Group designs (e.g. pre-post studies or RCTs) help give a generalised picture of response to a particular condition/intervention/treatment. However, it prevents the investigator from making any conclusions about how specific individuals react. This is where single case (or small-N) designs come in useful, a tool that can be used by practitioners to monitor progress of their own work. The idea of single case designs flourished in the 1970s as a means of evaluating the outcomes of one's own practice.

Such designs may take on a simple A-B makeup. A practitioner planning to carry out such an experiment with an individual would firstly need to get a baseline measure of the response or behaviour under investigation (A) (stability of measures at baseline is advised before proceeding with the intervention, although this may not always be possible). Once this has been established, the intervention is then delivered, changing conditions and measuring how this affects the participant's response (B). The intervention is then withheld, with the participant's response measured once again.

An A-B-A design is an improvement on the above, adding a reversal stage to the design, removing the intervention and then taking another measurement. If the variable being examined is effective, a return to baseline scores should ensure. When this does not occur, the practitoner is left needing to explain what other events could have brought about a change or prevented any change from resulting. The problem with such a design is that practitioners will have reservations about removing a treatment when they are supposed to be helping someone. In an attempt to try and overcome ethical problems related to treatment withdrawal, an A-B-A-B design could be tried, in which an individual is given then withdrawn an intervention and then given it once again. This process can be repeatedly followed creating an A-B-A-B-A-B design, increasing the investigator's confidence in the relationship between the intervention and outcome.

An example might be treating childhood enuresis (bed-wetting). A parent might be advised to alternate a period of fulsome praise for a dry bed with a period when dry beds are treated as normal and no fuss is made. Some children respond better to the first approach, others to the second. The outcome measure would be the number of nights without bed wetting, which would suggest whether one treatment was more effective than another.

The single case may not necessarily be an individual; it could also be an individual unit (a school class, a specific service). The main focus of such research is that repeated measures of the same individual (or unit) are taken at specific times, usually pre and post intervention.

Benefits of single case studies

1. These types of experiments are often consistent with routine practice and so are not time consuming or costly.
2. The information generated can help fill in the details about exactly 'what works for whom' that are often left unexamined in group experiments.
3. As accountability in practice becomes more important, this is a useful method of demonstrating outcomes, and a means of linking research and practice.
4. It allows for clinical self-monitoring and practice evaluation.
5. Data from single case designs can be used to supplement group study designs since these often fail to take account of within group or individual differences.

Limitations of single case studies

1. Conflict can arise between research requirements (fixed interventions phases, predetermined & limited goals, standardised intervention methods) and practice requirements (flexibility, responsiveness to client needs).
2. Practical constraints: time limits, premature termination, change of client focus/ life events. But perhaps these practical constraints could be overcome through replication of studies and the accumulation of data over time.
3. Replication of single case studies is essential to allow for the generalisation of findings, yet debates ensue over the extent of reliability and generalisability. To what extent is a single case study sufficient evidence to modify theory? Perhaps these studies can best be regarded as pilots for group designs.