Dunkin_Types_of_Errors - Review of Edited Research Summer...

Info iconThis preview shows pages 1–6. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 6
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Review of Edited. Research Summer wild, Vol. so. No. 4. PP. 37-9.? Types of Errors in Synthesizing Research in Education Michael J. Dunklrt The University ofNew South Wales Nine types oferrorr occurring in three stages ofthe process of synthesizing research ore described and illustrated with exornplesfrom :1 recent synthesis (Hog-on. i 992.! of research on teacher professional growth. Errors can occur intbe initialidennjficorion ondcollern'on ofrcports ofraseorch. in tits analysts of documents. and in the final stage of reaching generalizations about the whole body of research. Types oferrors include the exclusion of relevant literature. wmegly reportng details til-It'll as sample size. erroneously attrib- uting findings to studies. and storing unwmnred conclusions about the research reut'eued. implications for reviewers and users of reviews are considered. In the last 2‘3 years a large body of literature on ways of synthesizing research in education has developed (Du nkin. 199l4t'tlr'albe1'g. 1986}. Approaches to glean- ing the accumulated findings of that research have varied from the narrative through vote counting or box scores to meta—analysis. Some of these approaches make more demands on the conceptual and interpretative skills of the synthesizer titan others and. therefore. contain more scope for error and bias than others. although all approaches are subject to the fallibility of the syndiesieers and those upon whom tltey necessarily rely. It is important that the validity of all syntheses be tested. for they are the main ways in which assessments can be made about the accumulation and development of research-based knowledge. Syntheses of research are influential in regard to subsequent research. policy. and practice. They provide the empirical bases for applications for research grants. for higher-degree dissertations and theses. and for individual and institutional research. They are used by policymakers in designing strategies for development. and they are used to guide practitioners in the enhancement of professional activity. They provide the contents of highly regarded publications in handbooks. encyclopedias. and textbooks and become the best known statements of the state of knowledge on the topics to which they are addressed. The processes by which syntheses of research are conducted and disseminated are. therefore. crucially important. because they determine the quality of the syntheses and which syntheses are available publicly for the above purposes. Il'the synthesizers allow systematic biases to affect their selection of studies to review. The author would like to thank the anonymous refetees and the editors for their helpful advice in bringing this paper to the stale ot‘puhtishability. They are not in any way responsible for any errors that might be found in it. B”? fliltfit'fll if they for. to recognize that some authors of the original sources were wrong in annnuncjng their findings. if they allow their own priorities to affect the findings they report to the exclusion of contrary findings. then there is the potential for a synthesis that is seriously flawed. The likelihood that a poor synthesis would survive the rigorous refcrccing process employed by prestigious scholarly jour— nals is undoubtedly very small. Nevertheless, the consequences of such a mistake warrant contemplation. for they could be whole programs of misguided research. policy. and practice. The purposes of this article are to suggest the stages at which a synthesizer is at risk of making mistakes. to present a typology of errors that can be made. and to illustrate the typology w't'th errors made in a recent synthesis—namely. Kagan‘s “992) review of studies on professional development among preservice and beginning teachers. Stages in the Occurrence of Error There are three stages at which synthesizers might make errors. The first. the primary stage. is when the synthesizer searches the literature and selects from it the items judged relevant to the topic of the review. Errors made at this stage result in bias that might lead to conclusions that represent the findings of only part of the research and omit the findings of the rest. or that give equal status to the findings of good and poor research. At the secondary stage. the reviewer analyzes the literature selected in order to identify context. methods. and the findings of each study included. This is the stage at which the variety of errors made is greatest. As detailed below. error in identifying facts about contests and methods leads to the misclassilication of studies. and errors in identifying and reporting findings introduce error into the nest stage of the synthesis. This neitt stage. the tertiary stage. occurs when the synthesizer accumulates the findings identified in the previous stage in order to reach generalizations about the topic under investigation. Errors brought forward from the primary and secondary stages have their fullest impact at this stage. for they can lead to invalid generall- rations. Types of Reviewer Error Primary Stage Errors Type l: Unexplained selectivity. These are errors in which the reviewer ex- clndes research which comes within the declared scope of the review without explaining or justifying the exclusion. The outcome of this error is that the conclusions of the reviewer cannot be held to apply to the whole defined field of concern. It is possible. for example. that the findings of the body of the. excluded research contradict those of the included research. and that reliance on the latter alone produces a biased picture of the state of knowledge in the field. Contempo- rary facilities for conducting literature searches and developing bibliographies are so advanced that these errors can seldom be explained in terms of understandable reviewer ignorance of the existence of the excluded work. in her critique of Kagan‘s “992) review. Grossman “992] specified If: studies that came within the declared boundaries of the review but had been ignored. She BE Errors in .. .esi'zirtg Research also drew attention to the fact that a surprisingly high proportion of the papers analyzed had been written by just two authors. which suggests that relevant papers by other authors had been ignored. if Grossman was right. then the review cannot be regarded as adequately representing the state of research knowledge on teacher professional growth. because of the presence of Type 1 errors. I 1 Type 2: fact: of discrimination. Not all research on the-same topic ts of equal quality. Much of the literature available to a reviewer consists of tentative repeats in the icon of conference papers presented by authors seeking feedback prior to preparation of final drafts. which may or may not be submitted for publication to refereed scholarly journals. Cine test. therefore. of the qpality of a conference paper is whether or not it subsequently appears in one version or another. perhaps under a different title. in a refereed journal. In their efforts to be up—to‘date. reviewers often do not wait that long. and rely on conference papers themselves. Clearly. this is a risky practice which can result in preliminary findings lacing accorded the same stanis as findings contained in journal articles that have survived rigorous refereeing and editing processes. This is not to say that all conference papers are defective or that all journal articles are free of weaknesses. in either case. reviewers should be vigilant that they distinguish between good and poor research. lest they give equal status to both. _ ‘ Grossman [1992} accused Hagan {1992} of having committpd this-type of error but did not specify examples of poor research reviewed. This is possibly the most difficult type of error to identify. because criteria for evaluating research are usually controversial. and judges sometimes disagree on the difference between good and poor research. Most of the studies included in Kagan's review had been published in refereed journals. but It were cited as conference pitpers presented at Annual Meetings of the American Educational Research Association. Three of those papers could not be obtained for present purposes. l-lollingswortb (personal communication. September 29. IEtElrll reported that she did not coauthor the paper cited as “Lidstoue and Hailingst (19%)." and attempts to contact the other cited author were unsuccessful. The paper was not listed in the. main. printed program of the meeting at which it was said to be presented. and it seems not to have been published elsewhere. The paper cited as ‘Wecndpl “939)” was also unavailable for this analysis. It did not appear in the main. pnnted program of the meeting concerned. and repeated attempts to locate the paper were unsuccessful. Finally. the paper by Cochran-Smith (1939} was revised and subsequently pub- lished [Cochran-Smith. 199]} under a different title- During the years covered by Kagan's review {193? through [Still 1. the present author was the editor of Taught ing and Teacher Education: An international Journal of Research and Studies. Three of the papers reviewed by Kagaii were submitted for publication in that journal and were subsequently rejected.I I I I L The unavailability andt'or nonpublication of some papers in their original form. the form which was used in Hagan {1992}. may well indicate something about their quality and give justification to Grocsman‘s “992) criticism. Secondary Stage Error! As mentioned above. Grossinan “991] criticized Kagan fl???) mainly for making two types of errors. which have been classified as pnmary stage errors here. She said little. however. about the types of entire described below that also 89 Drurkt'u were found in Kagan‘s synthesis. Type 3: Errorreotrs detailing. These errors consist of incorrect statements of the sarupling. methods. designs. procedures. and contexts of the studies reviewed. In some cases. the reports reviewed are about pans of larger projects and contain descriptions of the larger projects as well as the relevant parts. Details of the whole are sometimes thought to apply to the part when. in fact. they do not. In other cases. tltc reviewer accepts an opening statement of sample size without recogniz- ing that attrition occurred and that the data actually came from a smaller sample. Errors in the reporting of methods. procedures. designs. and contexts can lead to the misclassification of studies so that they do not share the essential characteris- tics that are supposed to make them comparable with other studies. In Kagan {I992} the sample sizes of four studies were reported inaccurately. In one case {Calderhead r9. Robson. I991] the sample size was said to he l1. when it was '1". in another (McDaniel. l991] it was said to he 22. when it was 3; in a third stMthotko et al.. I99|]. it was said to bellil. when it wasjust I: and in the fourth {Wildmam Niles. i'vlagliaro. d: McLaughlin. I939] it was said to be IS. with inv depth profiles of 4. when in fact the id were the suhiect of an earlier report (Whitman. Magliaro. Niles. 3t McLaughlin. I933]. and only the 4 case studies were reported in the publication reviewed. The main problem in these cases was that the papers reviewed were sometimes reports of parts of larger studies. The larger sample sizes were true of the full-scale studies. but the pans of these studies being dealt with in the review had smaller sample sizes. Thus. all four errors were in the direction of larger sample size. so that the reader was led to believe that the accumulated findings of those four studies came from a total of 8? novice teachers. when in fact they came from a total ofjuat [5! Other Type 3 errors were made in reporting data gathering procedures. For example. concerning the study by Aitken dc lvlildon {1991]. it was claimed in Hagan “992] that only interview were used. when in fact there were workshop presentations. written preparations for these workshop presentations. self-evalua‘ tions. peer evaluations. written autobiographies. written responses to workshops. and researchers' field notes. A more serious en'or concomng these types of details led to the complete- miselassification of three studies. The LaboSkey III99l] study was classified in Hagan {I992} as being concerned with the “image of self as teacher “ tp. 1415]. Laboskey made no reference to self-image. and nowhere in the paper was any- thing said about the “central role“ that selfuimage was said by the reviewer to play. In the section headed "Requisites for Growth During Practice and Student Teach~ ing." the review summarized six studies which. it was claimed. “examined how candidates‘ knowledge of teaching changed during a practicum. student teaching. or the course of an entire preservice program" tp. 140]. In fact. one of the studies {Gore 3: Zeichner. 199i] did not investigate change at all. Then. the study by Chamberliu 3c Vailanee {l99l]. included in the group supposed to be “compre- hensive evaluations of practica or student teaching experiences.” was said to involve student teachers in spending half of each day in a 9-hour "block course" in classrooms. In fact. the student teachers in this study spent only half a day per week for 9 weeks in classrooms. The other studies reviewed in that section involved either I or 2 semesters or I year of full-time student teaching and were. therefore. clearly not comparable to the misclassified study. Needless to say. the 9t] Errors in brunest'zI-Ifi Research findings of that study could not legitimately be claimed to contribute to any accumulation of evidence that might be perceived in the rest of the group. Type 4: Double counting. This error consists of listing different reports from the same project as providing additional confirmation of the same finding. The risk of this error occurring seems to be present particularly when there is multiple reporting of results of the one project. This can be difficult to detect. especially given that titles and lists of authors are sometimes changed. While it might be expected that these multiple reports would contain acknowledgements of each other's existence. this does not always happen. Reviewers. therefore. have to be especially vigilant not to assume that independent studies have been reported. Kagan [I992] included one very interesting Type 4 error involving double counting. This involved the case studies reported by Bullough and his colleagues. Etulluugh. Knowles. and ICer [1939] reported case studies of three beginning teachers named Bonnie. Lyle. and Helena. Then Huilough U990] wrote about Helena alone. but called her Heidi. without mentioning that they were one and the sarrte. Next. Bullough and Knowles {1990] wrote about Lyle alone, without mentioning the earlier write-up on Lyle. Finally. Bullough and Knowles [1991] wrote about Bonnie alone. except that this time Bonnie was called Barbara: it was revealed in an endnote that Barbara “chose to have her real name used“ to. 139]. This endnotc was not signaled at any place in the text where the name "Barbara" was used. but. rather. was added to another. quite unrelated note. Only by accident could a reader discover the information about the change of name. Kagan was apparently unaware of these strange occurrences and so treated these publications as though they were all reporting independent studies. The effect of this Type 4 error was that Hagan t t 992] claimed that six cases had been found. when there were in fact only three. To those who consider such matters as the replicability or frequency of occurrence of findings to be important. this error is significant. Type .5: Nonrecognirfon of faulty author conclusions. Authors of original reports of research do not always represent their findings fully in their statements of conclusions. [1" reviewers uncritically accept such statements. they risk continua ing the misrepresentation. The occurrence of this error can be contributed to by the original author. who may be biased to the extent of selectively incorporating only expected or hoped for findings in statements of conclusions. One example of a Type 5 error that occurred in Kagan {1992] involved the report by Weinstein {1990]. Weinstein studied all 33 student teachers enrolled in one section of an introductory course required for formal admission to a teacher education program in the northeastern United States. On the basis of responses to an initial questionnaire which asked "How well do you think you will do during student teaching?." [5 were selected for interview. Students had also been asked to list the strengths and weaknesses they considered when answering that once. tion. The I 5 were then interviewed to explore topics in depth and answer questions about ways in which their thinking had changed during the semester. Of the 15. I1 agreed to be interviewed in the following semester. The questionnaire was administered in the autumn semester on the first and last days of class. Interviews were conducted in the following spring. After reporting the results of her analyses of the data. Weinstein concluded. “What is most striking about the data reported here is the lack of change that DI I..." .... .- manned r. g the semester" (p. 285}. In Hagan {1992}. Weinstein's conclusion regarding change was repeated as follows: "Despite coursework and field experi- ences. the candidates‘ beliefs about teaching and themselves as teachers remained unchanged throughout the semester" (p. MD). in fact. Weinstein's findings were as follows: [ii Tltere was a significant decrease in optimism {p s: .4115}. “although students remained extremely optimistic" tip. 281}. {2} “Explanations for the self-ratings given at the end of the semester were similar to those given at the beginning of the semester. Students continued to stress caring (32%} [down from 45%] and enthusiasm {40%} lap from 32%}. However. non—teaching experiences with children were rarely mentioned idfi‘ii: at the beginning of the semester versus 1.6% at the and (p c .flflfl I }: instead. 13% of the students {in contrast to an initial 3%]: now cited a knowledge of teaching gained from the course or the associated field experience (p = .03" (p. 233). {3} “Conceptions of a ‘really good teacher‘ remained largely the same: how— ever. there were some interesting changes. Ability. to maintain discipline was now mentioned by Sil'ib of the students {p = .135} [up from Fifth ]. as was enjoyment or enthusiasm for teaching (p 4: .ill} [up from 23%]. Fewer subjects cited ability to motivate students [p = .ti'Sti [from 34% to l 8%] and willingness to spend extra time and effort in = .flB] [from 32% to 16%]. There was also an increase in responses dealing with the ability to meet the diverse needs of individual students {from 5% to 19%. p = .fl'i'l" (pp. 234—4235). Weinstein also found that there was a decrease in mentions of “professional behavior" from 10% to 0% (p c .95]. but this was not mentioned in the text of her report. None of these findings by Weinstein regarding change was mentioned in Kagan {i991}. and so the conclusion was a gross misrepresontatiorn of the facls. probably due to uncritical acceptance of Weinstein‘s clearly unwarranted conclusion. Type t5: Unwarmnrea' attributions. This error consists of reviewers claiming that studies yield findings or reach conclusions that they do not. In an extreme form. this error can even consist of attributing findings to studies when the design limitations of the studies do not permit such findings to be reached. or when the studies do not even set out to investigate the subject of the attributed finding. Due glaring Type IS error was found in Kagan's (1992} treatment of the Hollingsworth {IQEQJ report. in the review. it was said that Holiingsworth “939) identified four factors that appeared to affect the acquisition of classroom knowledge by the novices: Ea} their images of themselves as learners: (bi an awareness that they needed to temper initial beliefs and come to terms with classroom management: {cl the presence of a cooperating teacher who was a role model that facilitated growth: {d} placement with a cooperating teacher whose ideas and practices were some- what different from the student teacher's beliefs. Modeling seasoned teach- ers was not sufficient to promote conceptual change: cognitive dissonance was needed to force novices to confront and modify their personal beliefs. (p. 145} This is what l-iollingsworth {I989} wrote: For those who did [reach a balanced managerial style}. there seemed to be at least four explanatory factors that helped them acquire that knowledge: {a} a role image of themselves as learners and critics of teaching. which allowed 91 Errors in 31' 't‘zt'ng Res-torch for error and change: [bi an awareness that they needed to change their initial beliefs to come to terms with classroom organization: {c} the cooperating teacher andt'or university supervisor as role models and facilitators of that change: and {de a notion of having something worth teaching that demanded student cooperation. (p. W4] Attributing to Hoilingsworlh “939) the finding concerning cognitive dissonance was a Type ti error in Kagan [1991}. Type F": Suppression of contrary findings. lClriginal reports sometimes contain findings that are actually contradictory of the generalizations which a reviewer claims they support. It is not necessarily the casein these instances that findings consistent with the reviewer's generalization are not contained in the reports; it is just that other. contradictory findings that are reported are ignored. Avoidance of Type '3" errors does not necessarily demand that reviewers acknowledge every single finding of a study. but it does require acknowledgement of every single finding that is contrary to a generalization a reviewer intends to make about the findings of the body of research concerned. I Grossman “9923 hinted that there was a Type 7 error in Kagan [1992] when she argued that in the pan of her study {Grossman. 1939] “not directly addressed" by the review there was no evidence that preservice teachers complained that a particular course was "too theoretical” (p. 174]. Perhaps the most severe Type 1" error. however. was referred to above concern- ing the Gore and Zeichner {i539 l} study. lGore and Zeichner were concerned with levels of reflective thought exhibited by student teachers. Theirs was a case study concerned with action research and reflective teaching in preservicc teacher education. The authors reported an analysis of the written reports of the action research projects conducted by lb student teachers during the 1933—1989 aca- _ demic year “to explore the extent to which action research seemed to be contrib- uting to reflective teaching practice as we have defined it: that is. reflection within all three domains of rationality“ (p. 129}. The three domains were technical. practical. and critical rationality [Van l'vlanen. 19TH. Student teachers kept journals: conducted some formal observations of other classes; prepared. taught. and evaluated a unit of work; and conducted action research. Gore and Zeichrter [Willi discovered three broad groups of action research projects: {a} a small number in which there was “clear concern for moral and political issues as integral to the project" (p. 129}. which they categorized as displaying critical rationality: lb} a larger number in which there was “some concern for these issues but [the writer] did not develop the ideas" Ep. [29}. which they care goriaed as displaying practical rationality: and (cl projects more than half of which "revealed no explicit concern for moral and political issues at all“ (p. lEQi. which they catcgorizedas displaying technical rationality. In Kagan {I992} it was said that there” was “little evidence of reflection; what little they did find consisted of technical rationality. the lowest level” (p. I42}. This was not true. At least 6 of the 15 cases studied (Jo. Bruce. Melinda. Helen. Leslie. and Annette} were classified as displaying either critical or practical rationality. the two levels of reflection ranked above technical rationality. By denying the actual findings obtained and thereby committing a Type 'i error. the reviewer drew unwarranted conclusions about the influence of preservice teacher education upon student teachers“ cognitive performance. 9'3 flittrl'r'tt Tertiary Stage Errors The tertiary stage of a synthesis is that at which the reviewer seeks to assemble the evidence of the individual studies according to the main topics or issues investigated. in order to see whether meaningful and justifiable generalizations {syntheses} can be stated about them. The questions asked are. Do titey add up‘?. and. If so. to what'il Errors at this stage can lead to the statement of invalid generalizations and to the failure to recognize valid ones. ltheorursc. errors made at the primary and secondary stages seriously threaten the validity of generaliza‘ tions at the tertiary stage. but it is also possible that errors can emerge at this stage for the first time. type d: C'nnsequential errors. These are generalizations that are flawed as a consequence of errors made at earlier stages. Errors ofany of the first seven types during the primary and secondary stages affect the validity of the findings of the body of studies that the reviewer attempts to synthesize at the tertiary stage. Colloquially. this problem has been captured in the WEll‘kJ'tD'ttt-tl'l. phrase “garbage in. garbage outl“ Valid generalizations cannot be reached from erroneous conclu- sions about the individual studies that are part of a review. The Kagan [E9921 synthesis addressed the question "Do preserviee candidates change their personal beliefs and images duri ttg the course of a teacher education program?" [13. I56.) and constructed the generalization that “the personal beliefs and images that preserviee candidates bring to programs of teacher education usually remain inflexible." lp. |54i. Later. it was claimed that “all but one study ' indicated that personal beliefs remained stable" (p. lid}. However. it has been demonstrated (Bunion. I995} that the reverse was the ease and that most of the studies cited did find substantial change. It seems that the review‘s representation offourofthe studies (McDaniel. I99| '.h'chaughlin. 199] :1 Pigge dc Marso. 1939; Weinstein. I99th contained Type "l errors. and that Type 6 errors had been made wittt another two {Calderhead tit Robson. 199i: Gore 3: Zeichner. 199i). neither of which reported findings about change or iaclt of it. As a consequence of these errors. a Type 3 error was made in Kagao [1992] in the genemliration about change in preserviee teachers' beliefs. Type 9: Failure to marshal? all evidence relevant to a generalization. When a reviewer. in the process of formulating conclusions. fails to recognize that a study contains evidence relevant to a generalization. he or she cotnmits this type of error. This is different from a Type l error. It is not that an entire study is omitted from the review but that one or more of the study's findings are not included in the assembling of evidence bearing upon the generalization in question. In Kagan (1992]. Type 9 errors were made when the reviewer failed to assemble evidence provided by Bullough et al. {1939}. Grossman {[9139}. and Levin and Ammon [I992] concerning the generalization that university courses are not sufficiently relevant to the needs of student teachers and that such courses fail to provide novices with adequate procedural knowledge of classrooms. in the re- view. it was claimed that there was extensive support for that proposition {p. 162). but contrary evidence provided in the three studies cited above. was ignored. The review failed to report Levin and Ammon's finding that university courses were effective in securing growth in student teachers‘ pedagogical conceptions. Fur- thermore. the review did not mention that Grossman had shown what could 94 Errors in S) . ..:slzt'ng Research happen in the absence of teacher education courses and had concluded that “teacher education coursework can help prospective teachers acquire knowledge about what students are lilter to find difficult in a particular subject. and a realistic sense of students' interests. abilities. understandings. and misconceptions con- ' oerning specific topics" tp. ace}. Moreover. the review did not refer to the following conclusion reached by Bullough et at.: art three teachers had adequate tl-teoretical knowledge about teaching. expo- sure to and practice of appropriate teaching skills: indeed. they completed the same preserviee teacher education program together. 1|t‘t’hat they initially lucked were useful understandings of the contests in which they would work and. particularly for Lyle. consistent. grounded. and accurate understandings of themselves as teachers. lip. 23” Bullough et al. argued that teacher educators could do more to help prospective teachers answer the question “Who am I?" but that the understanding of school contents was primarily the responsibility of school districts and principals. The failure to bring the above findings by Bullaugh et al. (1939). Grossntsn {[939}. and Levin and Ammon {1992) to bear on this generalization constituted a Type 9 siren.-l Conclusions The main concern to arise from this presentation of the types of errors facing reviewers is the utttstworthiness of their syntheses. It has been argued that all more types of error were present in just one synthesis {Kagam 1992}. If that is the case. it seems clear that no reliance should be placed on that synthesis. But does it say anything about other syntheses? Surely. readers of these works cannot go to the trouble of the detailed Scrutiny required to check the validity of every synthesis before they decide whether or not to rely on them. blot even referees or editors can be expected to subject syntheses in manuscript form to the painstaking inquiry process required to establish their validity. in this respect. a synthesis of research is probably no different from the individual studies included in the review. all of which. themselves. are subject. to error at every stage of their eonstntction. The only feasible. systematic approach to quality assurance in regard to synthe- ses of research is to educate educational researchers. all of whom conduct their own syntheses of research whenever they write a dissertation. a grant proposal. or a research report. to look for the types of errors identified here. and in the more general requirements of good scholarship. Moreover. potential users of syntheses should be entouraged to develop a healthy skepticism toward them. The availabil~ ity of a typology of synthefizer errors should assist in both procesSes. Notes 'lt would be a breach of editorial ethics to disclose the identities of the authors of those papers. I 2In Kagan {I992} it was mistakenly claimed that McLaughlin (199” had been published in the Alberta Journal ofErfttc‘atr'onof Research. I ‘A. detailed analysis of errors found in Kagan H992) concemtng research on first- year and beginning teachers is to be found in Dunkin {in press}. CH Dunhin Reierenccs Aitken. J. L.. lit. Mildon. D. [I991]. The dynamics of personal knowledge and teacher education. Curriculum inquiry. 2i. HIE—I52. Borko. H.. Eisenhatt. M.. Undertt'tll. R. CL. Brown. C. A... Jones. CL. St Agard. P. C. “99L April}. To teach mathematics for conceptual or procedural itrtorvieclge.‘ t‘l dilemma ofleaming to teach in the "new world order" ofrnatirernorics education refirmr. Paper presented at the Annual Meeting of the American Educational Research Association. Chicago. IL. Bollough. R. lt'.. Jr. {I99U}. Supervision. mentoring. and self-discovery: A case study of a first-year teacher. Journal of Curriculum and Supervision. 3. 333450. Bullough. R. "IA. Jr.. lit Knowles. J. G. {I990}. Becoming a teacher: Struggles of a second-career beginning teacher. international Journal for Qualitative Studies in Education. 3. lDl-l l2. Bullouglt. R. 1It'.. Jr.. & Knowles. J. C. tl99ll. Teaching and nurturing: Changing conceptions of self as teacher in a case study of becoming a teacher. international Journalfirr Qualitative Studies in Education. 4. Ill—Hi]. Bullough. R. ‘tl.. Jt.. Knowles. J. EL. 5: Crow. N. A. “9139). Teacher self-concept and student culture in the first yearoitcaching. Teachers College Record. 9t. 299—233. Calderhead. 1.. dc Robson. M. (199”. Images of teaching: Student teactters‘ early conceptions of classroom practice. Teaching and Teacher Education. ?. I—B. Chamberlin. C.. A 1|tlallance. J. tI99ll. Reflections on a collaborative scltool~hased teacher education project. Alberta Journal ofEducotiotrai Research. 3;". I4 I -I 56. Cochran-Smith, M. { I939. March}. Ofattestions. trot ansrvers: Tile discourse ofs'tudent teachers and their school and university mentors. Paper presented at the Annual Meeting of the American Educational Research Association. San Francisco. CA. Cochran-Smith. M. tl99l J. Learning to teach against the grain. Harvard Educational Revietr'. til. ITEl—dlil. Dunkin. M. J. {1994}. Teaching. synthesizing research on. In T. Husen and T. ht. Pcstlethwaite {Eds}. The international encyclopedia of education: Research and studies {End ed.. Vol. It}. pp. 6235—6240}. flaford. England: Pergatnon Press. Dunkin. M. J. “995). Synthesising research in education: A case study of getting it wrong. Australian Educational Researcher. Hi | i. I'l—33. Dunkin. M. J. [in press]. Synthesising research in education: A case study of getting it wrong. Part 2. Asia-Pacific Journal of Teacher Education. Gore. 1. M.. St. Zeichner. K. M. il'il‘Ell]. Action research and reflective teaching in preservice teacher education: A case study from the United States. Teaching and Teacher Education. 3'. I Iii—lilo. Grossrnan. P. L. {1939]. Learning toteactt without teacher education. Teachers College Record, 9i, 192—2133. Grossman. P. L. [1992}. Why models matter: An alternate view on professional growth in teaching. Review of Educational Research. 62. Iii—1T9. Hollingsworth. S. {1939}. Prior beliefs and cognitive change in learning to teach. American Educational Research Journal. 2d. loll—139. Kagan. D. M. “9921. Professional growth among prescrvice and beginning teachers. Review of Educational Research. :52. [19—169. lahosltey. V. K. {1991. April}. Case studies of two teachers in a reflective teacher education program: "How do you know" Paper presented at the Annual Meeting of the American Educational Research Association. Chicago. its. Levin. E. 3.. a Ammon. P. t I 992}. The development of beginning tcachers' pedagogi- ctti thinking: A longitudinal analysis of four case studies. Teacher Education Quarterly. “Pf-4]. IEl-J'l. McDaniel. J. E. (199]. April}. Close encounters: Harv do student teachers matte sense 96 Errors in Syria sing Research of the social foundations? Paper presented at the Annual Meeting of the American Educational Research Association. Chicago. tL. McLaughlin. H. J. iin]. The reflection on the blackboard: Student teacher self. evaluation; international dour-traitor Qualitative Studies in Education. 4. l4I-l 99. Figs. F. L.. & Marso. R. N. t t9li9. March]. A longitudinal assessment of the afiecuve impact ofprcservice training on prospective teachers. Paper presented at the Annual Meeting of the American Educational Research Association. San Francisco. CA. 'v'an Manen. ltd. tl9"l'i}. Linking ways of knowing with ways of being practical. Curriculum inquiry. t5. EDS—223. I I 1tl-lttlloerg. H. J. {I936}. Syntheses of research on teaching. In M. C. thtrocit {Ed}. Handbook of research on reaching [3rd ed.. pp. 2 Isl—29.9}. New York: Macintllan. Weinstein. C. S. {1990]. Prospective elementary tcachers' beliefs about teachtng: Implications for react-tar education. Teaching and Teacher Education. 6. 279499. Wildman. T. M.. Magliaro. 5. G.. Niles. J. A.. dc McLaughltn. R. A. [l9BE.IAprtl}.. Sources of teaching problems and the ways beginners solve then-r: An analysts of the first two yea rs. Paper Inesented at the Annual Meeting of the American Educational Research Association. New Ctrieans. LA. I I Wildman. T. M.. Niles. J. A.. Magliaro. S. 6.. St McLaughlin. R. A. [I939]. Teaching and Icaming to teach: The two roles of beginning teachers. Elementary School .iou rntti. dd. it'll—4'93. Author MICHAEL J . DUNKIN is Professor. School of Teacher Education. 1111: University of New South Wales—St. George Campus. PLO. Box 33. GalleerSW. Australia 2213*. m.dut1kin @unsw.cdu.au. He specializes in research on teaching and teacher educa- tion. Received March 1. 1995 Revision received January Id. 19% Accepted March [9. 1996 9'? ...
View Full Document

This note was uploaded on 02/07/2010 for the course C&I CI550 taught by Professor Markdressmen during the Fall '07 term at University of Illinois at Urbana–Champaign.

Page1 / 6

Dunkin_Types_of_Errors - Review of Edited Research Summer...

This preview shows document pages 1 - 6. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online