International Journal of Science Education Vol. 30, No. 14, 17 November 2008, pp. 1945–1969 RESEARCH REPORT Does Practical Work Really Work? A study of the effectiveness of practical work as a teaching and learning method in school science a b Ian Abrahams * and Robin Millar 8 0 20 aBishop Grosseteste University College, UK; bUniversity of York, UK r e mb TT1I0O2T000Dian09000aSanrrt.500yiy.eE.1 ga070rIll0Doionab-0n8nr0rar0_a a0tA6a&2Alih/ 9no00bAa _3dnF09rm2r aar75(Ft7hlasip0 rc4n@aJra0l8oimcenn6i9bucst9s4ir)is0n/s.h1s7 aLgo4l0 mpt61odg47f .-4Sa59cc23.i8ue09kn5 c(eo nEldinuec)ation e c e D 9 5 1 : 7 0 : Many within the science education community and beyond see practical work carried out by t A students as an essential feature of science education. Questions have, however, been raised by ] m iu some science educators about its effectiveness as a teaching and learning strategy. This study t or explored the effectiveness of practical work by analysing a sample of 25 ‘typical’ science lessons s on involving practical work in English secondary schools. Data took the form of observational field C k notes and tape-recorded interviews with teachers and students. The analysis used a model of effec- n Li tiveness based on the work of Millar et al. and Tiberghien. The teachers’ focus in these lessons was - AL predominantly on developing students’ substantive scientific knowledge, rather than on developing E [H understanding of scientific enquiry procedures. Practical work was generally effective in getting y: students to do what is intended with physical objects, but much less effective in getting them to use B d the intended scientific ideas to guide their actions and reflect upon the data they collect. There was e ad little evidence that the cognitive challenge of linking observables to ideas is recognized by those o nl who design practical activities for science lessons. Tasks rarely incorporated explicit strategies to w Do help students to make such links, or were presented in class in ways that reflected the size of the learning demand. The analytical framework used in this study offers a means of assessing the learning demand of practical tasks, and identifying those that require specific support for students’ thinking and learning in order to be effective. Introduction In many countries, one of the features of science education that sets it apart from most other school subjects is that it involves practical work—activities in which the students manipulate and observe real objects and materials. In countries with a *Corresponding author. Department of Education Studies, Bishop Grosseteste University College, Lincoln, LN1 3DY, UK. Email: ian.abrahams @bishopg.ac.uk ISSN 0950-0693 (print)/ISSN 1464-5289 (online)/08/141945–25 © 2008 Taylor & Francis DOI: 10.1080/09500690701749305 1946 I. Abrahams and R. Millar tradition of practical work in school science (such as the UK), practical work is often seen by teachers and others (particularly scientists) as central to the appeal and effec- tiveness of science education. The House of Commons Science and Technology Committee (2002), for example, commented that: In our view, practical work, including fieldwork, is a vital part of science education. It helps students to develop their understanding of science, appreciate that science is based on evidence and acquire hands-on skills that are essential if students are to progress in science. Students should be given the opportunity to do exciting and varied experimental and investigative work. (para. 40) The influential Roberts (2002) report, on the supply of people with science, technol- ogy, engineering, and mathematics skills, highlights the quality of school science laboratories as a key concern. These, it argues, ‘are a vital part of students’ learning experiences … and should play an important role in encouraging students to study 08 [science] at higher levels’ (Roberts, 2002, p. 66). It goes on to recommend: 0 2 r be that the Government and Local Education Authorities prioritise school science … labo- m ce ratories, and ensure that investment is made available to bring all such laboratories up e D to … a good or excellent standard … by 2010: a standard which is representative of the 9 5 world of science and technology today and that will help to inspire and motivate 1 : 7 students to study these subjects further. (Roberts, 2002, p. 66) 0 : t A There is also evidence that students find practical work relatively useful and enjoy- ] um able as compared with other science teaching and learning activities. In survey i t or responses of over 1,400 students (of a range of ages) (Cerini, Murray, & Reiss, s n o 2003), 71% chose ‘doing an experiment in class’ as one of the three methods of C nk teaching and learning science they found ‘most enjoyable’. A somewhat smaller i L - proportion (38%) selected it as one of the three methods of teaching and learning L A HE science they found ‘most useful and effective’. In both cases, this placed it third in [ y: rank order. B d Despite the widespread use of practical work as a teaching and learning strategy in e d oa school science, and the commonly expressed view that increasing its amount would l n ow improve science education, some science educators have raised questions about its D effectiveness. Hodson (1991), for example, claims that: ‘As practiced in many schools it [practical work] is ill-conceived, confused and unproductive. For many children, what goes on in the laboratory contributes little to their learning of science’ (p. 176). From a similar viewpoint, Osborne (1993) proposes and discusses a range of alternatives to practical work. Wellington (1998) suggests that it is ‘time for a reappraisal’ (p. 3) of the role of practical work in the teaching and learning of science. This article presents findings from a study of the effectiveness of practical work as it is typically used in science classes for 11-year-old to 16-year-old students in maintained schools in England. The research question the study addressed was, essentially: How effective is practical work in school science, as it is actually carried out, as a teaching and learning strategy? The study looked at both cogni- tive and affective outcomes of practical work; this article focuses on cognitive Effectiveness of Practical Work 1947 outcomes—the effectiveness of practical work in enhancing students’ knowledge and understanding, either of the natural world or of the processes and practices of scientific enquiry. Throughout we will use the term ‘practical work’, rather than ‘laboratory work’ or ‘experiments’, to describe the kind of lesson activity we are interested in. An ‘experiment’, particularly in philosophy of science, is generally taken to mean a planned intervention in the material world to test a prediction derived from a theory or hypothesis. Many school science practical tasks, however, do not have this form. And while many practical lessons are undertaken in specifi- cally designed and purpose-built laboratories (White, 1988), the type of activity we are interested in is characterized by the kinds of things students do, rather than where they do them. A Framework for Considering the Effectiveness of Practical Work 8 0 20 Practical work, as several authors have pointed out, is a broad category that encom- r be passes activities of a wide range of types and with widely differing aims and objec- m e c tives (Lunetta & Tamir, 1979; Millar, Le Maréchal & Tiberghien, 1999). It does not e D 9 make sense, therefore, to ask whether practical work in general is an effective teach- 5 1 ing and learning strategy. Rather, we need to consider the effectiveness of specific : 7 0 examples of practical work, or specific practical tasks. To develop an analytical frame- : t A work, the present study started from a model of the processes involved in designing ] m u and evaluating a practical task (Figure 1) proposed by Millar et al. (1999). i t or Figure 1.MTodel of thhe proceess of d esigns and tevaluaation rof a ptractiicaln taskg point (Box A) is the teacher’s learning objectives—what he or she s n Co wants the students to learn. This might be a specific piece of substantive scientific nk knowledge or a specific aspect of the process of scientific enquiry (about, for example, i L - L A E H [ : y B (cid:0) (cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:2)(cid:6)(cid:7)(cid:8)(cid:9)(cid:10)(cid:11)(cid:2)(cid:4)(cid:12)(cid:13)(cid:14)(cid:2)(cid:8)(cid:15)(cid:16)(cid:5)(cid:3)(cid:12)(cid:12)(cid:5)(cid:2) d e ad (cid:8)(cid:12)(cid:17)(cid:18)(cid:2)(cid:19)(cid:12)(cid:8)(cid:3)(cid:6)(cid:2)(cid:13)(cid:19)(cid:12)(cid:2)(cid:19)(cid:18)(cid:2)(cid:18)(cid:12)(cid:9)(cid:20)(cid:2)(cid:3)(cid:6)(cid:19)(cid:21) o l n w o D (cid:1) (cid:22)(cid:2)(cid:8)(cid:13)(cid:23)(cid:19)(cid:24)(cid:2)(cid:3)(cid:12)(cid:17)(cid:6)(cid:2)(cid:8)(cid:9)(cid:24)(cid:12)(cid:3)(cid:8)(cid:25)(cid:26)(cid:18)(cid:2)(cid:12)(cid:3)(cid:13)(cid:20)(cid:8)(cid:9)(cid:24) (cid:4)(cid:9)(cid:19)(cid:12)(cid:2)(cid:27)(cid:12)(cid:15)(cid:16)(cid:5)(cid:3)(cid:12)(cid:8)(cid:12)(cid:17)(cid:18)(cid:2)(cid:19)(cid:12)(cid:8)(cid:5)(cid:3)(cid:14)(cid:2)(cid:12)(cid:9)(cid:18)(cid:9)(cid:21) (cid:30)(cid:24)(cid:24)(cid:2)(cid:4)(cid:12)(cid:13)(cid:14)(cid:2)(cid:19)(cid:2)(cid:8)(cid:8) (cid:30)(cid:24)(cid:24)(cid:2)(cid:4)(cid:12)(cid:13)(cid:14)(cid:2)(cid:19)(cid:2)(cid:8)(cid:8) (cid:31)(cid:2)(cid:14)(cid:2)(cid:20)! (cid:31)(cid:2)(cid:14)(cid:2)(cid:20) (cid:2) (cid:28)(cid:5)(cid:3)(cid:12)(cid:12)(cid:5)(cid:2)(cid:8)(cid:12)(cid:17)(cid:18)(cid:2)(cid:19)(cid:12)(cid:8)(cid:3)(cid:4)(cid:12)(cid:17)(cid:3)(cid:20)(cid:20)(cid:29)(cid:18)(cid:9) (cid:3) (cid:28)(cid:5)(cid:3)(cid:12)(cid:12)(cid:5)(cid:2)(cid:8)(cid:12)(cid:17)(cid:18)(cid:2)(cid:19)(cid:12)(cid:8)(cid:3)(cid:4)(cid:12)(cid:17)(cid:3)(cid:20)(cid:20)(cid:29)(cid:20)(cid:2)(cid:3)(cid:6)(cid:19) Figure 1. Model of the process of design and evaluation of a practical task 1948 I. Abrahams and R. Millar the collection, analysis, or interpretation of empirical evidence). Once this has been decided, the next step (Box B) is to design (or select) a practical task that might enable the students to achieve the desired learning objectives. The next stage of the model (Box C) asks what the students actually do as they undertake the task. For various reasons, this may differ to a greater or lesser extent from what was intended by the teacher (or the author of the practical task). For example, the students might not understand the instructions; or they may understand and follow them meticulously but be prevented by faulty or inadequate apparatus from doing or seeing what the teacher intended. Even if the task is carried out as intended, and the apparatus func- tions as it is designed to do, the students still may not think about the task and the observations they make using the ideas that the teacher intended (and perhaps indeed expected) them to use. We can think of this as a matter of whether or not students do the things the teacher intended with ideas; that is, their mental actions as distinct from their physical actions. The final stage of the model (Box D) is then concerned with 8 0 20 what the students learn as a consequence of undertaking the task. This model there- r be fore distinguishes two senses of ‘effectiveness’. We can consider the match between m e c what the teacher intended students to do and what they actually do (the effectiveness e D 9 of the task at Level 1); and the match between what the teacher intended the students 5 1 to learn and what they actually learn (the effectiveness of the task at Level 2). ‘Level : 7 0 1 effectiveness’ is therefore concerned with the relationship between Boxes B and C : t A in Figure 1, while ‘Level 2 effectiveness’ is concerned with the relationship between ] m u Boxes A and D. i t or In the discussion above, we have already alluded to a further dimension—the kind s n Co of action (physical or mental), and hence learning, involved. The fundamental nk purpose of practical work in school science is to help students make links between i L L- the real world of objects, materials and events, and the abstract world of thought and A E H ideas (Brodin, 1978; Millar et al., 1999; Shamos, 1960). Tiberghien (2000) charac- [ y: terizes practical work as trying to help students make links between two ‘domains’ of B ed knowledge: the domain of objects and observables (o) and the domain of ideas (i) d a o (Figure 2). l n ow Figure 2.PrSactical woork: lminking two edomai ns (frosm Tibcerghihen, 200o0) ol science practical tasks deal only, or mainly, with the domain of D observables; others involve both domains. Combining the two-level model of effec- tiveness with this two-domain model of knowledge leads to the analytical framework presented in Table 1 for considering the effectiveness of a given practical task. This framework can apply equally to practical tasks in which the focus is on students’ learning of substantive scientific knowledge or on learning about some aspect of scientific enquiry procedures. Domain of observables (o) Domain of ideas (objects, materials and (i) phenomena) Figure 2. Practical work: linking two domains (from Tiberghien, 2000) Effectiveness of Practical Work 1949 Table 1. Analytical framework for considering the effectiveness of a practical task Effectiveness Domain of observables (o) Domain of ideas (i) A practical task is … the students do with the … whilst carrying out the task, effective at Level 1 (the objects and materials provided the students think about their ‘doing’ level) if … what the teacher intended them actions and observations using to do, and generate the kind of the ideas that the teacher data the teacher intended. intended them to use. A practical task is … the students can later recall … the students can later show effective at Level 2 (the things they did with objects or understanding of the ideas the ‘learning’ level) if … materials, or observed when task was designed to help them carrying out the task, and key learn. features of the data they collected. 8 0 20 The four cells of Table 1 are not independent. It seems unlikely, for example, that r be a task could be effective at Level 2:i unless it was also effective at Level 1:i, and m e c perhaps in turn at Level 1:o. And we are more likely to be interested in evidence of e D 9 successful learning at Level 2:o if the task has been effective at Level 1:o (in other 5 1 words, the actions and observations that the students recall are the ones we wanted : 7 0 them to make). Despite these interdependencies, this framework provides a useful : t A tool for analysing examples of practical work in school science. Table 2 shows how it ] m u might apply to a practical task in which the students are investigating electric currents i t or in parallel branches of an electric circuit, where the teacher’s aim is that students s n Co should develop their understanding of the scientific model of current as moving nk charges. If the teacher’s focus were instead on developing students’ understanding of i L L- how to deal with ‘messy’ real data, then domain-o thinking would focus on the actual A E H observations and data collected, whereas domain-i thinking would see these as an [ y: instance of a more general phenomenon, measurement error (or uncertainty). B d e d a lo Table 2. Indicators of the effectiveness of a practical task involving an investigation of electric n ow current at each level and domain D Effective Domain of observables (domain o) Domain of ideas (domain i) Level 1 (the Students set up the parallel circuit Students talk and think about the circuit ‘doing’ level) correctly from a given diagram and and the meter readings using the idea of are able to insert an ammeter electric current (charges flowing through correctly and read with sufficient wires, and the flow dividing and accuracy to obtain the pattern of recombining at junction points). readings intended by the teacher. Level 2 (the Students are able later to set up a Students show understanding of electric ‘learning’ level) parallel circuit, and can recall that current as a flow of charges, and can the sum of the ammeter readings in apply this idea to circuits with parallel two parallel branches is equal to branches, for example to explain why the the reading on an ammeter placed sum of the branch currents is equal to before or after the branch. the current before or after the branch. 1950 I. Abrahams and R. Millar A possible objection to this theoretical framework is that all observation is ‘theory- laden’, so there is no clear distinction between observables and ideas. Hanson (1958) argues that even basic observation statements that report sensory experience are dependent upon the theoretical framework within which the observer operates (for examples of this in science education contexts, see Gott & Welford, 1987; Hainsworth, 1956). Feyerabend (1988) goes further, asserting that ‘observation statement[s] are not just theory-laden … but fully theoretical’ (p. 229; original empha- sis). He argues, however, that a pragmatic distinction can nonetheless be made between observational and theoretical statements. A statement can be regarded as observational, Feyerabend suggests, if it is a ‘quickly decidable sentence’; that is: [a] singular, nonanalytic sentence such that a reliable, reasonably sophisticated language user can very quickly decide whether to assert or deny it when he is reporting on an occurrent situation. (Feyerabend, cited in Maxwell, 1962, p. 13) 8 00 The distinction that we draw in this study between the domain of objects and observ- 2 er ables and the domain of ideas (and hence between statements about these domains) b em is a pragmatic one, along these lines. We accept that all observations are, at some c e D level ‘theory-laden’, but would argue that the extent of their ‘theory-ladenness’ 9 5 differs considerably, and that the theory with which a given statement is ‘laden’ is 1 : 07 often not at issue or under test in the context in which the statement is being : At asserted. The distinction between observables and ideas is, we believe, a valuable and ] m important one in analysing the effectiveness of practical tasks. u i t r o s n o C Research Strategy and Methods k n i L L- Large-scale quantitative studies of school science practical work in the UK, the most A E H recent of which is now over 20 years old (Beatty & Woolnough, 1982; Kerr, 1964; [ y: Thompson, 1975), have provided insights into the views of teachers and students. B d These studies did not, however, compare expressed views on practical work with e d oa observations of actual practice. They might therefore be seen as studies of the rheto- l n ow ric of practical work, rather than the reality. It has been suggested by Crossley and D Vulliamy (1984) that questionnaire-based surveys are unlikely to provide accurate insights into the reality of teaching within its natural setting but are more likely to reproduce existent rhetoric. An interview study is open to the same objection (Cohen, Manion, & Morrison, 2000; Hammersley & Atkinson, 1983). In contrast, this study sought to explore critically the reality of practical work in the school labo- ratory. This requires a strategy that brings the researcher into closer contact with teachers and students as they undertake practical work, collecting data in the teach- ing laboratory, focusing on observation of actual practices augmented by interviews conducted in the context of these observations. Such a strategy may achieve a higher degree of ecological validity (Bracht & Glass, 1968); that is, external validity and generalizability to other settings. When an interviewee is aware that the interviewer has observed the practice being discussed, responses are more effectively anchored to realities, and less likely to be ‘rhetorical’ in nature. Effectiveness of Practical Work 1951 For these reasons a case-study approach was chosen. There are a number of precedents for the use of such a strategy to explore, in a critical manner, the relation- ship between rhetoric and reality within an educational context (see, e.g., Ball, 1981; Sharp & Green, 1976). To avoid what Firestone and Herriott (1984) term the ‘radi- cal particularism’ of the traditional single in-depth case study, we used a multi-site approach, involving a series of 25 case studies in different settings, similar in scale to those undertaken by Firestone and Herriott (1984) and Stenhouse (1984). Schofield (1993) suggests that ‘the possibility of studying numerous heterogeneous sites makes multi-site studies one potentially useful approach to increasing the generalizability of qualitative work’ (p. 101). Eight schools were approached and the head of the science department asked for permission to observe one or more science lessons at national curriculum Key Stage 3 or Key Stage 4 (students aged 11–14 and 15–16, respectively) that involved some student practical work, to talk to the teacher about the lesson, and perhaps also 8 0 20 to talk to some of the students. In some science lessons in English schools, students r be are assessed on their performance of a practical investigation, and this contributes to m e c their national test score at age 14 and their grade in the General Certificate of e D 9 Secondary Education at age 16. We asked that the lessons observed should not be of 5 1 this kind (indeed, we thought that schools were unlikely to give us permission to : 7 0 observe these, as a researcher’s presence could have been an unnecessary distrac- : t A tion). Some possible consequences of this are discussed below. All the schools ] m u approached were maintained state comprehensive schools, in a variety of urban and i t or rural settings. Some of their characteristics are presented in Table 3; the school s n Co names listed are pseudonyms. As a group they were broadly representative of nk secondary schools in England. i L L- We had limited control of the content or subject matter of the lessons actually A E H observed in each school. Typically, a date was agreed for the observation visit, and [ y: a number of lessons with different teachers were offered as possible options when B ed the researcher arrived. Choices were made on the basis of practical considerations d a o of timing to allow pre-lesson and post-lesson teacher interviews, and with the aim, l n ow as the study proceeded, of achieving reasonably even coverage of the five school D Table 3. School sample School Location Size Age range (years) Education authority Derwent Urban 500 11–16 A Foss Urban 1480 11–18 A Kyle Urban 1550 11–18 B Nidd Rural 890 11–18 B Ouse Rural 630 11–18 B Rye Rural 720 11–18 C Swale Rural 670 11–16 B Ure Rural 1280 11–18 C 1952 I. Abrahams and R. Millar Table 4. Sample of lessons observed by science subject and Key Stage Number of lessons observed Key Stage (student age) Biology Chemistry Physics Total 3 (11–14 years) 2 6 7 15 4 (15–16 years) 1 3 6 10 years in Key Stages 3 and 4, and ensuring that the sample included biology, chem- istry, and physics topics. The distribution of the lessons observed across Key Stages and science subjects is presented in Table 4. The lower number of biology lessons observed is a reflection of the number of student practical tasks that appear to be carried out by students in biology lessons as compared with chemistry and physics. The lesson observations later in the sequence seemed to raise the same 8 0 20 issues as earlier ones, suggesting that data saturation had been achieved by this r be point. The content of the 25 lessons observed is summarized in Table 5, along m e c e D 9 5 Table 5. Practical tasks and teachers observed 1 : 7 0 Task content Teacher Key Stage : t A m] 1 Food tests—test results Mrs Ugthorpe 3 u ti 2 Heart beat/pulse—numerical equivalence Mrs Risplith 3 r o s 3 Chemical reactions—how to identify Mr Dacre 3 n o C 4 Separation—sand and pepper Mr Fangfoss 3 k in 5 Separation—iron, salt, and sand Mr Keld 3 L - L 6 Chromatography—separation of inks Miss Nunwick 3 A E [H 7 Cooling curve—characteristic plateau Mr Oldstead 3 y: 8 Chromatograph—separation of inks Mr Saltmarsh 3 B d 9 Heat absorption—colour as a variable Mr Drax 3 e d oa 10 Electric circuits—current conservation Mrs Duggleby 3 l wn 11 Electric circuits—current conservation Ms Ferrensby 3 o D 12 Electromagnets—factors effecting strength Dr Kepwick (female) 3 13 Electromagnets—factors effecting strength Mrs Kettlesing 3 14 Pulleys and levers—factors affecting Miss Kilburn 3 15 Magnetic permeability of materials Mr Overton 3 16 Starch production—factors that effect Mr Sewerby 4 17 Acid + base = salt + water Mr Drax 4 18 Electrolysis—increase in cathode mass Mr Ulleskelf 4 19 Electrolysis—cathode deposits Mr Rainton 4 20 Lenses and eyes—similarities Mr Normanby 4 21 Refraction—ray paths Mr Normanby 4 22 Current in series and parallel circuits Mrs Uckerby 4 23 Voltage in parallel circuits Mrs Ramsgill 4 24 Work done in raising mass Miss Sharow 4 25 Current and voltage in series circuit Dr Starbeck (male) 4 Effectiveness of Practical Work 1953 with details of the teacher and the age of the students involved. The teachers’ names are all pseudonyms; the initial letter of their surname matches that of their school (see Table 3). Field notes were taken in each lesson observed, and tape-recorded interviews were carried out with the teacher before and after the lesson. The pre-lesson interview was used to obtain the teacher’s account of the practical work to be observed and of his or her view of the learning objectives of the lesson. The post-lesson interview collected the teacher’s reflections on the lesson and on its success as a teaching and learning event. Where possible, conversations with groups of students during and after the lesson were also tape-recorded. These were used primarily to gain insights into the students’ thinking about the task that were not apparent from observation alone, or to confirm the impression gained from observation. 8 0 20 Findings r e b m Introduction e c e D 9 The analytical framework presented in Table 1 was used in analysing the data, and 5 1 will also be used here to structure the discussion. We will begin by considering the : 7 0 effectiveness of tasks at Level 1 (in getting students to do what the teacher : t A intended), and then go on to consider effectiveness at Level 2 (in promoting the ] um learning the teacher intended). Throughout this discussion, each teacher is given a i t or pseudonym. In extracts from interviews with students, each is identified by a code s n o consisting of the first and last letters of the teacher’s surname (to identify the lesson C nk involved) and a number. i L - First, however, one general point should be made. In all the lessons we observed, L A HE the teacher’s focus appeared to be firmly (indeed almost exclusively) on the substan- [ y: tive science content of the practical task. There was almost no discussion in any of the B d lessons observed of specific points about scientific enquiry in general, or any exam- e d oa ples of use by the teacher of students’ data to draw out general points about the l n ow collection, analysis, and interpretation of empirical data. In some lessons where there D were clear opportunities to do this, they were not exploited. So, in the discussion below, our focus is largely on the use of practical work to develop students’ under- standing of substantive science ideas—not because our framework excluded other aspects of learning, but because this reflects what we actually observed. Readers familiar with the English national curriculum for science might see this as a conse- quence of our decision not to observe lessons in which students were being assessed. Donnelly et al. (1996), in a detailed study of the ‘Scientific Enquiry’ component of the English national curriculum (Attainment Target Sc1), found that extended, and more open-ended, investigative practical tasks were rarely used to teach students about specific aspects of scientific enquiry, but almost entirely to assess their ability to conduct an empirical enquiry ‘scientifically’. It would seem, therefore, that an unintended consequence of the introduction of Attainment Target Sc1 may be that teachers overlook opportunities that arise in the course of illustrative practical work 1954 I. Abrahams and R. Millar (i.e., practical tasks primarily intended to let students observe a phenomenon, or to help them understand a scientific idea or explanation) to highlight and discuss the rationale for the design of the task, or issues about data analysis and interpretation thrown up by the data actually collected—seeing this as a distinct strand of the science curriculum with which they deal on other occasions. What Students Do with Objects and Materials (Level 1:o) The practical work observed was, in most cases, effective in enabling the majority of students to do what the teacher intended with the objects provided—that is, success- fully to ‘produce the phenomenon’ (Hacking, 1983). Various factors contributed to this; in particular, the widespread use of ‘recipe style’ tasks (Clackson & Wright, 1992; Kirschner, 1992). In many of the lessons observed, teachers focused their efforts on ensuring that students understood the procedure they had to follow. 8 0 20 A particular piece of practical work (often the central feature of a lesson) was likely r be to be considered successful by the teacher if the students had managed to produce m e c the desired phenomena and make the desired observations. e D 9 Many teachers in the study, particularly those teaching outside their subject 5 1 specialism, explained their choice of the practical task observed by referring to a : 7 0 departmental scheme of work, as in the following excerpt: : t A m] Researcher: Why did you choose to do this as a practical? u ti Mrs Ramsgill: It was part of the new scheme of work [a commercially produced r so scheme that the department had recently purchased] we are now using. n Co Researcher: So it wasn’t really your choice? nk Mrs Ramsgill: No, no, it wasn’t. i -L Researcher: Is that the same for the work sheets? L EA Mrs Ramsgill: Yes, they are part of the same scheme. H [ y: This moved responsibility for the choice of question to be addressed and/or B d phenomenon to be produced (as well as other issues relating to the task) on to the e d oa author(s) of a published or departmentally produced scheme of work, and portrayed l n ow their own responsibility primarily in terms of ‘delivering’ an activity judged appro- D priate by others. Fourteen of the 25 teachers observed said they were following a scheme of work that included the practical activity observed. Nine teachers used worksheets that were part of such a scheme. Use of both was greater among teachers for whom the lesson was outside their science specialism. Table 6 shows that four (of nine) teachers teaching in their subject specialism were following a scheme of work, compared with 10 (of 16) teachers teaching outside their subject specialism. Similarly, while only two (of nine) teachers teaching within their subject specialism used worksheets, this rose to seven (of 16) for those teaching outside their subject specialism. While the sample size (n = 25) is too small to generalize with confidence from these data, the pattern is consistent with the findings of other research (e.g., Hacker & Rowe, 1985) that teachers working outside their specialist subject tend to rely more on routine and controllable activities, which reduce the likelihood of unexpected events or questions.