Converting mathematics tasks to learning opportunities: An important aspect of knowledge for mathematics teaching Peter Sullivan Monash University

Abstract: As part of a research and professional development project which focused on the opportunities and constraints provided by different kinds of mathematical tasks, a group of 67 primary and 40 secondary practising teachers of mathematics were asked to complete a survey focusing on their use of tasks. In this article, we discuss responses to one particular item which sought teachers ideas on taking a fraction comparison task (which is larger: 2/3 or 201/301?) and converting it into a mathematics lesson in the middle years of schooling. Drawing upon a number of components of mathematical knowledge for teaching as a framework, we attempt to examine those aspects of mathematical knowledge that are involved in making such a conversion. Our recommendation following this analysis is that in professional development settings greater emphasis is necessary on taking a potentially useful task and converting it into a worthwhile mathematics learning experience for students. Doug Clarke Australian Catholic University Barbara Clarke Monash University

Plenty of Interesting Ideas but Where Are the Interesting Lessons?

One of the paradoxes facing those whose task it is to support mathematics teachers is that, while the availability and accessibility of interesting teaching ideas is increasing steadily, the challenge of converting those ideas to successful student learning is as substantial as ever. We have each worked for some time on developing tasks that can be used as the basis of mathematics lessons, and have conducted numerous sessions on interesting tasks for both prospective and practising teachers. There has been an implicit assumption in some of our work that teachers can convert tasks to lessons easily. Clearly this is an aspect of the knowledge that teachers must have in order to teach well. This article examines some perspectives on this knowledge. Whether it is the availability of interesting ideas and tasks or teachers capacity to enact them, it seems that professional development programs have had limited impact on Australian classroom teaching in the middle years of schooling (Years 5 to 8), as evidenced by results of the Third International Mathematics and Science Study (TIMSS) Video Study. This study aimed to investigate and describe Year 8 mathematics and science teaching practices in a variety of countries (Hollingsworth, Lokan, & McCrae, 2003). The researchers videotaped and analysed 638 Year 8 lessons from seven participating countries, including Australia. Altogether, 87 Australian schools with one teacher in each school were randomly selected in a way that was representative of all states, territories, school sectors, and metropolitan and country areas. Each teacher was filmed for one complete lesson. Although the report detailed the international findings, the following specifically related to Australia: 1

- More than three-quarters of the problems used by teachers were rated as low in procedural complexity (requiring four or fewer steps to solve). There were more problems low in complexity than in any other country, and significantly more than for teachers from Japan (17%). - Seventy-six percent of all problems were repetitions of one or more problems students had done earlier in the lesson. - The majority of problems involved emphasis on correct use of procedures. - One-third of problems per lesson, on average, were solved publicly by giving the answer only. - Just over a quarter of problems were set up with use of real-life connections (42% in The Netherlands). - More than 90% of problems were presented to students as having only one solution. (Hollingsworth, Lokan, & McCrae, 2003, pp. xviii xxi) The authors noted that Australian students would benefit from more exposure to less repetitive, higher-level problems, more discussion of alternative solutions, and more opportunity to explain their thinking (p. xxi). They noted that there is an over-emphasis on correct use of the correct procedure to obtain the correct answer. Opportunities for students to appreciate connections between mathematical ideas and to understand the mathematics behind the problems they are working on are rare (p. xxi). The authors commented on a syndrome of shallow teaching, where students are asked to follow procedures without reasons (p. xxi). We do not argue that teaching using procedures should not happen, but wonder why there does not appear to be more variety in the approaches that teachers adopt. We are seeking to explore this apparent anomaly in which it seems to us that there are many interesting tasks available to teachers, yet it does not appear that teachers are taking advantage of these tasks in an effective way. In this article, we examine the connection between tasks and teaching and the knowledge needed to convert tasks to lessons,. We report responses to a particular question which sought teachers views on what they interpreted an illustrative task to involve mathematically, and how the task might be used as the basis of a lesson. The research questions, presented below, address what teachers see as the content of the illustrative task, and what sorts of lessons they might create from the task.

Tasks and Learning

Mathematical tasks are important for teaching, and the nature of student learning is determined by the type of task and the way it is used. It has been argued that the task set and the associated activity form the basis of the interaction between teaching and learning (Christiansen & Walther, 1986), and that instructional tasks and classroom discourse moderate the relationship between teaching and 2

learning (Hiebert & Wearne, 1997, p. 420). It has been found that when teachers pose higher order tasks, students give longer responses and demonstrate higher levels of performance on mathematical assessments (Hiebert & Wearne, 1997), and that greatest gains on performance assessments including questions that required high levels of mathematical thinking and reasoning were related to the use of instructional tasks that engaged students in doing mathematics or using procedures with connection to meaning (Stein & Lane, 1996, p. 50). There are some general characteristics of effective tasks. Christiansen and Walther (1986), from a mathematics education perspective, argued that non-routine tasks, because of the interplay between different aspects of learning, provide optimal conditions for cognitive development in which new knowledge is constructed relationally and items of earlier knowledge are recognised and evaluated. Ames (1992), from a motivational perspective, argued that students should see a meaningful reason for engaging in a task, that there needs to be enough but not too much challenge, and that variety is important. Fredericks, Blumfield, and Paris (2004), in a comprehensive review of studies on student engagement, argued that engagement is enhanced by tasks that are authentic, that provide opportunities for students sense of ownership and personal meaning, that foster collaboration, that draw on diverse talent, and are fun. Our sense is that there is a variety of sources from which teachers can choose tasks that have at least some of these characteristics (see, e.g., Curriculum Corporation, 2009; Lovitt & Clarke, 1988). The data presented below are from the Task Type and Mathematics Learning1 (TTML) project, which is investigating the best ways to use different types of mathematics tasks, particularly in Grades 5 to 8.

The Challenge of Converting Tasks to Lessons

There is a range of teacher actions necessary to plan and deliver lessons utilising potentially interesting tasks. In other words, there are specific actions that teachers must take to transform tasks into effective lessons. One perspective on the process of converting tasks to lessons has been described as the intended curriculum, the implemented curriculum, and the attained curriculum (e.g., Robitaille, Schmidt, Raizen, McKnight, Britton, & Nicol, 1993). Similarly, Gehrke, Knapp, and Sirotnik (1992) described the planned, enacted and experienced curriculum, while Burkhardt, Fraser, and Ridgway (1990) referred to the ideal, adopted, implemented, achieved, and tested curriculum. Our interest within the project is mainly in the implemented, or enacted, curriculum the ways in which a teacher takes a syllabus or curriculum guidelines or standards or particular
1 TTML is an Australian Research Council funded research partnership between the Victorian Department of Education and Early Childhood Development, the Catholic Education Office (Melbourne), Monash University, and Australian Catholic University.

mathematics tasks and enacts them in the classroom although this article focuses on the intended, or planned, or adapted, curriculum. There are some attempts to scaffold teachers in the process of converting tasks to lessons. Stein, Grover, and Henningsen (1996), for example, presented one model of task use. Their model described how the features of the mathematical task as set up in the classroom, and the cognitive demands it makes of students, are informed by the mathematical task as represented in curriculum materials, and influenced by the teachers goals, subject-matter knowledge and knowledge of students (p. 461). Hashimoto and Becker (1999) described a planning process for lessons involving the open approach, and Watson and Sullivan (2008) presented four templates that can be used for different types of mathematics lessons. We note that conventional wisdom would suggest that primary-level teachers have well developed language about pedagogy, and also that the mathematical knowledge of many is perhaps restricted. Likewise it is conventionally assumed that secondary teachers are strong mathematically but have less sophisticated language to describe pedagogy. Even though the project participants were from active clusters with a professional development orientation, there was no reason for assuming they were any different from teachers generally.

Categories of Teacher Knowledge

There are two major categories of knowledge needed to convert tasks to lessons, and for the teaching of mathematics generally: subject matter knowledge; and pedagogical content knowledge. Hill, Ball, and Schilling (2008) described diagrammatically components of these two types of knowledge. When converted to text, these categories and sub categories were: Subject Matter Knowledge common content knowledge specialised content knowledge knowledge at the mathematical horizon Pedagogical Content Knowledge knowledge of content and teaching knowledge of content and students knowledge of curriculum Under the category of Subject Matter Knowledge, the authors described three sub-categories. The first, common content knowledge, is the mathematics needed to solve a task. This is the type of knowledge that someone who is not a teacher but who is good at mathematics might have.

The second, is specialised content knowledge, or the knowledge that allows teachers to engage in particularly teaching tasks, including how to accurately represent mathematical ideas, provide mathematical explanations for common rules and procedures, and examine and understand unusual solution methods to problems (Hill et al., 2008, p. 378). This would also include recognising that a task can be solved in different ways. Specialised content knowledge is also described as the mathematical knowledge and skill uniquely needed by teachers in the conduct of their work (Ball, Thames, & Phelps, 2008). In looking for patterns in errors made by students or in considering whether a student-generated solution strategy could be generalised to other tasks, a teacher is drawing upon this knowledge. The third sub-category, knowledge at the mathematical horizon, is described by Ball et al. (2008) as an awareness of how mathematical topics are related over the span of mathematics included in the curriculum (p. 403). The second main category, Pedagogical Content Knowledge (PCK), was first described by Shulman (1986) as: an understanding of what makes the learning of specific topics easy or difficult: the conceptions and preconceptions that students of different ages and backgrounds bring with them to the learning of those most frequently taught topics and lessons. If those preconceptions are misconceptions, which they so often are, teachers need knowledge of the strategies most likely to be fruitful in reorganizing the understanding of learners. (pp. 910) Ball et al. (2008) also delineated three sub-categories of this. The first, knowledge of content and teaching, includes an understanding of how to sequence particular content for instruction, of how to evaluate instructional advantages and disadvantages of particular representations, and of the knowledge required to make instructional decisions about which student contributions to pursue and which to ignore or save for a later time (p. 401). The second sub-category, knowledge of content and students, is knowledge that combines knowing about students and knowing about mathematics (p. 401). This includes, for example, anticipating students cognitive and affective responses to particular tasks, what they will find easy and hard. They must also be able to hear and interpret students emerging and incomplete thinking as expressed in the ways that pupils use language (p. 401). Ball et al. (2008) argue that these two sub-categories coincide with the two central dimensions of Shulmans PCK: the conceptions and preconceptions that students of different ages and backgrounds bring with them to the learning of the most frequently taught topics and lessons, and

the ways of representing and formulating the subject that make it comprehensible for others (p. 9). The third sub-category, knowledge of curriculum, is slightly different in that it relates closely to Shulmans (1986) Curricular Knowledge, in which curriculum is represented by the full range of programs designed for the teaching of particular subjects and topics at a given level, the variety of instructional materials available in relation to those programs, and the set of characteristics that serve as both the indications and contraindications for the use of particular curriculum or program materials in particular circumstances (p. 10). We use these sub-categories to summarise and interpret the responses of the teachers who responded to our questionnaire items. We note that there are various ways of categorising or described the knowledge needed for teaching mathematics, but feel that this description provides adequate delineation of the elements of this knowledge for our purposes. The particular research questions we explore in this article are: What do teachers see as the content focus of an illustrative mathematics task? What sorts of hypothetical lessons do teachers create to include the illustrative task? Are there differences between responses of secondary and primary level teachers? We are also interested in whether the Ball et al. categories are helpful in describing different aspects of the knowledge that teachers might need to convert tasks to lessons.

The Illustrative Task

As part of a survey given to teachers, one prompt sought insights into the extent to which teachers could describe the content of a particular task, and the ways in which they might convert the task to a lesson. The survey was given initially to 41 primary teachers and 6 secondary teachers from the TTML project. To allow better primary/secondary comparisons, the survey was also completed by 26 willing experienced primary teachers who were participating in a mathematics teacher learning session, and 34 further secondary mathematics teachers who were in the same session. The task was presented to teachers as follows: The following is a description of an idea that might be used as the basis of a lesson: 2 201 Which is bigger 3 or 301 ? Readers are encouraged to take a moment to consider how they might decide which of these fractions is larger. Although many people state the correct answer quickly, they often have difficulty

articulating their reasons for the conclusion they draw. Strategies which emerged when we have previously posed this task to prospective and practising teachers include the following: - Finding common denominators and comparing 602/903 to 603/903 (or comparing 602 to 603 after cross multiplying) - Using a calculator to divide numerator by denominator to convert to decimals and then comparing 0.6666 to 0.66777 - Realising that 2/3 is the same as 200/300, and then considering the shift from 200/300 to 201/301. The reasoning is often along two lines, one correct, one incorrect. The correct reasoning takes the form of Im adding the same to the numerator as the denominator, but proportionally more to the numerator, so the ratio is increased. Some try this with simple cases to convince themselves that doing so will increase a fraction. That is, by going from 4/5 to 5/6 to 7/8, you are increasing the fraction, as it is getting closer to 1 each time, as the small part to be added to take the fraction to 1 is decreasingthis has been termed residual thinking (see Clarke, Sukenik, Roche, & Mitchell, 2006). This will work for all fractions between 0 and 1. The incorrect reasoning is that we are adding 1 to the numerator and 1 to the denominator, so that there is no real change, and the two fractions are the same. - Adding 199 to the numerator and 298 to the denominator, because the ratio of these two numbers is slightly more than 2/3 (0.66778), the effect must be to increase the fraction overall. - Some calculate the residual needed to build 200/300 and 201/301 to the whole, respectively. They determine that the first fraction requires 100/300 to do so, and the second 100/301, and because the 301ths are smaller parts, the residual needed is smaller. - A teacher in the United States used a basketball analogy to give a convincing alternative strategy. He argued that we could think of 200/300 as a basketball players free throw success rate, as 200 successful throws out of 300. In moving to 201/301, the basketball player has had one more throw, which was successful. His average must therefore have improved, and so 201/301 must be larger. - Similar to this reasoning is what some Chinese educators called the sugar theory. If the numerator represents sugar, and the denominator the whole, then adding another unit of sugar will increase the sweetness. These various approaches or strategies are intended to illustrate to the readers that it is the diversity of possible approaches to solving the task that is the power or potential of this task. While any one of these strategies might be described as common content knowledge, awareness of the range of possible tasks is better considered as specialised content knowledge.

What do Teachers Identify as the Content of the Task?

The first prompt to which teachers were invited to respond was: If you developed a lesson based on this idea, what mathematics would you hope that the students would learn? In our view, the fundamental mathematical concept is comparing fractions, with the task offering opportunities for students to seek alternate or intuitive strategies as well as considering formal approaches to comparisons such as finding a common denominator or converting to decimals. The teachers responses to the survey question were inspected, common responses were identified, initial categories were proposed, responses categorised, and then some categories combined where this did not detract from the integrity of the categories. In what follows, the categories are presented, along with examples of the responses of teachers that were so categorised, and then some comparative data on the responses are considered. Responses were categorised as Students learn various ways of comparing if we considered that teachers appreciated various ways of solving the task, and that they sought for students to learn more than a single approach. Some illustrative responses from the teachers were:
Understanding size of fraction, some ideas of equivalence, ordering fractions, changing fractions into decimals (possibly) Fractional understandings what is the size of the pieces? how many pieces are we talking about? what proportion of the whole is each fraction? Knowledge of equivalent fractions, ability to cancel down, place value of decimal numbers

In each case, such teachers have identified multiple aspects of the task, and the aspects they described are indeed important for solving the task. Responses were categorised as A single specific concept (not comparing) if they referred to only one concept. A number of responses used only the words of a concept such as
Equivalent fractions Place Value Fractions

There were also more descriptive responses, although still focusing on just one concept:
Looking at concepts related to fractions equivalence Concept of equivalent fractions Students would learn to compare two fractions the method I would use would be changing fraction to a decimal

The place value response is puzzling, but perhaps this implies an assumption of converting to decimals, although even with this it is hard to see the key concept as place value. It is also difficult to see the essence of the task as equivalence. Even though converting to denominators of 300 or 903, does indeed involve equivalent fractions, solving the task still requires more. We suspect that 8

either these teachers used language to describe concepts loosely, or they did not identify the relevant concepts in the task. Responses were coded as General and only vaguely related multiple concepts if they mentioned a number of concepts, but which were only vaguely related to the task. Examples of responses so categorised were:
What is a fraction? What does the notation mean? What do we mean by bigger? Meaning of numerator/denominator/ equivalent fractions What a fraction is. Fractions are just another way to write a number where they are used in everyday life. Concept of larger smaller

These seem to be vague and general and sufficiently removed from the task to suggest that those teachers would have difficulty focusing a discussion with students on the key elements of the task. Table 1 presents the number of teachers whose responses were coded in those three categories, broken down into TTML (the project) and other (non project) teachers, and further split into primary and secondary. Table 1: Categorisation of teachers responses describing the focus of the task (n = 107) TTML Primary Students learn various ways of comparing A single specific concept (not comparing) General and only vaguely related multiple concepts 8 22 11 TTML Secondar y 4 1 1 Other Other Primary Secondary 7 7 12 19 9 6

To express these data another way, overall 58% of the secondary teacher respondents and 22% of the primary teachers not only identified the essence of the task as comparing fractions but also that students might well be exposed to different ways of comparing. This means that about two-fifths of the secondary teachers and over three-quarters of the primary teachers choose a single content area (that was not comparing), or a range of vaguely related ones. It raises the possibility that some of these teachers were not able to identify readily the focus or potential of this mathematical task. It should be noted that the prompt for the teachers was not limiting and that considerable space was provided for the answer. We recognise that these are teachers responses to a particular item, completed in the midst of a larger survey, and so may tend to underestimate the possible responses of the teachers. Nevertheless, as Ball and Bass (2001) argued, in order to help students reason mathematically, the 9

teacher must do the following: uncover the students current base of common knowledge; establish and extend the students base of common knowledge; and model and guide the construction of acceptable mathematical arguments. Taking these steps clearly requires the teachers own common content knowledge to be well established. It also seems that a reasonable number of these teachers evidenced limited specialised content knowledge, and one could suspect that these teachers would have difficulty matching tasks to curriculum statements, or matching resources to particular tasks. This identification or deciphering of the mathematics content that a task is likely to elicit goes beyond common content knowledge. It requires specialised content knowledge in that the teachers are required to analyse and articulate the content in more detail than is required in simply solving the problem. It is also noted that, in order to communicate a response, teachers need the language and concepts that would be associated with knowledge of content and curriculum. This includes knowing what the terms equivalent fractions and fraction comparisons mean and do not mean. It is possible that some limitations in the teachers responses may be connected to inadequate understandings of the meaning of such key terms.Until now, this perspective has also been underemphasised in many teacher professional learning sessions that we have led.

Designing a Lesson
The second related prompt on the teacher survey was as follows: Describe, briefly, a lesson you might teach based on this idea We had hoped to see descriptions of lessons that would allow students to see that there were various ways to solve the task, and ideally that teachers would plan specifically to allow students to explore the task in their own way at some stage, with a concluding review of the students strategies.. As before, the responses overall were inspected, categories developed, the responses allocated to categories, inspected, and then categories and allocations adjusted. The resultant categories are presented, along with examples of responses of teachers that were so categorised. Some quantification of the responses follows. The first category was described as student centred, perhaps using a meaningful example, but emphasising student generation of strategies and discussion. This category would include teachers who seek to engage students in their learning and who remain open to the possibility of students generating their own approaches, including fostering discussion. The following are some examples of responses categorised in this group:
Take in counters or use student nos in class say you want 2 out of 3 in one pile and 201 out of 301 in the other. Which is the larger proportion? Get students to answer qn demonstrate answer to class encourage diversity of answers praise all attempts even those that are wrong as long as they are thinking and prepared to write down their thinking

10

2/3 or 201/301 if you were going to explain the answer to someone choose either multi media, poster concrete material to do so. In your explanation include written language as well as pictorial Pose the problem brainstorm processes we could use to determine the solution have groups attempt to solve the problem using their choice of methods Have the group answer the question and explain their process going over the solution discuss any errors in comprehension practice some other problems

These can be taken as a reasonable representation of the reform approach, and would certainly allow the possibility of the type of intended solutions that were described above. The second category was described as a teacher-centred lesson, incorporating a specific strategy for teaching the task. This category would include teachers who adopt a direct teaching approach, and who see mathematics as a set of procedures to be learned effectively. Such teachers might focus on a specific approach and teach it explicitly. The following are some examples of responses categorised in this group:
Change both factions to decimal numbers and find which one is bigger Firstly fractions what and how they are used, start with simpler fractions and build up you could do other examples of close but not quite the same fractions 6/7, 7/8 extend

While the latter response suggests an intent to build understanding, the phrasing implies teacher direction with specific teacher intent for their learning. Perhaps this could be described as good traditional teaching. Given that the teachers identified an appropriate focus for the task, such lessons have a chance of producing useful learning for their students. Nevertheless, such teachers are still missing key opportunities that this question offers. A third category was real-life exemplars but only vaguely related to the concepts. There is considerable emphasis in both curriculum documents and teacher support advice on the importance of using relevant examples (e.g., National Numeracy Review, 2008). The responses in this category seem to prioritise this aspect over the mathematical concepts:
If I wanted to focus on fractions I would begin having the kids identify life situations where fractions are used. From this I would physically have them compare their ideas to see if they could rate them from smallest to largest. A man who orders a pizza asked the chef to cut the pizza into 6 slices because he wont think he can eat the 8. I use Dove soap. Since I am allergic to moisturising cream, I use of the soap and throw the other away

These seem vague, and it is difficult to know how the contexts would clarify the mathematical concept that was the intended focus. The fourth category was teacher-centred but with a general strategy not specific to the task. The following are some examples of the response categorised in this group:
Converting between fraction decimals and percentage ratio another ways of writing ratio when doing drug calculations

11

A lesson taught would involve place value and fractions dealing with whole numbers, this would require the use of concrete material to enable students a visual prompt Give students flashcards with fractions written on them. Ask them to place themselves in order from smallest to largest, followed by whole group discussion on what they need to know to complete the task

Such descriptions seem to address only vaguely the general topic of fractions, but have not communicated that they are in any way addressing the hypothetical focus. A fifth and catch all category is described as nothing or dont know. There was a range of responses categorised as dont know. Some of these teachers wrote the words, and others gave no response at all. Note that it is not inferred that this is due to fatigue in completing the survey, since all teachers completed subsequent questions including some that required open responses. There were also responses such as the following in this category:
I feel it would be a very dry lesson unless I could find a hands on investigative approach. By looking at this problem I lack confidence so would probably shy away from it

It is noted that this teacher rated his/her knowledge of mathematics as 5 on a 10 point scale, and his/her confidence at teaching mathematics at 4 on a similar scale. A key development need for this teacher is clearly improved common content knowledge. Table 2 presents the number of teachers whose responses were coded in these five categories of hypothetical lessons, broken down into TTML and other teachers, and further split into primary and secondary. Table 2: Number of teachers in each category of response on the hypothetical lessons (n=104) TTML TTML Other Other Primary Secondary Primary Secondary Student-centred, perhaps using a meaningful example, emphasising student generation of strategies and discussion (or the process) A teacher-centred lesson, incorporating a specific strategy for teaching the task Real-life or concrete examples but only vaguely related to the concepts Teacher centred but with a general strategy not specific to the task Nothing or dont know Totals 5 1 4 4

7 16 6 6 40

3 1 0 1 6

4 7 0 13 28

6 14 3 3 30

About 13% of the lessons (that is 14 out of the overall total of 104) were of the reform style, with similar percentages for both primary and secondary teachers. Assuming that the teachers implement 12

their hypothetical lesson effectively this would be likely to foster learning, utilising the potential of the task. Around 19% of the lessons (that is 20 out of the overall total of 104) could be described as good traditional teaching, with a higher proportion of secondary teachers (25%) than primary (16%). Given that the teachers identified an appropriate focus for the task, such lessons have a good chance of producing useful learning for those students. On the other hand, it is less likely that important opportunities would be taken up in these lessons than in student-centred lessons. The other 70 lessons described do not create confidence that these teachers can transform a basic idea into an effective mathematics lesson. For example, 35% of the hypothetical lessons could be described as meaningless use of relevant or real-life examples. This is not what the emphasis on relevance is intended to achieve. Taken on face value, it could be suspected that most students would not learn effectively in such lessons. While there is variation in the distribution of responses, the responses of the primary and secondary teachers are similar, and the TTML teachers were similar to the others. The knowledge necessary for converting tasks to lessons is clearly part of Pedagogical Content Knowledge, and would contain elements of knowledge of content and teaching, and knowledge of content and students. It seems that only some teachers were able to do this the way we expected. It is noted that if the teachers do not have the common content knowledge, then aspects of the relevant PCK may not be able to be enacted (Thames, 2006). It is important to note that we recognise that this is a challenging problem for primary students, and that we did not encourage the teachers specifically to indicate how they would adapt the task for the grade levels they teach.

After One Year Working in the Project

As part of a subsequent survey of TTML teachers, we posed the same prompts after the teachers had been working on our project for one school year. Because we were asking for other sensitive information as well, we did not ask for names or level of teaching so it was not possible to exactly match responses. Nevertheless the overall responses were analysed in the same way. Table 3 compares the responses of the TTML teachers at the start and the end of the first year of our project. Table 3: Comparisons of teachers responses describing the focus of the task TTML start n = 47 Students learn various ways of comparing 13 26% TTML end n = 34 29%

A single specific concept (not comparing) General and only vaguely related multiple concepts

49% 26%

15% 56%

There were fewer teachers in the second administration of the survey. The main change seems to have been that those who had earlier given a single specific response, other than comparing, now are more likely to give a list of general and only vaguely related multiple concepts. There was no specific focus on developing a language of teaching and so it was not anticipated that there would be much change in the teachers responses to this prompt. Table 4 compares the responses the teachers gave to the prompt about creating a lesson. Table 4: Comparison of responses of project teachers at different time on teacher-created lessons TTML start n = 45 Student centred, perhaps using a meaningful example, emphasizing student generation of strategies and discussion (or the process) A teacher-centred lesson, incorporating a specific strategy for teaching the task Real-life or concrete examples but only vaguely related to the concepts Teacher centred but with a general strategy not specific to the task Nothing or dont know 11% 22% 38% 13% 16% TTML end n = 34 24% 21% 3% 53% 0

In general it seems that the teachers gave longer answers than previously. For example, an example of a response to the prompt on creating lesson that was categorised as Teacher-centred but with general strategy not specific to the task was:
Objective would be to illustrate equivalence and also consolidate fraction sense in terms of collections. Begin with a fraction wall and illustrate how 1/4 is the same as 2/8. Look at other examples, starting with fractions from a whole. Then show simple equivalent fractions using collections e.g. (4 out of 8) and (2 out of 4) challenging student to come up other examples. Ask many questions like how did you do that? and provide opportunities for share time.

The main change from one year to the next seems to have been that the teachers moved away from inappropriate use of contexts or materials, but seemed instead to have created somewhat vague lessons. Since the project was about creating lessons from tasks, and the teachers may have even written up descriptions of lessons they had taught, they might have been expected to provide more focus. Their difficulties might suggest that teachers need to identify the purpose of the task before being able to design an appropriate lesson. Indeed, it seems that to do this teachers need reasonable familiarity with all aspects of both Subject Matter Knowledge and Pedagogical Content Knowledge. 14

It could be noted in this context that six out of the eight teachers who described a student-centred lesson had identified the task as being about comparing, and their lessons consider different aspects of this. On the other hand, 16 out of the 18 teachers who did not identify the main concept as comparing described a general but non-specific lesson. This tends to suggest that both common content knowledge and specialised content knowledge are pre-requisitesas distinct from being corequisites (National Council of Supervisors of Mathematics, 2007)for Pedagogical Content Knowledge.

What Might Make this Difficult?

We also asked teachers in the second survey to indicate what might make creating a lesson from this task difficult. Although the teachers were asked to respond to a hypothetical lesson, it appears that they have responded instead as if asked what might make teaching difficult generally. Many, for example, commented on the constraints that aspects of the students readiness might present. The following are some of their responses: Nine teachers suggested that inadequate prior knowledge may be a barrier, with comments such as:
Childrens lack of understanding of simple fractions &/or place value concepts.

There were five teachers who commented on the difficulty of the specific concept, for example:
Conceptually hard - start with easier 1/2 of 2/3

There were five others who commented on specific misconceptions:

Student misconceptions around the meaning of fractions, e.g. 'larger is bigger' could create the need for focused teaching for some students.

There were three teachers who referred to student attitudes, for example:
Children may feel concept is too challenging - especially as it involves fractions

One commented on class size:

Too many students.

Two commented on equipment limitations:

Not sufficient equipment e.g. models of fractions

These comments all seem to relate to student difficulties, when the item was about teachers difficulties in creating a lesson from this task. As teachers expectation of student abilities and level is clearly one of the factors influencing set up, it is not surprising that teachers thought about creating a lesson for someone as distinct from hypothetically. In doing so, teachers may well seek to reduce the demand of the task. If so, this could have implications for the ways that teacher learning sessions are conducted, with a need for more specific attention to the particular needs of the participating teachers. 15

Revisiting Subject Matter Knowledge and Pedagogical Content Knowledge

One of the purposes of writing this report was to use the data as a way of clarifying different aspects of knowledge for teaching mathematics. Unless a teacher is planning to just throw the fraction comparison task to the class without having solved it personally, we believe that it is clear that a teacher needs relevant common content knowledge to determine which of 2/3 and 201/301 is larger for his or herself. Our observations during the completion of the questionnaire indicated that this was problematic for a number of teachers. Given our belief that an understanding of relevant common content knowledge is a necessary condition for successful translation of such tasks into classroom use, this may explain the inadequacies of a number of the responses. Conversations with teachers around this task indicated that many are not aware of the wide variety of strategies that students might bring to it, aside from attempting to find common denominators or using a calculator to divide the respective numerators by their denominators. Such awareness draws upon specialised content knowledge. Although it could be argued that the content would be challenging for the students of those Grade 5 and 6 teachers who completed the survey, one would nevertheless hope that the teachers possessed sufficient knowledge at the mathematical horizon to discuss how the problem might be used in a junior secondary classroom. We are also somewhat concerned that many teachers seemed unable to describe the mathematical content in terms which indicated that they realised this task was actually about fraction comparison. This connects directly to knowledge of content and curriculum. The responses call into question the sense teachers make of curriculum documents including syllabuses (i.e., the intended curriculum), when this is a difficulty. It is possible that knowledge of content and students had the effect of limiting the vision that teachers had for the use of the task. We are aware that the question was posed in the context of a larger survey and that the requirement to describe a lesson may have been unfamiliar to many of the respondents. Nevertheless, converting mathematics questions to mathematics lessons is the daily work of these teachers, and so the responses and the analysis can be taken on face value.

Implications for Professional Development for Teachers of Mathematics

It is clear that many teachers found translating the fraction comparison task into a worthwhile learning experience for middle school students difficult, or at least had great difficulty in 16

articulating how they might do so. It also seemed that primary and secondary teachers were equally likely to create student-focused investigative type lessons. It became clear to the project team that, particularly when considering topics that both students and teachers find difficult (such as rational number), professional development leaders need to take the time to focus on all six components of knowledge for teaching mathematics. It is also clear that we need to commit more time to discussing how such content might be addressed across the middle years. Importantly, we should not take for granted that all or even most teachers can necessarily translate a good idea or task into a worthwhile learning experience for students, without considerable professional-development support.

References
Ames, C. (1992). Classrooms: Goals structures and student motivation. Journal of Educational Psychology, 84(3), 261-271. Ball, D. L., & Bass, H. (2001). What mathematical knowledge is entailed in teaching children to reason mathematically? In Mathematical Sciences Education Board (Eds.), Knowing and learning mathematics for teaching (pp. 26-34). Washington, DC: National Academies Press. Ball, D. L., Thames, M. H., & Phelps, G. (2008). Content knowledge for teaching: What makes it special? Journal of Teacher Education, 59(5), 389-404. Burkhardt, H., Fraser, R., & Ridgway, J. (1990). The dynamics of curriculum change. In I. Wirszup & R. Streit (Eds.), Developments in school mathematics around the world (Vol. 2, pp. 3-30). Reston, VA: National Council of Teachers of Mathematics. Christiansen, B., & Walther, G. (1986). Task and activity. In B. Christiansen, A.G. Howson, & M. Otte (Eds.) Perspectives on Mathematics Education (pp. 243-307). Holland: D. Reidel. Clarke, D. M., Sukenik, M., Roce, A., & Mitchell, A. (2006) Assessing fraction understanding using task-based interviews. Proceedings of the Conference of the International Group for the Psychology of Mathematics Education, (Vol. 2 337-344). Prague. Curriculum Corporation (2009). Maths 300. Downloaded on April 8 from http://www1.curriculum.edu.au/maths300/ Fredericks, J. A., Blumfield, P. C., & Paris, A.H. (2004). School engagement: Potential of the concept, state of the evidence. Review of Educational Research, 74(1), 59-110. Gehrke, N. K., Knapp, M. S., & Sirotnik, K. A. (1992). In search of the school curriculum. Review of Research in Education, 18, 51-110. Hashimoto, Y., & Becker, J. (1999). The open approach to teaching mathematics creating a culture of mathematics in the classroom: Japan. In L. Sheffield, (Ed.), Developing mathematically promising students (pp. 101 120). Virginia, NCTM. Hiebert, J., & Wearne, D. (1997). Instructional tasks, classroom discourse and student learning in second grade arithmetic. American Educational Research Journal, 30(2), 393425. Hill, H. C., Rowan, B., & Ball, D. L. (2005). Effects of teachers mathematical knowledge for teaching on student achievement. American Educational Research Journal, 42(2), 371406. Hollingsworth, H., Lokan, J., & McCrae, B. (2003). Teaching Mathematics in Australia: Results from the TIMSS Video Study (TIMSS Australia Monograph No. 5). Camberwell, Victoria: Australian Council for Educational Research. 17

Lovitt, C., & Clarke, D. (1988). Mathematics Curriculum and Teaching Program: Activity bank Volumes 1 and 2. Canberra: Curriculum Corporation. National Council of Supervisors of Mathematics (2007). Improving student achievement by leading sustained professional learning for mathematics content and pedagogical knowledge development (The NCSM Improving Student Achievement Series, No. 2). Golden, CO: NCSM. National Numeracy Review. (2008). National numeracy review report. Canberra: Human Capital Working Group. Robitaille, D. F., Schmidt, W. H., Raizen, S., McKnight, C., Britton, E., & Nicol, C. (1993). Curriculum frameworks for mathematics and science (TIMSS Monograph No. 1). Vancouver: Pacific Educational Press. Shulman, L. S. (1986). Those who understand: Knowledge growth in teaching. Educational Researcher, 15(2), 4-14. Stein, M. K., & Lane, S. (1996). Instructional tasks and the development of student capacity to think and reason and analysis of the relationship between teaching and learning in a reform mathematics project. Educational Research and Evaluation, 2(1), 50-80. Stein, M. K., Grover, B. W., & Henningsen, M. (1996). Building student capacity for mathematical thinking and reasoning: An analysis of mathematical tasks used in reform classrooms. American Educational Research Journal, 33(2), 455488. Thames, M. H. (2006). Using math to teach math: mathematicians and educators investigate the mathematics needed for teaching (Critical Issues in Mathematics Education Series, Volume 2). Berkeley, CA: Mathematical Sciences Research Institute. Watson, A., & Sullivan, P. (2008). Teachers learning about tasks and lessons. In D. Tirosh, & T. Wood (Eds), Tools and resources in mathematics teacher education (pp. 109-135). Sense Publishers: Rotterdam.

Peter Sullivan Faculty of Education Monash University Wellington Rd Clayton 3800 Victoria, Australia [email protected] Doug Clarke Australian Catholic University Melbourne Campus 115 Victoria Parade Fitzroy 3065 Victoria, Australia [email protected] Barbara Clarke Faculty of Education Monash University Wellington Rd Clayton 3800 Victoria, Australia [email protected] 18