Read, Talk, Write: Running a CER So Students Argue Before They Write
A claim-evidence-reasoning task asks a student to do three hard things at once: understand a text, decide what it shows, and defend that decision in writing. Read, Talk, Write splits the work into three moves in a fixed order. Students read alone with a pencil moving, talk the argument through with a partner, and only then write the CER on their own. The talk is the rehearsal. A student who has already said "I think the plants wilt because the soil water is saltier than the cells" has most of a claim and the start of the reasoning before the writing begins.
Where the two pieces come from
The CER framework comes from Katherine McNeill and Joseph Krajcik, who built it over years of classroom research with middle school science students. Their version has four parts, not three: claim, evidence, reasoning, and a rebuttal that answers an alternative explanation. In one study of 331 seventh graders working through an eight-week chemistry unit, students whose written supports were faded over time wrote stronger reasoning on later tests than students who kept the full supports throughout. The lesson for a high school teacher is to start with sentence frames and take them away as the year goes on.
Read, Talk, Write is a close cousin of Nancy Motley's Talk, Read, Talk, Write, a routine widely used in classrooms with English learners. No trial has tested the routine by name. Its support comes from research on each of its parts, and that research is strong enough to build on.
Arguing from evidence is how science decides
Outside a lab, "argue" means a quarrel with a winner and a loser. Inside one, it means making a claim, showing the evidence, and inviting colleagues to find the weak spot. Scientists argue at lab meetings, in peer review, at conferences, and in the replies to each other's papers, and a claim that survives that pressure is one people start to trust. The National Research Council's framework for K–12 science names "engaging in argument from evidence" as one of the eight practices of science and engineering, alongside asking questions and analyzing data. It is part of the work, not a detour from it. Jonathan Osborne and Alexis Patterson draw the line that matters for a CER: an explanation says why something happens, while an argument defends whether a claim is right against another one. A student who can do both is doing what scientists do. For how independent lines of evidence build that kind of confidence, see The Chain and the Rope: How Science Gets Confident Without Getting Proof. For the classroom version of scientists checking each other's work, see Peer Review in the Classroom.
Read: a page worth arguing over
Talk only works if the reading gives students something to disagree about. A paragraph of definitions ends a partner conversation in thirty seconds. A good CER reading runs about one page, 600 to 750 words, and carries four things: a real question, numbers a student can point at, a place or person close to home, and vocabulary students have already met. The CER readings in the Biology section are built to that length on purpose.
Reading-centered science instruction has held up in rigorous tests at the high school level. In a randomized trial in high school biology, teachers who took ten days of Reading Apprenticeship training saw their students outscore controls on state tests in biology, English and reading (Greenleaf and colleagues, 2011). A second randomized trial in ninth-grade biology, Project READI, found that students who read several texts to build evidence-based models of a phenomenon improved in comprehension and in modeling new phenomena (Goldman and colleagues, 2019). Both depended on sustained teacher training, not on one routine alone.
- Read alone. Underline each claim about the real world and circle each unfamiliar science word as you go.
- Talk with a partner. What is the claim? Which two pieces of evidence are strongest? Does anything in the reading not fit?
- Write alone. One claim, two pieces of evidence, and the science rule that connects them.
Talk: rehearse the argument out loud
Unstructured talk drifts. Talk with ground rules does not. Neil Mercer and colleagues found that teaching students explicit rules for discussion, such as giving reasons and asking a partner for theirs, increased reasoned "exploratory talk" and improved reasoning. Jonathan Osborne, Sibel Erduran and Shirley Simon showed that science teachers given argument-based materials and training raised the quality of their students' arguments. Leema Berland and Brian Reiser add a warning worth keeping: students readily use evidence to make sense of something, but rarely try to persuade anyone, because ordinary classroom routines give them no one to persuade. The partner in Read, Talk, Write is that someone.
Stems help students start:
- I think the claim is…
- The strongest evidence is… because…
- That does not fit with…
- What would change your mind?
For English learners, the talk step is where new vocabulary gets used out loud before it has to be written. Okhee Lee, Helen Quinn and Guadalupe Valdés argue that science practices done through language grow both science learning and language learning. The results for current English learners on state tests are more mixed than that argument suggests, so the honest claim is that talk gives these students a real chance to use the language, not that it closes gaps by itself.
Write: the CER, scored against a rubric students have seen
| 3 · Meets | 2 · Developing | 1 · Beginning | |
|---|---|---|---|
| Claim | Accurate, complete, in science terms | Partly accurate or vague | Inaccurate or missing |
| Evidence | Enough relevant data from the reading | Some relevant evidence, but thin | Little or none |
| Reasoning | Explains why the evidence supports the claim | Connects evidence and claim, but the science is weak | No connection, or a misconception |
Score the reading before you score the students. A reading with no numbers caps every student at a 2 in Evidence, and a reading that defines its key idea two different ways sets students up for a misconception in Reasoning. Reviews of rubric research find that rubrics score more consistently when paired with exemplars, and that students improve most when they see the rubric while they work, not only after (Jonsson and Svingby, 2007; Panadero and Jonsson, 2013). Show students a 3 and a 2 on the same reading before they write their own.
When class time is short
Few teachers can give a CER a full period every week. The protocol splits cleanly:
- Read at home. One page on the iPad the night before. A little reading at home is reasonable.
- Talk in class. Five to eight minutes with a partner, while you listen in. This is the step that needs a teacher in the room.
- Write in FLEX. The CER, finished in the Word file and turned in.
A Word file saves as a student types on the iPad, so a CER started in class can be finished in FLEX or at home. Canvas Student Annotation seals each attempt, which makes it the wrong tool for work that spans two sittings.
When the reading came from AI
Readings built with ChatGPT or another AI tool need one more habit, and students can own it. Before a reading counts as evidence, students trace each claim to a source, flag unfamiliar terms, count the numbers and their units, and check that the key definition means the same thing every time it appears. The prompts and those four checks are at scienceonepagers.org/cer. The nearest research is on lateral reading: professional fact-checkers leave a page to check its source elsewhere, and a district-wide study found high school students taught that habit got significantly better at judging online sources (Wineburg and McGrew, 2019; Wineburg and colleagues, 2022). No classroom study yet tests high school students checking AI-written science text, so treat the four checks as sound practice rather than proven method.
What the research does not say
- No trial has tested Read, Talk, Write or Talk, Read, Talk, Write as a named routine.
- The core CER studies are from middle school, on researcher-built tests; none shows CER raising SOL scores.
- Critics argue CER as usually taught is explanation more than argument (Osborne and Patterson, 2011). The rebuttal step and a real partner to persuade move it toward argument.
- A rubric alone does not make scoring fair; exemplars and shared practice scoring do.
Sources: McNeill, K. L., & Krajcik, J. S. (2012). Supporting Grade 5–8 Students in Constructing Explanations in Science: The Claim, Evidence, and Reasoning Framework for Talk and Writing. Pearson (ERIC). McNeill, K. L., Lizotte, D. J., Krajcik, J., & Marx, R. W. (2006). Supporting students' construction of scientific explanations by fading scaffolds in instructional materials. Journal of the Learning Sciences, 15(2), 153–191 (doi). Motley, N. (2019). Talk, Read, Talk, Write. Language Magazine (link). Greenleaf, C. L., et al. (2011). Integrating literacy and science in biology. American Educational Research Journal, 48(3), 647–717 (ERIC). Goldman, S. R., et al. (2019). Explanatory modeling in science through text-based investigation. American Educational Research Journal, 56(4), 1148–1216 (doi). National Research Council. (2012). A Framework for K–12 Science Education: Practices, Crosscutting Concepts, and Core Ideas. National Academies Press (doi). Mercer, N., Wegerif, R., & Dawes, L. (1999). Children's talk and the development of reasoning in the classroom. British Educational Research Journal, 25(1), 95–111 (doi). Osborne, J., Erduran, S., & Simon, S. (2004). Enhancing the quality of argumentation in school science. Journal of Research in Science Teaching, 41(10), 994–1020 (doi). Berland, L. K., & Reiser, B. J. (2009). Making sense of argumentation and explanation. Science Education, 93(1), 26–55 (doi). Osborne, J. F., & Patterson, A. (2011). Scientific argument and explanation: A necessary distinction? Science Education, 95(4), 627–638 (ERIC). Lee, O., Quinn, H., & Valdés, G. (2013). Science and language for English language learners in relation to Next Generation Science Standards. Educational Researcher, 42(4), 223–233 (doi). Llosa, L., Lee, O., et al. (2016). Impact of a large-scale science intervention focused on English language learners. American Educational Research Journal, 53(2), 395–424 (doi). Jonsson, A., & Svingby, G. (2007). The use of scoring rubrics. Educational Research Review, 2(2), 130–144 (ERIC). Panadero, E., & Jonsson, A. (2013). The use of scoring rubrics for formative assessment purposes revisited. Educational Research Review, 9, 129–144 (ERIC). Wineburg, S., & McGrew, S. (2019). Lateral reading and the nature of expertise. Teachers College Record, 121(11) (doi). Wineburg, S., Breakstone, J., McGrew, S., Smith, M. D., & Ortega, T. (2022). Lateral reading on the open Internet. Journal of Educational Psychology, 114(5), 893–909 (doi).