Australian Association for Research in Education Annual Conference, Fremantle, Western Australia 22nd-25th November, 1993. Distinguishing between expert and non-expert problem-solving strategies in student teachers. Richard G. Berlach Department of Teaching & Curriculum Studies Edith Cowan University, Churchlands Campus Pearson St, Churchlands, Western Australia, 6018. Ph: (09)3838402 Fax: (09)3877095 ABSTRACT In an attempt to ascertain whether teacher education students differ in domain-specific problem-solving behaviour, 49 second year teacher education students (N=24 school-leavers; N=25 mature-age) were asked to read and respond to three classroom- oriented problem-solving vignettes. Criteria for inclusion into the category of expert or novice were established from available expert/novice and master-teacher literature. Responses were coded and verified using factorial and cluster analysis. Results indicated that expertise needed to be defined in terms other than merely experience. A cognitive processing model based on schemata differentiation was proposed to explain the difference in problem-solving style. Implications for pre-service Education courses are considered. Introduction Investigations conducted by De Groot (1965) and Chase and Simon (1973) with chess experts set the stage for an unprecedented foray into the realm of expert knowledge. Whereas early research tended to categorise expertise as general heuristic processing, more recent efforts have tended to see it as "domain-specific knowledge that is especially relevant to vocational education" (Glaser, 1986). Areas of concern have been as broad as medicine (Einhorn, 1974); language arts (Krabbe, 1989); mathematics (Leinhardt, 1989); business (Yekovich, Thompson, & Walker, 1991); genetic science (Kinnear & Simmons, 1990); and environmental issues (Tudor, 1992). Experts are typically distinguished from non-experts (novices) by performance criteria related to problem-solving tasks. Expertise in the field of general education has also received considerable attention, with expert pedagogues being identified in areas such as classroom management and instruction (Sabers, Cushing, & Berliner, 1991); mathematics (Ropo, 1987); physics (Chi, Feltovich, & Glaser, 1981; Mestre & Touger, 1988); social science (Gudmundsdottir & Shulman, 1987); and physical education (van-der-Mars, 1991). A search of the ERIC database for the years 1983-1993 revealed that expertise in general education has been investigated in numerous countries, within different contexts, with different age groups, using both or either gender, across a variety of curricula, with different personnel, and within a variety of school-based roles. It has only been in the last decade or so that expertise in teacher education has received serious attention. Reported classroom-based studies have tended to define pre-service individuals (students) as novices and in-service individuals (teachers) as experts (Clark, 1984; Stoddart, 1990; Vollmer, 1987). Put simply, the expert has been defined by status and experience. The danger with such a definition is that it tends to ignore inherent characteristics or traits, thereby causing one to conclude that (i) given sufficient time, anyone can become an expert and (ii) expertise is a purely developmental phenomenon and as such cannot be enhanced. The first of these conclusions seems to be counter-intuitive while the second doubtful on the basis of recent postulant (ie. emerging expert) studies (eg. Berliner, 1986; Clarridge, 1989; Sabers, Cushing & Berliner, 1991). What appears to be lacking in expertise research is an agreed upon definition which characterises an expert. The recognised work of Berliner (eg. 1986, 1987, 1988; Carter, Sabers, Cushing, Pinnegar, & Berliner, 1987) in developing criteria for differentiating between expert and novice practitioner has aided this cause in part, however, the question of definition is still very much a moot point. Shanteau (1988) exemplifies this by showing that an expert has been variously defined in the literature as one who has experience, competence, effectiveness, recognition, status, academic credentials, job titles or self-designation. Such a broad taxonomical approach makes it difficult to substantiate claims related to construct validity. One could be forgiven for asking the question, so what is an expert really? The position taken in this paper is that domain-specific difference between expert and novice problem-solver is qualitative rather than quantitative in nature; that is, experts do things faster, better and smarter than do novices, regardless of chronological age differences. This is by no means a novel idea. Numerous researchers have shown that students can possess the same qualities as do those who have been practitioners for a considerable period of time (Brehmer, 1980; Stallings, 1986; Welker, 1991). Results, however, are often confounded by failing to control for experience. The current research has attempted to control for this variable by studying students only and selecting subjects who are at the same point in their training. Expertise is thus considered here as a within- group rather than between-group phenomenon. Such an approach allows possible generic traits to surface while controlling for teaching experience. The research question, then, is twofold. First, do student teachers within a similar cohort differ regarding the degree of expertise they exhibit? Second, are mature-age students more expert than school-leavers? The design will emphasise what Cohen and Manion (1980) term a multi-method approach, that is, one which combines both qualitative as well as quantitative approaches. Method Sample Twenty four school-leavers and 25 mature-age (N=49) second year B.A. (Education) students from the Churchlands campus of Edith Cowan University comprised the sample. School-leavers were defined as those who entered university the year following the completion of Year 12 and mature-age students were those who had been away from formal education for a period of at least five years prior to entering university. Female students only were selected as too few males were undertaking the course to provide a representative sample. Subjects were taken from both the primary and early childhood education streams. Second year students were chosen for the following reasons: (i) the researcher did not teach this cohort, (ii) being halfway through their course, they had some knowledge of how classrooms operate, (iii) classroom experience was too limited for first years and third years were too busy to act as subjects. Participation was on a voluntary basis. Procedures The first task involved identifying the attributes associated with expertise. The principles advocated by Miles and Huberman (1984) and Strauss and Corbin (1990) for evaluating qualitative data from a grounded theory perspective were adopted. The first task related to what Strauss and Corbin call conceptualising the phenomena and consisted of a literature search for developing a set of defining attributes (concept indicators) for the classification "expert teacher". This was undertaken by scanning the ERIC database for the years 1983-1993 using the descriptors "EXPERT with NOVICE and TEACHER". Of the 52 articles examined, 38 identified significant attributes ascribed to expert problem-solvers. These articles were examined in detail and the defining attributes listed (see Appendix 1). --------------------------------------- Appendix 1 about here --------------------------------------- Attributes appeared to cluster around 4 general categories (evaluation procedures, metacognitive processes, problem-solving events, and person/material presentation). The three most frequently appearing attributes in each category, which were relevant to the problem-solving task to be considered, were selected as being representative of that category. Dimensional profiles (criteria) for interpreting each attribute as it applied to the present problem-solving task were devised (see Appendix 2). ------------------------------------ Appendix 2 about here ------------------------------------ An article X attribute matrix was constructed and point loadings were assigned on the basis of attribute frequency as identified from the 38 articles under consideration. Results were as follows: 4 points - attributes identified in approx 85% of the articles (1 occurrence out of 12 attributes = 4 points) 3 points - attributes identified in approx 60% of the articles (1 occurrence out of 12 attributes = 3 points) 2 points - attributes identified in approx 30% of the articles (3 occurrences out of 12 attributes = 6 points) 1 point - attributes identified in approx 15% of the articles (7 occurrences out of 12 attributes = 7 points) The rationale here being that in order to fall into the expert category a subject would have to score high in the categories which are worth the fewest number of points. It was possible for subjects to obtain a total of 20 points. Three vignettes describing general classroom-related problem- solving situations were constructed. To keep the problems as broad as possible, the first dealt with a classroom management issue where the subject was a participant in the situation; the second dealt with an academic issue where the subject was an observer in the situation; and the third involved a social/emotional issue where the subject had a vicarious interest in the situation (see Appendix 3). Subjects, who were informed that their responses would be audio-tape recorded, were asked to make themselves available for a half hour session. A roster was established so that data could be collected on an individual basis. Subjects were instructed that they would be given 5 minutes to read through the first vignette during which time they would be free to make notes if desired. they were then asked 'how would you go about solving the problem'? The same procedure was followed with the second and third vignettes. Debriefing took the form of thanking subjects for their participation and informing them that, once their audio-taped protocols had been transcribed, they would be welcome to come and discuss their statements if they so desired. Analysis Protocol analysis was undertaken by firstly scanning the entire transcript pool for significant words and phrases. This was done in order to acquire a "feel" for the data. Open coding was used and consisted of a line-by-line analysis with concept indicators (attributes) being assigned using code notes on the transcripts. A summary matrix for school-leavers and mature-age subjects was compiled (Appendix 4 & 5 respectively). Data were analysed using the SPSS software package. A principal components factor analysis was selected to assess the inter-relationship of the variables. Eight mature-age and eight school-leavers scored 17 points or above (85%), these were designated as experts. A cluster analysis was undertaken on the raw data sets as a means of validating the cut-off score at the 85% (17 points) level (squared Euclidean was the similarity measure used and the clustering method selected was average linkage). Results Factor Analysis Looking at an orthogonal solution, it appears that factor one is defined by variables A1, A3, B1, B2, B3, C2, C3, and D3; factor two appears to relate to variables B2, D1, D2, and A2; with variable C1 being unrelated to any of the above (see Table 1). Twenty seven percent of the total variance is accounted for by factor one and 13% by factor 2 (see Table 2). The results indicate that attribute C1 did not contribute to either factor and as such is inadequate for classifying expertise. This result would suggest that variable C1 does not discriminate sufficiently between subjects (see Appendices 4 & 5). Factor two variables may be related in that of all 12 variables, B2, D1, D2, and A2 would tend to be the most open to subjective interpretation during coding. As such, they may be suspect in determining expertise. A closer examination of these variables suggests that they may be grouped by knowledge of the content (procedural or "knowing that"-type knowledge), which has been shown to be related to novice forms of problem-solving (Chi, Glaser, & Rees, 1982). Results seem to indicate that the eight factor one variables may be grouped by language expression and knowledge that is more declarative (or "knowing how"-type knowledge) in nature. Qualities such as these have been shown elsewhere to define expertise (Yon, 1991). As such, it is argued that factor one variables are adequate for differentiating between expert and non-expert problem-solvers. Table 1 Orthogonal Transformation Solution - Varimax Attribute Factor 1 Factor 2 ___________________________________________ a1 .54 .18 a2 .11 .30 a3 .47 .17 b1 .71 -.11 b2 .43 .54 b3 .73 -.29 c1 -.22 .03 c2 .72 .26 c3 .69 .21 d1 -.23 .74 d2 .05 .71 d3 .54 .19 _____________________________________ Table 2 Eignvalues and Proportions of Original Variance Magnitude Variance Prop. _________________________________ Value 1 3.26 .27 Value 2 1.61 .13 _______________________________________________ Cluster Analysis Four clusters can be identified from the dendrogram (see Appendix 6). Subjects 3 and 31 seem undifferentiated and no explanation can be offered for their nesting within the cluster, although the observation can be made that the performance of subject 3 on variable A1 seems atypical for someone with a relatively high total score. The bottom 17 subjects identified in the dendrogram (excluding subjects 3 & 31) form a definite cluster. This cluster contains all 16 who had been designated as experts using the open coding method. It is therefore contended that the factor being identified in this group is expertise. Such a designation finds support in the responses of this group, for example, ..........I would also look at the children's likes and dislikes, try and establish what sort of things they like doing, what they don't like doing, and try to implement that into a programme so that you've got something built around their likes........... (subject 27, mature-age, response to vignette 1) .........Um this kid, he's he's um good with his social skills and that's obviously why he has been pushed along because they thought that um because he's socially adaptable that this might increase his sort of cognitive processes, so I'd I'd use social skills and I'd use partner work.......... (subject 5, school-leaver, vignette 2) The top 23 subjects identified in the dendrogram tend to nest together and on the basis of their points tally as displayed in Appendix 4 and 5, could be designated non-experts (novices). Comments indicative of novice responses include, ...........obviously you would have to establish their attention, you'd have to get their attention before you said anything, um, so I would either not say anything at all until they shut up out of curiosity or um that's where I stopped, and um. (subject 36, mature-age, vignette 1) ............Um, I might, I'm not sure but I might start a clapping game like, and then maybe try and take, you know, just a few kids might pick that up, and um, they seem to, sort of, I've done this in a few, few time and they seem to like, you know like, just like doing that and and trying to get it done............ (subject 19, school-leaver, vignette 1) With regard to the expert/novice distinction, subjects 11 through 30 seem to be on the cusp (postulants) and with further experience may cross over to the expert category. Again, such a designation appears to be warranted in the light of their coding points tally (see Appendix 4 & 5). Discussion It would appear from the results that there is more to expertise than simply the passage of time. The fact that approximately the same number of school-leavers as mature-aged subjects emerged as experts from similar-sized age differentiated samples, tends to suggest that chronological progression and concomitant experience are not in and of themselves sufficient conditions for the acquisition of classroom-related problem-solving expertise. Experts of varied ages tended to be more familiar with subject- appropriate language; evaluated from a more "situationally oriented" perspective; exhibited greater self-reflexivity; possessed more logically-structured thinking processes; treated a situation holistically prior to analysing discrete details; utilised extraneous information to advantage; made better problem-solving connections; and presented more information than did novices or postulants. Such results tend to support the notion that experts within a similar cohort have a quality about their thinking processes that is not present in novices. The question of experience, however, ought not to be discounted as this factor would no doubt have an affect on further refining the qualities displayed by experts as well as allowing postulants to progress into the expert category. In the parlance or Kuhn (1970) "paradigm shifts" may be difficult to achieve but are by no means impossible. Both past studies (Clarridge, 1989; Sabers, Cushing & Berliner, 1991) as well as the present one have identified a group on the cusp who, with greater insight, may make the necessary shift into the arena of expert problem-solving. It is at this point that the task of the teacher educator becomes crucial. Evidence here as well as elsewhere (Doyle, 1985; Griffin, 1985) suggests that educators would be well served in encouraging students to operate in what Vygotsky (1978) termed the "zone of proximal development". Such an approach may prove to be especially beneficial to those identified as postulants. There may be a further implication for the way in which classes are structured and courses taught. Should those, for example, who cluster on the basis of specific traits be taught as a group and should teaching of such a group consist of a trait-specific forms of instruction? This begs the larger question of how one instructs student teachers who possess domain-specific expertise. Why do experts exhibit superior problem-solving strategies? In an effort to answer this question, numerous models of expert thinking processes have been proposed in the field of cognitive psychology. Information processing models have emphasised aspects such as organisation (Joyce, 1979); integration (Shavelson & Stern, 1981); and complexity (Calderhead, 1983; Peterson & Comeaux, 1987). With regard to teaching expertise, individuals displaying these and similar traits have been described as possessing more cognitively rich schemas than their novice counterparts (eg. Livingston & Borko, 1990; Westerman, 1991). It is suggested here that these traits ought not to be seen as being discrete but rather, as being imbedded in the environment in which they operate. Such an environment may consist of components such as knowledge (content), intuition (creativity), and intelligence (capacity). What needs greater investigation is the nature of the interplay between traits and environment. What is the relationship, for example, between low creativity and cognitive complexity. An answer to questions such as this may provide greater insight into expertise. Conclusion To summarise, it is suggested that expertise needs to be defined more broadly than merely experience. Also, it would appear that the eight factors as described are valuable for identifying problem-solving expertise. Further, it is contended that the instrument as described is valid for differentiating between expert and non-expert student teachers. It is suggested that future studies be broadened to include males as well as subjects from different years within the teacher training programme. Classroom-related vignettes appear to be accurate indicators of expert performance and verbal protocols an effective way of gathering such information. Replication studies would further validate this method of investigating expertise in student teachers. References Berliner, D.C. (1986). In pursuit of the expert pedagogue. Educational Researcher, 15(7), 5-13. Berliner, D.C. (1987, April). Expert and novice interpretations of classroom data. Paper presented at the annual meeting of the American Educational Research Association, Washington, DC. Berliner, D.C. (1988). The development of expertise in pedagogy. Paper presented at the annual meeting of the American Association of Colleges for Teacher Education, Washington, DC. (ED 298 122). Calderhead, J. (1983, April). Research into teachers' and student teachers' cognitions: Exploring the nature of classroom practice. Paper presented at the meeting of the American Educational Research Association, Montreal, Canada. Carter, K., Sabers, D., Cushing, K., Pinnegar, S., & Berliner, D. (1987). Processing and using information about students: A study of expert, novice, and postulant teachers. Teaching and Teacher Education, 3, 147-157. Chase, W.G., & Simon, H.A. (1973). Perception in chess. Cognitive Psychology, 4, 55-81. Chi, M., Feltovich, P., & Glaser, R. (1981). Categorisation and representation of physics problems by experts and novices. Cognitive Science, 5, 121-152. Chi, M., Glaser, R., & Rees, E. (1982). Expertise in problem solving. In R.J. Sternberg (Ed.), Advances in the psychology of human intelligence. N.J.: Lawrence Erlbaum. Clark, C. (1984, April). Research on teaching and the content of teacher education programs: An optimistic View. Paper presented at the annual meeting of the American Educational Research Association, New Orleans. Clarridge, P. (1989, March). Alternative perspectives for analysing expertise in teaching. Paper presented at the annual meeting of the American Educational Research Association, San Francisco,CA. Cohen, L., & Manion, L., (1980). Research methods in education. London: Groom Helm. De Groot, A.D. (1965). Thought and choice in chess. The Hague: Mouton. Doyle, W. (1985). Effective teaching and the concept of master teacher. The Elementary School Journal, 86(1), 27-33. Einhorn, H.J. (1974). Expert judgement: Some necessary conditions and an example. Journal of Applied Psychology, 59, 562-571. Glaser, R. (1986). On the nature of expertise. In F. Klix & H. Hagendorf (Eds.), Human Memory and Cognitive Capabilities. Amsterdam: Elsevier Science Publishers B.V. Griffin G.A. (1985). The school as a workplace and the master teacher concept. The Elementary School Journal, 86(1), 1- 16. Gudmundsdottir, S. & Shulman, L. (1987). Pedagogical content knowledge in social science. Scandinavian Journal of Educational Research, 31(2), 59-70. Joyce, B.R. (1979). Research into the teaching mind: A vital direction. Educational Research Quarterly, 3(4), 10-15. Kinnear, J.F., & Simmons, P.E. (1990). "Expert" and "novice" subjects' approaches to genetic problem solving in a computer-based setting. Research in Science Education, 20, 171-180. Krabbe, M.A. (1989, March). A comparison of experienced and novice teachers' routines and procedures during set and discussion instructional activity segments. Paper presented at the American Educational Research Association annual meeting, San Francisco, CA. Kuhn, T.S. (1970).The structure of scientific revolutions (2nd ed.). Chicago, Ill: University of Chicago Press. Leinhardt, G. (1989). math lessons: A contrast of novice and expert competence. Journal for Research in Mathematics Education, 20(1), 52-75. Livingston, C., & Borko, H. (1990). High school mathematics review lessons: Expert-novice distinctions. Journal of Research in Mathematics Education, 21(5), 372-387. Mestre, J., & Touger, J. (1988). Cognitive research: What's in it for physics teachers? Report sponsored by the National Science Foundation, Washington, DC. (ED 299 173). Miles, M.B., & Huberman, A.M. (1984). Qualitative data analysis. Beverly Hills, CA: Sage Publications. Peterson, P.L., & Comeaux, M.A. (1987). Teachers' schemata for classroom events: The mental scaffolding of teachers' thinking during classroom instruction. Teaching and Teacher Education, 3, 319-331. Ropo, E. (1987, April). Teachers' conceptions of teaching and teaching behavior: Some differences between expert and novice teachers. Paper presented at the annual meeting of the American Educational Research Association, Washington, DC. (ED 287 824). Sabers, D.S., Cushing, K.S., & Berliner, D.C. (1991). Differences among teachers in a task characterised by simultaneity, multidimensionality, and immediacy. American Educational Research Journal, 28(1), 63-88. Shanteau, J. (1988). Psychological characteristics and strategies of expert decision makers. Acta Psychologica, 68, 203- 215. Shavelson, R.J., & Stern, P. (1981). Research on teachers' pedagogical thoughts, judgements, decisions, and behavior. Review of Educational Research, 51, 455-498. Stoddart, T. (1990). Perspectives on guided practice. Report sponsored by the National Centre for Research on Teacher Education, Washington, DC. (ED 330 682). Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. London: Sage Pub. Tudor, M. (1992). Expert and novice differences in strategies to problem solve an environmental issue. Contemporary Educational Psychology, 17, 329-339. Van-der-Mars, H., et al. (1991). Novice and expert physical education teachers: They may think and decide differently...but do they behave differently? Independent report, Oregon. (ED 336 354). Vollmer, M. et al. (1987, February). The usefulness and accuracy of self evaluation of teaching competencies. Paper presented at the annual meeting of the American Association of Colleges for Teacher Education, Arlington, VA. Vygotsky, L.S. (1978). Mind in society: The development of higher psychological processes. Cambridge: Harvard University Press. Welker, R. (1991). Expertise and the teacher as expert: Rethinking a questionable metaphor. American Educational Research Journal, 28(1), 19-35. Westerman, D.A. (1991). Expert and novice teacher decision making. Journal of Teacher Education, 42(4), 292-305. Yekovich, F.R., Thompson, M.A., & Walker, C.H. (1991). Generation and verification of inferences by experts and trained non-experts. American Educational Research Journal, 28(1), 189-209. Yon, M.G. (1991). The relationship between a year-long student teaching experience and student teachers' schemata for classroom phenomena. Journal of Personnel Evaluation in Education, 5, 55-67. Appendix 1 Characteristics of Expert Problem Solvers: Eclectic representation of attributes gleaned from 52 articles ù use principles to a greater extent ù use more abstract conceptualisation's ù operate semantically rather than episodically ù operate more quickly, both on perception and recall ù demonstrate more automatisation of routines & control processes ù think more flexibly ù information is organised by chunking ù use more elaborate problem-solving schemas ù use time more efficiently & effectively ù are better organised ù have better knowledge of content ù exhibit better teaching strategies ù operate in several domains at the same time ù make correct interpretations about different students ù have better instructional interventions ù possess more information, both facts and patterns ù interpret classroom phenomena more accurately ù are more intuitive/tacit ù discern and prioritise the importance of events ù evaluate own performance more realistically ù integrate new information with prior knowledge ù are aware of behavioural cues from students ù have better articulation of knowledge ù teach more by analogy and representation ù use frameworks rather than facts ù show greater depth in reports and plans ù tend to ignore others' information about students ù are better able to evaluate understanding ù use language more precisely and avoid multiple meanings ù think differently when preparing to take over a new class ù have more effective routines for getting to know students ù differ in type and amount of info they remember about students ù exhibit more effective pedagogy ù exhibit better improvisational performance ù have more flexible working plans ù explanations are conceptual rather than procedural ù better integrate knowledge for the learner ù declarative knowledge is greater in complexity ù think in action and adjust approach accordingly ù hold multiple points of view and have a wider perspective ù frame problems in a more understandable fashion ù are proactive decision-makers ù have a greater repertoire of instructional strategies ù lessons are not objective-driven ù do not make more decisions, just qualitatively better ones ù exhibit more self-reflexive behaviours ù go for overall impression first rather than specifics ù shows evidence of using episodic memory Appendix 2 Categories, Attributes and Criteria for Expert Problem-Solvers A. EVALUATION PROCEDURES 1. exhibits subject-discipline appropriate language to express ideas on >2 occasions (up to 3 points). criterion: uses the language of classroom management : understands the language of the vignettes : uses to advantage trigger words and key phrases 2. evaluations are inferential rather than merely descriptive 1 point). criterion: displays analytical thinking : is able to make predictions : looks beyond surface characteristics of problem 3. evaluations are "situationally oriented" rather than "text- book" based (1 point). criterion: relevance is exhibited : flexibility is evident : evaluations are confidently made B. METACOGNITIVE PROCESSES 1. exhibits self-reflexivity in thinking processes on >1 occasion (up to 2 points). criterion: exhibits an empathy with the subject matter criterion: can reflect on own thinking processes criterion: draws on an established repertoire of knowledge 2. exhibits logically structured thinking processes on >1 occasion (up to 2 points). criterion: can sustain thought-flow criterion: utilises episodic and not just semantic memory criterion: intuition may take the place of fact 3. initially focuses on overall impression rather than specifics (1 point). criterion: shows evidence of chunking information criterion: is not phased by minor or irrelevant details criterion: adopts a holistic approach C. PROBLEM-SOLVING EVENTS 1. suggests >3 potential problem-solving strategies (up to 4 points). criterion: can offer alternative suggestions criterion: understands the problem configuration (management) criterion: can select appropriate solution or strategy 2. utilises extraneous information to advantage (1 point). criterion: uses declarative and not just procedural knowledge criterion: uses analogy and/or representation criterion: is more interpretive 3. makes problem-solving connections (1 point). criterion: looks for principles rather than specific examples criterion: identifies relationships between concepts criterion: can integrate new information with prior knowledge D. PRESENTATION 1. exhibited no obvious verbal or non-verbal angst (1 point). criterion: transcript of protocol is free of language suggesting psychological discomfort 2. better able to articulate strategies, ie. speech clarity (1 point). criterion: as determined by the transcriber of the verbal protocols. Must be in top 2/5 of Likert- type scale to score 1 point 3. length of presentation was greater than the average and notes were used to advantage (2 points) criterion: "average" was determined by word count (1 point) criterion: notes were either detailed, organised, or creative (1 point) Appendix 3 Vignettes used for Obtaining Verbal Protocols VIGNETTE 1: You have just been assigned to a class. The school is in a lower socio-economic area and so many of the children come from deprived backgrounds. Discipline in the home normally consists of a warning followed by some sort of corporal punishment for a repeated offence. Most of the teachers, including the principal, are at the school because it offered a promotional position or because they were simply assigned there. On entering the assigned class, you find it to be extremely unruly. Children are talking and pushing each other about and the greater majority seem to ignore you as you enter the room. One child is hitting another with what appears to be a sandshoe. Several children are adopting what can clearly be construed as a cowering or submissive posture. You clap your hands loudly and the general noise ceases a little. VIGNETTE 2: Compared with the rest of the class, John Adams seems well behind in most academic areas including mathematics, reading and spelling. He seems to enjoy art and P.E., but seems unable to cope with even the simplest tasks requiring cognitive processing. A check through the previous year's records indicated that he had considerable academic problems the previous year also. His teacher at that time seemed to put it down to lack of readiness and didn't really address, it seems, the academic question. It was at the end of the year suggested that he repeat year 1; however, a final decision was made to allow him to graduate with his cohort as his social skills were developing at a more than adequate rate for his chronological age and it was believed that both he as well as his parents would be devastated if he was "kept down". VIGNETTE 3: Sussie Yin comes from a Chinese background. Her parents migrated from Mainland China 12 years ago and Sussie was born some 2 years later. She is an only child, her mother having to have a hysterectomy shortly after Sussie's birth. Both of her parents are professionals - her mother an accountant and her father a successful general practitioner. Sussie is presently in a class belonging to a close friend of yours. Sussie is excelling academically. However, her social development appears to be retarded. She does not make friends easily, is shy and rarely puts up her hand to answer the teacher's questions. She loves working by herself but finds it very difficult to work in group situations where co- operation is required. At lunch times and during most breaks she tends to sit by herself and becomes absorbed in reading fiction. Note: Vignettes were presented to subjects on separate sheets of paper, one at a time, and were double-spaced.