Treatment integrity refers to implementing interventions as intended. Treatment integrity is critically important for experimental validity and for drawing valid inferences regarding the relationship between treatment and outcome. Yet, it is rarely adequately addressed in psychotherapy research. The authors examined barriers to treatment integrity…

Describes hypothesis validity as extent to which research results reflect theoretically derived predictions about relations between or among constructs. Discusses role of hypotheses in theory testing. Presents four threats to hypothesis validity: (1) inconsequential research hypotheses; (2) ambiguous research hypotheses; (3) noncongruence of…

Notes that studies in psychology often have low power because of inadequate sample size and that prevalent psychometric distributions are sometimes radically nonnormal. Demonstrates robustness of independent samples t-test with respect to Type I error. Shows that researchers may use power tables based on population normality without modification…

Reviews misuse of statistical tests in psychotherapy research studies published in "Journal of Consulting and Clinical Psychology" in 1967-68, 1977-78, and 1987-88. Focuses on inappropriate uses of null hypotheses tests and p values, neglect of effect size, and inflation of Type I error rate. Concludes with practical suggestions for rational…

After a concise overview of the technical characteristics that define brief psychotherapy, the current use and misuse of these methods in clinical research on the outcome and process of treatment is examined. Suggests that brief psychotherapy methods possess unique technical advantages to the researcher. (Author/BL)

Notes that psychotherapy process research has produced advances in measure development and substantive findings of process-outcome relations. Describes paradigm shift toward sequentially patterned, significant change episodes, emphasizing segmentation of process by meaningful patterns wherever they occur. Reviews theoretical, psychometric, and…

Discusses problems and issues of psychotherapy process research. Notes complexity of research that attempts to combine process and outcome adequately. Suggests that more than one therapeutic orientation be studied in specific project to include some type of control. Cautions that, although statistically significant correlations between process and…

Presents a statistical technique that yields consistent estimates for censored samples. Application of this technique to models related to selection bias and the presence of depression indicate that the prevalence of this disorder in mental health center populations approaches 44 percent rather than the 25 percent from a censored sample.…

Analyzes the attrition dilemma and reviews the methods used to compensate for data loss. Argues that attrition is not utlimately a problem of bias but a problem of lack of information. Offers a reformulation that integrates single-case studies with traditional group comparison methodology in an attempt to find optimal causal relations of…

The results from the present study indicate that in a sample of psychotherapy patients, cooperative and less cooperative research subjects are distinguishable on a number of characteristics including background and psychotherapy variables. These results caution against generalizations based merely on cooperative participants. (Author)

Questions assumptions that process variables have fixed meanings, discretely contribute to outcomes, and have a decontextualized net worth, evaluated by their correlations with outcomes. Suggests conceptual framework that integrates theory-driven and discovery-oriented strategies into cyclical, ecologically oriented model. (Author/NB)

Child abuse research has provided information on psychological sequelae of sexual molestation, with findings helpful in informing social policy and guiding mental health practice. Because of recency of interest, costs, and time investment associated with more rigorous longitudinal research, many studies have used correlational designs and…

Identifies questions in child psychopathology and delinquency that can best be answered by using longitudinal data and discusses advantages and problems of longitudinal studies. Reviews methodological issues arising in longitudinal research on child psychopathology and proposes solutions to problems. Contrasts retrospective and prospective designs…

Demonstrates low probability that nonequivalence will produce erroneous inferences in small samples. Sees probability of erroneous inference in absence of true treatment effect as generally no greater than nominal Type I error rate. Seems unlikely that small samples have biased inferences drawn from past psychotherapy outcome research. Cites other…

Issues related to the selection of assessment procedures, the scaling of assessment devices, the computation of difference or change scores, and the translation of these scores into common indexes of outcome are explored. Suggestions are provided for refining the measurement of inferred states. (Author/BL)