Flashcards in The Science of Psychopathology (Exam 1) Deck (81):
What does the nature of science mean?
a way of knowing
how we know what we know
What does science value?
empiricism, objectivity, and replicability
What does science demand?
rigorous standards of proof
What are the means of science?
a means for testing hypotheses and theoretical claims
Is science value free?
as a human enterprise, it can be value laden
Case studies: narratives
record in detail and in narrative form what happens during therapy (stories)
Usefulness of case studies
-useful for rare cases
-used to generate research questions and hypotheses
Main problems of case studies
-results do not generalize
-not convincing data for casual modeling
-does not control for outside factors
-no manipulation of variables
Bias in case studies
-interviewer's bias for soliciting selective info
-patient's bias for recollecting selective information
taking data and counting numbers
new case rate per unit of time
total number of cases in certain time frame (base rate)
subject to bias, patient may not recall
keep track with population
subject to bias, expensive
-prospective studies -> longitudinal study
Why is it essential to know prevalence rates (base rates)?
-can study changes over time
-can try to follow to see if problem remits naturally (ex: depression after 5 years untreated)
associations but not cause
What does correlational research look at?
looking at a relationship between two or more variables
What does "Do they co-relate?" mean?
What does "Do they vary together?" mean?
one measure goes up, other goes up
ex: relation of hours worked and tips received
one measure goes up, other goes down
ex: relation of practice time on an instrument and less errors in recital
knowing value of one measure does not allow you to predict value of the other measure
-r or R
-range: -1.0 to 1.0 (can't go over this range)
What does the correlation coefficient mean?
-used to summarize whether two measures vary together
What is the strongest correlation? (test question)
-2.6, -0.8, 0, 1.2, 0.7?
Does correlation mean causation?
No, correlation does NOT mean causation
-measures can be correlated, but doesn't mean that one caused the other
mediator and moderators
Direction of Casualty: Consider case where X and Y are correlated. Why? (Test question)
-X might cause Y or Y might cause X
-X might be correlated with Y, because of Z (3rd variable)
Can spurious relationships exist?
What do correlations show?
patterns, not causes
demonstrating casual relationships
Begin with a hypothesis (hypothesis formation)
-some variable causes something to happen
-develop a specific and testable research question
identify and manipulate an independent variable (X) to look at its impact on dependent variable (Y)
What is identifying a comparison group called?
-variable manipulated by experimenter
-must involve at least 2 conditions (ex: violent/nonviolent show)
-has levels (ex: percentage)
behavior that is measured or observed (ex: aggressive behavior)
How is the hypothesis tested?
observing effect of manipulating independent variable (ex: # of acts of aggression caused by exposure to different levels of violence)
a comparison condition in which subjects are not exposed to the same treatment as in the experimental condition
Does the control condition receive the independent variable?
Can there be confounding variables?
Confounding variable examples
-can change systematically with the independent variable
-will alter ability to make casual statements
(ex: impact of cognitive therapy for depression that includes a large group interaction component)
random assignment to conditions
-equal chance of being assigned to control or treatment condition
-allows comparison between groups
How do you stratify participants?
based on knowledge of characteristics before participants are assigned
assesses whether differences between groups is due to influence of IV or some other factor
What is the active ingredient of change?
mechanism of change
reflects a therapy effect due to expectations of the person rather than true treatment effect (large threat to internal validity)
What is the purpose of a double-blind procedure?
control for placebo effects
neither participants nor experimenter know which group the participant is in (led to believe they are receiving a treatment)
extent to which the results of an experiment can be generalized beyond the conditions of the experiment
Clinical Efficacy Trails
concerned primarily with internal validity (why and how does the treatment work)
Clinical Effectiveness Trials
concerned more with external validity (where and with whom does the treatment work, given that it seems to work)
larger design with manipulated IV and DV
statistical significance testing
evaluating the dependent variables mathematically (are the differences between groups attributable to chance?) (mathematically difference, meaning)
clinical significance testing
testing if participants have moved from here to there (have participants moved from one group to another?)
ex: depressed -> nondepressed
What are the limits to group designs?
-can't control for everything
-hard to generalize from a very controlled study to outside world
-may not always yield info we are after with treatment outcome studies
may not always yield info we are after with treatment outcome studies (limits to group designs)
average patient/client treated by average therapist
What is the key to single subject designs?
repeated measurement and evaluation of variability, level, and trend
What does a single subject design look at?
in depth, scientific study of one person at a time
What does a single subject design emphasize?
measurable, observable, meaningful change
What can a single subject design show?
true experimental control OR true accountability of treatment effect (by the clinician)
What does a baseline need in order to determine sufficient data?
(some data will be easier to hold at a baseline)
What is the ABAB Design?
another type of single subject design
A (ABAB design)
What does a baseline measure look at? (ABAB design)
-record behavior without intervening
-no treatment time
-want to observe regular pattern
B (ABAB Design)
treatment (record changes in behavior)
2nd A (ABAB Design)
What happens when you remove treatment? (ABAB design)
-see if you have a return to baseline
-naturally done with termination of treatment in psychotherapy
2nd B (ABAB Design)
reinstate treatment (see if behavior returns to level of 1st B or maintains)
What does a multiple baseline approach look at?
looks at baselines across
What does a multiple baseline approach show?
shows experimental control at n=1 level of analysis
What are the strengths of a multiple baseline approach?
implements treatments staggeringly and look at the effects
ex: teaching social skills in multiple settings to a socially anxious person one at a time
Example of a multiple baseline approach
person feels anxious in social settings -> lacks social skills (hypothesis) -> teaching social skills in different contexts (work, dating, family)
Where do we see single subject designs being used?
-individual psychotherapy (private practices, HMO based practices)
-teachers and educators (targeted behaviors of interest, special education)
-developmental disabilities (autism and other PDDs)
What is the Human Subjects Internal Review Board (HSIRB) for ethics in research?
-analyzes components of experiment
-decides whether experiment can be conducted ethically and legally with minimum risk to subjects
Is deception allowed during experiments?
no, deception is frowned upon and is not used in experiments very often
Are participants informed about the study after completion?
yes, participants are debriefed of the purpose of the study after completion
Do participants give informed consent?
yes, they give informed consent to participate