Single-Case Research: Procedures & Data Collection

General Procedures in Single-Case Research

• This chapter outlines key considerations for planning, conducting, and interpreting data in single-case research to enhance the rigor of such studies
• Critical reflection on research and practice helps identify shortcomings and is key to becoming an effective practitioner
• Issues in data collection, types of data, and determining meaningful change are considered to enhance the rigor of research

Data Collection

• The strength of single-case research relies on the quality of data collected through interviews, self-report measures, observation, and physiological testing

The Assessment Phase

• Single-case research determines the utility of an intervention designed to bring change to a participant who presents with an issue to resove
• Researchers must decide on the baseline data during the assessment phase
• Standardized psychometric instruments, such as the Competitive State Anxiety Inventory-2 (CSAI-2), can be used to collect baseline data
• Collecting baseline data involves monitoring variables of interest identified in the assessment phase
• Interviewing the participant is a frequently used approach in single-case research
• Interviewing those close to the participant can strengthen research and offer different perspectives unless privacy is a concern
• It is crucial to collect sufficient data to provide an accurate picture of the issue
• Triangulation of data using complementary methods outlines a more complete picture than relying on a single measure

Observation

• Observation is an essential part of the assessment phase and monitoring behaviour
• Defining the target behavior to be observed is imporatnt, clarity is crucial in how the behavior will be measured
• A clear definition of the variables enables research papers to be compared and studies to be replicated

Limitations of Operational Definitions

• Operational definitions may not cover all behaviors
• There must be intent on behalf of the player to harm the victim either psychologically or physically.

Hawthorne Effect

• Single-case research designs should have clear "real-world" applicability that facilitates communication of the findings to the general public
• This effect suggests that observation changes the behaviour under scrutiny
• Researchers can counteract this effect by being aware it exists
• Hawthorne studies were done in the 1920s to determine the culture and experience that the working class was different from today
• It would be hard to fathom in today's environment that professional level atheletes would have their behaviour altered by being observed in someone else's

Defining Research Aims

• Variables of interest must be identified
• Consider the dependent variables
• Clearly define the dependent variable(s) and to determine what change in the dependent variable(s) is expected and desired.
• The assessment and subsequent change of a desirable or undesirable behaviour is perhaps the most common approach in single-case research
• Focus may be on a particular outcome
• Psychological contructs may also be the variable of interest
• Self-report inventories including psychometric tests may be used

What Changes is Important?

• In applied contexts, significant differences may not be appropriate or meaningful in determining the change desired
• In determining the change desired, consider not only whether a difference is observed but also whether or not that change is meaningful
• Researchers should consider not only whether a difference is observed but also whether or not that change is meaningful
• Detecting a meaningful change is not a statistical procedure, rather it is based on observing a change that has a tangible outcome
• Determining what is a meaningful change in behaviour or performance outcome is possible with any measure but there are particular challenges with self-report inventories

The Baseline

• In order to determine change (statistical, meaningful or both) it is necessary to collect baseline data
• Baseline collection of data will also inform the intervention to be delivered
• Considers the stability and its length
• The smaller the variability the easier it is to detect changes
• The bigger the variability the easier it is to not detect change.
• Average number of runs can help eliminate the variabilty in the data to facilitate comparison with post-intervention data
• Variability in baseline can make determining a meaningful change after the intervention difficult

Problems Measuring Baselines

• Data are inherently variable may help to average them and to present them in blocks
• Guidelines for visual analysis suggest that overlap between data points makes drawing conclusions through visual analysis difficult
• It Depends on the extremity of any variability or trend,Analysis of the data can consider the trend line and reversing an increase in an undesirable outcome could of course be seen as a success.

Baseline Length

• Baseline length is the tension between the strongest possible research design and the demands of working in an applied setting where practical circumstances may upset a researcher's plans
• Participants are already trying to seek help from an expert and may be unwilling to do a lengthy test
• The Purpose of the baseline is to know the standard and know the level before.

Determining the right Baseline

• Researchers has determined the behavioural changed
• Researcher should be aware of natural variation in interpreting changes
• One particular type of variation pertinent to single-case research is regression, which is the tendancy for a behaviour to revert to normal.

Assessing Long Term Change

• The degree in which change is observed is a chance that the results were accurate
• If the participant's behaviour is related to the intervention that was useful.
• Underlying issues are likely to arise if the underlying issues are present.

Graphical Representation

• Visual inspection of graphfical data is widely used and easily undetstood method of data analysis for single-case research -Key information to consider: different phases should be clealry delineated, data paths using geometric forms should be clearly delineated, graphs should be labelled clearly and logically and the scaling system easy to read