Strengths and Limitations

Attachment L:
Strengths and Limitations of the SingleSubject Multiple Baseline Design
Research on the Efficacy and Feasibility of
Essentials for Parenting Toddlers and Preschoolers
Division of Violence Prevention
National Center for Injury Prevention and Control
Centers for Disease Control and Prevention

Strengths
Despite their limited use, single-subject research designs have significant methodological advantages
over large-sample research designs that could be used to facilitate more rapid clinical research advances.
First, such designs can demonstrate clear causal relations between intervention and behavior change
with much more efficiency than large-sample designs (Nock, Michel, & Photos, 2007). Indeed, largesample studies typically require enormous amounts of time, financial resources, and staff support to
implement. In contrast, any adequately trained researcher or clinician with a modest amount of resources
and effort can use single-case research designs (Nock et al., 2007). Given this advantage, researchers
and clinicians can use single-subject research designs in rapid succession to develop and evaluate
individual or multiple versions of interventions, and thus can do in a matter of weeks what could take
years (and millions of dollars) using large-group designs (Nock et al., 2007).
Second, single-case research designs offer much more flexibility in the implementation and evaluation
of interventions than large-sample designs (Nock et al., 2007). Indeed, with a focus on maintenance of
consistency of procedures across many subjects, group designs do not allow for a tailoring of the
manipulation or intervention to the individual. The opportunity to modify interventions as needed
provides greater research and clinical options and can lead to more innovative treatment development.
Using variations on the single-subject experimental designs as described previously, researchers can
flexibly complete efficacy or effectiveness studies, dismantling studies, parametric studies, or any other
evaluations that can be addressed using large-sample designs—and can do so within or between subjects
(Nock et al., 2007).
Third, the assessment methods used in single-subject research designs provide for the evaluation of
individual change patterns in the data. Large sample designs most often employ only pre- and posttreatment assessment, precluding evaluations of how and why individuals change over the course of
treatment (Kazdin & Nock, 2003). The use of continuous assessment and multiple experimental phases
in single-subject research designs allow for detailed examinations of patterns of change and the temporal
relations between manipulations and their effects over time.
Fourth, although the widespread use of large sample designs grew largely out of advances in inferential
statistical methods, these very methods have come under attack due to several problems with the way
they are most often used (Cohen, 1990; Krueger, 2001; Loftus, 1996; Nickerson, 2000). Most of these
statistical or methodological problems are avoided or remedied through the use of single-case research
designs.
Limitations
Of course, single-subject research designs also have clear limitations that must be considered. The
limitation most often cited in discussions of single-subject research designs is a lack of generality of
obtained effects. Indeed, interventions shown to be effective for a single individual may not be effective
with other individuals, and these effects may not even replicate when re-administered to the same
individual at a later time. Although this is a clear limitation of single-subject research designs, two
caveats should be kept in mind. First, the use of large-sample designs does not preclude the occurrence
of such problems. Indeed, what is needed for generality is the evaluation of obtained effects using
different populations, conditions, or settings. Effects obtained using a homogeneous sample of
individuals (which is the rule rather than the exception in large-sample studies) also may suffer from a
lack of generality.

Second, the use of multiple and heterogeneous individuals within and across studies, can be
implemented to demonstrate generality. For the current project, we are aiming to recruit a heterogeneous
group of 200 parents, and we will assign them to different conditions (guided or natural navigation) and
order of conditions (for the guided navigation group). We are also planning to use a multiple baseline
design, which will help with generality. Given that several participants will be assigned to each
condition and we are using a multiple baseline design across individuals, we will be able to examine
effects in a “single group” of parents. In this way, we will be able to determine that the intervention is,
on average, effective (or not) for the small group of parents as well as for individual members in the
“group”.
Another limitation cited for single-subject designs is related to testing. In the current study, it is likely
that exposure to some of the measures can affect “scores” on other measures or repeated exposure to a
measure can lead to socially desirable responding or reactive responding. Strategies to reduce or
eliminate these influences have been included. For example, in single-subject research, the repeated
assessment of the dependent variable(s) across phases of the design can help identify this potential
threat. Replicating the effect across multiple individuals at various points in time also helps to reduce the
plausibility of a claim that repeated assessment per se accounted for the intervention effect or that some
external influence resulted in the change.
For additional information on the technical aspects of single-subject designs, please consult technical
documentation compiled for the What Works Clearinghouse (Kratochwill et al., 2010):
http://ies.ed.gov/ncee/wwc/pdf/reference_resources/wwc_scd.pdf.
References
Kratochwill, T. R., Hitchcock, J., Horner, R. H., Levin, J. R., Odom, S. L., Rindskopf, D. M. & Shadish,
W. R. (2010). Single-case designs technical documentation. Retrieved from
http://ies.ed.gov/ncee/wwc/pdf/wwc_scd.pdf.
Nock, M. K., Michel, B. D., & Photos, V. (2007). Single-case research designs. In D. McKay (Ed.),
Handbook of research methods in abnormal and clinical psychology (pp. 337–350). Thousand
Oaks, CA: Sage Publications.