2006 POSTGRADUATE CONFERENCE Searching for a methodology to bridge the gap between ideas and practice: An investigation of the sensorimotor mode of learning John Haynes, Dr Judith Miller and Professor John Pegg Abstract The processes underpinning a specific ‘idea’, namely, how the assessment of the quality of movement within the sensorimotor mode of learning can be undertaken within a Structure of Observed Learning Outcomes (SOLO) framework, is described. The sensorimotor skill under examination is the ‘forward roll’ in gymnastics. Firstly, a review of the pertinent literature was undertaken. This involved a search for the relevant fields of study and reporting the trends and findings, as well as the identification of a conceptual framework. Secondly, data were collected, in the form of digital videotapes, for three cohorts: children, young adults and older adults (N = 117) as they performed the forward roll. These data were analysed, which resulted in the development of a new framework, for the purpose of measuring the quality of the sensorimotor skill under investigation. Thirdly, the underlying construct of the new framework was verified by applying the Australian Council for Educational Research ‘s (ACER) Quest statistical technique, commonly known as Rasch. Finally, using the underpinning principles established during data analysis, a re-examination of the data from a SOLO perspective is currently being undertaken. Introduction The entire purpose of the human brain is to produce movement. Movement is the only way we have of interacting with the world. All sensory and cognitive processes may be viewed as inputs that determine future motor outputs. Wolpert et al. (2001, p. 478) This study is concerned with the way people move. The initial question is: ‘How can one assess the quality of movement, within the sensorimotor mode of learning, for individuals, across the lifespan?’ The purpose of this paper is to inform the reader about the work in progress aimed at attaining an answer to this question within a Structure of Observed Learning (SOLO) model of learning. As a preliminary step in this investigation, an initial framework illustrating the ‘research process’ was envisaged. This framework is outlined in Figure 1. 130 UNIVERSITY OF NEW ENGLAND Figure 1: Research Guide By following the direction of the arrows provided in Figure 1, the steps used to guide the research are outlined. A preliminary guide is useful, because it gives both direction and form to the investigative process. In addition, a starting point emerges. The search for information and knowledge pertaining to the proposed pronged approach, delineated in Step 1 in Figure 1. Reference lists investigation commenced with a three- of several well known, and ‘apparently’ various combinations of these, to name but a few, were ‘fed into the ether’. The ird source of information comprised the advice and knowledge offered by the two persons given over to the task of supervising this thesis, whose advice was ‘write something!’ relevant (to the first author) texts were scrutinised. From these texts, lists of references (perceived to be relevant to the question) were constructed. Once the texts were procured, information applicable to the research theme was gathered. This step, whilst selective, was in the latter part of the research endeavour, a random process given the plethora of information and knowledge pertaining to the chosen theme. Secondly, the Internet, via a variety of search engines, was scrutinized. For example, words such as, outcomes, learning, measurement, and th 131 2006 POSTGRADUATE CONFERENCE From this intensive process the issue related first topic of the investigation evolved. The quest was to assemble information pertaining to learning theories in general, learning theories associated with human movement and measurement. Background information related to learning theory was necessary to gain an understanding of, and knowledge about, what ‘was’ and compare and contrast it with the present ‘state of knowledge’, i.e. what is now! Where to begin? A constructivist approach to learning emerged as a starting oint, and what better way to start than with the apparent ‘father’ of constructivism, Jean Piaget. plies that knowledge does not exist without the individual attaching ry motor aspects of development, but did not pursue an vestigation of its role in the development of the individual. ng the ensorimotor organisational processes, thus suggesting, the child developed sensorimotor records resulting e neo-Piagetian model lopment. It is noteworthy that the model’s foundational mode, i.e. the ensorimotor mode is thought to be available across an individual’s lifespan. Their SOLO model comprises a p Preliminary Preparation: Learning Theories Piaget (1952) considered that sensorimotor skills could be observed from a developmental perspective. The assertion was that the sensorimotor stage of learning was the precursor to the cognitive development of an individual (child). Piaget (1952) promulgated a constructivist’s view of learning, i.e. a view that purports a learner’s view of the world, is founded on the concept that the individual constructs meaning about various events and experiences. This view im meaning to that knowledge. Both Dewey (1938) and Vygotsky (1997) supported the concept of constructivism. Dewey (1938) contributed to this approach by providing input regarding the importance of the actual learning environment in the learning process. Vygotsky (1997) considered that past learning was significant for future learning. Both authors recognised the importance of the senso in Piagetian theory was based on the concept that the learner ‘passed through’ a number of stages on the way to becoming a fully developed individual. There have been several more recent, in chronological terms, stage theories. For example, Case (1985) provided a model of learning, which included a sensorimotor stage. Fischer (1980) implied that Piaget (1952) placed too much emphasis on the maturational factors affecting the learning process to the detriment of the apparent importance of the environment. Mounoud (1986) considered the importance of genetic inheritance in the development of the individual, includi s from interactions with the environment. Generally learning theories equate improvements in learning with age. However, th proposed by Biggs and Collis (1980) contended that the age at which an individual (child) attained a certain stage was not necessarily a true indicator for that stage. Biggs and Collis’ (1980) SOLO model highlighted four possible paths of deve s number of modes of learning that are hierarchal. 132 UNIVERSITY OF NEW ENGLAND The SOLO modes of learning are termed: the sensorimotor, ikonic, concrete symbolic, formal and post rmal. Biggs and Collis (1980) identified differences in performances within these modes and these es, especially in the reas of mathematics and science education, but have not been explored for the sensorimotor mode. that motor development occurs as a result of an e (1970), Kephart (1971) and Willingham (1998). the idea that the human body is comprised of a elf-organising systems, which respond to environmental task demands to produce a movement. d significance amongst researchers including Ulrich, Roberton, Langendorfer and for the explanation of movement model, but used different terminology to describe the process. Vereijken (1991) proposed a three-stage fo differences were termed ‘levels’. Levels demonstrate an hierarchy in the learning of a task from an incomplete to an expert performance. The levels are termed, unistructural (U), multistructural (M) and relational (R). Biggs & Collis (1980) referred to the progression of an individual, through these levels as a ’learning cycle’. Cycles have been identified and investigated within a number of mod a Preliminary Preparation: Motor Learning Perspectives Preliminary preparation, whilst keeping in mind the ‘big picture’ associated with learning theories, required an examination of motor learning perspectives. What was the literature saying about theoretical perspectives pertaining to motor learning? Firstly, the maturationist’s perspective, espouses the view unfolding of a ‘ master plan’ existing within the genetic structure of the individual. This theory was developed from the work of Hall (1904), Gesell (1928) and McGraw (1935), but was later modified to take the environment into account as a factor in motor development. Secondly, there have been a number of cognitive theories relating to the attainment of motor skills. These theories professed that motor learning resulted from cognitive events, and that the brain played a major role in an individual’s motor development and movement choices. Notables amongst the realm of cognitivists ar Adams (1971) Schmidt (1991), Connolly Thirdly, the concept of Dynamic Systems Theory (DST) was supported by a number of researchers, including Bernstein (1967), Kelso (1982) and Kugler and Turvey (1986), as a way of explaining movement outcomes. The basis of this approach to motor development involved number of s This theory has gaine Gallahue (personal communications 2003) as an acceptable theory responses. How do these learning theories translate into practicality? Fitts and Posner (1967) addressed questions pertaining to motor skill learning through the use of a three phase model. In contrast, both Adams (1971) and Gentile (1972) proposed two stage models, which were similar in concept to the Fitts and Posner (1967) 133 2006 POSTGRADUATE CONFERENCE model, which implied that a learner attempted to manage the dynamics of the movement to solve a movement problem. Differing from the previous models, Graham et al. (1998) took a ’teaching approach’, developing a four stage model involving observable characteristics of generic levels of skill proficiency. Further exploration of the literature revealed that some researchers explored the psychomotor domain of the Bloom et al’s (1956) Taxonomy as a way of explaining movement. Bloom et al. (1956) did not provide details lating to this domain, however, descriptive models were devised by others such as Dave (1975), Corbin e relating to learning theories and how motor learning and skill cquisition takes place; attention was now refocused on the issues surrounding measurement. Those n the ore. This ethod of scoring has been accepted as the standard for competition performances in gymnastics ederation 2002). What emerged from an analysis of the literature, however, was that a number of other perspectives have been based roll, based upon biomechanical principles as described by George (198 model of prof levels were re (1976), Harrow (1972) and Simpson (1972). Preliminary Preparation: Measuring Movement Having attained a good working knowledg a procedures employed to measure the sensorimotor mode movement, i.e. human movement, were selected for further investigation. Issues surrounding the origin and history of testing and measurement, as well as assessment and test theory were also examined. However, space precludes a full discussion of these issues. Notwithstanding, the main emphasis, in this paper is placed on the measurement of movement. One reason for choosing the forward roll in gymnastics as the main object for observation was based on the author’s prior knowledge and experience as a qualified gymnastics coach and Physical Education Specialist teacher. The second, and probably more valid reason, was that it is one of the few motor skills that has been validated in terms of its developmental characteristics (Williams 1980). Her work was based o developmental sequences of Roberton & Halverson (1984). Measurement perspectives: The Forward Roll The process employed by gymnastics judges to score performances was familiar to the principal researcher. The method consists of deducting points for performance errors, from a hypothetical maximum sc m (Australian Gymnastics F used to assess the performance of the forward roll (and other motor skills). Firstly, there was a skill- description of the ideal forward 0). Secondly, a developmental stage model was proposed by Gallahue and Ozmun (1998). This latter was founded on three stages of development, the initial, elementary and mature. Thirdly, a level iciency model, based on four levels of ability, was reported by Graham et al. (1998). The 134 UNIVERSITY OF NEW ENGLAND termed the precontrol, control, utilisation and proficiency. Lastly, the phases of the forward roll, were derived from the component approach of analysis of motor skills initiated by Roberton and Halverson (1977). Preliminary Preparation: Research Questions and Initial Framework nt of the basic foundations of the task, i.e. the ideas of educational theorists, motor re established: • Research Question 1: Are the observable components for the forward roll the same for children, young adults and older adults? • Research Question 2: Is the SOLO model an appropriate tool for assessing subjects’ performances, representing a diverse developmental range, as they demonstrate competency in the forward roll? • Research Question 3: Can SOLO provide an assessment framework for the forward roll? The search for an appropriate initial conceptual framework on which to base the research ended when Knudson and Morrison’s (1997) model was found to be both appropriate and useful for the qualitative analysis of human movement. Initial Framework The Knudson and Morrison (1997) model includes a task list, namely, preparation, observation, evaluation/diagnosis, and intervention. Furthermore, it is a comprehensive, integrated model that lists issues associated with each of these four tasks. All of the model’s tasks were retained, i.e. ‘Intervention’. This was eemed necessary as it relates to the applicability to the sensorimotor mode of learning. Figure 3 shows the initial framework use With the establishme learning perspectives, and how the measurement of movement, specifically for the forward roll was undertaken, the next step was to formulate the research questions, and gather and analyse data. Research Questions Prior to data gathering the research questions we d d to scaffold the research. 135 2006 POSTGRADUATE CONFERENCE Figure 3: A Comprehensive, Integrated Model of Qualitative Analysis This model shows four phases. Undertaking th Preparation as outlined in the research guide (Figure 1), into Step 2 the ese moved the research forward from Step 1, the Preliminary Pilot Study, which aligns with the odel. chnical sues pertaining to data collection, namely video recording procedures. The following paragraphs describe, these issues. ‘normally’ perform such skills (Miller 2001). The purpose of trialling data research skills at Beachside was threefold. Firstly, it was to ensure the eventual target group understood the questions (Borg & Gall 1989, p. 445), secondly to avoid exposing that group to second stage of the Knudson and Morrison (1997) m Pilot Study: Preparation Step Step 2, of the research involved a number of issues including, having knowledge of the elements of the forward roll, the ability to identify the critical features, as well as identifying and solving a number of te is in attenuated form Ethics approval from UNE was obtained prior to data collection and trailing data gathering techniques. ‘Beachside’8 was the site chosen to practice these techniques, such as videotaping, interviewing and questioning. This site represented and ‘ecologically valid’ environment, i.e., one where children would 8 Beachside is the pseudonym used for a gymnastics club in the Coffs Harbour area of NSW 136 UNIVERSITY OF NEW ENGLAND the actual questions, and thirdly, to trial camera positions whilst the subjects were performing the forward roll. Camera angles and positioning were largely exploratory. Interviewing participants (supervised by an experienced researcher JM9) permitted information to be gathered concerning a number of issues, including what each subject thought about just prior to their performance, what they considered to be the characteristics of a forward roll, plus what might be the characteristics of a well executed forward roll. This data has the potential to assist with the exploration of the modes of the rvation and Data Gathering tep 3 involved data collection from three cohorts from a broad age range. Cohort 1, comprised children as transcribed and emerging themes noted (Cohen & Manion 1992). dditional information, including the name, age and time-span since each subject had previously performed a tape data for Cohort 1 (children) occurred, on a number of occasions, during the time SOLO model. Obse Step 3 of the research guide includes observation of the subjects and data gathering. The two primary data collection sites included the rural township of Whitestone•, population 9,000) where data pertaining to children were collected. The rural city of Hilltop• was the data site for young adults and older adults. Data Gathering S whose age was less than 17 years. Cohort 2 included young adults, aged from 17 to 22 years. Cohort 3 comprised older adults whose age was over 23 years. Performances of the forward roll were recorded in the first instance using a hand-held Panasonic digital video camera. All the videotaped materials were converted to compact disc (CD), MPEG which permitted individual ‘frame by frame’ observation of the subjects, with the added benefit of having a time reference calibrated in tenths of a second. In addition to videotape recordings, audiotapes of interviews of selected young adults and older adults were conducted. The interview data w A forward roll, was gathered. Collection of the video period when the subjects were involved in their normal gymnastics program at the Whitestone Gymnastics Club. 9 JM (PhD Physical Education) • Not real name 137 2006 POSTGRADUATE CONFERENCE Cohort 2 (young adults) consisted of three sub-groups. One sub-group included University students enrolled athering data from older adults employed the same procedures as for the young adult cohort. Data was collected from older adults, termed Cohort 3. This took place over several days, and involved a number of eparate sub-groups. Each sub-group from this Cohort was identified. One sub-group included those duals representing each cohort, one from each level of quality ere analysed in depth (n = 9), using the four assessment perspectives. Namely, gymnastics approach, in the first semester of their first-year of study. The second sub-group included first-year University students enrolled in the second semester. The third sub-group included students enrolled in an elective Physical Education unit of study. All subjects from these sub-groups were enrolled in a Bachelor of Education/Bachelor of Teaching degree at the University. It was considered, that these three sub-groups would be representative of the range of movement experiences for the young adult cohort. G s subjects who attended a residential school during one particular year. The second sub-group consisted of residential school participants from the following year. Students attending residential classes usually reside outside the immediate Hilltop• area and originate from locations both within Australia and occasionally overseas. These attending students were provided with brief outlines of the proposed study and subsequently invited to participate. Data Analysis Using George’s (1980), gymnastics approach to assessment of the forward roll, all subjects (N = 117) were analysed, and subsequently divided into three groups representing low, medium and high quality performances. Case Studies From each of the three groups, three indivi w Gallahue and Ozmun’s (1998) stages of development method, Graham et al’s. (1998) levels of proficiency, and Roberton and Halverson’s (1977) phases perspectives. Following this in-depth analysis some shortcomings were noted when the criteria for each perspective were applied across the different cohorts. Table 1 summarises the findings of three of the case study examples. 138 UNIVERSITY OF NEW ENGLAND The information contained in Table 1 shows that, for each perspective, there was a discrepancy between the measurement criteria and the performance of the forward roll. For example, for the ‘stages of development’ perspective subjects were judged to be ‘between’ stages or the instrument did not cater for a particular individual. A similar scenario existed for the levels of proficiency model. The ‘phases’ model shows individuals operating at different positions within a phase. l for determining the quality of an individual’s performance of the forward roll. These cues were rmed, indicators and descriptors. The indicators are the position of the various body parts (components) entiate the range of movement quality. Four indicators emerged from the data analysis for the beginning sequence, with two to four descriptors for each indicator. There are two indicators for the bridging sequence, which have three and four descriptors respectively. For t See Figure 4 for an example of a descriptor and indicators for the eginning sequence. In order to overcome the perceived shortcomings of the various perspectives, further analysis of all subjects was undertaken. This process led to the emergence of three hypothesised sequences within the rolling action, namely, beginning, bridging and end. A refereed paper providing more details of these processes has been accepted for publication (Haynes & Miller 2006). In addition, observational cues were developed, which were usefu te within each sequence. The descriptors are the subdivisions within each indicator used to portray and differ he end sequence there are three indicators with two to four descriptors. ‘hand position’ for the b 139 2006 POSTGRADUATE CONFERENCE The following Pictures 6.1(i), (ii) and (iii) in Figure 4 illustrate the three descriptors for indicator 1 of the beginning sequence of the forward roll. Figure 4: Hand Position Indicator Descriptors Figure 4 shows the first indicator, namely, the position of the hands on the surface, i.e. their location lateral to the sagittal plane. The sagittal plane is an imaginary line lengthwise through the body running from front to back. The body is therefore divided into left and right sides (Tortora & Anagnostakos 1990, p. 9). Pre Step 5 in the research includes an exposé of the possible findings that emerged from the data analysis. To this e d from the data analysis, which illustrated the mechanism by which ork used to determine the quality of movement for the forward liminary Findings nd, firstly, a new framework emerge an investigation of the quality of movement could be obtained across all cohorts. By following the steps outlined in this framework, one is able to determine the quality of the skill under scrutiny, namely, the forward roll. Secondly, a new and innovative way of presenting the output from Quest (Adams & Khoo 1993) was devised, using pictorial data. The presentation of data using pictures, being substituted for numbers, allows the reader to be able to see what the performance of an individual at a pre-determined level of quality looks like. However, when the pictures are substituted none of the veracity of the statistical information is compromised. This means that the placements of the pictures on the statistical map appear in the same locations, as did the numbers. Figure 5 provides a description of the Framew roll. 140 UNIVERSITY OF NEW ENGLAND Figure 5: A Model for Assessing Movement Quality of the Forward Roll (Haynes et al.2005) Figure 5 presents an overview of the instrument derived from the qualitative analysis of the forward roll. The formation accompanying each arrowed link informs the process of analysis and provides a broad template r assessment. Working through the model, commencing with the three sequences of the forward roll, the beginning, bridging or end, by followin and then to the descriptors for each in fo g the arrows to the indicators sequence, one can begin to determine the quality of a performance. 141 2006 POSTGRADUATE CONFERENCE A more detailed account, of the Framework presented in Figure 5, and for the Rasch (1960) analysis, is brief, ‘Rasch measurements are particularly suited to investigations in the wide range of human sciences’ g the ason why people and specifically selected items behave as they do. ulty is independent of the set of persons used to calibrate the item. ). lore dditional aspects of group performance, such as differences between the qualities of movement for each ariate techniques. quality and ems. Persons and items are located on the map according to their movement quality and difficulty available elsewhere (Haynes et al. 2005). However, it should be noted that the Framework, which emerged from the data represents an assessment instrument, which is sufficiently fine-grained to be applied across all cohorts. Statistical Analysis: Rasch In (Bond & Fox 2001, p. 189), which according to these authors is the only technique generally available for constructing such measures. The software package used to calculate the Rasch (1960) scales is termed Quest, (Adams & Khoo 1993) which is useful for gaining an understanding of the processes underlyin re When a variable indicating a single particular construct has been identified, within a targeted population, the measurement of the subject’s ability is independent of the set of items that were administered, and the item diffic Snyder & Sheehan (1992, p. 88 ‘Movement quality’ is the underlying construct of the Framework presented in Figure 5. Output from Quest (Adams & Khoo 1993), using the partial credit form, which specifically incorporates the possibility of having differing numbers of steps for different items on the same test (Masters 1982) revealed that both items and persons were shown to behave in a predictable manner. It is realistic, therefore to suggest that the applicability of the new Framework is worthy of continued investigation within the field of movement studies. Thus, based on the acceptable levels of the fit statistics, the case estimates could be used to exp a cohort, using multiv Innovative Data Presentation The output from Quest (Adams & Khoo 1993) is normally portrayed in what are termed ‘maps’. The measurement unit of these maps is the logit, which is the common unit for both person movement it estimates, respectively. Logit information is displayed down the far left hand side of the map. Because it is a logit scale, i.e. an interval scale, the equal distances up and down that scale have equal value. Down the centre-left of the map ‘Xs’ represent the distribution of case (subject) estimates on the logit scale. Each ‘X’ represents the estimate for one subject, which signifies a 50% probability the subject will be able to achieve 142 UNIVERSITY OF NEW ENGLAND that item (descriptor) at the same position on the logit scale. Numerals on the right hand side of the map show step difficulties. Notwithstanding, the opportunity existed to present the map in pictorial format. This method allows the reader to quickly scrutinise the ‘form’, in this case, of the person performing the forward roll. Figure 6 presents an example of the pictorial format, for one indicator, namely, hand position and the rd roll. Figure 6: Item-person Fit Map igure 5 shows a segment of the item person fit map, generated using Quest (Adams & Khoo 1993). These The distance ’. For example, the picture at the bottom of igure 6 represents the lowest quality descriptor for the indicator of hand position, the middle picture better descriptors which are used to discriminate the quality of a performance, for the beginning sequence of the rwafo 3 0.08 0.06 0.04 0.02 2 4 2 1 0.02 0.02 min 2 0.08 min 4 0.08 0.06 2 0.08 0.06 0.0 0.0 0.08 0.06 0.04 zero 0.08 0.06 8 0.04 min 1 0.08 0.06 0.04 0.02 0.08 0.06 0.04 0.02 min 3 0.06 0.04 0.02 0.04 Below Scale F data are generally shown as a map, however, pictures are employed to replace numbers. In summary, the numerals shown in the left hand column of Figure 6 are termed logits. between each picture of a gymnast, illustrating hand position descriptors, can be calculated in terms of the number of logits between them. Numbers have replaced ‘Xs F 143 2006 POSTGRADUATE CONFERENCE quality and the top picture the highest quality performance (the bottom picture is actually below the logit he important implication is that the number of logits between each picture represents the degree of difficulty e middle position to the top position than it is to move from the lowest position to the middle position. all indicators, descriptors for each equence was presented at the International Council for Health, Physical Education, Recreation, Sport & ata. An w Framework, showed that a SOLO perspective of the forward roll is feasible. Thus far, s et al. an ‘idea’. duced some edifying results, and contains a number of features, which differentiate it from rior investigative studies. The use of three cohorts from differing age groups is a feature that has not been igure 5) that explains the pathways involved in the analysis of the forward roll is unique to this study. The ments has not been employed in research concerning the investigation of levels of quality of sorimotor scale, as generated by Quest, (Adams & Khoo 1993) but is included for illustrative purposes). T of moving from one descriptor for the hand position to the next. In this case it is much easier to move from th With reference to the above Figure the remainder of the pictorial data for s Dance (ICHPER-SD) (Haynes 2006). Future Findings: SOLO From a SOLO perspective, the descriptors also form the basis for coding the movement quality d examination of the SOLO cycles in the sensorimotor learning mode is currently under investigation, i.e., Step 6 in the research process. It is sufficient to say that, preliminary analysis using the information gleaned, from the use of the ne exemplary case studies have been forthcoming from which SOLO cycles have been tentatively examined. In addition, Rasch (1960) analysis, as well as intra- and inter-rater reliability checks have been conducted. Conclusion In searching for a methodology to bridge the gap between ideas and practice a Framework (Hayne 2005) emerged which is instrumental in this endeavour. A good mixture of the processes, introduced via the search guide, with the steps outlined in Knudson and Morrison’s (1997) initial framework, led to there creation of an artifact that answered the original question; the question that arose from This study has pro p attempted in earlier research concerning the quality of movements for the forward roll. The Framework (F use of Rasch (1960) modelling, via Quest (Adams & Khoo 1993) software, to analyse the quality of ovem performance for the forward roll. With further research and analysis a SOLO explanation of the sen mode of learning may be forthcoming. If the findings either confirm or deny a SOLO explanation for this mode, there is the potential to add information of a highly significant nature to the conceptual framework of that model. 144 UNIVERSITY OF NEW ENGLAND REFERENCES Adams, J. A. 1971, A closed loop theory of motor learning, Journal of Motor Behavior, vol.3, pp.111 49. - 1 Adams, R. J. & Khoo, S. 1993, Quest - The Interactive Test Analysis System, ACER, Hawthorne, ictoria. V Australian Gymnastics Federation 2002, A History of Australian Gymnastics 2002, Retrieved from http://www.gymnsw.org.au/history_australia.html Bernstein, B. 1967, The Co-Ordination and Regulation of Movements, Pergamon, Oxford UK. Biggs, J. & Collis, K. 1980, 'The solo taxonomy', Education News, vol.17, no.5, pp.19 - 23. The Cognitive Domain, McKay, New York. lbaum Associates, Mahwah New Jersey. ischer, K. W. 1980, 'A theory of cognitive development: The control and construction of hierarchies of skills', Psychological Review, vol.87, no.6, pp.477-531. Fitts, P. M. & Posner, M. I. 1967, Human Performance, Brooks-Cole, Belmont CA. Gallahue, D. L. & Ozmun, J. 1998, Understanding Motor Development Infants, Children, Adolescents, Adults, 4th edn, McGaw-Hill, Boston. Gentile, A. 1972, 'A working model of skill acquisition with application to teaching', Quest, vol.17, pp.3-23. Bloom, B. S., Englehart, M. D., Furst, E. J., Hill, W. H. & Krathwohl, D. 1956, Taxonomy of Educational Objectives: Bond, T. G. & Fox, C. M. 2001, Applying the Rasch Model; Fundamental Measurement in the Human Sciences, Lawrence Er Borg, W. R. & Gall, M. D. 1989, Educational Research: An Introduction, 5th edn, Longman, New York. Case, R. 1985, Intellectual Development, Birth to Adulthood, Academic Press. Cohen, L. & Manion, L. 1992, Research Methods in Education 3 ed., London: Routledge, Orlando Florida. Connolly, K. J. 1970, Mechanisms of Motor Skill Development, Academic Press, London. Corbin, C. B. 1976, Becoming Physically Educated in the Elementary School, Lea & Febiger, Philadelphia. Dave, R. H. (ed) 1975, Developing and Writing Behavioral Objectives, Educational Innovators Press, Tuscon AZ. Dewey, J. 1938, Experience and Education, Macmillan, New York. F 145 2006 POSTGRADUATE CONFERENCE George, G. 1980, Biomechanics of Women's Gymnastics, Prentice Hall, Englewood Cliffs, NJ. Gesell, A. 1928, Infancy and Human Growth, Macmillan, New York. Graham, G., Holt/Hale, S. A. & Parker, M. 1998, Children Moving: A Reflective Approach to Teaching Education, 4th edn, Mayfield Publishing Company, Mountain View, CA.. all, G. S. 1904, Adolescence, Vol. 1 & 2, Appleton-Century-Crofts, New York. 1972, A Taxonomy of the Psychomotor Domain, David McKay Co., New York. H Harrow, A. J. Haynes, J. 2006, Objective measurement using pictorial data, Proceedings of Fusion Down-under Recipes for Movement: Challenging perspectives and constructing alliances, ICHPER-SD, Wellington NZ. Haynes, J. & Miller, J. 2006, 'Differences and similarities between the performance of a fundamental skill in gymnastics for three cohorts: Children, young adults and older adults', Proceedings of Australian Association for Research in Education. HAY 06512, AARE, Adelaide SA. Haynes, J., Miller, J., Callingham, R. & Pegg, J. 2005, 'Applying item response modelling to confirm the underlying construct of a new process instrument in gymnastics', Refereed Paper at the Australian Association for Research in Education, www.aare.com.au HAY 05720, Parramatta. Kelso, J. A. S. (ed) 1982, Human Motor Behaviour: An Introduction, Erlbaum, Hillsdale, NJ. Kephart, N. C. 1971, A Slow Learner in the Classroom, Merrill, Columbus, OH. Knudson, D.V. & Morrison, C. S. 1997, Qualitative Analysis of Human Movement, Human Kinetics, Champaign, IL. Kugler, P.N. & Turvey, M.T. 1986, Information, Natural Law, and Self-Assembly of Rhythmic Movement, Erlbaum, Hillsdale, NJ. Masters, G.N. 1982, A rasch model for partial credit scoring. Psychometrica, vol.47, no.2, pp.149- McGraw, M.B. 1935, Growth: A Study of Johnny and Jimmy, Appleton-Century, New York. 174. Miller, J. 2001, The product and performance of the two handed sidearm strike for primary school- aged children: The interrelationship of coordination, age and gender, Unpublished Doctoral Thesis, New England, Armidale. Mounoud, P. 1986, Similarities between developmental sequences at different age periods, in Stage and Structure: Reopening the Debate, ed I. Levin, Ablex, Norwood NJ, pp. 40-58. Piaget, J. 1952, The Origins of Intelligence in Children, International Universities Press, New York. Rasch, G. 1960, Probabilistic Models for some Intelligence and Attainment Tests Expanded edition, University of Chicago Press, Chicago MI. 146 UNIVERSITY OF NEW ENGLAND Roberton, M. A. & Halverson, L. 1977, The developing child — his changing movement, in Physical Education for Children— A Focus on the Teaching Process, ed B. Logsdon, Lea & Febiger, Philadelphia. Roberton, M Teachers, L . A. & Halverson, L. 1984, Developing Children - Their Changing Movement A Guide for ea & Febiger, Philadelphia. Journal of Early Intervention, vol.16, no.1, pp.39-53. Tortora, G. J. & Anagnostakos, N. P. 1990, Principles of Anatomy and Physiology, Harper and Row, Schmidt, R. A. 1991, Motor Learning and Performance: From Principles to Practice, Human Kinetics Books, Champaign IL. Simpson, E. J. 1972, The Classification of Educational Objectives in the Psychomotor Domain, Gryphon House, Washington. Snyder, S. & Sheehan, R. 1992, The rasch measurement model: An introduction, New York. Vereijken, B. 1991, The Dynamics of Skill Acquisitio, Unpublished PhD, Free University Netherlands, Amsterdam. Vygotsky, L. S. 1997, The Collected Works of L S Vygotsky, Springer, New York. Williams, K. 1980, 'Developmental characteristics of a forward roll', Research Quarterly for Exercise and Sport, vol.51, no.4, p.703. Willingham, D. B. 1998, 'A neuropsychological theory of motor skill learning', Psychological Review, vol.105, no.3, pp.558- 587. Wolpert, D. M., Ghahramani, Z. & Flanagan, R. J. 2001, 'Perspectives and problems in motor learning', Review TRENDS in Cognitive Sciences, vol.5, no.11 November, pp.487-494. 147