Skip to content
ascd logo

Log in to Witsby: ASCD’s Next-Generation Professional Learning and Credentialing Platform
April 1, 2009
Vol. 66
No. 7

How to Start Academic Conversations

An innovative technique draws young English language learners into academic discussions.

premium resources logo

Premium Resource

"Why did the author write this?""To teach us about courage.""Yeah, the guy was brave.""OK. What do we do now?"
Such student conversations were the norm when we began our action research project with English language learners in 4th grade classrooms in our northern California school district. As mentor teachers with the New Teacher Center, we worked with teachers who noticed that their students lacked the skills they needed to focus, deepen, and extend conversations about academic topics. At the urban elementary school we focused on, 73 percent of students were English language learners and 88 percent qualified for free and reduced-price lunch. All students struggled with academic English.
In the years leading up to this project, we taught and observed many lessons at various grade levels. We found that English language learners (ELLs) had very limited opportunities to engage in extended, meaningful talk in school, a conclusion that other research supports as well (Nystrand, 1997; Staarman, Krol, & van der Meijden, 2005). English learners need to produce meaningful linguistic output to develop oral proficiency (Swain, 1985), but most whole-class discussions limit the amount of time each student gets to talk, and responding in front of many others often intimidates ELLs.
Many classroom activities, such as think-pair-shares or vocabulary games, elicit short bursts of student output. But we wanted to teach students to engage in extended discussions that involved constructing academic ideas with others (Cazden, 2001).
We calculated that paired conversations would enable the most talk per minute among these young ELLs: Half of the class could talk concurrently. Yet most of the think-pair-shares we observed were short and shallow. They offered students little chance to negotiate meaning or make decisions about the direction or depth of a conversation. Even when teachers gave students extra time in pairs, students didn't automatically do the things proficient speakers and experts do to have powerful conversations (Zwiers, 2008). We predicted that equipping students with conversational skills would make meaningful academic conversations during class less of a rarity over time.

What Makes a Good Conversation?

We set out to analyze the features of a good conversation. We began by analyzing ineffective conversations we had observed in schools and in our own lives, as well as great discussions we'd had about books and movies. Then we looked at features of good academic conversations among 4th graders. Using Goldenberg's (1992) features of effective whole-class discussions as a starting point, we analyzed what was happening in students' paired conversations. We observed 12 student pairs and participated in 25 short one-on-one conversations with students about fiction and nonfiction texts, recorded these conversations, and analyzed the transcripts for features, prompts, and discourse moves students used that extended and deepened their mutual thinking.
Six of the most useful and teachable features—initiating a worthwhile topic, elaborating and clarifying, supporting one's ideas, building on or challenging another's ideas, applying ideas to life, and paraphrasing/summarizing—became our target conversational skills. As we taught these six features, we came up with prompts that students could use to initiate each feature and respond to it in conversation, as well as visual symbols and hand gestures for each feature (see fig. 1). The visual symbols reflect a comparison between constructing a good conversation and building a house of meaning.

Figure 1. Academic Conversation Features


Scaffolding Conversation Skills

Our students required major scaffolding to use these features effectively to construct more meaningful exchanges. When we showed students a poster of Figure 1 and asked them to practice using these features to prompt a better conversation, they zipped straight through it as if it were a worksheet. Students needed to understand the recursive nature of conversations: Ideas often keep emerging, needing fresh elaboration, support, and application.
To scaffold this cycle of ideas, we had students make visual reminder cards. On one side of each card, they drew the symbol we had created to represent each conversational feature. We required students to memorize at least one prompt to start using the feature each symbol represented (for example, they might memorize the phrase, "Can you elaborate on that point?" for the symbol of elaboration). On the back of each card, students wrote possible prompts for responding to the feature in question (for example, "It means that…").
Teachers modeled how to use the cards recursively during a conversation, returning to one of the features when conversation lagged or veered to a nonacademic topic. To reduce the dependence on cards, we taught students hand motions that they could use as they prompted each feature. For example, they pulled their hands apart while saying, "Can you elaborate?"

A Typical Lesson and Conversation

With our guidance, the teachers explicitly taught each conversation feature. Here's a typical lesson. One of the teachers, Karen, points out on the poster the highlighted feature students will work on as they converse that day: elaboration. Students look at their cards and practice the hand motion and prompts for elaboration. Karen then reads a story about Columbus's sailors arriving in the Caribbean, stopping at times to elicit students' comments and questions. As students offer ideas and interpretations, she encourages them to elaborate.
At one point Karen asks, "Why was the boy afraid?" When Elia responds, "Because the guy just touched their gold," Karen asks, "Can you elaborate?" Elia answers, "I think the boy got all worried because that guy, Columbus, only wanted gold they were wearing. In the picture he looked mean." Karen uses Elia's response to create a pair-share prompt for the upcoming discussion, asking students to tell their partners whether they agree or disagree with Elia and why. She models using the sentence starter, "I agree with Elia's interpretation because…"
Karen leads a short whole-class discussion to brainstorm themes that came up in the story. This provides students with ideas for their conversations. Students write their top choices for themes and jot down any examples from the text that might support each theme. She reminds students that this writing will help them have better discussion sessions.
Karen gives a minilesson in which she acts as one conversant and the whole class acts as the other. Students ask her in unison, "Why do you think the author wrote this story?" Karen answers, "Perhaps she wrote it to teach readers that it is important to listen to children." She waits a few seconds and then asks, "Now what might you ask me? Did I say enough?" Several students respond, "Can you elaborate?" Karen replies, "Well, the adults didn't listen to the boy's warnings about the visitors and their greedy actions. Later it turned out that the boy was right, but it was too late."
Karen reminds students of the prompt that Elia offered and then pairs students to begin academic conversations. Students first take out their symbol cards and review them, testing one another on the prompts associated with each symbol.
Karen moves around the room listening, interjecting at key moments, but letting students facilitate their conversations. She notices Juan and Ana using the starter phrases, cards, and gestures to extend their conversation and encourages them to connect ideas to their lives and to try new vocabulary.
Juan: I think it was about greed.
Ana: Can you elaborate that?
Juan: Like, Columbus only touched the gold that they were wearing and not their skin. That maybe means the people don't matter, just the gold.
Ana: [using the symbol for building on an idea] I add to that the idea that Columbus's people took over the islands and made the boy's people into slaves. They probably wanted to steal all the gold and kill people, like pirates. What do you think?
Juan: Yeah, but pirates mostly attack other ships.
Ana: [thumbing through her cards] How can we apply this to our lives?
Juan: [laughing] I don't know. Maybe we shouldn't be pirates.
Ana: Or maybe we shouldn't be greedy.
Juan: Yeah, we shouldn't think that because we have more guns and ships, or that we are bigger, that we have the … uhhh…
Ana: The right?
Juan: Yeah, the right to take over other people and take their land.
Karen: Can you elaborate with some modern examples?
Juan: Like at school there are bullies, and they shouldn't beat up others and take their money.
Ana: And what about when armies go in to take a country … for oil or land? I hear that still happens. But I wonder, should they fight back?
Juan: We get in trouble when we fight back at school. Sometimes the fights get worse. … [finding the "summarize" card] How can we summarize our conversation?
Ana: We can say we thought the story teaches us that people are more important than money, that greed is bad and bullying isn't right.
Karen: Another term for not right is unjust.
After conversing, all the pairs share their academic synthesis statements with the class, and each pair writes an "exit ticket" synopsis of their conversation. Karen points out that Juan and Ana's discussion uncovered a question that comes up throughout history. She encourages students to write down any big questions that remain. Finally, Karen has students reflect on the process and self-assess with a kid-friendly checklist based on the rubric available atwww.ascd.org/ASCD/pdf/journals/ed_lead/el200904_zwiers_rubric.pdf
These young language learners' conversation focused on a meaningful theme—greed and its effects on others. They connected this theme to real-world situations, found examples in the text to support the theme, constructed interpretations, generated a shared synthesis, and posed a question for future discussion. After four months of practice sessions like these, students began to use the conversation features more automatically.

An Overall Enhancement

  • Students improved at extending and deepening conversations. By June, these 4th graders were discussing meaningful themes in texts and applying them to their lives, rather than retelling parts of the story.
  • Students began using new vocabulary to communicate big ideas, not just to create disconnected sentences or fill in the blanks.
  • Students became more independent thinkers and talkers, shaping their conversations on their own.
  • Whole-class discussions improved as students used many of the prompts from their cards during group discussions. Instead of depending on the teacher to mediate comments, students built their responses on others' ideas without "popcorning out" unrelated thoughts.
We suspect that enhanced academic conversations also contributed to other positive changes. Students showed improvement in writing (giving more evidence to support ideas), critical thinking, and using academic vocabulary to answer questions. Teachers noticed more student participation. In June, students engaged in more minutes per hour of on-task talk than they did in February. One student commented, "It sounds weird, but I feel like we've done something important after a good conversation." The quality of discussions during history and science lessons also improved. The following academic year, many students asked their 5th grade teacher when they were going to start having academic conversations.
English language learners need accelerated language development. That acceleration is fostered by experiences that allow students to share ideas, support them with evidence, and construct new knowledge with other students. This action research suggests that paired academic conversations can provide such experiences, equipping students with communication and thinking skills needed in school and beyond.

Cazden, C. (2001). Classroom discourse: The language of teaching and learning. Portsmouth, NH: Heinemann.

Goldenberg, C. (1992). Instructional conversations: Promoting comprehension through discussion. Reading Teacher, 46(4), 316–326.

Nystrand, M., with Gamoran, A., Kachur, R., & Prendergast, C. (1997). Opening dialogue: Understanding the dynamics of language and learning in the English classroom. New York: Teachers College Press.

Staarman, J., Krol, K., & van der Meijden, H. (2005). Peer interaction in three collaborative learning environments.Journal of Classroom Interaction, 40(1), 29–39.

Swain, M. (1985) Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In S. Gass and C. Madden (Eds.), Input in Second Language Acquisition(pp. 235–253). Rowley, MA: Newbury House.

Zwiers, J. (2008). Building academic language: Essential practices for content classrooms. San Francisco: Jossey-Bass.

Jeff Zwiers is a senior researcher at Stanford University.

Learn More

ASCD is a community dedicated to educators' professional growth and well-being.

Let us help you put your vision into action.
From our issue
Product cover image 109026.jpg
Supporting English Language Learners
Go To Publication