Scientific Communication — Translate empirical findings for non-scientific audiences - Eppp Skills

SECTION 1

Historical Context & Motivation

The challenge of communicating scientific findings to lay audiences is as old as science itself, but the formal study of scientific communication became a distinct discipline only in the twentieth century. In behavioral health, the stakes of poor translation are especially high: when clients, families, or policymakers misunderstand research findings, treatment decisions suffer and public trust in evidence-based practice erodes. The evolution of science communication in psychology reflects broader cultural shifts—from an era in which clinicians were seen as unquestioned authorities to one in which informed consent, shared decision-making, and public mental health literacy are considered ethical imperatives.

1948 • World Health Organization Founded

The WHO's establishment foregrounded the need for clear public health messaging, including mental health, across diverse global populations with varying literacy levels.

1969 • Community Mental Health Movement

Deinstitutionalization demanded that clinicians communicate treatment rationale directly to clients and families rather than relying on institutional gatekeepers, catalyzing the need for accessible language.

1992 • Evidence-Based Practice Emerges

The EBP movement, pioneered by Sackett in medicine, required practitioners to integrate research evidence with clinical expertise and client values—necessitating that findings be translated for non-researchers.

2006 • APA Presidential Task Force on EBP

The American Psychological Association formally endorsed evidence-based practice in psychology, explicitly calling for practitioners to communicate research findings to clients in understandable terms as part of ethical practice.

2020s • Infodemic & Health Literacy Crisis

The COVID-19 pandemic exposed catastrophic consequences of poor science communication, intensifying focus on translational skills for all health professionals, including behavioral health practitioners.

This historical trajectory reveals a persistent question that the EPPP expects competent practitioners to address: How can behavioral health professionals accurately convey what research shows—including its limitations—without oversimplifying, distorting, or intimidating the people who most need to understand it? The sections that follow equip you with principles, frameworks, and practical skills to answer that question.

SECTION 2

Core Principles of Translating Empirical Findings

Effective translation of empirical findings rests on a set of interrelated principles that balance accuracy with accessibility. These principles are not mere rhetorical strategies—they are competencies embedded in the EPPP's expectation that practitioners maintain a scientific orientation to practice while serving diverse stakeholders. Each principle addresses a distinct barrier to comprehension: jargon, statistical confusion, cognitive bias, cultural mismatch, or emotional reactivity.

Fidelity to Data

The translated message must preserve the direction, magnitude, and uncertainty of findings. Simplification must never become distortion. Report what the evidence does and does not support.

Audience Calibration

Adjust vocabulary, analogies, and detail depth based on the audience's health literacy, cultural context, educational background, and emotional state. A judge in a forensic case needs different framing than a grieving parent.

Jargon Translation

Replace or explicitly define technical terms. For example, 'statistically significant' becomes 'the difference was unlikely due to chance alone,' while 'comorbidity' becomes 'having more than one condition at the same time.'

Contextual Framing

Situate individual findings within the larger body of evidence. A single study rarely warrants definitive claims. Help audiences understand where a finding sits on the continuum from preliminary to well-established.

Actionability

Translation should empower the audience to make informed decisions. Connect findings to concrete implications: 'Based on this research, here is what it might mean for your treatment options, and here are the remaining uncertainties.'

✦

KEY TAKEAWAY

Think of translating research like translating between languages: a skilled translator preserves the meaning and emotional tone of the original while using words the listener actually knows. A word-for-word translation (e.g., dumping raw statistics) confuses; a loose paraphrase (e.g., 'science says this works') distorts. The goal is a faithful, fluent rendering that honors both the data and the audience.

SECTION 3

The Translation Pipeline — From Data to Audience Understanding

The following diagram illustrates the Translation Pipeline—a conceptual framework showing how empirical findings move from raw research output to audience-ready communication. Each stage represents a deliberate transformation in which the practitioner serves as a bridge between the language of science and the language of everyday life.

The Translation Pipeline moves from raw data (Stage 1) through message distillation (Stage 2), audience assessment (Stage 3), and delivery (Stage 4), with a critical feedback loop that verifies the audience's understanding. The practitioner iterates until both scientific fidelity and audience comprehension are confirmed.

Notice that the pipeline is not purely linear. The feedback loop at the bottom—implemented through techniques like the teach-back method—ensures that translation is a dynamic, iterative process rather than a one-shot lecture. In clinical settings, this might mean asking a client to restate what they heard about the evidence supporting a treatment option, and then refining your explanation based on their response. In forensic or policy settings, it might mean providing a written summary and asking the audience to identify any points of confusion before you finalize your recommendations.

SECTION 4

Translating Statistical Concepts for Non-Scientists

One of the most challenging aspects of scientific communication involves translating statistical findings. Non-scientific audiences rarely have the training to interpret p-values, confidence intervals, or effect sizes. The practitioner must convert these abstract quantities into concrete, meaningful statements without introducing inaccuracy. Below are the key statistical constructs you will encounter and frameworks for translating each one.

Translating p-Values

STATISTICAL SIGNIFICANCE

p < .05 → "The result was unlikely due to chance alone."

A p-value below .05 means that if the treatment truly had no effect, we'd see results this extreme less than 5% of the time. For audiences: "We can be reasonably confident this wasn't a fluke." Importantly, do not say "the treatment definitely works"—statistical significance ≠ clinical importance.

Translating Effect Sizes

COHEN'S d (STANDARDIZED MEAN DIFFERENCE)

d = (M₁ − M₂) / SD_pooled

Cohen's d expresses how far apart two group averages are in standard deviation units. For audiences: d = 0.2 → "a small difference, like the height difference between 15- and 16-year-old girls"; d = 0.5 → "a medium difference, noticeable in everyday life"; d = 0.8 → "a large difference, clearly visible to most observers." The key translation principle: convert abstract units into concrete, relatable comparisons.

Translating Confidence Intervals

95% CONFIDENCE INTERVAL

95% CI [lower bound, upper bound]

For audiences: "If we repeated this study 100 times, about 95 of those results would fall within this range. So the true value is very likely somewhere between [lower] and [upper]." This translation preserves the notion of uncertainty as a feature, not a flaw of the evidence.

Number Needed to Treat (NNT)

NUMBER NEEDED TO TREAT

NNT = 1 / (rate_control − rate_treatment)

NNT is exceptionally useful for lay audiences because it is inherently concrete. For example, NNT = 4 means: "For every four people treated, one additional person improves who wouldn't have improved otherwise." This bridges the gap between abstract probability and tangible human impact.

⚠

Common Pitfall

Avoid the phrase "research proves" when communicating to non-scientific audiences. Research supports, suggests, or indicates—it rarely "proves" in the colloquial sense. Using the language of certainty where uncertainty exists violates the principle of fidelity to data and can undermine trust if the audience later encounters contradictory findings.

SECTION 5

Audience-Specific Translation Strategies

As a behavioral health practitioner, you will communicate research findings to several distinct audiences, each requiring a tailored approach. The diagram below maps the most common audiences along two critical dimensions: scientific literacy (horizontal axis) and emotional investment (vertical axis). Where an audience falls on this grid determines your communication strategy.

Each audience occupies a distinct region of the Literacy × Emotional Investment grid. Clients and families (upper left) have high emotional stakes but often low scientific literacy—requiring maximum plain language and empathic framing. Peer reviewers (upper right) have both high literacy and high investment in methodological rigor. Policymakers occupy the center, needing concise, action-oriented summaries.

Audience	Key Strategy	Language Example
Clients / Families	Use analogies, emotional validation, and teach-back. Limit numbers; emphasize what findings mean for the individual.	"Studies show that this type of therapy helps about 7 out of 10 people with similar concerns feel significantly better within 12 weeks."
Policymakers	Lead with bottom-line implications. Provide cost-effectiveness data. Use bullet-point executive summaries with full reports appended.	"Investing $1 in early intervention programs yields approximately $7 in reduced crisis care costs, based on three large-scale studies."
Judges / Courts	Be explicit about certainty levels. Define all technical terms. Distinguish between scientific consensus and your individual opinion.	"The research literature strongly supports that this condition impairs decision-making capacity. The confidence in this conclusion is high, based on 15 replicated studies."
General Public	Use narratives and visuals. Avoid hedging so excessively that the message is lost, but don't overstate certainty.	"New research suggests that regular physical activity can reduce symptoms of depression by about 25%, comparable to the effect of some medications."
Interdisciplinary Teams	Use shared professional vocabulary where it exists. Clarify discipline-specific terms that may differ across fields (e.g., 'assessment' means different things to a psychologist vs. a social worker).	"The meta-analysis—a study that combines results from many studies—showed a moderate effect size of d = 0.55, which clinically translates to a meaningful improvement in daily functioning."

Audience-specific strategies for translating empirical findings in behavioral health contexts

SECTION 6

Worked Example — Translating a CBT Meta-Analysis for a Client

Imagine you are a clinical psychologist discussing treatment options with a client experiencing moderate generalized anxiety disorder. You have reviewed a recent meta-analysis of cognitive-behavioral therapy (CBT) for GAD that reports the following: k = 41 studies, N = 2,843 participants, overall effect size g = 0.82 (95% CI [0.65, 0.99], p < .001), moderate heterogeneity (I² = 52%), and a Number Needed to Treat (NNT) of 3. Your task is to translate these findings for your client.

Translating a CBT Meta-Analysis for a Client with GAD

1
Step 1 — Identify the Core Message

Before speaking with the client, distill the meta-analysis to its bottom line: CBT has a large beneficial effect on GAD symptoms, supported by a substantial body of research. The finding is statistically robust (p < .001) and clinically meaningful (large effect size, NNT = 3). However, there is moderate variability across studies, meaning not everyone responds equally.

Core message: CBT is strongly supported for GAD, though individual results vary.
2
Step 2 — Assess the Audience

Your client has no research training. They are emotionally invested—they are anxious and seeking reassurance but also fearful of committing to a treatment that might not work. They are a college-educated professional, so they can follow structured reasoning but will not understand statistical jargon. Their primary question is: "Will this actually help me?"

Audience profile: high emotional investment, moderate general literacy, low statistical literacy.
3
Step 3 — Translate the Effect Size

Convert the Hedges' g of 0.82 into language the client can grasp. Say: "Researchers combined results from 41 separate studies with nearly 3,000 people. They found that CBT produced a large improvement in anxiety—on average, people who received CBT felt substantially better than those who received no treatment. The effect wasn't small or moderate; it was in the 'large' category." Then use the NNT: "To put it another way, for every 3 people who receive CBT, 1 additional person gets significantly better who wouldn't have improved on their own."

Translated: "Large improvement; for every 3 people treated, 1 extra person gets significantly better."
4
Step 4 — Address Uncertainty Honestly

Acknowledge the moderate heterogeneity without undermining hope: "Not everyone responds the same way. Some people see dramatic improvement, while others see more modest gains. The research can't predict exactly how you personally will respond, but the overall pattern is very encouraging. We'll monitor your progress closely and adjust our approach if needed." This framing honors the I² = 52% without introducing a number that would confuse the client.

Translated uncertainty: individual variation acknowledged with collaborative monitoring plan.
5
Step 5 — Verify Understanding (Teach-Back)

Ask the client: "I want to make sure I explained that clearly. In your own words, what did you take away about what the research says about CBT for anxiety?" If the client's response accurately captures the core message—that CBT is well-supported but not guaranteed—you have achieved a successful translation. If they overstate certainty ("So it'll definitely fix my anxiety") or understate the evidence ("So it might not even help"), gently redirect.

Feedback loop confirms fidelity + comprehension.

SECTION 7

Common Strengths and Pitfalls in Science Translation

Effective scientific communication is a skill that can be developed through deliberate practice, but practitioners commonly fall into predictable traps. Understanding these pitfalls—and their corresponding best practices—prepares you to communicate with both accuracy and empathy. The table below contrasts frequent errors with recommended alternatives.

Common Pitfall	Why It's Problematic	Best Practice Alternative
Jargon dumping	Using terms like 'comorbid psychopathology' or 'regression to the mean' without definition alienates the audience and creates an illusion of understanding.	Replace with plain-language equivalents. If a technical term is essential, define it immediately using everyday language.
False certainty	Saying 'research proves X' when studies support but don't definitively prove. This undermines trust when future findings nuance or contradict the claim.	Use calibrated language: 'strong evidence suggests,' 'most studies indicate,' 'the current best evidence supports.'
Excessive hedging	Over-qualifying every statement ('It might possibly, under some circumstances, potentially help some people') leaves the audience with no actionable information.	Match hedging to actual uncertainty. Well-replicated findings deserve confident language; preliminary findings warrant more caution.
Ignoring base rates	Reporting relative risk reduction (e.g., '50% reduction in risk') without base rates makes small absolute changes sound dramatic.	Use absolute numbers or NNT: 'The risk dropped from 4 in 100 to 2 in 100' is more honest than '50% reduction.'
Ignoring cultural context	Assuming all audiences interpret concepts like 'mental illness,' 'therapy,' or 'medication' similarly can cause miscommunication and reduce engagement.	Ask about the audience's framework for understanding distress. Integrate culturally resonant language and metaphors.

Common pitfalls and best practices in translating empirical findings

✦

KEY TAKEAWAY

The sweet spot of scientific communication is what researchers call the Goldilocks Zone of certainty: not so confident that you overstate the evidence, and not so hedged that you render the evidence useless. Think of it like a weather forecast—a good forecaster says "there is a 70% chance of rain, so bring an umbrella" rather than "it will definitely rain" or "we can't really predict weather." Your job is to give the audience a clear, honest forecast of what the evidence suggests, along with the right 'umbrella' of actionable guidance.

SECTION 8

Ethical and Professional Standards for Science Communication

Translating empirical findings is not merely a communication skill—it is an ethical obligation deeply embedded in professional standards. The APA Ethics Code, EPPP competency frameworks, and informed consent principles all converge on the requirement that practitioners communicate scientific information responsibly. Understanding these standards is critical for both examination preparation and clinical practice.

Standard / Principle	Relevance to Science Communication
APA Ethics Code 2.04 — Bases for Scientific Judgments	Psychologists base their scientific and professional judgments on established scientific and professional knowledge. When communicating findings, you must accurately represent the state of the evidence, not selectively cite studies that support a preferred conclusion.
APA Ethics Code 3.10 — Informed Consent	Informed consent requires that clients understand the nature of the proposed services, including their empirical basis. This is impossible if findings are communicated in inaccessible jargon.
APA Ethics Code 9.01 — Bases for Assessments	In forensic and evaluation contexts, psychologists must communicate assessment results and their empirical support clearly to legal decision-makers, who typically lack statistical training.
EPPP Domain 1 — Scientific Orientation	The EPPP explicitly assesses the ability to apply scientific reasoning in practice, which includes translating research findings for stakeholders as part of evidence-based decision-making.
Beneficence & Nonmaleficence (Principle A)	Poor translation can cause harm—a client might decline an effective treatment due to misunderstood risk data, or a policymaker might defund a program based on a misleading summary. Accurate translation serves beneficence.

Ethical standards relevant to translating empirical findings

Looking ahead, the field is increasingly recognizing that science communication competency should be a formal training requirement, not merely an assumed skill. Graduate programs in clinical and counseling psychology are beginning to incorporate health literacy assessment and science communication practica into their curricula. For EPPP preparation, the essential point is that translating findings accurately is not optional—it is an ethical and professional competency that intersects with virtually every domain of practice.

SECTION 9

Practice Problems

Question 1: CONCEPTUAL

A psychologist tells a client: "The meta-analysis demonstrated a Hedges' g of 0.76 with an I² of 47%, suggesting moderate heterogeneity across the included RCTs." Identify at least three specific problems with this communication and explain why each is problematic for a non-scientific audience.

Show answer

Three key problems: (1) The term 'Hedges' g' is undefined statistical jargon—the client has no framework for interpreting this number or knowing whether 0.76 is small, medium, or large. (2) 'I² of 47%' introduces a second undefined statistic and the phrase 'moderate heterogeneity' uses technical vocabulary that means nothing to a layperson. (3) The acronym 'RCTs' (randomized controlled trials) is used without explanation, leaving the client unclear about what kind of studies were included. The overall communication violates the principles of jargon translation and audience calibration. A better version might be: 'Researchers combined results from many high-quality studies and found a meaningful improvement—most people who received this treatment felt noticeably better, though the amount of improvement varied somewhat from study to study.'

Question 2: BASIC APPLICATION

A study finds that a new anxiety intervention reduces symptom severity scores from 24 to 16 on a standardized measure (scale range: 0–40, clinical cutoff: 20). The control group scores changed from 23 to 21. Translate these findings into one paragraph suitable for a client considering this treatment.

Show answer

A suitable translation might read: 'In this study, people who received the new treatment saw their anxiety scores drop from above the clinical threshold—meaning levels that typically interfere with daily life—to below it. On average, they moved from what clinicians would call the 'clinically significant' range into the 'normal' range. People who didn't receive the treatment showed only a small change and generally remained above the threshold. This suggests that the treatment produced a meaningful, real-world difference—not just a statistical one—by moving people from a level of anxiety that causes problems to one that most people would consider manageable.' This translation preserves the direction and clinical significance of findings while avoiding raw numbers that lack context for the client.

Question 3: INTERMEDIATE

You are presenting to a school board about the effectiveness of a school-based social-emotional learning (SEL) program. The research shows: d = 0.30, NNT = 8, p < .01, and the program costs $150 per student per year. The comparison condition is no SEL program. Draft a two-paragraph executive summary appropriate for this audience, and explain the translation choices you made.

Show answer

Executive summary: 'Research involving thousands of students shows that this social-emotional learning program produces a meaningful, though moderate, improvement in students' emotional skills, behavior, and academic performance. For every eight students who participate, one additional student shows clinically important gains who would not have improved without the program. At $150 per student per year, this means each significant student improvement costs approximately $1,200—a figure that compares favorably to the estimated $2,500–5,000 annual cost of disciplinary interventions for struggling students. The results were statistically robust, meaning they are very unlikely to be due to chance.' Translation choices: (1) The effect size d = 0.30 was characterized as 'meaningful, though moderate' to avoid both overstating and understating a small-to-medium effect. (2) NNT was preserved because it is inherently concrete and decision-relevant for policymakers. (3) Cost per improvement was calculated to connect scientific findings to budget decisions. (4) Statistical significance was described in plain terms without citing the p-value itself.

Question 4: APPLIED

A parent of a child with ADHD asks you: "I read online that medication works better than behavioral therapy. Is that true?" You know that the MTA study found medication superior to behavioral treatment alone for core ADHD symptoms (d = 0.56) but that combined treatment was superior to medication alone for functional outcomes, and that long-term follow-ups show attenuated differences. Craft a response that is accurate, balanced, culturally sensitive, and empowering for the parent.

Show answer

A well-crafted response: 'That's a really important question, and I'm glad you're looking into the research. The answer depends on what outcomes we're focusing on. The largest study on this—which followed hundreds of children—found that medication was more effective than behavioral therapy alone for reducing the core symptoms of ADHD, like difficulty focusing and hyperactivity. However, when researchers looked at the bigger picture—things like how children got along with peers, how families functioned, and how children felt about themselves—the combination of medication and behavioral strategies worked best. There's also an important long-term finding: after several years, the differences between groups narrowed, suggesting that no single approach maintains its advantage forever without ongoing adjustment. What this means for your child is that there isn't a one-size-fits-all answer. The strongest approach is usually a combination, tailored to your child's specific needs. I'd like to help you think through what combination might work best for your family.' This response demonstrates fidelity to data (reporting both the short-term medication advantage and the long-term attenuation), audience calibration (validating the parent's research, using accessible language), contextual framing (positioning one study within broader patterns), and actionability (offering collaborative next steps).

Question 5: CRITICAL THINKING

A colleague publishes a press release about their new study stating: "Our groundbreaking research proves that mindfulness cures depression." The actual study was a single RCT (n = 48) comparing an 8-week mindfulness intervention to a waitlist control, finding a statistically significant reduction in PHQ-9 scores (p = .03, d = 0.61). Conduct a detailed critique of the press release, identifying at least five specific distortions, and write a revised version that accurately represents the findings.

Show answer

Five distortions: (1) 'Proves' implies definitive causal certainty, which a single study with 48 participants cannot establish—'suggests' or 'provides initial evidence' would be appropriate. (2) 'Cures depression' implies complete symptom elimination, whereas a significant reduction in PHQ-9 scores means symptoms decreased, not that they were eliminated. (3) 'Groundbreaking' implies novelty that may not exist—mindfulness for depression has been studied extensively, and one more small trial does not constitute a breakthrough. (4) The waitlist control is a weak comparator that inflates effect sizes due to expectancy and attention effects—the press release should note that the comparison was to no treatment, not to an alternative active treatment. (5) The sample size (n = 48) is small, meaning the findings are preliminary and the confidence interval around the effect size is likely wide, which goes unmentioned. Revised press release: 'A new study of 48 adults with depression found that an 8-week mindfulness program was associated with a moderate reduction in depression symptoms compared to receiving no treatment during the same period. These findings add to growing evidence that mindfulness-based approaches may be helpful for some people with depression, though larger studies comparing mindfulness to established treatments are needed to confirm these results and understand who benefits most.' This revision preserves the positive finding while accurately representing the study's limitations.

SUMMARY

Summary — Translating Empirical Findings for Non-Scientific Audiences

Summary

Translating empirical findings for non-scientific audiences is a core competency within the EPPP's scientific orientation to practice. Effective translation follows the Translation Pipeline: identify the core message, assess your audience's literacy and emotional state, translate using plain language and concrete analogies, and verify understanding through the teach-back method. Five core principles guide this process: fidelity to data, audience calibration, jargon translation, contextual framing, and actionability.

Key statistical concepts like effect sizes, confidence intervals, and Number Needed to Treat should be converted into concrete, relatable language—numbers become stories, probabilities become proportions, and uncertainty becomes honest acknowledgment of what we know and don't know. This competency is not merely a communication nicety; it is grounded in APA ethical standards including informed consent (3.10), scientific bases for judgments (2.04), and the principle of beneficence and nonmaleficence. Remember the Goldilocks Zone: communicate with enough confidence to be useful, enough caution to be honest, and enough empathy to be heard.