2006 POSTGRADUATE CONFERENCE Mathematical Understanding in a Vocational Context: An Exploration in Applying a Stimulated Recall has proved valuable in gathering such uthentic data useful in identifying mathematical cognition in plumbing trade apprentices. Intr Comp also extra ompassing collectives, therefore impacting on respective orkforces. Cornford and Beven (1999) stated that a skilled workforce is the key to national wealth and for effective participation in work, one of ich being the use of mathematical ideas and techniques. In terms of education and training, the ‘practical Technique Using Video Michaela Inglis, Dr David Paterson Abstract Extracting data containing information about cognitive processing in workplace settings can be a difficult exercise, particularly as cognitive functioning cannot be accessed directly but only through observation of action. When the task being observed is of a highly practical and individual nature, concurrent verbalisation of action is not normal practice. This makes accessing the thoughts of the participant particularly difficult and requires thoughtful and targeted methods to provide authentic data to inform cognitive processing. This article presents the derivation a four-step stimulated recall technique which a oduction etence in a chosen vocation is of importance to the majority of the population. Such competence polates beyond that of the individual enc w economic ascendancy; an idea that has been adopted by various governments in Australia, leading to calls for a ‘knowledge nation’ (Chifley Research Centre 2001). Mayer (1992) listed a number of fundamental competencies required wh or functional application and use of mathematics’ has been identified as a key area (National Centre for Vocational Education Research 2001 p.2). Such a focus for research has not been limited to Australia, and previous studies have covered a wide range of contexts. Some of which include: the mathematical knowledge of carpenters (Milroy 1992); street-vendors (Carraher et al. 1985); shoppers (Lave et al. 1984); understanding of floor plans (Clarke & Helme 1993); carpet layers (Masingila 1992); paediatric nurses (Noss et al. 2000); bankers (Noss & Hoyles 1996); technological workers (Sträßer 1999), technical drawing of engineers (Sträßer & Bromme 1992); book makers (Schliemann & Acioly 1989); building industry workers (Eberhard 2000; Fioriti & Gorgorio 2001); retailers (van der Zwart 2000); reading of building site plans (Bessot 2000); and jewellery makers (Hahn 2000). 168 UNIVERSITY OF NEW ENGLAND Predominantly, research has been concerned with the identification of what mathematics is used. Whilst these inspections are worthy and important to the field, research which has moved ahead from this search for ‘how much’ to looking at the quality of responses are recent and rare. Studies such as those by de Aguero (2003) into the differences in painters’ schemas for solving the same workplace problem, have begun to delve deeper into the notion of the ‘quality’ of workers’ mathematical behaviour. Such behaviour in workplace ontexts is intrinsically related to actions such as planning, making judgements, and problem solving, as r the quality of workers’ mathematical ideas and techniques on a cognitive level, wo difficulties were the major influence on choosing to use a Stimulated Recall o collect the data, and are therefore explored in the following discussion. Workers do irst difficulty that workers often do not re fact that workplace mathematics is first instan hidden within artefacts/t mathematic yet “practitioners’ epistemologies of quantity, space and time n c opposed to just listing the mathematics in use. All of these actions require a certain level of mental functioning, or cognitive performance. Consequently, it seems reasonable to propose that cognitive theory could provide a mechanism by which levels of quality could be identified and analysed (Inglis & Pegg 2003). Methodological problems in uncovering mathematical behaviour In attempting to uncove researchers face at least two major difficulties. The first of these is that workers do not often recognise the mathematics they are using. The second is that such cognitive functioning is not directly measurable (Biggs & Collis 1982). These t Technique t not recognise the mathematics they are using. There are two possible explanations for the f cognise the mathematics they are using, both of which are related to the embedded in context (e.g. Noss et al.2000; Sträßer 1996, 1999). In the ce, it may be that workers often do not recognise the mathematics they are using because it is echnology (Sträßer 1996, 2003). The second is that the most easily recognisable s is that oriented to schooling, and can differ fundamentally from those anticipated from a school mathematics orientation” (Noss et al. 2000 p.33). Noss et al. (2000 p.32) label these epistemologies as situated abstractions, defined as the ‘practitioner’s conceptions of the mathematics they use at work.’ These situated abstractions of workers are generally limited to the four major mathematical operations of addition, subtraction, multiplication and division (Foyster 1990; Milroy 1992), or standard arithmetic algorithms (Wedege 2002). Mathematics in use may be neither important nor accessible to the worker. From the workers’ point of view, the importance of knowing that mathematics exists is likely to be minimal where mathematics is interwoven within workplace problems. They are more likely to see occupational and professional concerns as being more important (Noss et al. 2000), and hence afford more attention to those than to overt mathematics. Instead, the worker uses strategies to minimise time taken and maximise monetary gain. The workers’ access to mathematics is also less likely in circumstances where efficiency implies automaticity. Cognitio 169 2006 POSTGRADUATE CONFERENCE that is automated is generally not accessible to the individual (Ericsson & Simon 1980), and hence it is not Fo he What ndesirable response effects leading to invalid and/or unreliable data on ac l Ac ss l challenge faced by researchers attempting to xplore the quality of worker’s mathematical cognition is that cognitive behaviour cannot be measured directly ve perspective. The most common method to gain such access is by terviewing the participant. White and Gunstone (1992 p.65) stated: stance is a deep probe of the student’s understanding of a single concept, cher is often immersed in the workplace, with explanations given by e worker during the task performance. This procedure has not been without some criticism, particularly and downgrade performance.’ However, despite the limitations, uch strategies have the benefit of providing the participant with the opportunity to explicate practice in the surprising that workers do not recognise the mathematics that they are using. r t se two reasons, minimal importance and inaccessibility, research which asks the worker ‘ mathematics do you use?’ may give u tua workplace practice. ce to cognitions. The second major methodologica e by the researcher, and must instead be gathered by observing the actions and gaining access to the thoughts of the participant. Consequently, in order to assess the quality of workers’ mathematical cognitions, the researcher needs not only to gather information using techniques such as field notes and observations, but also to access the worker’s cogniti in An interview about an in that checks whether the student can not only recognise whether the concept is present in specific instances but also whether the student can explain his or her decision. The explanation reveals the quality of the student’s understanding. Interview methods are commonly used to obtain participants’ experiences and points of view. In studies of workplace mathematics, there appear to be three main types of interview method: (i) participant verbalisations of practice in ‘real-time’, (ii) participant reflections on action after the event, and (iii) participant reflection on action after the event stimulated by reflective prompts. In the first type of interview, the resear th because of the potential for concurrent verbalisation to cause ‘an unacceptable degree of interference with task performance’ (Marland 1984 p.157). Thinking-aloud strategies have also been documented as placing additional strain on working memory and impeding ‘best-practice.’ Studies such as those by Dickson et al. (2000) proposed that using concurrent verbalisation to study cognitive processes in task execution could have ‘reactive effects on task processes s presence of salient cues in real-time. 170 UNIVERSITY OF NEW ENGLAND The second type of interview involves obtaining reflections on a lived event. These data are collected after the event to potentially provide insight into the participants’ point of view. Generally, such interviews are emi-structured and seek to clarify issues which arose during the observation (see, for example, Hogan & s. Initially, the task is performed and then followed by interview session; static art c verbal probes are on hand to provide the cues and stimulate recall. In pp er (2000) stated that stimulated response type research was superior to tra io for this being that the richness of the data collected llowed the researcher the potential to uncover the hidden mathematics of performance. ons of the activity, with supplementary information provided in the form of familiar resources and visual displays: e.g. copies of blood pressure charts, screen shots of s Morony 2000 p.104). What such interviews cannot guarantee is accurate recall, due the absence of appropriate stimuli. The third type of interview involves the participant providing thoughts on or in action after the event, but stimulated by appropriate cue efa ts (such as photos) and targeted su ort of such methods, Sträß dit nal interview and survey studies. The reasons a An example of such a study examined the mathematics in use by investment bank employees, paediatric nurses and commercial pilots (Noss et al. 2000). In relation to the data-collection methodology they stated the following: A series of task simulation interviews was devised, based on a selection of workplace activities and contexts. The majority of the simulations recreated the breakdown episodes witnessed during the ethnographic observation. Each scenario was made as ‘real’ as possible, by the use of context-rich descripti instrumentation (p.21). These interviews provided insight into the participants’ reasoning whilst allowing the researchers to manufacture novel situations. However, this type of method does not provide information regarding actual workplace practice, as such ‘simulations’ cannot be said to provide salient cues in real time to the participant. Contrived episodes of simulated workplace practice are to some extent decontextualised: devoid of the affective factors influencing performance in an actual real-time task. For interviews that rely on participant reflection on a certain task, the quality and authenticity of the data gathered is directly related to the appropriateness of the cues provided to stimulate recall. The accuracy of either explanations of practice, or thought-in-action, rely heavily on the provision of salient cues to stimulate such recall. One way of providing salient cues to stimulate recall is by using videotaped recording of task performance. As much as possible, these perspectives of the video should reflect the participants’ line of sight to maximise possibility of accurate recall in the stimulated recall interview sessions. 171 2006 POSTGRADUATE CONFERENCE Traditionally, video recordings have been used by researchers in triangulation with other data sources such as field notes and artefacts, providing additional real-time data in context. Video data have the ability to: • preserve subtle aspects of performance (eye gestures, body posture, intonation, environmental trinsic bias • Altho of resear ssed. Using video-data of the lived event as a stimulus for an interv ome minimisation of resea h erate data from vocational contexts is stimulated oural sciences as interpersonal process recall (Kagan & minantly involves the use of mechanically recorded data (audio or video) consisting of ‘reactive effects on task processes and downgrade performance.’ To maintain the internal validity of the noises and subsequent interactions) • allow for repeated examination of data, • have no in are flexible with regard to setting types. ugh such data increases the validity of analysis when used within mixed-methodological studies, issues cher subjectivity must still be addre iew however, can provide data from the participant’s point of view, enabling s rc er subjectivity in analysis. Literature Review One possible data gathering methodology proposed to gen recall. Stimulated recall (also referred to in behavi Kagan 1991)) is a technique designed to provide access to the cognitive processes of human problem solvers during performance tasks (Marland 1984). First pioneered by Bloom in 1956 to study the thought processes involved in university students’ discussions, the most typical instances of stimulated recall as a defined genre of method have been to examine the ‘in-flight’ (Paterson 2000 p.44) thoughts of teachers (e.g. Marland 1984; Paterson 2000; Stough & Palmer 2003) and the reflective practice in the teaching and learning of physical activities (e.g. Lee 2002; Parker & Pittney 2003; Solomon & Lee 1997). Although these areas of research are not exclusive, these two authors have found no overt uses of this technique in existing literature on mathematics in vocational contexts. Stimulated recall predo a participant’s behaviour (visual and audio) whilst they are carrying out a particular task, to ‘stimulate recall of simultaneously occurring thought processes’ (Marland 1984 p.156). The common underlying rationale for research using a stimulated recall method is that it potentially provides the participant with appropriate stimuli to be able to relive an original situation. Marland (1984 p.157) described the use of stimulated recall in circumstances where think-aloud methods would cause ‘an unacceptable degree of interference with task performance.’ More recently, Dickson et al. (2000 p.220) cautioned against the use of concurrent verbalisation regarding in-task execution as it may have 172 UNIVERSITY OF NEW ENGLAND data, these issues must be taken into account, particularly when the day-to-day execution of tasks being observed are executed without such concurrent verbalisations. Stimulated recall is a powerful methodology for data collection (Paterson 2000); however, special care must • the availability of interactive thoughts for recall. can be used to direct the participants to what is required from the interview. In rder to reduce distortion of the participants’ verbal data, the probes must remain neutral. This makes it less ected, providing in-flight thinking at the time of the event. f optimum recall conditions require that the recorded data must be as close as is only “tangentially related to actual thinking during the recorde d added that it is more likely to be ‘reflection-on- action’ rathe ibing what they see themselves doing, a be taken in its application. To date, no singular optimum way for applying the technique has been determined, with individual research circumstances dictating the particular methodology required. Keith (1988 p.8) raised three questions which a researcher should pay attention to regarding adoption of a stimulated recall technique: • possible bias in the questioning • the nature of the stimulus Bias in the questioning. Bias in the questioning appears to manifest mainly within the nature of the probes being directed at the participants during the interview. These probes may cue the participants to certain areas that the researcher may wish to focus on, and must be carefully monitored to reduce response effects. Keith (1988) suggested that one way of reducing these response effects was to use unstructured interviewing techniques. Questions such as ‘stop the video when you want to discuss what you see’ or ‘tell me what is going through your mind’ o likely that the participant is prompted and their self-reporting data distorted from what they otherwise would have said (Yinger 1986). The nature of the stimulus. Marland (1984) made two assumptions regarding aural or visual stimuli. Firstly that such stimuli should provide sufficient cues to “stimulate accurate and near-complete recall of mental processes” (p.158) and secondly that the presence of the recording hardware should not change the nature of the performance in a research setting. To address the first point, stimulated recall proponents generally assume that the up-close and uninterrupted record of performance prompts participants to recall as accurately and realistically as could be exp Additionally, provision o possible to the perceived view of the participant. Despite these assumptions regarding the methodology, the extent to which retrospective reports on in-flight thinking can be considered accurate sources of data fluctuates depending on the researcher. Critics such as Yinger (1986 p.273) argue that at best, stimulated recall d event and at worst be entirely fabricated,” an r than ‘reflection-in-action.’ Such terms relate to whether the participant is descr s opposed to reflecting on what they remember thinking at the time. However, close 173 2006 POSTGRADUATE CONFERENCE examination of the transcripts of the interviews can provide useful clues to filter such data. For reflection on alistic; this an be done by: he • making sure that the participants understand the focus of the research clearly iewing neself. Keith (1988) and Yinger (1986) both stated that such an experience was likely to have an impact on erm memory which is ollis 1982; Paterson 2000), but observed by associated processes. As such, terviews carried out by stimulated recall using videotape potentially suggest that participants are i for the worker’ (Sträßer 2000 p.245). Additionally, criticisms of stimulated recall techniques that action, these clues would include verbalisations in present tense, whereas reflection on action would tend to be exhibited through past tense. The second point to be raised is associated with the stimuli, and refers to the effect of the videotaping itself on the initial task performance. Marland (1984) stated the importance of keeping the setting natur c • establishing a rapport with the participants prior to the data-collection sessions whereby making the participants feel at ease with the researcher • establishing a comfortable precedence with the presence of the video camera by recording t everyday goings on in the context • that the videotaping does not unduly disrupt the naturalistic setting. An additional complication with regards to the nature of a video stimulus deals with the effect of v o the authenticity of reports of perceived in-flight thinking. Availability of interactive thoughts for recall. As mentioned earlier in this paper, when performance becomes automated, these behaviours are often unavailable for verbalisation on a cognitive level. This is supported by Keith (1988 p.12) who stated that in studies of classroom teachers, tasks of an automated nature were ‘not available for verbal reports because [participants] are not aware of them’, therefore ‘it may well be impossible for teachers to describe their highly automated interactive decisions and the cues used to make them, even if they wanted to do so.’ Therefore, stimulated recall interviews need to be carried out no more than 48 hours following the event. The theory underlying this restriction is related to short and long-t beyond the call of this paper; however, the reader can consult Ericsson and Simon (1993) and Ericsson and Kintsch (1995) for examples of this work. The availability of interactive thoughts for recall is closely related to the idea that cognitive processes cannot be measured directly (Biggs & C in ‘responding not only to their memory of the viewed situation, but also to a set of cues supplied’ (Yinger 1986 p.269). Therefore, data can only be interpreted as a combination of recall and construction of thought, as the participant is not likely to be able to discern between the two (Yinger 1986 p.270). Such criticisms can be overcome by accumulating knowledge of the event prior to the main data collection in ‘order to present the “right” stimul 174 UNIVERSITY OF NEW ENGLAND propose that the participant responds primarily to the stimulus and secondarily to the thoughts that occurred ved ixed-methodological constructs (Corrigan 2001; Ethell & McMeniman 2000). Triangulation of data directly related to the nature of the ulus. As mentioned previously, there are many factors that influence the ability of the stimulus to provide • was the level of rapport between the interviewer and participant sufficient to enable verbalisation? at the time of the event (Yinger 1986) can be partially addressed by careful analysis of the discourses collected. Calderhead (1981) stated that it was unlikely that stimulated recall would be used as the sole data-collection instrument in a research design. Recent research methodologies using stimulated recall have invol m collection techniques (such as researcher field notes and observations, stimulated recall techniques and semi-structured interviews) have strengthened the validity of the data collected as being authentic representations of thought-in- (or -on) action. Evolution of a four-step stimulated recall procedure The preceding review of the literature on the use of stimulated recall using video, identified three points requiring careful attention when used to collect rich and authentic data. These are bias in the questioning, the nature of the stimulus, and the availability of interactive thoughts for recall. A fourth point not identified in the earlier review relates to the analysis of the data obtained. In the present study, where the focus is on mathematical cognition, the theoretical framework for analysis is the Structure of the observed Learning Outcome (SOLO) Model. To reduce incidents of bias in questioning, the use of probes such as ‘can you tell me what you were thinking here’ and ‘what thoughts were going through your mind’, were employed. As intended, the participants were then able to give responses which were from their point of view as opposed to what they thought the researcher wanted. This technique also ensured that the mathematical focus of the study remained hidden from the participants, allowing collection of data which could also determine whether the participants were overtly or covertly using mathematics in their solution strategy. As such, their responses in the stimulated recall interview sessions were varied and often did not contain references to mathematical concepts. An example of this relating to issues important to them, such as how much ‘easier it would have been to have the use of hydraulic tools’. Another point in achieving a rich and authentic source of data was stim sufficient cues to stimulate accurate recall of in-flight thinking by the participants. Of particular interest in this pilot study, was the solution to three main questions: • was the setup and choice of task appropriate? • was the video a suitable stimulus for the participant to enable verbalisation? 175 2006 POSTGRADUATE CONFERENCE The participants appeared to have little trouble deciphering the Job Sheet provided and the direction on the task. The direction given was that they had to complete the task as closely as they would if they were at work. The first two participants both commented that if they were on the job site, they would just copy an xisting downpipe, or that they had an angle that they used for all downpipes; rarely would they use a is so that you can save time tomorrow and just fit them straight to the gutter. This add the third participant, and who then could place this task in a real-l nal and contextualising element of task completion was an important, yet previ dy. The p rtant element the use of stimulated recall. The presence of the video camera in the TAFE workshop on the morning of n being carried ut apart from the class cohort, two of the three participants were anxious at beginning the task. Such ted once the participants began to concentrate on task execution. tation of various taping positions, the participant’s direct line of sight was ruled out due to It wa embarras a low profile within and around both the p ic leaving t omplete the task independently, and apart from the class cohort and researcher presence. e prescribed angle for fabrication. Therefore, for the third participant, additional context was added: Your boss has got a job in progress out of town. Job specifications have stated that the angle of the downpipes have to be set at 120°. Due to time and vehicle constraints, he has asked you to stay at the workshop today and make up the other three from the Job Sheet provided. This itional element of context made sense to ife setting. This motivatio ously unidentified, component in obtaining valid data and as such, was added in the Main Stu reparation of a video record which would provide an appropriate stimuli was another impo in the visit was a novelty initially, with the apprentices interested in seeing themselves on tape. This recording of the class as a whole proved highly important; it dispelled much of the anxiety felt by the individual performance being recorded. With only one day in which to carry out this data-collection for each of the three participants, the lack of comfort with the presence of the video camera was apparent when the participant was videoed in the afternoon individually performing the given task. Despite the task executio o anxiety soon dissipa Following experimen the dynamic nature of the task. A camera angle directly in front of the participant from the chest down was considered the most appropriate. The camera was mounted on a tripod in such a way as to show all the required workspace. The participants were asked to look through the camera so that they knew the limitations of the field of view. s important show respect for the participants in their environment. To minimise the possibility of causing sment or anxiety to the participants, the researcher maintained art ipants and the cohort of the class during the data-gathering phase. Anxiety was further reduced by he participant to c 176 UNIVERSITY OF NEW ENGLAND The inter prior to th e feelings of anxiety mentioned earlier. Highly influential to thi flow of s e participant was a difficult balance of probes and general conversation, as ell as interest from the researcher in the vocation of the participants and the immediate task performed. authentic data was the availability of interactive thoughts for call. The open-ended interviews using the video as the stimulus was a successful approach in providing a level ed descr ely relate ng commentary of what they saw on the video. These actions cannot be discounted entirely as they are onnected in some way to the participants’ thoughts due to the automated nature of the task (see Fitts 1964). is study, such verbalisations are important for understanding cognitive functioning and decision-making. Such data, whilst not the primary aim of a stimulated recall technique, ‘will better provide understanding of the information processing skills and styles that characterise … performance’ (Marland 1984 p.158). Such verbalisations were not discounted and enriched the data for analysis using the proposed cognitive framework. Gathering data useful in gaining insight into the participants’ mathematical understanding relating to the task proved to be the most difficult factor in this study. As alluded to earlier, the participants were not aware that the focus of the research was on mathematical knowledge and skills. Responses to the stimulated recall interviews by all participants gave important information regarding and explanations of actual practice, providing some information of the participants’ mathematical understandings related to this task. However, there were particular sections during the fabrication of the downpipe which the researcher deemed useful to enable mathematical coding; one example being the need for the second bend to be the same as the first so that the two vertical arms of the downpipe would be parallel. The fourth point is the analysis of the data collected using a cognitive framework. The cognitive framework utilised in the present study was the SOLO Model (Biggs & Collis 1982). This model provides a structure that can be used to measure human knowledge, by identifying ‘how well’ (qualitative) something has been learned, as opposed to the more traditional approach of measuring ‘how much’ (quantitative). SOLO can viewer/participant rapport was highly important in the interviews. Firstly, this relationship, nurtured e task execution, was important to overcom s end were assurances of confidentiality of performance. Secondly, verbalisation of thoughts and a free uch information from th w The third point necessary in assuring rich and re of verbalisation of thinking or explanations about the task for these participants. All those interview ibed something about their thinking processes, but most of the content of the responses was clos d to explanations of why certain actions were taken. In many cases, participants gave a runni c Indeed, despite the ‘pure’ vision of a stimulated recall technique assuming that subjects recall only what they were thinking at the time of the event by viewing a stimulus, more liberal stimulated recall researchers ‘adopt the position that what people think influences directly what they do’ (Marland 1984 p.158). As such, these apprentices’ explanations or reflection-on-action (as opposed to reflection-in-action) could have been a direct influence on their performance at the time. In terms of a stimulated recall methodology for th 177 2006 POSTGRADUATE CONFERENCE provide a system by which a response can be classified not only in terms of how much content knowledge a regarding a particular concept, but also the depth of knowledge possessed. The scope of this sing the SOLO Model as the theoretical framework, provides a mechanism by which a k can be made between the data collected and mathematical cognition. With lengthy practical tasks tages were points in the performance where the participants made choices about the next step, but were not d angle between the first and second bends 5. hese indicators identified important markers of mathematical elements of the task. Such markers allowed points that were missed in a semi-structured interview on completion of the stimulated call interview. This enabled data to be collected containing the participant’s point of view without response dure to enhance the quality of data onceptual s’. Such indicators of the mathematical components of the Mechanical recording (in this case video) of the task in accordance with stimulated recall guidelines person has paper does not lend itself to a discussion of the model in depth, however the reader is directed to work by Biggs and Collis (1991), Pegg (e.g. 1992), and Inglis and Pegg (2003) for further information. Discourse analysis u lin underpinning the data collection, it was necessary to identify Key Performance Stages of the task. These s in themselves in the present case, inherently mathematical. These Key Performance Stages were: 1. The set-up of the task prior to initiating fabrication, i.e., the scale drawings required to establish actual measurements for the downpipe 2. Making the first bend 3. Establishing the offset, or measurement an 4. Making the second bend Underlying knowledge of the job as a whole, i.e., possible cause and effect scenarios. T the researcher to identify whether the participant had mentioned these in the interview. If not, the researcher could ask about the re effects, whilst the semi-structured interview added insight regarding mathematical cognitions. Discussion he Pilot Study resulted in the design of a specific data-gathering proceT collected. The following procedure potentially provided avenues by which to gather data from the participants’ points of view, whilst also to providing data rich enough to inform a predetermined c analysis. Specifically, this four-step stimulated recall procedure consists of: • Derivation of ‘Key Performance Stage task performance need to be pre-determined by some means. • (Marland 1984). 178 UNIVERSITY OF NEW ENGLAND • Collection of verbal data comprising in-flight thinking or explanations of action by using a stimulated recall interview. At this stage the researcher needs to note whether the participant has addressed e participant has not addressed in the Key stimulated recall that the technique does offer fruitful means of exploring cognitive concomitants of ppropriate for ust be set up carefully in order to assure the provision of appropriate cues to stimulate recall. all points in the Key Performance Stages. • A semi-structured interview based on what th Performance Stages. Although questions regarding internal validity of this method provide the main barriers to its use, the authors concur with Marland (1984 p.164) who stated: In spite of … problems and limitations, there is a generally held view among those who have used human behaviour in the classroom and other task settings, and of developing new insights into important enterprises … Initial findings based on the use of this proposed stimulated recall protocol in the setting described were romising. The four-step stimulated recall procedure produced rich qualitative sources of data ap analysis using a cognitive framework, whilst maintaining the authenticity of the participant’s point of view. Overall, the stimulated recall method provides a tool to gain insight into workers’ mathematical cognitions, but m 179 2006 POSTGRADUATE CONFERENCE REFERENCES essot, A. 2000, 'Geometry at work: Examples from the building industry', in Education for iggs, J. & Collis, K. 1982, Evaluating the Quality of Learning: The SOLO Taxonomy, Academic iggs, J. & Collis, K. 1991, 'Multimodal learning and the quality of intelligent behaviour', in loom, B. (1956). Taxonomy of Educational Objectives: The Classification of Educational Goals. ecall: A method for research on teaching', British Journal of ducational Psychology, vol.51, pp.211-217. ematics in the streets and in the schools', ritish Journal of Psychology, vol.3, pp.21-29. larke, D. & Helme, S. 1993, 'Context as construct', Proceedings of the 16th Annual Conference of ornford, I. & Beven, F. 1999, 'Workplace Learning: Differential learning needs of novice and more inz, pp.68-77. berhard, M. 2000, 'Forms of mathematical knowledge relating to measurement in vicational training ricsson, K. & Simon, H. 1980, 'Verbal reports as data', Psychological Review, vol.87, no.3, pp.215- B Mathematics in the Workplace, eds A. Bessot & J. Ridgeway, Kluwer, Netherlands, pp.143-158. B Press, New York. B Intelligence: Reconceptualization and Measurement, ed H. Rowe, ACER, Melbourne, pp.56-76. B David McKay, New York. Calderhead, J. 1981, 'Stimulated R E Carraher, T., Carraher, D. & Schliemann, A. 1985, 'Math B Chifley Research Centre. 2001, Knowledge Nation Taskforce Report, Chifley Research Centre. C the Mathematics Education Research Group of Australasia, Queensland. C experienced workers', Australian and New Zealand Journal of Vocational Education Research, vol.72, pp.25-54. Corrigan, G. 2001, Conceptual Development, Investigative Skills and Decisions about Processes. Unpublished PhD, University of Sydney, Sydney. de Aguero, M. 2003, 'Studying everyday mathematical working', in Proceedings of the 10th International Conference on Adults Learning Mathematics, Strobl, Austria, eds J. Maasz & W. Schloeglmann, Universitätsverlag Rudolf Trauner, L Dickson, J., McLennan, J. & Omodei, M. 2000, 'Effects of concurrent verbalisation on a time-critical, dynamic decision-making task', Journal of General Psychology, vol.127, no.2, pp.217-222. E for the building industry', in Education for Mathematics in the Workplace, eds A. Bessot & J. Ridgeway, Kluwer, Netherlands, pp.37-52. Ericsson, K. & Kintsch, W. 1995, 'Long-term working memory', Psychological Review, vol.102, no.2, p.21. E 250. 180 UNIVERSITY OF NEW ENGLAND Ericsson, K. & Simon, H. 1993, Protocol Analysis: Verbal Reports as Data, MIT Press, Cambridge, MA. Ethell, R. & McMeniman, M. 2000, 'Unlocking the knowledge in action of an expert practitioner', Journal of Teacher Education, vol.51, no.2, pp.87-97. Fioriti, G. & Gorgorio, N. 2001, 'Geometry at work: Or Pythagoras will never fail you!' Proceedings of the 25th Annual Conference of the International Group for Psychology of Mathematics Education, ed itts, P.1964, 'Perceptual-motor skills learning', in Categories of Human Learning, ed A. Melton, oyster, J. 1990, 'Beyond the mathematics classroom: Numeracy on the job', in Being Numerate: hat counts?, ed S. Willis, ACER, Melbourne. Hahn, C. 2000, 'Teaching mathematics to shop-assistant apprentices exploring the content of didactical situations', in Education for Mathematics in the Workplace, eds A. Bessot & J. Ridgeway, Kluwer, Netherlands. Hogan, J. & Morony, W. 2000, 'Classroom teachers doing research in the workplace', in Education for Mathematics in the Workplace, eds A. Bessot & J. Ridgeway, Kluwer, Netherlands. Inglis, M. & Pegg, J. 2003, 'A possible cognitive framework able to mend the school-vocational mathematical divide', Proceedings of the 10th International Conference on Adults Learning Mathematics, eds J. Maasz & W. Schloglmann, Universitätsverlag Rudolf Trauner, Linz. Kagan, N. & Kagan, H. 1991, 'Interpersonal process recall', in Practical Guide to Using Video in the Behavioural Sciences, ed P.W. Dowrick, Wiley, New York, pp.221-230. Keith, M. 1988, 'Stimulated recall and teachers' thought processes: A critical review of the methodology and an alternative perspective', paper presented at the 17th Annual Meeting of the Mid-South Educational Research Association, Louisville, Kentucky. Lave, J., Murtaugh, M. & de la Rocha, O. 1984, 'The dialectic of arithmetic in grocery shopping', in Everyday Cognition: Its Development and Social Context, eds B. Rogoff & J. Lave, Harvard University Press, Cambridge, MA, pp.66-94. Lee, A. 2002, 'Promoting quality school physical education: Exploring the root of the problem', Research Quarterly in Education and Sport, vol.73, no.2, pp.118-124. Marland, P.1984, 'Stimulated recall from video: Its use in research on the thought processes of classroom participants', in Video in Higher Education, ed O. Zuber-Skerritt, Nichols Publishing Company, New York, pp.156-165. Masingila, J. 1992, 'Mathematics practice in carpet laying', Proceedings of the 16th Psychology of Mathematics Education Conference, vol. 2, pp.80-87, eds W. Geeslin & K. Graham, University of New Hampshire, USA. Mayer Committee. 1992, Employment-related Competencies: A Proposal for Consultation, Mayer Committee, Melbourne. M.V.D. Heuval-Panhuizen, Freudenthal Institute, Netherlands. F Academic Press, New York. F W 181 2006 POSTGRADUATE CONFERENCE Milroy, W. 1992, 'An ethnographic study of the mathematical ideas of carpenters', Journal for Research in Mathematics Education, vol.5. National Centre for Vocational Education Research. 2001, Review of research: Literacy and ining. Retrieved 14 January 2002 visibility of meanings: Modelling and the mathematics of banking', ournal of Computers for Mathematical Learning, vol.1, no.1, pp.3-31. les, C. & Pozzi, S. 2000, 'Working knowledge: Mathematics in use', in Education for atical knowledge developed at work: The cognition of ractice versus the contribution of schooling', Cognition and Instruction, vol.6, no.3, pp.185-221. ee, A. 1997, 'Development of an instrument to access cognitive processes in Research Quarterly in Education and Sport, vol.68, no.2, pp.152-160. pulsory Education and Training, vol. 1, pp.55-63, Workplace, eds A. Bessot & J. of experienced draughtsmen', Proceedings of the 16th Psychology of Mathematics Education numeracy in vocational education and tra Noss, R. & Hoyles, C. 1996, 'The International J Noss, R., Hoy Mathematics in the Workplace, eds A. Bessot & J. Ridgeway, Klewer, Netherlands. pp.17-35. Parker, J. & Pittney, W. 2003, 'Are content and competencies enough? Strategies to encourage reflective practice among athletic-training students', The Journal of Physical Education, Recreation and Dance, vol.74, no.2, pp.46-50. Paterson, D. 2000, Teaching in Inclusive Classrooms in Secondary Schools: A Study of Teachers' Inflight Thinking, Unpublished PhD, University of Alberta. Pegg, J. 1992, 'Assessing students' understanding at the primary and secondary level in the mathematical sciences', in Reshaping Assessment Practices: Assessment in the Mathematical Sciences Under Challenge, eds M. Stephens & J. Izard, ACER, Melbourne, pp.368-385. Schliemann, A. & Acioly, N. 1989, 'Mathem p Solomon, M., & L physical education classes', Stough, L. & Palmer, D. 2003, 'Special thinking in special settings: A qualitative study of expert special educators', Journal of Special Education, vol.36, no.4, pp.206-222. Sträßer, R. 1996, 'Mathematics for Work: a diadactical perspective', 8th International Congress on Mathematical Education: Selected Lectures, eds C. Alsina, J. M. Alvarez, B. Hodgson, C. Laborde & A. Pérez, S.A.E.M. 'THALES', Seville, pp.427-441. Sträßer, R. 1999, 'On the disappearance of mathematics from societal perception', Proceedings of the 7th International Conference on Post-com Griffith University, Gold Coast, Queensland. Sträßer, R. 2000, 'Conclusion', in Education for Mathematics in the Ridgeway, Kluwer, Netherlands, pp.241-246. Sträßer, R. 2003, 'Mathematics and Work: Adults and Artefacts', Proceedings of the 10th International Conference on Adults Learning Mathematics, Strobl, Austria, eds J. Maasz & W. Schloeglmann, Universitätsverlag Rudolf Trauner, Linz, pp.30-37. Sträßer, R. & Bromme, R. 1992, 'Description of solids in technical drawing: Results from interviews 182 UNIVERSITY OF NEW ENGLAND Conference, eds W. Geeslin & K. Graham, vol. 3, pp.43-50, University of New Hampshire, USA, pp.43-50. van hematics for learners of low abilities', in Education for Mathematics in the Workplace, eds A. Bessot & J. Ridgeway, Kluwer, Netherlands, pp.87-100. Wedege, T. 2002, 'Mathematics - that's what I can't do': People's affective and social relationship with White, R , Falmer Press, London. Ying ological critique on research n interactive teaching', Teacher and Teacher Education, vol.2, no.3, pp.263-282. der Zwart, P.2000, 'Working mat mathematics, Literacy and Numeracy Studies, vol.11, no2, pp.63-78. . & Gunstone, R. 1992, Probing Understanding er, R. 1986, 'Examining thought in action: A theoretical and method o 183