Experimental Clinical Psychology
,Week 1 2
Chapter 2 and 3 and 13 2
Lecture 1 5
Week 2 7
Chapter 4, 5, and 6 7
Workgroup 1 12
Lecture 2 13
Week 3 15
Chapter 7 15
Lecture 3 18
Workgroup 2 20
Week 4 21
Chapter 8 and Chapter 12.6 and 12.7 21
Observational research, randomized trials, and two views of medical science
(Vandenbroucke, 2008) 23
Lecture 4 24
Week 5 27
Chapter 14 and 17 27
Lecture 5 30
Week 6 35
Chapter 15 and 16 35
Lecture 6 38
, Week 1
Chapter 2 and 3 and 13
The plausible rival hypothesis addresses competing interpretations that might be posed to
explain the findings of a particular study.
Four different types of validity are;
● Internal validity ; are there any extraneous influences that can impact the results?
● External validity ; can the results be generalized?
● Construct validity ; what aspect of the intervention was actually causally related to the
results?
● Data-evaluation validity ; is the data manipulated in a way that the results are safely
relevant?
Together they convey the considerations that investigations should have before designing an
experiment.
Internal validity
Some factors that can threaten internal validity are the history, maturation, testing,
instrumentation, or attrition.
Instrumentation as a threat to internal validity is any instance in which differences (e.g., between
one assess-ment occasion and the next) might be attributed to a change in the instrument or to a
change in the criteria (response shift) used to complete that instrument. It is important not to
confuse testing (a threat addressed previously) with instrumentation. Both include measurement,
and both relate to changes from one occasion to the next.
Statistical regression is also an additional threat to internal validity, but it can be helped out by
-Assigning participants randomly
-Use measures with high reliability and validity
-Test everyone on pretest twice and only use those that had extreme answers on both occasions
(regression occurs from first to second assessment).
Diffusion occurs when the control group adopts a new lifestyle/habits because of being part of
the experiment. It is important that participants in each condition receive only that condition.
When participants are aware they are part of the control group, they might try to compensate for
that.
External Validity
, External validity refers to the extent to which the results of an investigation can be generalized
beyond the conditions of the experiment to other populations, settings, and circumstances. Some
threats to external validity are;
● Sample characteristics ; extent to which the results can be extended to other
people/participants. For example, can the results of studies about animals be extended to
humans. Besides this, there are still many underrepresented groups and minorities.
● Narrow Stimulus Sampling; extent to which the results are restricted to some materials or
conditions
● Reactivity of experimental arrangements or assessment; extent to which knowing you are
in an experiment or being tested influences your response
● Multiple-Treatment Interference
● Novelty effects
A cohort refers to a group born or gathered in a particular period that is studied at a particular
point in time and followed. A result found in one cohort may not generalize to another.
However, external validity is not always a concern, like for example in proof of concept studies.
The internal validity of an experiment is usually more important/has priority over external
validity.
Construct validity
Construct validity addresses the presumed cause or the explanation of the causal relation
between the intervention or experimental manipulation and the outcome. Which facet of the
intervention is responsible for change?
Confounds are these other facets that can interfere with the interpretation of the results. Some
threats to it are
● Attention and Contact Accorded to the Client
● Single operations and narrow stimulus sampling
● Experimenter expectancies
● Demand characteristics
Data-evaluation validity refers to the facets of the evaluation that influence the conclusions we
reach about the experimental condition and its effects. It can be seen from two standpoints.
● Understanding the tests and their bases (probability theory, distributions)
● The computational aspects of statistical tests (application of data sets, use of software)
Some examples of threats to data-evaluation validity are low statistical power, subject
heterogeneity, errors in data recording, or unreliability of the measures.
,Week 1 2
Chapter 2 and 3 and 13 2
Lecture 1 5
Week 2 7
Chapter 4, 5, and 6 7
Workgroup 1 12
Lecture 2 13
Week 3 15
Chapter 7 15
Lecture 3 18
Workgroup 2 20
Week 4 21
Chapter 8 and Chapter 12.6 and 12.7 21
Observational research, randomized trials, and two views of medical science
(Vandenbroucke, 2008) 23
Lecture 4 24
Week 5 27
Chapter 14 and 17 27
Lecture 5 30
Week 6 35
Chapter 15 and 16 35
Lecture 6 38
, Week 1
Chapter 2 and 3 and 13
The plausible rival hypothesis addresses competing interpretations that might be posed to
explain the findings of a particular study.
Four different types of validity are;
● Internal validity ; are there any extraneous influences that can impact the results?
● External validity ; can the results be generalized?
● Construct validity ; what aspect of the intervention was actually causally related to the
results?
● Data-evaluation validity ; is the data manipulated in a way that the results are safely
relevant?
Together they convey the considerations that investigations should have before designing an
experiment.
Internal validity
Some factors that can threaten internal validity are the history, maturation, testing,
instrumentation, or attrition.
Instrumentation as a threat to internal validity is any instance in which differences (e.g., between
one assess-ment occasion and the next) might be attributed to a change in the instrument or to a
change in the criteria (response shift) used to complete that instrument. It is important not to
confuse testing (a threat addressed previously) with instrumentation. Both include measurement,
and both relate to changes from one occasion to the next.
Statistical regression is also an additional threat to internal validity, but it can be helped out by
-Assigning participants randomly
-Use measures with high reliability and validity
-Test everyone on pretest twice and only use those that had extreme answers on both occasions
(regression occurs from first to second assessment).
Diffusion occurs when the control group adopts a new lifestyle/habits because of being part of
the experiment. It is important that participants in each condition receive only that condition.
When participants are aware they are part of the control group, they might try to compensate for
that.
External Validity
, External validity refers to the extent to which the results of an investigation can be generalized
beyond the conditions of the experiment to other populations, settings, and circumstances. Some
threats to external validity are;
● Sample characteristics ; extent to which the results can be extended to other
people/participants. For example, can the results of studies about animals be extended to
humans. Besides this, there are still many underrepresented groups and minorities.
● Narrow Stimulus Sampling; extent to which the results are restricted to some materials or
conditions
● Reactivity of experimental arrangements or assessment; extent to which knowing you are
in an experiment or being tested influences your response
● Multiple-Treatment Interference
● Novelty effects
A cohort refers to a group born or gathered in a particular period that is studied at a particular
point in time and followed. A result found in one cohort may not generalize to another.
However, external validity is not always a concern, like for example in proof of concept studies.
The internal validity of an experiment is usually more important/has priority over external
validity.
Construct validity
Construct validity addresses the presumed cause or the explanation of the causal relation
between the intervention or experimental manipulation and the outcome. Which facet of the
intervention is responsible for change?
Confounds are these other facets that can interfere with the interpretation of the results. Some
threats to it are
● Attention and Contact Accorded to the Client
● Single operations and narrow stimulus sampling
● Experimenter expectancies
● Demand characteristics
Data-evaluation validity refers to the facets of the evaluation that influence the conclusions we
reach about the experimental condition and its effects. It can be seen from two standpoints.
● Understanding the tests and their bases (probability theory, distributions)
● The computational aspects of statistical tests (application of data sets, use of software)
Some examples of threats to data-evaluation validity are low statistical power, subject
heterogeneity, errors in data recording, or unreliability of the measures.
