Chapter 7: Study Design Flashcards
Types of Correlational Research Designs in Developmental Psychopathology Research (4)
(1) Cross-Sectional
(2) Repeated measures designs:
(3) Longitudinal Design
(4) Sequential Design
Cross-Sectional Design def
Surveying a group at one time point.
Cross-Sectional Design: Benefits (2)
(1) Compare cohorts of different ages to one another at a given time (e.g. diff between 20 and 21yo on self-esteem)
(2) Relatively cheap and practical
Cross-Sectional Design: Drawbacks
(1) Can’t learn about how individual people change with age/time
(2) AGE effects are confounded with COHORT effects: There might be something about the 21yo that might lead their self-esteem being naturally higher OR covid
Longitudinal Design benefits (2)
(1) Can make within-subject comparisons
(2) No cohort effects
Longitudinal Design: Drawbacks (5)
(1) Subjects drop out
(2) May be effects of repeated testing
(3) Requires foresight (and funding!)
(4) Time consuming
(5) Age effects confounded with time of measurement effects
-> E.g. COVID kids social dev -> That might NOT generalize to kids now who are going in in-person school
Sequential Design (aka Cohort sequential, accelerated longitudinal design) - def
Multiple cohorts that i’m following longitudinally. Can look at changes within pple and diff across cohorts when they’re diff ages.
Sequential Design: Benefit
Helps disentangle age effects from (1) cohort effects & (2) time of measurement effects
=> Adress limitations of both cross-sectional and longitudinal design.
Sequential Design: Drawback (3)
(1) SAMPLE SIZE: 3 times as big as longitudinal design (cuz it’s like 3 longitudinal designs together)
(2) Lot of planning
(3) Very time consuming, complex, & expensive!
-> BUT inferences we can draw are very powerful (main drawback = logistic)
Old guidelines for well-established treatments (Chambless et al. 1999) (2)
(1) A large series (>= 9) of single-case study designs demonstrating efficacy
OR
(2) At least 2 between group-design experiments
Today - how do we establish efficiency of treatments?
Shifting to systematic review/meta-analysis of the literature followed by a committee reviewing the evidence
Single-Case Experimental Designs def
Examine the effect of a treatment on a single child’s behavior
- Repeated measures of behavior
- Replication of treatment effects
A-B-A-B Reversal Designs
Single-Case Experimental Designs
A – baseline of behavior
B – intervention phase
A – return to baseline (remove intervention)
B – reintroduce intervention
Single-Case Experimental Designs: Advantages (2)
(1) Internal validity: Multiple baselines and intervention try-outs
(2) Temporal ordering: Can infer causality
Single-Case Experimental Designs: Disadvantages (4)
(1) External validity: might not generalize to other kids in diff situations from diff populations
(2) Can be hard to interpret the findings
(3) Stable change
(4) Ethics: is it ethical to remove the intervention?
Randomized Control/Clinical Trial (RCT): def
- Experimental and control conditions
- Random assignment, sometimes double blind
Why is random assignment a good thing in RCTs? (2)
(1) Internal validity: Bc they’re randomly assigned, it rly helps with validity
(2) Realistic differences between those groups/help to disperse diff types of demographics and characteristics in both of the groups
Diff levels of evidence that RCTs give (3)
(1) Powerful test of intervention efficacy (treatment vs control + random assignment + good causality)
(2) Powerful test of theory
(3) If designed carefully, can let researcher establish cause
Question asked about internal validity in RCTs
Is it my intervention that is causing the change in outcomes?
=> Control vs Treatment
Question asked about construct validity in RCTs
What about my intervention is causing the change in outcome?
=> What is it WITHIN the intervention that’s leading to the most change?
RCTS: What are the diff types of control groups? (5)
(1) No treatment control group
(2) Wait-list control group
(3) Placebo (in medication trial)
(4) Treatment as usual
(5) Another, effective treatment
RCTs: Disadvantages (4)
(1) External validity: Comorbidity adds noise into these trial which is why they’re excluded at the start
-> Maybe treatment doesn’t work for someone who has ADHD + Depression
(2) Drop out: Huge concern
(3) Attrition Bias: Pple with certain characteristics might be more likely to drop out than others
(4) Vast majority of trials are focused on EFFICACY (tightly controlled clinical trial)
Efficacy vs Effectiveness vs Efficiency
(1) Efficacy: Does this work in a tightly controlled clinical trial
(2) Effectiveness: Does this work in APPLIED clinical practice
(3) Efficiency: Does it contribute to more efficient use of resources?
Type of evidence for the treatment intervention (3)
Efficacy, Efficiency, Effectiveness
