Pragmatic Solomon Four-Group Design — Real-World Pretest Sensitization Control
Pragmatic Solomon Four-Group Experimental Design · Also known as: pragmatic S4GD, real-world Solomon four-group design, pragmatic pretest-control design, pragmatic Solomon design
The Pragmatic Solomon Four-Group Design combines the pretest-sensitization control logic of the classic Solomon (1949) four-group structure with the broad eligibility, flexible delivery, and real-world conditions characteristic of pragmatic trials. Four groups are formed: two receive the intervention (one pretested, one not) and two serve as controls (one pretested, one not), allowing simultaneous estimation of treatment effects and pretest sensitization effects under ecologically valid settings.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use this design when you need to (a) evaluate an intervention in real-world practice conditions rather than efficacy conditions, and (b) are concerned that administering a pretest may itself change how participants respond — a risk that is especially high when the pretest is a questionnaire that prompts reflection on the very behavior being targeted by the intervention. It is well-suited to educational, behavioral, and public health interventions delivered through routine systems such as schools, clinics, or workplaces. Do not use it when you cannot randomize at least four separate groups — the minimum requirement for the Solomon structure. Avoid it when the pretest is unlikely to sensitize participants (e.g., biological or physiological measures), as the added complexity yields no benefit. It is also inappropriate when the outcome must be measured longitudinally at multiple time points for individual participants, since Groups 3 and 4 have no baseline.
Strengths & limitations
- Directly estimates and controls for pretest sensitization, a bias that standard pretest-posttest designs cannot detect.
- Pragmatic framing maximizes external validity and real-world applicability of findings.
- The 2x2 factorial structure allows simultaneous estimation of treatment effects and pretest-by-treatment interactions.
- Non-pretested groups (3 and 4) provide a clean, uncontaminated posttest comparison that is free of reactivity bias.
- When the sensitization interaction is absent, pooling Groups 1 and 3 (and 2 and 4) increases effective sample size and statistical power.
- Requires four separate randomized groups, substantially increasing the required sample size compared to a simple two-group design.
- Groups 3 and 4 have no baseline measurement, making individual-level change scores unavailable for those participants.
- Greater logistical complexity in pragmatic settings where usual-care providers manage multiple patient streams.
- The pragmatic orientation reduces internal control, making it harder to attribute effects to specific active ingredients of the intervention.
Frequently asked
Why is the Solomon four-group design not used more often if it controls for sensitization?
The main barrier is sample size. Splitting participants into four groups rather than two roughly doubles the required N for equivalent power in the primary treatment comparison, making the design expensive. Researchers often judge the pretest sensitization risk as low enough to tolerate, especially when baseline measures are physiological rather than attitudinal. The design is most justified when the intervention targets the same attitudes or behaviors that the pretest directly measures.
What is the difference between a pragmatic Solomon four-group design and an explanatory Solomon four-group design?
Both use the same four-group structure to control for pretest sensitization. The pragmatic version applies broad eligibility criteria, delivers the intervention through routine practice channels without a scripted protocol, uses patient-relevant or system-relevant outcomes, and operates with a light compliance and fidelity monitoring framework. The explanatory version applies strict eligibility, highly controlled delivery protocols, and is primarily interested in efficacy under ideal conditions. The pragmatic version sacrifices some internal control to gain external validity and direct generalizability to real practice.
How do I analyze data from a pragmatic Solomon four-group design?
The standard approach is a 2x2 factorial ANOVA (or ANCOVA if covariates are included) on the posttest scores of all four groups: Factor A = intervention status (yes/no), Factor B = pretested (yes/no), and their interaction A x B. A significant interaction indicates that pretest sensitization occurred. If the interaction is non-significant, the simpler main effect of Factor A is the primary estimate of the treatment effect. Mixed-model or multilevel approaches are preferred when cluster randomization is used, as in most pragmatic settings.
Can I add a follow-up measurement to the pragmatic Solomon design?
Yes. A delayed posttest (e.g., 3 or 6 months after intervention) can be added to all four groups to assess sustainability of effects. However, by the follow-up time point, Groups 3 and 4 still have no baseline, so individual-level change analysis remains limited to Groups 1 and 2. The factorial ANOVA framework extends naturally to include the additional time point as a repeated measure.
Is the pragmatic Solomon four-group design suitable for cluster randomization?
Yes, and cluster randomization is often the natural choice in pragmatic settings where providers or sites — rather than individuals — deliver the intervention. Clusters (schools, clinics, wards) are randomized to one of the four cells. The analysis must account for within-cluster correlation using multilevel models or GEE; standard ANOVA will produce anti-conservative standard errors and inflated Type I error rates.
Sources
- Solomon, R. L. (1949). An extension of control group design. Psychological Bulletin, 46(2), 137–150. DOI: 10.1037/h0062958 ↗
- Thorpe, K. E., Zwarenstein, M., Oxman, A. D., Treweek, S., Furberg, C. D., Altman, D. G., ... & Chalkidou, K. (2009). A pragmatic–explanatory continuum indicator summary (PRECIS): a tool to help trial designers. Journal of Clinical Epidemiology, 62(5), 464–475. DOI: 10.1016/j.jclinepi.2008.12.011 ↗
How to cite this page
ScholarGate. (2026, June 3). Pragmatic Solomon Four-Group Experimental Design. ScholarGate. https://scholargate.app/en/experimental-design/pragmatic-solomon-four-group-design
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Cluster Randomized Controlled TrialExperimental design↔ compare
- Control Group Experimental DesignExperimental design↔ compare
- Pragmatic Randomized Controlled TrialExperimental design↔ compare
- Pretest-Posttest Experimental DesignExperimental design↔ compare
- Solomon Four-Group DesignExperimental design↔ compare