8.1 Selecting Reliable Evidence-Based Practices

Selecting Reliable Evidence-Based Practices

Quick Answer: An evidence-based practice (EBP) is an instructional or behavioral practice with a documented, replicated research base showing it improves outcomes for students with or at risk for disabilities — verified through sources such as the What Works Clearinghouse (WWC), peer-reviewed research, and the Council for Exceptional Children (CEC) Standards for Evidence-Based Practices, not through popularity, tradition, or a vendor's marketing claims. A practice only produces its documented effect when delivered with fidelity — as designed, at the intended dosage, in the intended sequence.

Every day, teachers are approached with claims that a new program, product, or technique will transform outcomes for their students with disabilities. The FTCE Exceptional Student Education K-12 exam tests whether a candidate can tell the difference between a practice that has actually earned that status through rigorous research and one that merely borrows the language of evidence-based or research-based as a marketing label. This distinction matters because students with disabilities have limited instructional time and cannot afford to have that time spent on approaches that do not work.

What Makes a Practice Evidence-Based

A practice earns EBP status when independent researchers have demonstrated, through a causal research design — not just a correlation or a single success story — that using the practice produces a meaningfully better outcome than not using it (or than an alternative practice), and that this effect has been replicated across multiple studies, settings, or research teams. This is different from:

• Promising practices — supported by some research, but not yet enough high-quality, replicated evidence to be classified as EBP.
• Practice-based evidence — insights from experienced practitioners' clinical judgment, valuable for informing decisions but not a substitute for a research base.
• Tradition or convenience — we've always taught it this way or the district already bought the materials are not evidence.

Where to Verify: Reliable Sources

The Evidence Hierarchy in Special Education Research

Special education relies heavily on single-case (single-subject) research design — including multiple-baseline, withdrawal/reversal, and alternating-treatments designs — because it can demonstrate a functional relationship between a practice and an individual student's behavior, which is essential given how heterogeneous and low-incidence many disability categories are. A single-case study is considered strong evidence only when a functional relationship is demonstrated through visual analysis of level, trend, and variability across at least three demonstrations of an effect, ideally replicated across studies or research teams.

Group experimental and quasi-experimental designs (including randomized controlled trials) provide stronger evidence for effects across a population and are common in general-education-oriented research reviewed by the WWC. Correlational studies can identify a relationship between two variables but cannot establish that one caused the other — useful for generating hypotheses, not for adopting a practice. Anecdote and testimonial — a single family's story, a single teacher's impression, a conference talk — sit at the bottom of this hierarchy and are never sufficient justification for selecting a practice for a student with a disability.

Recognizing Red Flags: Practices to Question

Some widely marketed approaches lack an independent research base or have been directly contradicted by controlled studies, including facilitated communication and the rapid prompting method as methods of establishing authorship for students who cannot otherwise communicate, patterning-based motor programs, and unverified sensory-diet claims presented as a stand-alone cure for academic or behavioral difficulty. Common red flags include:

• No independent, peer-reviewed replication — only vendor-produced or single-source research
• Reliance on testimonials or anecdotes rather than controlled comparison data
• Claims of dramatic, rapid, across-the-board improvement for every student regardless of need
• Marketed as a single cure-all rather than a targeted practice matched to a specific skill deficit
• Resistance to independent evaluation or refusal to publish methodology

Fidelity of Implementation: The Missing Half of Evidence-Based

Even a practice with a strong WWC or CEC rating will not reliably produce its documented effect unless it is delivered with fidelity — sometimes called treatment integrity or procedural fidelity. Fidelity means implementing the practice the way the research demonstrated it: the correct steps, in the correct sequence, at the intended dosage (frequency and duration), in the intended group size and setting, using the specified materials. A teacher who delivers only part of a validated reading intervention, skips steps, or reduces session frequency is not implementing the evidence-based version of that practice — the research base does not transfer to a modified, low-fidelity delivery.

Fidelity is monitored through direct observation with a fidelity checklist, coaching and performance feedback, self-monitoring logs, review of permanent products (worksheets, data sheets), and tracking of dosage (sessions delivered versus sessions planned). For the FTCE, remember that a scenario describing a teacher who selected a strong-evidence practice but implemented it inconsistently or partially is testing fidelity, not the validity of the practice itself — the fix is coaching and monitoring, not abandoning a sound practice.

Bringing It Together

Selecting a reliable practice for a student with a disability requires two connected judgments: verifying that the practice itself has a documented, replicated research base from a credible source such as the WWC, peer-reviewed literature, or CEC standards, and confirming that it is being delivered as designed. A practice that fails on either count — no real evidence base, or a strong evidence base delivered with low fidelity — cannot be expected to produce the outcome a teacher or family is hoping for.