Research Flashcards
Neuropsychological consideration in administering
- Validity
- Reliability
- Purpose/rationale (ethical)
- Standardisation (Norm for who?)
- Age range
- Time of administration
- Attention capacity
- Educational level
E.g., ID for MMPI (sams case)
Neuropsychological results interpretation
- Was test valid/reliable ^
- Was it scored right/ Was it administered correctly?
- Malingering? / conditions which they answers/took it? ^
Assuming ^ all okay then:
- Statistical significance (rainbow table, t-scores)
- BASE RATES (below 10?)
- Research: What about for specific age, culture, comorbidity, current stage
- Manuel
- Does this reflect or contradict achievement in life? Collateral? School perspective?
good internal consistency reliability, which shows that all of the items in a scale intercorrelate well and tap into the same construct. Cronbach’s alpha,
Factors affecting the reliability of a client’s data
● rapport
● distractibility, comprehension, memory for instructions
● environmental noise
● client factors: tiredness, attitude, malingering
● clinician factors: bias, attitude
Research consideration
- Sample size (power?)
- Participant characteristics (determinative of the population it will be administered on? Have diagnosis? Mild, moderate? – how were they ‘diagnosed’)
- How were participants recruited (sort of people to sign up already diff to pop)
- What was the study design? Correlation vs experimental
- Was there randomisation? Were the two groups similar?
- Blinding, double blinding? for bias
- Placebo/control to know it’s not just random
- Confounding variables? Do we know the results due to what is being tested?
- How were the results measured? When were they measured (pre, post, follow-up?)
- Statistical significance vs clinical significance – meaningful difference?
- Effect sizes
- Cost/benefit
- Was the procedure indicative of the research qs (CBT carried out properly?)
- Is this research replicable/reliable?
- Correlations:
strength and direction of association between 2 variables (-1 to +1)
- Effect sizes:
- Effect sizes: how meaningful the relationship between variables or the difference between groups is. It indicates the practical significance of a research outcome
o typically as Cohen’s d (0.2 = small, 0.5 = medium, 0.8 = large)
- Confidence intervals:
- Confidence intervals: mean of your estimate plus and minus the variation = how good an estimate is
o eg confident that upon retesting multiple times their score would land in interval 95% of time
- Odds Ratio:
- Odds Ratio: odds of something happening in one group V other group
o eg abused in childhood and depression, 1 = no diff,
o <1 less likely develop dx,
o >1more likely develop dx
Statistical significance vs clinical significance
- Clinical = observes dissimilarity between the two groups
- Statistical = implies whether there is any mathematical significance to the carried analysis of the results or not
Clinical effectiveness vs clinical efficacy
Clinical effectiveness vs clinical efficacy
- Effectiveness: Treatment intervention outcomes in real-world settings
- Efficacy: Treatment intervention outcomes under ideal and controlled circumstances
Reliability vs Validity
- Reliability = the consistency of a measure (whether the results can be reproduced under the same conditions)
- Validity = the accuracy of a measure (whether the results really do represent what they are supposed to measure).
- E.g., I’m sad, I’m upset, I’m not happy = RELIABLE all probably yes but not VALID of depression as it’s much more than that
Mediation vs moderation
Mediation: a third factor (B) explains the relationship between A and C eg stress - rumination - depression (mediator is explaining the correlation between 2 variables)
Moderation: when B impacts on the relationship between A and C (makes it stronger/weaker)
