Control Techniques

Chapter 9
Control Techniques ♣
Back to Brief Contents
 Introduction 
 Randomization 
 Matching 
 Counterbalancing 
 Control of Participant Effects 
 Control of Experimenter Effects 
 Likelihood of Achieving Control
9.0 Introduction
Back to Chapter Contents
 Goal of experimentation ― identify the causal

effect of the IV
 Must have internal validity to do this
 Internal validity requires control of confounding

variables
 Ways of achieving control
— Design of the experiment
— Statistical adjustments
— Incorporate control techniques into the research
design
9.1 Randomization -1
 Randomization is a statistical control technique

to equate groups of participants
 This is the most important and basic control
technique
 Random selection —selecting people at random
from a defined population
 Insures that the sample selected is representative of the
population
representative: sample P have the same characteristics
as the people in the population
 Studies seldom if ever do this because of expense, etc
9.1 Randomization -2
 Random assignment —randomly assigning

participants to treatment groups
 Provides maximum insurance that groups are equal
 Equates groups because every person has an equal

chance of being assigned to each group
 Accomplishes this by randomly distributing the
extraneous variables over the treatment groups
Fig 9.1 Tab 9.1

9.1 Randomization -3(end)
 Random Assignment Exercise

 Randomly assign 40 children to four different drug
conditions using the table of random numbers in
appendix D Exhibit 8.1 1 2
 Logon to www.randomizer.org and use the
randomizer in this site to randomly assign 40 children
to the four drug conditions
9.2 Matching -1
 Uses of any of a variety of techniques to equate

participants in the treatment groups on specific
variables
 Advantages of matching
 Controls for the variables on which participants are
matched
 Increases the sensitivity of the experiment
9.2 Matching -2(end)
 Matching by Holding Variables Constant 

 Matching by Building the Extraneous Variable into the
Research Design 
 Matching by Yoked Control 
 Matching by Equating Participants 
9.2.1 Holding Variables Constant
 Matches on the variable held constant

Fig 9.2
 Disadvantages
 Restricts the population size
 Restricts generalization to the type of participants in
the study
9.2
9.2.2 Building EV into Research Design
 Should be used only when you are interested in the

effect of the effect of the extraneous variable
Fig 9.3
9.2
9.2.3 Matching by Yoked Control
 Controls the temporal relationship between an

event and a response
 Brady (1958)
 Emotional stress → Ulcer
 Stress: Press a lever every 20-sec to avoid shock
 Control: receive the same temporal sequence of shock
9.2
9.2.4 Matching by Equating Participants
 Precision control —match case by case Fig 9.4

 Disadvantages
 Identifying the variables on which to match
 Matching increases as the number of variables on which
to match increases
 Some variables difficult to match
 Frequency distribution control —match on the

overall distribution of the selected variables Fig 9.5
 Disadvantage
 Combination of variables may be mismatched
(e.g.) Age-IQ: (E) Old-high IQ, Young-low IQ
(C) Old-low IQ, Young-high IQ
9.2
9.3 Counterbalancing
 Used to control sequencing effects
 Type of Sequencing effects Fig 9.6

 Order effect Tab 9.2
arising from the order in which the treatment conditions are
administered to P
 Carry-over effect Tab 9.3
occurs when performance in one treatment condition affects
performance in another treatment condition
 Counterbalancing procedures
 Intrasubject Counterbalancing: The ABBA technique 
 Intragroup Counterbalancing 
9.3 Counterbalancing - intrasubject
 Intrasubject or ABBA technique — counterbalances

on a case-by-case basis
 Controls only for linear sequencing effects Tab 9.4
 Nonlinear order effects can be controlled if you use the
ABBA plus BAAB counterbalancing Tab 9.5
 Can’t control nonlinear carry-over effects
9.3 ◄
9.3 Counterbalancing – Intragroup -1
 Intragroup Counterbalancing
(e.g.) ABC ACB BAC BCA CAB CBA
 Incomplete Counterbalancing (Latin square)
Participant Sequence
1 A B D C
2 B C A D
3 C D B A
4 D A C B
9.3 ◄
9.3 Counterbalancing - Intragroup -2
A B E C D D C E B A
B C A D E E D A C B
C D B E A A E B D C
D E C A B B A C E D
E A D B C C B D A E
9.3 ◄
9.3 Counterbalancing - Intragroup -3(end)
1,2,n,3,(n-1),4,(n-2),5,…
2,3,n+1,4,…
(e.g.) A B F C E D ( p. 282, 283 )
B C A D F E
C D B E A F
D E C F B A
E F D A C B
F A E B D C
9.3 ◄
9.4 Control of Participant Effects -1
 Double Blind Placebo Model

 Participants, Experimenter
 Blind: the treatment condition administered to P
 Deception ― giving the P a bogus rationale for the

experiment
 provide hypothesis : unrelated or orthogonal; false but
plausible
9.4 Control of Participant Effects -2(end)
 Perceptual Control, or Control of Participant

Interpretation
 Retrospective verbal report

 postexperimental inquiry
 Concurrent verbal report
 sacrifice groups: stopped at a different point
 concurrent probing: at the end of each trial
 think-aloud technique
9.5 Control of Experimenter Effects -1
 Control of Recording Errors

 Aware of the necessity: ensure the accuracy
 Kept blind
 Mechanical or electronic device
 Control of Experimenter Attribute Errors

 Interact with treatment effect? Tab 9.6
 Minimize: Control Attributes that correlate with DV
9.5 Control of Experimenter Effects -2(end)
 Control of Experimenter Expectancy Error

 The Blind Technique
 The Partial Blind Technique
 Automation

Control Techniques

Uploaded by

Copyright:

Available Formats

Control Techniques

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Control Techniques

Uploaded by

Copyright:

Available Formats

Chapter 9

 Goal of experimentation ― identify the causal

 Internal validity requires control of confounding

 Randomization is a statistical control technique

 Random assignment —randomly assigning

 Equates groups because every person has an equal

Fig 9.1 Tab 9.1

 Random Assignment Exercise

 Uses of any of a variety of techniques to equate

 Matching by Holding Variables Constant 

 Matches on the variable held constant

 Should be used only when you are interested in the

 Controls the temporal relationship between an

 Precision control —match case by case Fig 9.4

 Frequency distribution control —match on the

 Type of Sequencing effects Fig 9.6

 Intrasubject or ABBA technique — counterbalances

 Double Blind Placebo Model

 Deception ― giving the P a bogus rationale for the

 Perceptual Control, or Control of Participant

 Retrospective verbal report

 Control of Recording Errors

 Control of Experimenter Attribute Errors

 Control of Experimenter Expectancy Error

You might also like