Research Methods in Clinical Psychology 2019/20

Clinical significance can be assessed using Kendall's equivalency testing method.

Clinical significance is expressed when a client's post-intervention score represents abnormal functioning.

Clinical significance is expressed when a client's post-intervention score has moved outside the range of the dysfunctional population.

Clinical significance is reached when intervention and control groups differ significantly regarding the primary outcome.

Seligman (1995) identifies five properties of psychotherapy that are difficult to assess in controlled studies. Which of these properties are they?

How can the results of small-n-studies be generalized across studies?

An audit in clinic research is the basis of quality assurance to set up procedures in services.

An audit in clinic research is a concept that refers to treatment efficacy studies in outpatient services.

An audit in clinic research is an approach that is comparable (if not synonymous) to program evaluation.

An audit in clinic research describes a circle in progress, as is typically present in evaluation research.

In comparison to fMRI, EEG has a much higher spatial resolution.

PET is the only method to infer causal effects of differences in brain activity on behavioral outcomes.

Using TMS is not possible (and not warranted advisable) to reach subcortical structures.

Neuro-scientific methods assess only functional but not structural malfunctioning.

Outcome evaluation in program evaluation research includes the assessment of positive outcomes and benefits for the clients.

Outcome evaluation in program evaluation research includes the assessment of negative outcomes.

Outcome evaluation in program evaluation research includes the assessment of client satisfaction (with the service).

Outcome evaluation in program evaluation research includes the assessment of outcomes of some individuals (case tracking).

Before calculating a combined effect size, we need to ensure that all individual effect sizes have the same metric.

Meta-analyses are highly powered interval-based estimates of parameters based on replication principles.

Meta-analyses can be computed to generalize data across a number of small-n-design studies.

“Random effects” meta-analyses are a way to address and model the heterogeneity between studies.

The validity of a measurement is computed for metric data using Cronbach's alpha (internal consistency).

The validity of a measurement can be assessed by means of face validity (as provided by means of qualitative expert ratings).

The validity of a measurement is always (numerically) larger than the reliability of the measurement.

The validity of a measurement assesses the meaning of measurement (as specified in its definition).

“Random effects” is a term that refers to approaches that assume that parameter estimations are the results of random experiments but remain stable across different studies/participants.

“Random effects” are evident in some types of non-overlapping pairs approaches in single case designs.

“Random effects” are computed in some multi-level models.

“Random effects” models are computed to generalize the results of meta-analyses to the whole population of studies.

SCED refers to “single case experimental designs”.

SCED designs rely on only a few participants.

SCED have the advantage over larger scale clinical research studies to produce data that do not constitute a time series.

ABAB studies represent typical designs for SCED.

Using small-n pilot studies could lead to underestimations of statistical power in Randomized Controlled Trials.

ANCOVA models should be preferred over ANOVA models in Randomized Controlled Trials.

A PPF (pre-post-follow up) design is the best way to assess clinical outcomes and to evaluate the efficacy of a treatment in Randomized Controlled Trials

The term 'controlled' in Randomized Controlled Trials refers to the inclusion of a control group.

The number of studies that can be included in a meta-analysis is a limitation of meta-analyses.

Qualitative information is ignored in meta-analyses.

The quality of the results depends on the quality of the included studies in meta-analyses.

Effect sizes do not convey meaningful information in meta-analyses.

CONSORT is a Consolidated Standards of Reporting Trials.

CONSORT includes a flow diagram which displays the steps of statistical analyses to be conducted in RCTs

CONSORT provides a checklist to avoid biased estimates in analyses of RCTs.

CONSORT are guidelines how to report RCTs.

Client survey methods are a typical example of assessments in therapy process research.

Audio-tape recordings of therapy sessions are a typical example of assessments in therapy process research.

Therapist self-report measures are a typical example of assessments in therapy process research.

Observational methods are a typical example of assessments in therapy process research.

ITSACORR is a specific kind of time series analysis.

ITSACORR is known to be (statistically) flawed but can be used in SCED studies with only minimal corrections in the error terms.

ITSACORR is the most current type of multi-level analysis for SCED studies.

ITSACORR is specifically designed to deal with short time series.

Multilevel models (Multilevel modeling, MLM) are suggested to be computed in case of missing data and unbalanced designs.

Multilevel models (Multilevel modeling, MLM) represent the state of the art of statistical analyses in RCTs.

Multilevel models (Multilevel modeling, MLM) are sometimes just called mixed model regressions.

Multilevel models (Multilevel modeling, MLM) should not be applied in therapy process research.

Outcome domains in therapy process research are...

Klaus Grave identified therapeutic bond as a main mechanism of change in psychotherapy.

Klaus Grave identified 5 mechanisms of change particularly important in psychodynamic therapies (but not in cognitive behavioural psychotherapy).

Klaus Grave identified symptom shifts as typical change mechanisms in psychotherapy.

Klaus Grave conducted the first meta-analysis to identify typical change mechanisms in psychotherapy.

Small-n-studies are well known to lead to underpowered RCTs when they are used to estimate necessary sample sizes.

Small-n-studies are essential for establishing treatment efficacy in therapy research.

Small-n-studies can be reliably analyzed using graphs and diagrams (=visual data analysis).

Small-n-studies are best analyzed using multilevel random effects modeling.

In evaluation research, there is a distinction made in service evaluation between needs, demands, and supply. Explain what is meant by these concepts, and how they differ, using an example of your choice.

________ is a study which combines results from more than one quantitative empirical study into a single estimate of the parameter.

_________ represents a more powerful and precise statistical test than ANOVA when used for randomized treatment studies.

Internal consistency (Cronbach's Alpha) is the standard way of assessing the ________ of a scale that is composed of multiple similar items. The assumption is that the items are equivalent or parallel, that is, that they all aim to tap the same underlying construct.

It is important that covariates are measured ________ treatment begins; otherwise differences between conditions on the first assessment will be adjusted or equalled between conditions.

In clinical research, external validity refers to the generalizability of the study's findings to other populations, settings, and interventions. Explain what is meant by external validity in the context of clinical research, and how is it defined in this case?

When comparing different therapeutic interventions, what factors need to be considered to ensure a fair and accurate comparison? Describe some problems and pitfalls that need to be considered.

What are 3 different methods to assess the reliability of an observation? Which of these methods assumes nominal scales of the ratings?

What is “creaming”? At which step of a service evaluation could ‘creaming’ create a problem and for whom?

In clinical research, there is a distinction made between efficacy and effectiveness studies. Explain in your own words what these terms refer to.

Name four reasons why clinical psychologists (who are practitioners) might not engage in research, and explain one of these reasons by your own words.

Meta-analysis

Small-n-studies

Clinical Significance

Multilevel Model

Service Evaluation

Randomized Controlled Trial (RCT)

Single Case Experimental Design (SCED)

External Validity

ITSACORR

Validity

Reliability

Mixed Model Regression (Mixed Effects Model)

Effectiveness Studies

Efficacy Studies

ANCOVA

Control Group Design

Waitlist Control Group

Active Control Group

Outcome Evaluation

Bonferroni Correction

Face Validity

Cronbach's Alpha

Inter-rater Reliability

Creaming

Needs Assessment

Demand Assessment

Supply Assessment

Matched Control Group

Time Series Analysis

Mechanisms of Change