Research Methods

Einzelfallexperimente in der Sprachtherapie/Logopädie: Brückenbau zwischen Wissenschaft und Praxis

In diesem Beitrag beschreiben wir Einzelfallexperimente in der Sprachtherapie, ihre wissenschaftliche und praktische Relevanz, Definition, Klassifizierung, Gestaltung, sowie Herausforderungen bei der Anwendung und betonen die Bedeutung einer sorgfältigen Durchführung.

Single Case Design Research als Methode evidenzbasierter Sprachförderung und -therapie

Der Beitrag stellt die Idee hinter kontrollierten Einzelfallexperimenten beziehungsweise Single Case Research Designs (auch Single Case Experimental Designs) vor und arbeitet ihr Potential im Kontext evidenzbasierter Sprachtherapie und …

Open Science als Weg zu verlässlicher sonderpädagogischer Forschung

Outlines how open science practices can help special and inclusive education researchers to make the research process more transparent and communication with the research community more open and immediate. Offers a commentary on how this is nothing else than a revival of broadly accepted scientific values.

Statistical Power of Piecewise Regression Analyses of Single-Case Experimental Studies Addressing Behavior Problems

We conducted a Monte-Carlo study to investigate the influence of single-case research design chracteristics (phase length, initial DV level, intervention effect, trend) on the test power of piecewise Poisson-regression models. PLM have a high potential of adequately identifying intervention effects in single-case research designs. However, few measurement times, especially in phase A, and low initial DV levels make it impossible to detect even large intervention effects. Research designs with a high number of measurement times show robust power.

Validity and Judgment Bias in Visual Analysis of Single-Case Data

We had 186 pre-service teachers visually analyze single-case graphs, representing an intervention effect, a baseline trend, both, or no change at all. The presence of a baseline trend increased type I errors from 5% to 25%. Inter- and intra-rater reliabilities were low, particularly for graphs including a baseline trend.