In the year 2011-2012, we video-taped and transcribed four classroom discussions led by three teachers from the Grade 5 sample, as they participated in the Talk Science professional development program and enacted the Inquiry Project curriculum, both for the first time. The teachers were Ms. Carson, Ms. Silvia, and Ms. Bates (teacher names appearing here are pseudonyms). During the approximately ten-week long professional development program, teachers used web-based resources to develop their practice in promoting students’ science learning: they studied video cases depicting exemplary classroom discussions to become familiar with four types of science discussions — elicitation, data, explanation, and consolidation discussions — and with various academically productive talk moves (APT moves) to lead the discussions. Teachers also studied scientist cases to understand the science more deeply, and to understand how scientists reason and talk about phenomena.

The findings presented here focus on teachers’ use of the APT moves to facilitate various science discussions in their classrooms. We conceptualized the APT moves as talk tools for teachers to utilize in leading classroom discussions. In conceptualizing the APT moves as tools, the underlying expectation was that with support, teachers would begin to use the moves strategically to suit the emergent teaching and learning needs in their classrooms. We did not expect that teachers would use the various APT moves uniformly in the discussions, or consistently increase or decrease their use of particular types of talk moves over time, or even use the moves in each of their turns at talk. Instead, the professional development program was designed to help teachers develop their practice by identifying appropriate talk moves to incorporate into their facilitation while guiding discussions. We examined four discussions led by each of the three teachers to understand how they used various talk moves in facilitating their classroom discourse, and how their students participated in and reasoned about the science during the discussions. The discussions were video taped and transcribed and occurred during two lessons early in the curriculum (Investigations 5 and 6), and two lessons late in the curriculum (Investigations 16 and 17).

Our initial research plan was to study temporal changes in classroom discourse as teachers participated in the Talk Science program. However, a careful examination of the classroom discussion transcripts revealed that the four lessons selected for analysis each lent itself to a different type of discussion. For example, the discussion for Investigation 5 required students to formulate claims and provide evidence from their measurement data, whereas the discussion for Investigation 6 required students to propose initial ideas regarding the process of evaporation from their classroom observations and prior experiences instead of measurement data. Therefore, to understand better teachers’ facilitation and students’ participation during these discussions, one needed to know the science investigation framing the discussion, and the learning goals of the discussion. Each of the four lessons involved a different kind of investigation and learning goals, and these differences appeared to have shaped the emerging classroom interactions. The transcripts showed that teachers’ facilitation and students’ reasoning varied according to the learning goals and science investigations guiding the discussions. In other words, a simple temporal analysis of changes in teachers’ facilitation over time was no longer appropriate owing to the differences in the nature of the discussions. Hence, in analyzing the data, we focused less on overarching temporal changes and more on how teachers used particular talk moves and how students reasoned about the science by considering carefully the investigations and learning goals shaping the discussions.

As stated before, the Talk Science program provided video cases on four broad types of science discussions: Elicitation, Data, Explanation, and Consolidation Discussions. The general purpose of these discussions was to enable students to make meaning of their classroom investigations and experiences, but each discussion also had a particular focus. The four foci of the discussions were as follows: eliciting students’ initial ideas; interpreting data; generating explanations; and consolidating understanding. The reader should note that the four discussions we selected for video recording and included in this analysis mapped on to one of the four types of discussions that were presented in the video cases. Below is a table with contextual information on the four discussions we selected for analysis from the Inquiry Project curriculum, and how these relate to the four types of discussions depicted in the Talk Science video cases.

Grade 5 Lesson	Purpose of the Lesson Discussion/Learning Goals	Discussion Type and Focus
Investigation 5: What changes & what stays the same when salt dissolves in water? Students record measurement data (weight and volume) to track the presence of salt dissolved in water.	Students use the weight and volume measurement data they have collected to make claims addressing the investigation question, and to provide evidence for the idea that the weight of a substance stays the same as it dissolves in water, and that tiny things have weight and take up space.	Data Discussion This type of discussion occurs after students collect data to help them connect the investigation question with their data; grapple with discrepant or anomalous data; identify data that can serve as evidence to support a claim; and link data to a representation.
Investigation 6: What happens to the water? Students observe evaporation of water from a paper towel and surface of a plastic cup.	Students use their observations of water evaporating from the paper towel and plastic cup, their prior experiences and reasoning to propose initial ideas about the process of evaporation.	Elicitation Discussion This type of discussion is conducted prior to instruction or at the beginning of a new unit to uncover students’ prior knowledge and experience; increase students’ awareness of their own relevant ideas and experiences; and help them expand and broaden their ideas by listening to others.
Investigation 16: What are some properties of air (3)? Students observe the effects of heating and cooling air through a class demonstration with soap and plastic bottle. Students also observe a computer-based particle magnifier model to understand the changes that occur at the particle level when air temperature changes.	Students use the particle magnifier model to explain in terms of the motion of air particles the expansion and contraction of air they observed during a class demonstration.	Explanation Discussion This type of discussion builds on students’ analysis of the data to help them identify evidence and explain the reason(s) it justifies or supports a claim; and to describe a scientific principle or reasoning that explains the findings.
Investigation 17: What’s the story behind the graph? This investigation occurs towards the end of the curriculum. Students annotate their graphs describing changes in their mini-lakes over several weeks.	Students tell the story of the transformations in their mini-lakes by first describing and accounting for the changes occurring in their mini-lakes at the macroscopic, visible level (the weight of the mini-lakes). Then they use the particle model and describe the changes occurring at the microscopic level to account for the changes at the macroscopic level.	Consolidation Discussion This type of discussion is conducted at the wrap-up of an investigation or when connecting to the one that follows to ensure that students can describe what they did, why they did it, and what they found out; and to replicate what a class did in words, including giving a rationale for their methods and describing their findings

We developed a coding scheme for this analysis through a preliminary examination of the transcripts, and by modifying the coding schemes for classroom discourse from our previous work. The coding scheme examined multiple aspects of classroom discussions: teachers’ use of academically productive talk moves (APT moves) to guide students’ reasoning; students’ attempts at co-co-constructing science understandings with peers; and students’ attempts at reasoning about the science.

We coded teachers’ turns at talk for the following talk moves.

Teachers’ Facilitation of Science Discussions:

• a.  Expand Moves (Say More, Revoice, Time to Think) (e.g., “Okay. Can you say a little more about that?”; “So you think the amount of space these take up depends on the room that they’re in or the house that they’re in?”; “Let’s all take a minute to think about that”)
• b.  Listen Moves (Who can Restate/Repeat) (e.g., “Can someone repeat what Avery said in their own words?”)
• c.  Dig Deeper Moves (Press for Reasoning/Why, Challenge) (e.g., “What is your evidence?”; “How do you know it didn’t rise? Did you measure it?”)
• d.  Think With Others (Add On, Who Can Explain, Do you Agree/Disagree) (e.g., “Oh! Hmm..What do we think? Anyone want to, maybe want to revise Mario’s idea, maybe change it, add to it?”; “What do we think that means? What do you think Amalia means when see says it causes physical breakdown?”; “Anyone disagree with that?”)

We coded students’ turns at talk for the following attempts at co-constructing ideas with their peers, and at reasoning about the science.

Students’ Co-construction Moves:

1. Agree (e.g., “I kind of agree with Daniel, because when you’re going to kick a soccer ball of course if it’s deflated you can’t really kick it that far.”)
2. Disagree (e.g., “No. But I disagree with what Daniel said with the salt being hot.”)
3. Ask for Clarification (e.g., “What do you mean when you say..?”)
4. Clarify Other (e.g., “I think what Shareen is trying to say, the water might dissolve into the salt and the salt might get dissolved, so the water level might go down a little bit.”)
5. Challenge (e.g., “But aren’t you pumping hot air into the ball? Because if you blow up, like a balloon, sometimes you pump in hot air and after that it starts making it rise kind of, and after that.”)
6. Add-On (e.g., “Um I also wanted to add on to Louie’s..”)
7. Restate Other (e.g., “She said there’s more space in the air particles. I mean when the particles are pushed like — yeah, pushed.”)

Students’ Reasoning:

Sense-Making Attempts: This category captured students’ efforts at making sense of the science.

1. Revise their own thinking (e.g., “Yeah, I agree that it’s Lela. Because after the — well, at first I thought it was Fern because I didn’t know that air had weight. Then after my education, I learned that Lela is probably correct.”).
2. Raise a related question (e.g., “I have a question. Where does the water go when it evaporates?”).
3. Propose test/thought experiment (e.g., “When Christiani said that if you put a cup and air in, but we’re not talking about a cup, we’re talking about a ball. So, if you have a scale and we fill it up with air, and it would not stay on zero. It would go — it would, um, go out between two, three — “).

Reasoning With Core Science Ideas: This category captured instances of students’ reasoning by drawing on core science ideas (classroom science investigations and scientific principles from the curriculum).

1. Reference to Classroom Science Investigations — Students referred to quantitative data and/or observations from previous and/or present science curriculum units (e.g., “Yeah. And they weighed the same, but then we kept one of the balloons not inflated and then we blew up the other one. And when we put it on that side was a little farther down, so that means it was heavier when it had air in it.”).
2. Reference to Scientific Principles (Principles such as air is matter; matter has weight and takes up space; the particulate nature of matter, etc.) - Students referred to scientific principles, ideas from the particle model (e.g., “I respectively disagree with Kiaja because I do think air has weight and that I agree with Layla and that the inflated soccer ball weighs more than the flat one. “).

Reasoning Without Core Science Ideas: This category captured instances where students drew on ideas outside of formal scientific understandings.

1. Reference to Outside Experience — Students described experiences from everyday life (e.g., “I think that Tomas is right, because it’s the same. I don’t have a soccer ball, but I do have a football. And, when the football gets flat, it is heavier. But, um, but when, um, air goes into the soccer ball, um, it makes it lighter because of all the gravity around”).
2. Presenting Assertions/Opinions — These were instances where students presented assertions that were either opinions or facts that may have been accurate or inaccurate with respect to canonical science (e.g., “Well, Claire is the most right, but the soccer ball would probably be a little heavier, because air is like .000000000001 more heavier, and the flat ball is the same exact thing as the actual soccer ball, but it just doesn’t have any air in it, so it’s pretty much the same.”).
3. Analogy — This code captured instances where students drew similarity to other hypothetical situations (e.g., “ I have something- I agree with Ryan because if you take an air mattress out it would feel heavy and then when you blow it up it would feel easier to carry and lighter.”).
4. Logical Train — This code captured “if...then” statements expressing axiomatic reasoning and counterfactual thinking (e.g., “But if you think that the air has weight, like if it adds weight to it, then if you put a scale in the middle of the room right here there would probably be at least a pound showing on it.”).

The analysis of the Grade 5 Inquiry Project curriculum discussions reveals how teachers guided students’ reasoning, and the various ways in which students engaged in the discussions. As stated before, the four lessons selected for videotaping and analysis each had a different kind of investigation and learning goals that shaped the emerging discussion. To better understand the findings, one needed to know the context of the discussions with respect to the underlying science investigations and learning goals. Therefore, we report here on each of the four discussions separately. Furthermore, within each discussion, we describe each teacher’s practice separately, documenting her facilitation and her students’ talk to identify similarities and variations among teachers’ practice, and to enable us to relate teachers’ practice to their students’ participation. Here we present a summary of the findings across the four discussions and three teachers from our analysis. Please refer to the NSF report for more details on the quantitative data from this analysis.

The findings show that teachers drew on various academically productive talk moves (APT moves) to guide their classroom discussions. They used the talk moves often to encourage students to expand their ideas, and to deepen their reasoning. The teachers also used talk moves to encourage students to listen and respond to their peers’ ideas, albeit to a much lesser extent. Overall, the findings indicate that while participating for the first time in the Talk Science professional development program, the teachers took on board the APT moves and incorporated various talk strategies into their classroom practice.

Further, in the case of two of the three teachers (Ms. Bates and Ms. Carson) examined in this analysis, their use of Expand and Dig Deeper sets of talk moves was aligned with the underlying purpose of the discussions, and varied with the different types of discussions. When the purpose of the discussion was to elicit students’ initial ideas about the process of evaporation (Investigation 6), the teachers utilized strategies to draw out students’ preliminary thinking (Expand Moves), and probed students less often to provide evidence and explanations (Dig Deeper Moves). On the other hand, when the purpose of the discussion was to encourage students to construct scientific explanations, the teachers utilized talk moves often to help students deepen their reasoning with the help of evidence and scientific principles. These findings suggest that the two teachers may have understood the different types of science discussions (elicitation, data, explanation, and consolidation) they were introduced to in the Talk Science program, and used talk strategies differently in leading their discussions to address the underlying learning goals. The present analysis, nevertheless, provides a quantitative overview of the classroom discourse; a more detailed examination is needed to clarify how the teachers adjusted their use of talk moves to the moment-to-moment interactions to accomplish differing goals of these discussions.

Our analysis revealed that students participated actively in various ways to make meaning of the science. To reason about the science, they drew often on their classroom investigations, referring to the experimental procedures and the data and observations gathered during the investigations. They included also formal scientific principles from the curriculum in generating their explanations. Additionally, students drew on ideas from outside the curriculum, incorporating everyday experiences, facts or opinions, analogies, and so forth as they tried to understand the science.

Furthermore, in two of the three classes (Ms. Bates’ and Ms. Carson’s), students’ reasoning with the help of ideas from within and outside the curriculum was fairly consistent with the type of discussion and the investigation framing the discussion. When the students did not have measurement data and the particle model to generate explanations (Investigation 6), they tended to invoke their everyday experiences, facts or opinions, and analogies to formulate their ideas. On the other hand, when students were provided with measurement data and a computer-based particle model (Investigation 5 and Investigation 16 respectively), they recruited these resources more often to construct explanations and reasoned less with the help of ideas and experiences from outside the curriculum.

Our examination of teachers’ practice showed also considerable variation among the teachers. One of the teachers (Ms. Silvia) used talk moves consistently less often than the other two teachers (Ms. Carson and Ms. Bates) (see Figure 1).

Further, there was variation in teachers’ use of different types of talk moves, with Ms. Bates using the Think With Others and Listen moves more often than the other two teachers. These differences notwithstanding, the general pattern was that teachers used more often talk moves designed to help mainly individual students explicate their ideas and deepen their own reasoning (Expand and Dig Deeper set of moves), and used much less often talk moves designed explicitly to foster active listening and co-construction of ideas by prompting students to attend carefully and respond to their peers’ thinking (Listen and Think With Others set of moves). (see Figure 2)

The low emphasis on using the Listen and Think With Others talk moves should be noted because overall the students also made few attempts at co-constructing science understandings with their peers. Further, students’ co-construction was generally consistent with the teachers’ use of Listen and Think With Others talk moves. Specifically, we observed that students tended to actively restate and respond to their peers’ ideas in discussions where the teachers utilized strategies to encourage them to engage with their peers’ thinking. Our findings suggest that students may need explicit teacher guidance to make meaning of the science collectively with their peers through a communal exchange of ideas. Students’ co-construction and teachers’ use of the Listen and Think With Others talk moves are critical in bringing about a dialogic exchange of ideas, where students go beyond sharing out their own ideas to building on their peers’ thinking, and work toward critiquing and refining the emergent shared understanding of the science within their classroom community.

Overall, our findings indicate that teachers’ participation in the Talk Science professional development program for the first time laid a foundation for developing their practice by including certain types of talk strategies into their teaching. Future research and design of professional development needs to explore ways to increase teachers’ facility at leading a more dialogic discourse to promote students’ scientific reasoning.