in the same way during the experiment, except for administration of the IV
• Why is this important?
• Control during the experiment– Counterbalancing within-participants
designs
– Controlling for participant effects– Controlling for experimenter effects
Counterbalancing
• Used to control for sequencing effects in a repeated measures (aka within-subjects) design
• Sequencing effects occur when participants participate in more than one condition
• Two types of sequencing effects– Order effect– Carryover effect
Counterbalancing
• Order effect– “Arises from the order in which treatment
conditions are administered to participants”– Treatment/experiment exposure can influence
performance on subsequent tasks and measures– Most common:• Practice effect• fatigue
Counterbalancing
• Carryover effect– Performance in one condition is affected by the
condition that precedes it– Example: Participant receives active drug before
the placebo, and the residual effects are still present during placebo condition
• One strategy: “wash-out” period
Counterbalancing Techniques
• Randomized counterbalancing• Intrasubject counterbalancing
• Complete counterbalancing• Incomplete counterbalancing
individual
group
Types of Counterbalancing
• Randomized counterbalancing– Sequence order is randomly determined for each
individual– Just like random assignment to conditions– You do not decide the sequence, must use a random process to decide
order
Types of Counterbalancing
• Intrasubject Counterbalancing– When each participant receives all levels of the IV
more than one time– Have participants take conditions first in one
order, then again in the reverse order
– Disadvantage: Participant burden is increased• Must complete each condition more than once
Types of Counterbalancing
• Complete and Incomplete counterbalancing– Group counterbalancing– Determine possible sequences– Randomly assign to sequence such that sequences
are distributed across groups rather than individuals
Participant Effects
• Demand characteristics– Cues in the experiment that might influence
participant behavior
• Positive self-presentation– Motivation for participants to present themselves
in a positive light
Control of Participant Effects
• Deception– Giving participants a bogus rationale for the
experiment
• Can range from minor deceit to more elaborate schemes
Classic example: Milgram studies
Control of Participant Interpretation
• Previously discussed methods provide good control for demand characteristics of study
• But how do we know what participants’ perceptions of our study are?– Ask them!
Control of Participant Interpretation
• Retrospective Verbal Reports: after experiment– Disadvantage: Participants might forget
perceptions by the end of the study
• Concurrent Verbal Reports: during experiment– Solomon’s Sacrifice Groups– Concurrent probing– Think-aloud technique
Control of Experimenter Effects
• Experimenter effects– The biasing influence that can be exerted by the
experimenter
• Data Recording errors--control– Be careful– Multiple observers and data recorders– Keep experimenter blind to participants’ conditions– Electronic or mechanical data recording*
Control of Experimenter Effects
• Experimenter Attribute Errors– Some experimenters, because of their attributes,
produce more of an effect than other experimenters
• Control technique:– Experimenters should run all conditions– Experimenters same on characteristics that might
affect DV
Control of Experimenter Effects
• Experimenter Expectancy Errors– Experimenter’s expectations about the study
influence participant responses
Control techniques:• Blind technique• Partial blind technique• Automation
Ideal:Control Participant AND Experimenter Effects
• Double-Blind Placebo Method– Participant and experimenter blind to condition– “Devise manipulations that appear essentially
identical to research participants in all conditions”– Example: Compare drug to identical sugar pill
(placebo)