Introduction
Basic Descriptive Statistics
Central Tendency
Mean
Median
Mode
Dispersion
Standard Deviation
Variance
Normal Curve Deviations
Kurtosis
Skewness
Classical Test Theory (CCT)
Obtained Score
Weights Items Equally
Reliability
Correlation Coefficient
Test-Retest (Temporal)
Internal Consistency
Homogeneity among items
Average intercorrelations among all items
Chronbach's coefficient alpha
Upper limit
Lower limit
Interrater
Percentage of agreement
Vulnerable to chance agreements
Vulnerable to ranking/level of scores
Interclass correlation coefficient
Kappa coefficient
Standard Error of Measurement (Confidence Intervals)
Cutting Scores
Stability of Test Patterns or Profile Scores
MMPI
Pattern stable in fewer than 50%
More research needed
Rorschach
Generally trait stable
Generally state unstable
Redundancy of Test Scores
Validity
Limits
Various Comparisons
Abstract vs Construct
Convergent Validity
Discriminant Validity
Incremental Validity
Clinical Decision Making
Prevalence (Base Rate) Data
Base Rates in General Population
Variable by Diagnosis
Mood Disorders
Major Depressive Disorder
Dysthymic Disorder
Bipolar Disorder
Suicide
Anxiety Disorders
Panic Disorder
Obsessive-Compulsive Disorder
Post-Traumatic Stress Disorder
Generalized Anxiety Disorder
Social Phobia
Agoraphobia
Specific Phobia
Schizophrenia
Eating Disorders
Anorexia Nervosa
Bulimia Nervosa
Binge-Eating Disorder
Personality Disorders
Antisocial Personality Disorder
Avoidant Personality Disorder
Borderline Personality Disorder
Base Rates in Clinical Settings
Less information on frequency
Frequency statistics influenced by a clinical setting
Clinic Focus Clinic Specialization
Good Decisions
General Prevalence Limitations
Local Norms
Challenge
Clinical versus Statistical Judgment
Statistical Judgment
Clinical Judgment
Meta-Analysis Findings
Accuracy Variance
with the type of decision being made
with the setting in which data are collected
with the type of statistical formula used
with the amount of information available to the clinician
and the amount of data available for inclusion in the formulas
Accuracy Increase
Statistical increases over Clinical Judgments was Modest
Statistical judgments 13% to 10% more accurate than clinical judgment
Accuracy Equal
Clinical judgment equaled the accuracy of statistical judgment in many instances
and was even more accurate than a formula.
Complementary Approaches
Balanced Manner
Enhance Accuracy
Variables to Use
Ideal Goal
Often Pitted Against Each Other
Risk Factors
Odds Ratios
Lifetime Prevalence
Heterogeneity in Patients within the Same Diagnostic Group
Alcoholics
Borderlines
Homogeneity myth
Heterogeneity truth
Polythetic Diagnostic Criteria
Patients with Various Disorders
Commonly occurring overlap in symptoms
Lack of firm boundaries between disorders
Examples
Borderline Personality Disorder
Patient A
Criterion 1
Criterion 2
Criterion 3
Criterion 4
Criterion 5
Criterion 6
Criterion 7
Criterion 8
Criterion 9
Patient B
Criterion 1
Criterion 2
Criterion 3
Criterion 4
Criterion 5
Criterion 6
Criterion 7
Criterion 8
Criterion 9
Major Depressive Disorder
Patient C
Criterion 1
Criterion 2
Criterion 3
Criterion 4
Criterion 5
Criterion 6
Criterion 7
Criterion 8
Criterion 9
Patient D
Criterion 1
Criterion 2
Criterion 3
Criterion 4
Criterion 5
Criterion 6
Criterion 7
Criterion 8
Criterion 9
Appropriate to use
Diagnostic Efficiency Statistics
Cutting Scores
Impediments to Accurate Decisions
Vividness (Saliency)
Confirmatory Bias
Lack of Feedback on the Accuracy of Interpretations
Lack of Awareness of Relevant Variables
Task Complexity
Incomplete Information
Concluding Comments