Using the Concept of Zone of Proximal Development to Explore the Challenges of and Opportunities in Designing Discourse Activities Based on Practical Work

Practical Work

Lunetta, Hofstein, and Clough (2007) define practical work as “learning experiences in which students interact with materials or with secondary sources of data to observe and understand the natural world.” This definition establishes learning and understanding as the main purposes of conducting practical scientific activities in school. However, several classroom studies have noted several challenges related to students’ learning outcomes from practical work. Johnstone (1991) emphasizes the overload problem in some activities, in which no mental capacity remains for students’ interpretations and development of ideas because of complex procedural and technical challenges. Sere (2002) stresses that there are often many different objectives embedded in a single practical activity. She suggests a need for targeted activities in which the learning objectives are better adapted to the situation. Other researchers have focused on problems related to the implicit assumption that ideas can “emerge” directly from observations (Hodson, 1993; Millar, 2010). The overall picture from classroom studies is that teachers and students primarily focus on the practical aspects (e.g., following recipes and handling an apparatus) of conducting a particular activity (Abrahams & Millar, 2008; Hodson, 1993; Lunetta et al., 2007; Tiberghien, Veillard, Le Maréchal, Buty, & Millar, 2001).

Despite these challenges, we believe that practical work has considerable potential for anchoring and deepening students’ learning and understanding. Lave and Wenger (1991) claim “abstract representations are meaningless unless they can be made specific to the situation at hand” (p. 33). Dewey (1916) emphasized that experience, as an empirical situation, is a necessary initial stage in students’ thinking. Practical activities that are intended to illustrate natural phenomena can offer such a situation, thus providing students with perceptible stimulations that they can refer to in their thinking. According to Dewey (1910), all learning involves both inductive and deductive aspects. Through induction, learners generate interpretative ideas about observations, experiences, and information. Through the deduction of possible consequences, learners can test their newly developed ideas by determining consistency with observations and explanations by more competent individuals. By formulating their own interpretations, the learners can obtain feedback from other individuals that enable them to strengthen, modify, or refute their own knowledge and improve their tentative understanding.

Based on this perspective, students must be given the opportunity to not only work with materials or observations but also attempt to connect these elements to scientific ideas or theoretical models about the world by interpreting, explaining, or modeling their observations (Tiberghien, 2000). Consequently, connecting the world of objects and world of ideas (see Figure 1) might be seen as a fundamental purpose of concept-focused practical work and fundamental to general conceptual learning in science (Millar, 2010; Millar, Le Maréchal, & Tiberghien, 1999; Tiberghien, 2000).

Explaining Observations

Vygotsky (1986) claims that “communication presupposes generalization and development of word meaning” (p. 7). This implies that students could benefit from communicating their interpretations to others through social interactions, for example, in discussions, presentations, and written tasks. Students’ attempts to represent their observations and interpretations of an activity by using speech, written text, drawings, or gestures can help them construct meaning and understand interpretative ideas (Tiberghien, 2000). Consequently, the students’ active use of language, in the form of writing and talking about observations and ideas with other individuals, can be viewed as a crucial component of the practical activity (Keys, 1999; Knain, 2008). However, the emergence of a language that connects events in observations to explanatory ideas has been described by Roth and Lawless (2002) as a slow and evolutionary process. When newcomers sense and manipulate objects through hands-on activities, they begin with muddled, incoherent language that incrementally develops into a more mature language. This transition is mediated by object manipulation at the beginning and subsequently by deictic and iconic gestures that simulate the manipulations and movements of the objects. Based on this result, Roth and Lawless (2002) argue that students should have the opportunity to describe and explain phenomena in the presence of objects and events that can cause these phenomena.

Based on Bereiter and Scardamalia's (1987) ideas regarding knowledge-transforming writing, Keys (1999) argued that writing in scientific genres promotes learning by creating a reflective environment for learners engaged in scientific investigations. When students perform closed, prestructured hands-on activities to develop targeted scientific concepts, Keys (1999) maintains that the scientific genre of explanation is highly suitable. This view is supported by Roth and Lawless’ (2002) above recommendation for how to develop students’ language in the presence of objects and phenomena. This perspective is also consistent with a perspective that science learning is becoming increasingly competent at participating in discursive practices in which scientific language is used (Hamza & Wickman 2013). Explanation is the genre that scientists’ use to present an understanding of a concept, phenomenon, or observation by using scientific concepts and general ideas that elucidate the generation of an explanandum (Osborne & Patterson, 2011). The genre contains both descriptions that aim to identify the phenomenon and the sequence that aims to explain the phenomenon (Veel, 1997). Thus, this genre has been considered a relevant tool to help students to articulate connections between the world of objects (the phenomenon or explanandum) and world of ideas (the explanation or explanans).

Language Authenticity and the ZPD

In science class, we want students to master a scientific method of conceptualizing the events in activities (Lemke, 1990). We want students to use scientific language, which many indicate as challenging and alien (Brown, 2004; Evagorou & Osborne, 2010; Wellington & Osborne, 2001). Using a scientific manner to discuss and think presupposes a mastery of the generalized scientific vocabulary and logical connectives often used by the scientific community to express and connect abstract ideas. If the scientific language differs substantially from the language that students normally use to understand events, it will not function as a resource for the students’ understanding because of their inability to “populate it with own intention” (Bakhtin & Holquist, 1981). A scientific way to talk and think about events is, at the outset, outside the students’ ZPD. Consequently, a crucial dilemma is how to organize a social environment (Guk & Kellogg, 2007), in which the students can use a language within their own ZPD that helps them to make meaning from observations and simultaneously develop a more scientific language and manner of thinking.

In the past 15 years, many studies have focused on scaffolding students’ talking and writing in connection with practical activities. The Science Writing Heuristic developed by Keys, Hand, Prain, and Collins (1999), which combines interpretative writing with opportunities to negotiate meaning, has been an effective strategy to promote thinking and students’ integration of scientific ideas with observations from practical activities (Akkus, Gunel, & Hand, 2007; Grimberg & Hand, 2009; Lunetta et al., 2007). McNeill, Lizotte, Krajick, and Marx (2006) have observed that writing scaffolds that support students’ claim justifications can improve students’ reasoning when they argue in support of claims based on a practical activity. Nevertheless, it is unclear to what extent the students should be encouraged in their interpretative writing to express personal or canonical scientific views.

Other studies have considered the potential for allowing students to use everyday language and conceptions as intellectual resources (Ballenger, 1997; Warren, Ballenger, Ogonowski, Rosebery, & Hudicourt-Barnes, 2001) or as a starting point for the meaning-making process (Brown & Spang, 2008; Scott, Mortimer, & Aguiar, 2006). We see this as a method to allow students to use their own formulations to interpret observations. However, we have found Carolyn Wallace's (2004) concept of language authenticity more useful to characterize students talk when expressing personal meaning, well aware that a student-centered authentic language is often characterized as an everyday or vernacular language. Our understanding is that authenticity refers only to the ownership of language use (Wallace, 2004) and not to how it is characterized. According to Wallace, successful science learning implies that a student has become able to use a “scientific language to communicate personally meaningful scientific events” (p. 903). The student has then included the scientific vocabulary and manner of talking in their own authentic language, which implies that students express a correct scientific understanding using their newly extended authentic language.

Research Question and Hypothesis

According to Lijnse (2000), the task of a “developmental research” study is “to seek for essential improvement and scientific extension” of the teacher's didactical knowledge and experience (p. 312). The aim is to develop exemplary practices for teaching particular topics through cycles of intervention and improvements based on reflection and analyses of teaching and learning situations. In particular, a developmental research study should study the language and actions of students and teachers in interactive teaching situations (Lijnse 1995, p. 192). The analyses from the initial cycle of our study indicated that our didactical design did not function as intended. However, after the incorporation of two seemingly minor changes, which explicitly asked the students to generate their own explanations, we observed that the students were able to formulate meaningful interpretations of their observations to a much larger degree.

Didactical Design of the Intervention

The science teachers initially expressed their concern regarding the students’ mediocre learning outcomes from practical activities. With the cooperation of five different teachers, we wanted to address the above-discussed challenges and opportunities by designing targeted (Sere, 2002) science lessons that would encourage students to interpret and explain observations and thus use talking and writing to develop the ideas expressed.

A number of goals were formulated. To encourage focus on theoretical ideas and simultaneously avoid overloading the students’ working memory during the practical work (Johnstone & Al-Naeme, 1991; Kirschner, 2002), we decided to provide only a short introduction of a few scientific ideas prior to the practical work. Thus, we hoped to facilitate interplay between observations and ideas during the practical activities and subsequent discussions (Abrahams & Millar, 2008, p. 1965). To stimulate engagement and provide possible anchor points for new ideas (Ausubel, 1963), we designed practical work that would yield puzzling observations. To provide maximum opportunities for students to focus on observations and interpretations, we used simple experiments that did not involve “black boxes.” To facilitate the students’ development of a scientific language (Keys, 1999; Lemke, 1990; Vygotsky, 1986), a template was used to scaffold their writing together with explicit teaching regarding the purpose and structure of the explanation genre. By incorporating these well-known didactical notions, we attempted to address the worries expressed by the teachers.

Description of the Study

Case Selection

To address the two research questions, two cases from the first intervention cycle are described. In these cases, both the students’ oral talk (Cases 1, 3, and 4) and written texts (Case 2) are described and analyzed. Our focus was not on the differences between the two modes of articulation, talking, and writing (Halliday, 1993), but on how the teachers’ actions and students’ expectations supported or hampered the students’ ability to formulate their own interpretations. The two cases were selected because there was an absence of meaningful scientific talk despite the teachers’ strong efforts to help the students contribute with explanations. The analysis of the cases led to adjustments in the didactical design and to a second intervention cycle in which the students’ authentic talk was a larger component of the lesson.

The first case emphasized how expectations embedded in the triadic IRF (Initiation, Response, and Follow-up) dialogues (Sinclair & Coulthard, 1975) might have affected the students’ contributions. This type of classroom interaction is well documented in the literature (Edwards & Mercer, 1987; Fisher & Larkin, 2008; Lemke, 1990; Wellington & Osborne, 2001) and leaves little room for students’ explorations of their own interpretations (Fisher & Larkin, 2008). It was particularly notable how the students’ absence of prior knowledge regarding scientific content and the teacher's use of cues affected the IRF dialogue. The science teacher also used this type of dialogue in her other three lessons during the first cycle.

The second case was selected because this lesson triggered reflections leading to the changes in our design, which were implemented in the second cycle of our study. In the third case, a template and explicit instructions regarding the explanation genre were provided to the students to assist them in writing interpretations of their observations. Nevertheless, the students’ writing mainly focused on procedural descriptions. To support the claim regarding the effects of the changes, we characterized the students’ explanations in two of the five lessons subsequent to the implementation of the two design principle amendments. The first of these second cycle cases was selected because it was from the first lesson after the changes were implemented. The second case was selected because it involved identical types of changes in the students’ practices although it described a case from a different context (school, class, teacher, age).

Data

Data from the 21 lessons in the study included field notes, video recordings from seven of the lessons, sound recordings from four of the lessons (both from whole-class and group discussions), students’ lab reports, informal interviews with all the teachers after each lesson, and seven short interviews with the students. The student interviews were conducted with randomly selected students during the lessons and after two of the lessons. One interview was conducted 2 months after the lessons with two deliberately selected students because of their oral engagement in the Case 1 lesson. A lesson analysis from the first cycle of the study was primarily based on the researcher's (the first author) field notes. No video or sound recordings were performed in these initial lessons partially because permission from the students and the students’ parents was not obtained and partially because this method did not appear to be normal at such an early stage of collaboration with the teachers. Students’ written reports were used in Cases 1 and 2 as secondary sources of data with notes from follow-up conversations with the teachers and short student interviews after the lesson.

The analysis of the Case 3 lesson was based on the teacher's reformulation of the oral explanations of the students in the whole-class discussions. Field notes were used to describe how the lesson evolved. The Case 4 lesson analysis was primarily based on transcribed video and sound recordings from the section of the lesson when the students presented their explanations. A shorter interview with two of the students after the lesson and a longer video-recorded interview with the science teacher after the lesson were used as secondary sources of data.

Analysis

In our analysis of the four cases, we focused on the characteristics of the student's expressed explanations and contextual elements to characterize the language authenticity (Wallace, 2004) in the students’ talking and writing. Language authenticity creates an awareness of the fact that learning not only involves the acquisition of new concepts but also new methods of discussion (Lemke, 1990). Moreover, according to a Vygotskian understanding, concept and language learning cannot be separated because they develop as a whole (Vygotsky, 1986). This principle applies to scientific and general concepts. Thus, the focus on language authenticity indicated the difficulties that the students encountered when using a scientific discourse that differs considerably from the language used to express personal meaning in other nonscience contexts. We then investigated how the students’ opportunities to use their own language had consequences for their ability to participate within their ZPD. More precisely, our study aimed to understand how the teachers’ and students’ expectations of certain forms of talking and writing affected their ability to operate within their own ZPD. We defined the ZPD as language activities that students could authentically participate in under adult guidance, thus emphasizing that the students had to be an active partner with the adult.

Therefore, we focused on how certain expectations of the teachers and students regarding language use affected the manner in which the students spoke and wrote. The language expectations were occasionally stated explicitly by the teacher or the students, but we also used characteristics of the discourse activities to argue that teacher expectations were embedded in the structure of the talk. For example, the IRF dialogue (Sinclair & Coulthard, 1975) serves to both characterize the class talk and to control the responses provided by the students (Fisher & Larkin, 2008; Lemke, 1990). The IRF framework describes a triadic dialogue in which the teacher initiates the discussion with a question (I) aimed to obtain a response from the students (R), which is then followed up by the teacher (F) (F for follow-up). In the analysis of the first case, we used the IFR framework to identify patterns that were presented and discussed with the teacher in a follow-up interview. The analysis also aimed to characterize the students’ and teachers’ discourse activities to determine whether the students expressed their current understanding of the observations.

In the final three cases, the analysis characterized the students’ explanations. The students’ presentations of their own explanations from Case 4 were transcribed and coded using the software Atlas.TI. The aim of this lesson was to help the student connect scientific ideas to their observations; therefore, we used Tiberghien's modeling approach (2000) as a basis for our analysis. We coded statements in which the students expressed their observations and ideas. We also addressed how and to what degree the observations and ideas were connected in the students’ statements.

In our initial analyses, the difficulties involved in separating observation and ideas in the students’ explanations became evident. The major problem was that explicit formulations of more abstract or general ideas were nearly absent in the students’ explanations (Driver, Leach, Millar, & Scott, 1996; von Aufschnaiter & Rogge, 2010). However, our preliminary analysis indicated that the purpose of the students’ explanations was to make sense of the phenomenon that they experienced during the activity, which Osborne and Patterson (2011) described as a key feature of an explanation. This result motivated us to base our analysis on the characteristics of the explanation genre and, in particular, examine the structure and purpose of a scientific explanation as a basis for the analysis of the students’ explanations.

These codes were used as a basis to characterize the students’ explanations in Cases 3 and 4 and were also used to discuss the absence of phenomenon identification and explanation sequences in Case 3. In addition, the students’ use of scientific terms, language markers “sort of” and “kind of,” and gestures was also coded (in Case 4).

The First Cycle

In the first cycle of the study, we infrequently observed students engaged in group talk or a whole-class talk about their own observations and interpretations. Therefore, we present an analysis of two cases from this section of the study, which generated our hypothesis that contextual parameters inhibited the students from expressing their current understanding.

Second Cycle: Changing the Design. Allowing the Students Space to Work With Their Authentic Language

Case 3: Requiring and Emphasizing Students Own Explanations

This third case was a lesson that occurred immediately after the lesson described in Case 2. The identical activity (starch and saliva) was led by the identical teacher (the first author) in another class in the identical vocational program and addressed the identical theme (the digestive system). Five different student groups conducted the activity and based on a group discussion, recorded a tentative explanation on the template provided. In a subsequent whole-class sequence, all the groups orally presented their explanations and the teacher recorded them on the blackboard as close as possible to the students’ wording (Figure 2).

Three of the explanations in Figure 2, (1), (2), and (4), contain descriptions of the observed colors. All five explanations contained theoretical elements that were not directly visible for the students, and none of the five explanations contained procedural descriptions, which was in contrast to most of the explanations in Case 2. The first two explanations were similar and emphasized that the saliva dissolved the starch. The third explanation confirmed the idea of the preceding two explanations and added another explanatory idea. This explanation focused on the reaction between iodine and starch and the absence of a reaction between iodine and monosaccharaides. Additionally, the explanation explicitly stated that the polysaccharides were dissolved into monosaccharaides by the saliva, thus using more scientific concepts in place of starch and glucose. This explanation did not contain any descriptions of the observations, most likely because it was deemed unnecessary because the observations were described by previous groups. The fourth explanation contained a brief description of the observations and explanatory ideas presented earlier by Group 3 but stated that “the saliva broke down the starch.” In contrast to the previous groups, this group did not use the verb dissolved. This explanation was followed by the fifth group, which noted that the starch was broken down by enzymes. The use of “it” and “break down” indicated that this contribution was additional to the explanation from the fourth group.

The explanations also contained consequential conjunctions such as “because,” “thus,” and “therefore.” In Explanation 1, the conjunction “because” served as a linguistic resource that connected the observational description to the explanatory entities (Veel, 1997). In Explanations 3 and 4, the conjunctions “thus,” “because,” and “therefore” were used to connect different theoretical elements.

The teacher emphasized that the fourth and fifth explanations used the term “broke down,” whereas the previous three explanations used the term “dissolved.” When the teacher asked the class which term they preferred, several students responded that “broke down” was the most appropriate. When considering the fifth group's statement that “it is enzymes that breaks down,” the teacher asked for the name of the enzyme, and one of the students responded “diastase.” Nearly all of the students in the class participated in these dialogues.

Case 4: Changing the Context

Case 4 was obtained from a lesson conducted in the second project school. A 10th-grade class in a lower secondary school in Norway where 9 girls and 12 boys learned the concept of friction. The students were described by their teachers as easy to engage in whole-class discussions but did not prefer writing tasks. The science teacher had been a teacher for 28 years and was well respected by both her colleagues and students. This teacher had also been a former colleague of one of the researchers (the first author).

The concept of friction was introduced based on a teacher demonstration of different objects sliding down an inclined plane and a full-class discussion. The students then conducted a practical activity. Nine different groups fastened a coin between the drawer and case of an empty matchbox (see Figure 3). The students then tapped the matchbox repeatedly on a table (Activity 1). This action caused the coin to gradually disappear into the matchbox. Subsequently, the students held the box in the air while they hit it with a ruler (Activity 2). For each hit, the coin slowly rose from the box (see Figure 4). The students were informed that their task was to generate explanations for the two observations, write them on the template provided, and present them as input for a whole-class discussion.

During the activity, the teacher walked around and listened to and discussed the interpretations of the different groups. Altogether, the class used 21 minutes to conduct, discuss, and record a preliminary explanation for why the coin disappeared and rose in the two activities. Transcriptions of the oral presentations of these explanations were coded and categorized based on Veel's characterization of causal explanations. First, all nine groups presented an explanation for Activity 1, and then an explanation for Activity 2. This time the teacher recorded an extract of the students’ explanations on the blackboard.

First, similar to Case 1 (see p. 1064), we observed that the students were actively discussing their own interpretation of the observations during the activity. However, in contrast to what was observed in Case 1, the students did not ask the teacher to help them formulate a correct explanation on the template. Such demands for a correct explanation were often experienced in the first cycle of the study. Instead, the students repeated the activity simultaneously as they discussed and recorded the explanations.

Second, we observed several characteristics in the students’ explanations that support our experiences immediately after the design change. Most of the students’ explanations contained the identical main characteristics as Veel's causal explanations: phenomenon identification, causal entities, and connecting resources. In some of the explanations, the students focused on the event details and the movement of the objects involved in the phenomenon, which indicated that the explanations were based on careful observations of the phenomenon:

Martin: When you hold the match box in the air, and have the coin inside … and then you hit it …Then the match box goes down, … but simultaneously then sort of you squeeze it, or not squeeze but holds it. The only thing that goes down is sort of the match box, but the coin inside will sort of remain. Then it will go sort of … then … … sort of …

In other explanations, the students also used scientific terms and focused on object descriptions and interactions:

• (1)

  Malin: eeeh .. So when you knock the box against the table, then the coin goes down because the surface is smooth, like in that one,

• (2)

  Victoria: so it becomes friction

• (3)

  Malin: It is smooth, so that it …

• (4)

  Teacher: Becomes friction ..?

• (5)

  Malin: So … the surface on that thing

• (6)

  Victoria: The drawer sort of

• (7)

  Malin: Yes, it is smooth, so that the coin slides by …

Third, in nearly all the explanations, the students frequently used the language markers “sort of” and “like,” and they also accompanied their talk with iconic gestures that illustrated the form and movement of the objects. Their explanations contained hesitations and thought breaks, and the students seldom used scientific terms similar to a scientist (it becomes friction (2)). Altogether the students’ explanations consisted of approximately full sentences, similar to the genre structure and anchored in observations. There were dialogues and reasoning that consisted of chained utterances. However, the explanations were also faltering and unfinished utterances with hesitations, gestures, and a tentative use of scientific terms consistent with Roth and Lawless’ (2002) description of a tentative and explorative use of language. The explanations from the final two groups exemplified a more frequent use of scientific terms. This explanation and term usage was faltering and searching, but the students were also closer to the correct scientific use of the terms.

• (1)

  Jonathan: When you hit the box downwards [moves both hands downwards], then it stops like this. But the other one has still potential energy or what you call it. And it gets more force than the other and it shoves through the slit …

• (2)

  Karl Ove: Yes, so the whole thing then has potential energy … no, not potential energy but kinetic energy, and then the kinetic energy stops in the box, but not in the coin, it continues.

In a group interview after the lesson, three randomly selected students characterized the lesson as instructive. Primarily, the students claimed it was important to base their understanding on practical concepts instead of reading the textbook. One student formulated the explanation in the following manner (translated from Norwegian): “… or that you first do it, and then consider for yourself what we have observed.”

The interviewer (the first author) then asked whether a lesson that began with a practical activity directly followed by a teacher explanation would work. The identical student responded: “But, you don't get to consider yourself. Then it has been pointless, if you have not thought for yourself what you have performed, in a way.”

Two other practical activities were performed in two other classes using the identical didactical design and emphasizing the two previously mentioned design amendments. In these lessons, we observed the identical degree of engagement in the students when formulating and improving their own explanatory texts to interpret the observations.

How Did the Design Changes Cause Changes in the Discourse Activity?

After implementing the two changes, we observed a shift in the students’ discussions. First, instead of mainly presenting procedures and scientific facts, there were more attempts to formulate scientific explanations. These attempts were imprecise and somewhat incomplete, but such shortcomings indicated that the students were honestly expressing their current understanding of their observations in their own words as the teacher had requested (Change 1, p. 1067). Additionally, the students did not ask the teacher to help them formulate a correct response. By informing the students that their explanations would be written on the blackboard, whatever their flaws (Change 2, p. 1067), the students most likely realized that their own tentative ideas and personal language and expressions would be appreciated. This claim was supported by the student's interview following Case 4, which emphasized the importance of their own thinking and own consideration of what they observed. Second, the students’ explanation contained the identical genre characteristics as Veel's causal explanation, including phenomenon identification, a description of the explanatory entities, and the use of connecting words.

Third, the students’ explanations to the entire class were characterized by word-rich, faltering, and unfinished explanations and were accompanied by language markers, gestures, hesitation, and the unscientific use of scientific terms, which indicated that the students were testing their own interpretations and language use.

The final two cases did not involve students from the first cases, and the increased engagement in formulating explanatory ideas could be attributed to individual differences in the students’ ability, the class culture, or their learning history.