PSYC 2520 Chapter Notes - Chapter 8: Longitudinal Study, The Sequence, Internal Validity

6 views2 pages
1 Feb 2013
Department
Course
Professor
PSYC 2520: Introduction to Experimental Psychology
Beginning Behavioral Research: A Conceptual Primer (7th ed. 2012) Rosnow & Rosenthal
Chapter 8: Nonrandomized Research and Causal Reasoning
Prospective data - Data that will be collected in the future
Retrospective data - Data from past situations
The challenge is to try to emulate the causal reasoning of Mill's methods (Ch. 7) in order to arrive at causal hypotheses that are as sound as possible within
the limitations of retrospective data
How is causal reasoning attempted in the absence of randomization?
Covariation
e.g. that age is responsible for the correlation between children's shoe size and spelling ability
The concept of a rival explanation is known as the third-variable problem (that is, in nonrandomized research, a "third variable" that is
correlated with both X and Y can also account for the correlation between X and Y)
Exclusion of plausible rival explanations of the covarying relationship between X and Y
Requirements for causal inference:
How is the third-variable problem relevant?
Where O = observation or measurement and X = treatment or intervention
NR = nonrandomized, R = randomized
Group A
NR
O
X
O
Group B
NR
O
O
Nonequivalent-groups designs - A between subjects design in which the sampling units are allocated to the experimental and control groups by means
other than randomization
Group 1
R
O
X
O
O
Group 2
R
O
O
X
O
May overcome objections to a randomized design if the objections are based on the ethical cost of depriving control subjects of the benefits of
the treatment given to the experimental subjects
Those in group 2 are later given the opportunity to receive the treatment if a beneficial effect is observed
Information is gained about the immediate and delayed effect of the treatment as well as a replication of the immediate effect
Benefits:
Randomized design with a wait-list control group
"Type G Error" (for "group error")- Relevant extraneous factors exist that are characteristic of one group but uncharacteristic of the other group
This procedure, described as subclassification on propensity scores, reduces all of the variables on which the "treated" and "untreated" sampling
units differ to a single composite variable (this variable is called a propensity score)
In nonequivalent groups designs with large relevant subgroups, comparability of the "treated" and "untreated" subjects may be achieved by
subclassification on propensity scores
What is meant by subclassification on propensity scales?
In time-series designs, the defining characteristic is the study of variation across some dimension over time
Where O is the occurrence of a situation and X is the intervention
O
O
O
X
O
O
Interrupted time-series designs compare the "effects" of an intervention in a situation before and after it occurs (e.g., the Vienna subway suicide study)
Found experiments-- experiments that are essentially "found" (or discovered) in naturally occurring situations
What are time-series designs and "found experiments"?
Single-case experimental designs are a family of nonrandomized designs that is a mainstay of behavior modification research
They incorporate "treatments" (interventions) that are manipulated and controlled for within a repeated-measures design
Only one sampling unit is studied, or only a few sampling units are studied
Repeated measurements are taken of the unit (a within-subjects design)
Random assignment is rarely used
Characteristics of all single-case experimental designs:
Single-case experimental designs come in many different forms (e.g., A-B-BC-B and A-B-A-B); the unit of study may be an N of 1 (e.g., the study of Robbie)
or a few subjects (Skinner's study of superstition in pigeons) or several groups of individuals with one of the treatments randomized (the A-B-C study in
Box 8.3)
Behavioral baseline- a relatively stable pattern in the subject's behavior before the experimental treatment or intervention
In the A phase, no treatment (or intervention) is in effect
In the B phase, a treatment (or intervention) is operating
The first A is the baseline period
A-B design- the simplest of all single-case designs
What within-subjects designs are used in single-case experiments?
Ch. 8 - Quasi-Experimental Designs
Friday, December 14, 2012
8:59 AM
Textbook Notes Page 1
Unlock document

This preview shows half of the first page of the document.
Unlock all 2 pages and 3 million more documents.

Already have an account? Log in

Get OneClass Notes+

Unlimited access to class notes and textbook notes.

YearlyBest Value
75% OFF
$8 USD/m
Monthly
$30 USD/m
You will be charged $96 USD upfront and auto renewed at the end of each cycle. You may cancel anytime under Payment Settings. For more information, see our Terms and Privacy.
Payments are encrypted using 256-bit SSL. Powered by Stripe.