Lecture 9 single subject designs ii (in lab)
Download as PPT, PDF1 like697 views
The document discusses visual analysis as an approach to analyzing data in single subject research. It outlines key elements of visual analysis including changes in means, levels, trends, overlap, latency, variability, and reliability. Visual analysis is best suited for data with minimal variability, consistent patterns within phases, and clear changes between phases. Statistical analysis may be warranted for unstable baselines, new interventions, highly variable environments, or when small changes are important. Students are instructed to download datasets and follow instructions to graph and analyze data from an ABAB design and multiple baseline study using visual analysis techniques.
Lecture 9 single subject designs ii (in lab)
- 1. #9 Location In The Research Process Formulating Research Questions Previous Research Craft Knowledge Appropriate Design Appropriate Sampling Decisions Ensure Research Ethics Data Collection & Management Data Analysis Interpretation Dissemination Ongoing Research
- 2. Visual Analysis Only interventions which lead to marked changes in targets will be viewed as significant Clinical and experimental significance coincide more often when using visual analysis as compared with statistical analysis
- 3. Visual Analysis Changes in means Changes in average rates of performance when phases shift Changes in level Shifts or discontinuity in performance when phases shift
- 4. Visual Analysis Overlap How many data point values co-occur across phases Latency The amount of time between phase shifts and observed performance changes
- 5. Visual Analysis Changes in trend Changes in the systematic increase or decrease in behavior when phases shift Variability and stability How consistent the patterns of performance are within phases Reliability of data - An important consideration in visual analysis, as it is with all research - Inter Observer Agreement
- 6. Visual Analysis Data that have minimal variability, high consistency, relatively long phases, changes in means, levels, or trends with minimal latency most easily demonstrate causality Statistical Analyses are warranted in single subject research when: Baseline data are unstable New interventions are being investigated When the participants' environments have high variability When small changes in a target may be important
- 7. Lab Assignment Download the data set “ SOWK 468 SSD Data Sets,” and the file ”Visual Analysis Instructions” Open the data set and follow the instructions during this lab session to create a graph of the ABAB design data and the Multiple Baseline (MB) data Apply the elements of visual analysis to the following data