Ross, John A.; Rolheiser, Carol; Hogaboam-Gray, Anne
Effects of Collaborative Action Research on the Knowledge of Five Canadian
Teacher-Researchers.(teachers learning from other teachers) Elementary School
Journal v99, n3 (Jan, 1999):255 (1 pages).
COPYRIGHT 1999 University of Chicago
The key to professional growth is inquiry. For teachers this once
meant implementing the findings produced by expert researchers. Now
it means teachers becoming researchers, inquiring into their practices
for purposes of professional renewal. In this article we examine the
consequences for five teachers (the teacher-researchers) of joining
a school-university partnership to inquire into the routines of 13 of
their colleagues in the same district (the exemplary teachers). We
wanted to know whether studying peers helped teachers conduct
inquiries into their own practice. We especially wanted to know how
teacher-researchers used the findings from the original study to alter their
practice.
Theoretical Framework
Collaborative Action Research and Professional Renewal
We define collaborative action research as systematic inquiry into
teacher practice that is conducted by a team of teachers and
university researchers working as equal partners. We distinguish it from
unequal alliances such as school improvement research (in which
researchers have all the expertise and teachers are subjects of the inquiry)
and teacher research (in which teachers have all the knowledge that
counts and researchers have an ivory tower view of schools that needs
a reality check). In our view of collaborative action research there
is status equality. Each partner has a distinctive body of knowledge
that is complementary (permitting a shared framework for joint work)
and nonoverlapping (which fosters interdependence and makes
cooperation worthwhile).
The goals of collaborative action research are numerous (e.g.,
Noffke, 1997). We focused on two. First, action research helps teachers
improve their individual practice in several ways. Teachers buffeted
by reform movements launched from outside the school become more
powerful when they have access to research data and tools (Schensul &
Schensul, 1992). Teacher-researchers may be more willing to take
professional risks if their involvement in research leads them to feel in
greater control of their professional lives. Data obtained through
one's own efforts are more immediate and meaningful. Teachers are more
likely to use research findings productively if the research provides a
sense of ego involvement (Cousins & Earl, 1995; Cousins & Walker,
1995). Finally, action research encourages teachers to become
instructional theorists. When teachers conduct research, they articulate their
intentions, test assumptions, and make connections among elements of
their practice (Cochran-Smith & Lytle, 1990).
Case studies of action research have supported these expectations.
Teachers who engaged in action research indicated that they became
more reflective about their instructional practices during the inquiry
(Caro-Bruce & McCreadie, 1994). Some teachers reported that their
action research project convinced them to change their teaching, for
example, by giving more attention to prerequisite knowledge and skill
(Buckmaster, 1994), introducing strategies for promoting gender equity
(Stroeher, 1994), or integrating project-based activities into a
traditional science curriculum (Scott, 1994). Others reported that
action research gave them greater insight into students' thinking, for
example, concerning students' cognitions about their teacher's
assessment practice (Stuart, 1994). Still others have credited action research
with motivating teachers to change the context in which teaching
occurs, for example, by extending teacher collaboration within the
school (Simms, 1994) or motivating teachers to challenge assessment
policies that fail to acknowledge the out-of-school achievements of
minority students (Kester, 1994). In these studies action research
activities were described and the outcomes for teachers (and occasionally for
students) were identified, but the connections between the two were
typically implicit. In our research we wanted to bring action and
outcome closer together.
Our second purpose for focusing on action research is that it
contributes to knowledge about teaching by drawing on teachers'
experience to identify questions neglected by researchers and by enriching the
interpretation of findings with the tacit knowledge of teachers
(Lytle & Cochran-Smith, 1990). Evidence in support of this claim is not
well developed. Case studies describing specific forms of this new
knowledge have not appeared, although the professional development
schools movement continues to hold promise (Darling-Hammond, 1994; Valli,
Cooper, & Frankes, 1997).
Teachers are more likely to realize these two ends if they
participate in research partnerships with university-trained researchers
than if they do research on their own. Partnerships help overcome such
obstacles as teachers' lack of skill in research methods, a problem
affecting even teachers with formal training in conducting research
(Green & Kvidahl, 1990). Teacher involvement is also limited by lack of
time to do research, a problem that can be reduced if collaboration
with professional researchers brings additional resources to the
enterprise. In addition, cultural norms giving primacy to classroom
instruction over all other teacher roles constrain teachers' participation
in research. Frequent contact with professional researchers through
joint research may strengthen the image of the teacher as researcher
who generates and uses findings to improve practice (Huberman, 1995).
It takes time for school and university partners to learn new
skills, assimilate unfamiliar norms and beliefs, and negotiate complex
roles. The effects of collaboration on teachers' practice may not be
immediately visible. In this article we track the effects of one
collaborative action research project (Phase 1) on the subsequent
professional inquiries of the five teacher-researchers who conducted it (Phase
2).
What the Teacher-Researchers Did in Phase 1
The project focused on student evaluation (we used the terms
evaluation and assessment interchangeably, as did the teachers in the
study) because a district needs assessment had identified evaluation as a
top in-service training priority for teachers who were using
cooperative learning methods (hereafter CL). Evaluation is problematic for
all teachers, but teachers using CL have three additional challenges:
the need to disentangle individual from collective performances, the
dilemma of how to share control of assessment with students while
fulfilling accountability requirements, and the need to balance the
cognitive and social assessment demands of CL (Ross, Rolheiser, &
Hogaboam-Gray, 1998b).
Sample
The district's staff development officer formed the research team
in response to our invitation to engage in collaborative action
research around a theme of mutual interest. He identified five teachers
who had shown an interest in action research, cooperative learning, and
student assessment. All teachers were female and had been using CL
for at least 2 years. The officer selected teachers who represented
the four geographic areas of the district and a range of grades: grades
1-2 (the primary division), grade 6 (two teachers who worked
together closely in the junior division in the same school), grade 7
(intermediate division), and grade 12 (senior division). All the teachers
were university graduates and had taken one or more advanced
qualification courses (but did not have graduate degrees). At the beginning of
Phase 1 their years of teaching experience were 21, 6, 14, 26, and
20, respectively. These teachers (hereafter the teacher-researchers)
met with the three authors who focused their teaching, research, and
field development activities on CL, two principals (from two of the
schools of the teacher-researchers) for whom CL was their top school
priority, and the district curriculum consultant responsible for CL
in-service training. In creating the collaborative team, the staff
development officer chose a purposeful sample rather than one that was
representative of teachers in the district. The district was large
(60,000 students) with a high commitment to staff development. District
teachers had considerable autonomy within an outcomes-based curriculum
framework set by the province. Instructional innovation, especially
CL, was supported through an extensive array of staff development
programs and curriculum development initiatives.
The teacher-researchers selected teachers known to them (through
district in-service sessions, district curriculum documents,
newsletter reports, and personal contacts) as successful users of CL
techniques. After considering additional nominations from principals, district
support staff, and other members of the project team (all were based
on personal knowledge of the nominees), the teacher-researchers
selected 13 exemplary teachers representing a range of grades: three
teachers taught primary (grades 1-3), two taught junior (grades 4-6),
four taught intermediate (grades 7-8), and four taught secondary (grades
9-12). Nine teachers were female, and all had been teaching for 10
years or more. Each teacher had attended a full range of district
in-service sessions, and some teachers had been involved in their
delivery. These teachers were connected to CL teachers outside their
district, and many had attended conferences and CL institutes. Some were
heavily involved in the district's assessment initiatives, such as
portfolio assessment, but these activities were not a criterion in their
selection.
An anonymous reviewer expressed concern that neither the five
teacher-researchers nor the 13 exemplary teachers would be considered
experts in student self-evaluation. Self-evaluation was not identified
as a focus of action research inquiry until the second phase of our
study in response to the findings from Phase 1. The results we obtained
may not generalize to teacher-researchers who select as the focus of
their inquiry a topic about which they have considerable knowledge
and skill. It is also possible that the data that the
teacher-researchers collected would have been different if they had studied
teachers selected for their excellence in assessment rather than instruction
and may have affected the teacher-researchers' use of these data in
ways different from those we observed.
Inquiry Methods
The purpose of the inquiry was to find out what outstanding users
of CL did about student evaluation. The questions were based on
teacher-researchers' concerns and were pooled into an interview guide that
had four sections. The guide began by asking the exemplary teachers
to describe their use of CL: how they became involved, the training
sessions they had attended, the CL model(s) they used, and their
feelings about CL. The second section focused on beliefs about evaluation:
there were probes asking exemplary teachers to describe their
student assessment practices in CL (e.g., What do you assess during a CL
lesson? What instruments do you use? What works well? Does your method
of assessment differ in a CL lesson from other learning situations?).
Teachers were asked about specific issues (e.g., How do you judge
individual success on a group product? How do you adjust your
assessment strategies for exceptional students? How do you report CL
experiences to parents? How do you record CL data?). There were probes about
roles (e.g., Who is involved in student assessment? Do you tell
students how their work will be assessed? How do you ensure consistency
among teachers regarding evaluation in CL?). There were also questions
about when and how often exemplary teachers assessed students in CL.
The third section asked for examples that illustrated the teacher's
evaluation principles. The fourth section invited interviewees to give
their thoughts on any evaluation issues not probed earlier.
Each exemplary teacher was interviewed at his or her school for
60-90 minutes by two teacher-researchers, or by other members of the
district team, alternating between questioner and recorder. The
recorders made detailed notes during the interviews and prepared
interpretive notes identifying issues afterward. The whole team reviewed these
notes and identified themes using analytic induction (scanning the
data for categories and relations among them), constant comparison
within cases (e.g., comparing a response to one question to responses from
the same teacher to related questions), comparisons across cases
(e.g., looking for similarities and differences between elementary and
secondary teachers), and triangulation of interviews with artifacts
(such as evaluation instruments) provided by interviewees.
The second interview guide was generated by pooling the questions
posed by individual teacher-researchers concerning the five themes.
The same interviewers conducted the second interview with the
exemplary teachers in the same locations, 2 months after the first, for 60-90
minutes. The interviewers began by describing one aspect of the
teacher's assessment methods the interviewers particularly liked and then
asked if the interviewee had additional thoughts about evaluation
since the first encounter. The interview then probed five themes: (1)
linking teaching and evaluation (e.g., Was evaluation a factor when
you selected your approach to CL? Do you feel any tension between your
evaluation methods and your CL teaching?), (2) evaluation criteria
(whether a series of key words drawn from the first interview, such as
fairness and rigor, could be applied to the interviewee's evaluation
methods), (3) evaluation methods (interviewees indicated if and how
they used methods that involved peer evaluation, anecdotal records,
rewards, portfolios, etc.), (4) reflections on evaluation (e.g., What
are the next steps for you in developing your evaluation methods? Do
you have a support group with whom you can discuss evaluation?), (5)
special issues for intermediate / senior teachers (e.g., Do you have
different strategies for assessing in rotary classes than in nonrotary
classes? How do you balance individual with group accountability?).
The final section was unique to each interviewee and focused on
ambiguities in responses to the first interview (e.g., "In your last
interview I heard you say that CL contributed to final grades in two ways
... do you think students are aware of the direct and indirect
effects of CL on their final grades?"). Detailed interview notes were again
compiled.
Analysis
The first and second interviews were tape recorded and transcribed
verbatim. Categories for coding the transcripts were created by the
academics from interviewer notes and minutes of interpretive meetings
of the project committee. The teacher-researchers refined the codes.
The academics independently coded 10 pages of one transcript, and
the teacher-researchers used the results to finalize the coding scheme.
This scheme had four main categories (feelings and cognitions about
evaluation, evaluation principles, evaluation practices, and training
issues) with subcategories and sub-subcategories within each. Pairs
of teacher-researchers used a manual defining and illustrating each
category in the coding scheme to independently code eight examples
from one transcript. Discrepancies (there were few) were resolved
through discussion.
Pairs of team members began independently coding the transcripts
and negotiating discrepancies. Due to constraints on practitioners'
time, 80% of the transcripts were coded by one academic. The
transcripts and codes were entered into text-analysis software (Drass, 1986),
which sorted the data into the categories of the coding scheme.
Descriptive cross-case summaries were created for each coding
category. Although pairs of teacher-researchers wrote at least one
summary, most were written by the three academics and were revised
following feedback from the teacher-researchers. From these summaries the
full team drew a series of assertions, and evidence for each was
assembled.
Action Research Learning Activities
During Phase 1 the academics provided readings on research methods
and developed learning activities to help the teacher-researchers
acquire the knowledge and skill to participate in research decisions.
These learning activities were incorporated into monthly meetings from
February to December 1994 (8 half days, not including
between-session tasks such as interviewing in schools).
For example, in preparation for the October meeting, team members
read an account of how to analyze interview data (Merriam, 1990, pp.
123-140). At the meeting the second draft of the scheme for coding
the exemplary teacher interviews was reviewed. One academic modeled its
use by coding a page from one transcript. The academic divided the
page into utterances and defined an utterance as a complete thought,
usually of at least one sentence, relevant to one of the main
categories of the coding scheme. The academic coded each utterance on the
page by determining the category, subcategory, and sub-subcategory that
best applied. Teacher-researchers independently coded a
representative sample of eight utterances drawn from several transcripts. After
each coding, team members discussed the codes they had assigned with a
partner. At the end of the eight, the whole team discussed how they
had coded the passages. A draft manual giving an explanation of each
code and an example for each was distributed. Team members
independently coded four pages from one transcript (each pair had a different
transcript) and resolved discrepancies through discussion. The whole
team discussed the adequacy of the coding scheme and made changes to
the codes and to the coding manual. Between meetings, pairs of team
members used the revised scheme to code the remainder of the transcript
they had worked on at the meeting.
Results
Because the results of Phase 1 are described in Ross et al.
(1998b), only a brief summary is given here. There were three main
findings:
1. The exemplary teachers were less confident about student
evaluation than about other aspects of their teaching. Anxiety, guilt, and
uncertainty permeated teachers' talk about evaluation. These negative
feelings were in stark contrast with the exemplary teachers' pride
in their ability to use CL to enhance student learning.
2. The potential knowledge of the exemplary teacher group exceeded
the private knowledge of each individual. The transcripts provided a
host of practical suggestions about how a CL teacher could evaluate
students. These ideas were potentially more powerful when they were
assembled by the team. For example, exemplary teachers' strategies for
teaching students how to evaluate their work were assembled into a
four-stage procedure: (1) involving students in determining the
criteria used to evaluate their work, (2) teaching students how to apply
the criteria, (3) giving students feedback on their applications, and
(4) helping students develop action plans from their self-evaluations.
3. The exemplary teachers wanted to learn more about evaluation
methods, but few mechanisms for doing so were available to them. There
were no networks for sharing evaluation ideas and learning from
others. They wanted professional development sessions on evaluation
focused on the needs of CL teachers.
Effects of Phase I on Teacher-Researchers
The academics interviewed the teacher-researchers individually
about their experiences in Phase 1. The questions focused on what the
teacher-researchers had learned during the inquiry and how it had
affected their thinking about evaluation in CL (if it had). Interviews
were taped and transcribed. The teacher-researchers indicated that
participating in Phase 1 had increased their research skills and
confidence in their ability to conduct an inquiry on their own. Phase 1 also
influenced their perception of their abilities as teachers in three
important ways.
First, the exemplary teachers' descriptions of their strategies
stimulated the teacher-researchers to think about evaluation methods.
These reflections began during the interview ("As soon as others start
speaking about their experiences, an echo starts to go within you").
By talking to others, the teacher-researchers felt they were
learning about themselves. During the synthesis of the data they came to
view the assessment practices of CL teachers as problematic ("I didn't
feel there was a real problem until I did start to read those
summaries [of Phase 1 data]").
Second, the teacher-researchers were reassured to learn that even
teachers they viewed as exemplary were uncertain about the adequacy
of their evaluation methods. Learning this enabled the
teacher-researchers to talk more freely about their concerns in an area of
their practice about which they were uneasy. They could acknowledge any
deficiencies they were feeling without their overall competence being
threatened. "When I listened to the other teachers speaking about their
difficulties, what was happening is that they were articulating
feelings that I had not had enough time to think about and reflect upon and
put into an idea. So that talk between us probably helped me more
than it helped the teachers we were interviewing."
Third, in processing the data the teacher-researchers appreciated
the evaluation methods of the exemplary teachers. But the
teacher-researchers became convinced that they could do better than the
teachers they interviewed; they thought they had learned something beyond
what the individual exemplary teachers told them--synthesizing the
interviews created new knowledge. This confidence in being able to do
better was particularly strong with regard to using self-evaluation: "I
felt good that other teachers were having the same concerns about
[self-evaluation] that I was. Once we started talking about it in our
meetings, I really got excited. Well, if we trained them [the students],
could it be done? Is there a way we could come at this, perhaps in a
different way or come at it in a more professional way.... And
that's when [we] started talking about how we would train them."
What the Teacher-Researchers Did in Phase 2
Phase 1 ended with the teacher-researchers in a state of positive
dissonance. The knowledge they had acquired during the inquiry left
them dissatisfied with their current ways of evaluating students in
CL, aware of attractive new strategies, and confident in their ability
to make changes in their classrooms. The purpose of Phase 2 (January
to June 1995) was for each teacher-researcher to construct her own
interpretation of the Phase 1 findings and to use the findings and the
inquiry process that generated them to strengthen her classroom
practice. In designing and implementing her action research, each teacher
drew on the other teacher-researchers and the other members of the
Phase 1 team (three academics, two principals, district curriculum
coordinator).
Setting Individual and Group Goals
The teacher-researchers began by deciding whether they wanted to
work on one aspect of student evaluation as a group. The
teacher-researchers regarded five categories of strategies used by the
exemplary teachers as especially powerful: peer-evaluation, self-evaluation,
reporting to parents, combining individual and group accountability, and
using rubrics and benchmarks to record data. They concluded that
self-evaluation (to be combined with peer evaluation by two of the
teacher-researchers) could provide a common focus.
Their decision was influenced by two factors. The most important
influence was the belief that in pooling the student self-evaluation
strategies of individual exemplary teachers to form a four-step model,
the teacher-researchers had created powerful new knowledge. They
wanted to be among the first to use it. In addition, the academics made
an overt appeal for the study of self-evaluation. The academics
argued that self-evaluation plays a key role in fostering an upward cycle
of learning. Positive self-evaluations encourage students to set
higher goals and commit more personal resources to learning tasks
(Bandura, 1986; Schunk, 1996). Negative self-evaluations, however, lead
students to embrace goal orientations that conflict with learning, select
unrealistic personal goals, adopt ineffective learning strategies,
exert low effort, and make excuses for performance (Stipek, Recchia, &
McClintic, 1992). Without teacher involvement in student
self-evaluation, teachers have no direct knowledge about whether their students
are on an upward or downward path. The choice for teachers is not
whether students evaluate their own work (they will regardless of
teacher input) but whether teachers attempt to influence the process. The
academics presented a summary of previous research showing that
teaching students self-evaluation techniques increased students'
achievement (Arter, Spandel, Culham, & Pollard, 1994), self-regulation (Henry
1994; Schunk, 1994, 1996), motivation (Hughes, Sullivan, & Mosley,
1985), and use of mastery-oriented help-seeking and help-giving learning
strategies (Ross, 1995b).
The academics also argued that CL manuals (e.g., Bennett,
Rolheiser, & Stevahn, 1991) encourage self-evaluation and provide tools to
guide students' reflection on their progress. Only a few researchers
(reviewed in Ross, 1995a), however, have examined teachers' use of
these methods or students' perceptions of their value. The academics
tried to persuade the teacher-researchers that they could increase
student learning by teaching students how to evaluate their work.
Each teacher-researcher selected a personal action-research focus
within the group topic. The personal focus was based on the teacher's
past use (or nonuse) of self-evaluation. Sharon (grade 1/2) wanted
"to find out if the children were capable at this young age of
defining their own criteria" (stage 1 in the model for teaching
self-evaluation). Anne Marie (secondary school) decided that "the stage I had
always missed was the modeling" of criteria: "I knew it was the next
stage I should somehow get to, but I wasn't sure how to go about it"
(stage 2). Dianne (grade 7) was concerned about the credibility of
students' self-evaluations. "There was never that evidence.... It was more
of a judgment call. And I think that's why I felt a little uneasy
with it, because it was subjective" (stage 2). Michelle (grade 6)
thought self-evaluation was important, but she rarely had time for it.
"Periods just never seemed long enough by the time you explained it and
got the kids set up and they found their partners and they started
working ... there was no time left." Michelle wanted to increase her
use of self-evaluation, emphasizing the integration of self-evaluation
with evaluation by the teacher (stage 3) and setting goals (stage 4).
Cheryl (grade 6) thought that she and her students were not making
full use of self-evaluation data. "I would collect the sheets, and
then usually I would flip through the sheets. If there was something ...
[that] needed to be addressed ... then I would speak to the group,
and we would do some problem solving on how they could improve ... the
sheets would sit on my desk ... until the next time I tidied.... I
was not collecting the data and using it in a truly meaningful way."
Cheryl's goal was to devise a strategy for making better use of
self-evaluation data by combining them with other evaluation data (stage 3)
and by setting goals (stage 4).
Sources of Phase 2 Data
Each teacher-researcher made her own decisions regarding data
collection procedures. Data collection was continuous and included such
items as records of teacher plans and reflections, student reflective
journal entries, student surveys, student interviews, observations of
students, and achievement test scores. In addition to these unique
data sources, the whole team collected a common body of information
about each project.
Student surveys of attitudes toward evaluation. Before the
teacher-researchers implemented new approaches to self-evaluation, they
administered (in April 1995) a survey measuring students' attitudes
toward evaluation. The survey consisted of 16 Likert items measuring
students' beliefs that the evaluation they experienced was fair,
participatory, motivating, and meaningful (e.g., "After the evaluation, I know
what to work on"). A shorter version was prepared for grade 1/2
students. It consisted of 10 items with simpler wording and happy faces
as response options. The items were adapted from Paris, Turner, and
Lawton (1990) and Wiggins (1993). An earlier study (Ross, 1995a) showed
that the instrument had adequate reliability (alpha = .89), a
single-factor solution provided the best fit of the data, survey responses
were congruent with interview responses, and the items correlated
negatively with age, as predicted by previous research indicating that
student attitudes toward evaluation decline with school experience
(Paris, Lawton, Turner, & Roth, 1991). The survey was readministered 8
weeks after the teacher-researchers began to change their
self-evaluation methods. The academics anticipated that evaluation attitudes
would improve because the model of self-evaluation generated by the
teacher-researchers contained features that past research had shown to be
associated with improved student attitudes. Students view evaluation
more positively when they collaborate with teachers in assessment
(Mabry, 1992), if evaluation is frequent and open (Sarason, 1987), and
if it provides students with direction about how to improve
(Fredericksen & Collins, 1989; Moss, 1992).
Student focus group interviews. Scores on the pretest survey were
used to select two focus groups in each class. One focus group was
made up of the four students with the most positive attitudes to
evaluation (group means ranged from 2.37 to 3.59); the other consisted of
the four students with the most negative attitudes (group means were
4.59-5.27). The 20-minute interviews (in June 1995) included five
questions: What do you like/dislike about self-evaluation? Do you think
it is fair to ask you to evaluate your work? Do you have a chance to
say how you will be evaluated? Does evaluation help you do better in
school? Would you like to change anything about evaluation in your
classroom? All interviews were conducted by one of the nonteaching team
members, recorded, and transcribed verbatim.
Teacher-researcher interviews. The five teacher-researchers were
interviewed for 45-60 minutes in June 1995 by one of the nonteaching
team members. The interview was organized around the four stages of
self-evaluation (described previously). Five questions were posed for
each of the four stages: (1) Before interviewing the exemplary
teachers, how did you handle this stage (if you gave attention to it)? (2)
How did you handle this stage after you read the summaries of the
interviews and participated in the discussions in the committee? (3) What
were you trying to find out in your project with regard to this
stage? (4) What happened when you used your evaluation strategy? Did this
phase work the way you expected? Were there any surprises? Would you
do it this way again? (5) What did you learn from your project with
regard to this stage of teaching students their role in evaluation?
All interviews were recorded and transcribed verbatim.
Observation of team deliberations. The teacher-researchers met
with nonteaching team members of the committee for 3 hours on six
occasions from January to June 1995. (They were not observed during an
additional half-day of release time to work on the project at their
school.) At these meetings teachers planned their projects, identified
problems in implementation, engaged in collective problem solving, and
exchanged data on effects. The academics prepared notes from these
meetings and circulated them for discussion. For example, at the March
1995 meeting the teacher-researchers developed a parental consent
letter (after examining models provided by the academics), reviewed plans
for administering the student survey, and planned their individual
action research projects. At this meeting and the previous one, each
teacher-researcher met separately with one nonteaching member of the
team. The teacher-researchers described how their thinking had
progressed since the last meeting, developed instruments and procedures,
anticipated problems, and identified future tasks. The nonteaching
partner responded to these plans with constructive feedback. For example,
Sharon (grade 1/2) was concerned about how she was going to arrange
private conferences with each student. Her partner suggested
strategies such as asking another primary teacher to take a double class for
an activity, asking the preservice teacher scheduled to be in Sharon's
room to take the rest of the class, and increasing the productivity
of teacher-student conferences by having students practice peer
conferencing. In other meetings teacher-researchers borrowed ideas from
one another. For example, Diane incorporated Anne Marie's description
of videoconferencing as a strategy for modeling evaluation criteria
into her own plans.
Action research reports. The processes and products of each action
research project were reported in two related ways. The academics
wrote a short case describing each project in which they reported what
the teacher-researcher did, provided examples of the instruments or
procedures she developed, and described the data the teacher collected
and the results of the student surveys and interviews. This report
was based on data generated by each teacher-researcher (journals,
lesson plans, reflective notes, examples of student work, instruments),
combined with teacher interviews and meeting notes. Each
teacher-researcher, with support of nonteaching team members, constructed
narrative storyboards (a series of 12 graphics) that described her
understanding of what she did. (The Appendix displays one set of storyboards.)
The case reports were revised in response to feedback from the
teacher-researchers, and the storyboards were revised in response to
feedback from nonteaching members of the project team. The storyboards and
case reports are reported in Rolheiser (1996).
Analysis
For the quantitative data (student surveys), we prepared
descriptive (means and standard deviations) and inferential statistics (t
tests and effect sizes using Glass, McGaw, & Smith, 1981). For the
qualitative data, student and teacher interviews were recorded and
transcribed verbatim. Field notes of team meetings were made in vivo,
elaborated immediately after, and circulated to the teacher-researchers for
member checks. The qualitative data were interpreted using analytic
induction (scanning the data for categories and relationships among
them), constant comparison (checking responses against other data for
the same case), and triangulation of data sources (Erickson, 1992).
The data were organized around broad themes: what the
teacher-researchers were trying to do, what they did, and their reactions to
data they collected and to data collected on their behalf (student interviews,
student surveys, and teacher interviews). These data were used to
create the case reports (described previously) and the cross-case
analysis (which consisted of data matrices plotting themes against
persons). All interpretations were the outcomes of negotiations among the
three academics and the five teacher-researchers.
Results
How Phase 1 Findings Influenced Phase 2 Actions
Participation in collaborative research in Phase 1 influenced what
the teacher-researchers did in Phase 2. The teacher-researchers
occasionally adopted strategies the exemplary teachers had described ("As
they ruminated, I began to make connections to my own work and even
[found] a very quick solution for a class the next day"). More
commonly, the teacher-researchers used the strategies reported to them as
the basis for creating new methods that they thought would be more
effective. The variability in the use of Phase 1 data is illustrated in
the teacher-researchers' responses to their actions regarding step 1
of the model: involving students in defining the criteria for
evaluating their work. Our discernment of the teacher-researchers' actions
began to develop as we wrote up our field notes immediately following
our monthly meetings, but our understanding became much clearer when
we obtained feedback on drafts of the individual case reports and
helped the teacher-researchers construct storyboards of their inquiries.
The exemplary teachers used brainstorming to elicit student input.
In whole-class discussion the teacher reworked students' suggestions
to create a composite list of evaluation criteria. For example, one
exemplary teacher's grade 4 students created a set of indicators for
group presentations that included assessments of academic content
("shows main message about the province"), presentation style
("creativity/uniqueness; clear, strong voices--can we hear you?"), and
preparation ("well-rehearsed--everyone knows their part").
Two of the five teacher-researchers used these findings to confirm
their current practice. Cheryl (grade 6) decided to continue her
usual strategy because it involved more negotiation than the methods
developed by the exemplary teachers. Michelle (grade 6) decided not to
adopt exemplary teachers' strategies because she believed the teacher
alone should be responsible for defining evaluation criteria.
Participation in Phase 1 research increased these two teachers' confidence
about the worth of what they were doing. Two other teacher-researchers
(Dianne, grade 7 and Anne Marie, grade 12) decided that this stage
was important, but it was a lower priority for them than other aspects
of self-evaluation on which they wanted to work. Participation in
collaborative research helped these teachers to add an item to their
agendas for professional renewal and to determine when they would deal
with that item.
Sharon (grade 1/2) made student involvement in developing
evaluation criteria an action research goal because she saw it as a way of
generating grades for the social skills section of the new report card.
She took a strategy she was already using and transformed it by
thinking about the intentions implicit in the techniques of the exemplary
teachers for this stage. Before her project Sharon had students
complete T-charts for social skills. That is, students described what the
teacher should see and hear when students were performing a skill,
and the teacher recorded this information in chart form. Sharon
accepted answers that matched her list of criteria and rewarded student
responses, using her own language. In her action research, Sharon
continued with T-charts, but she "tried to use kid language; I wrote down
the exact words that the students gave to me as the criteria ... so
that when they self-evaluated, those words would be familiar to them."
Sharon organized students' responses into report-card categories. She
did not accept student responses as offered, nor did she impose her
own criteria. The process involved negotiation.
Sharon began by selecting one social skill (maintaining
self-control). She recorded students' comments about what the skill would look
like ("hands on lap when sitting on the floor, back straight, hands
to yourself, good behavior and manners, no fighting") and what it
would sound like ("listening when someone else is speaking, put your hand
up when you need to speak, use a quiet voice, being quiet when you
work"). Students' words became rows in an assessment form. In the
first column students recorded their self-evaluations: Six levels of
social skills performance on the report card were reduced to three levels
represented by happy, neutral, and sad faces. Sharon used the second
column to record her appraisal with the same indicators, and she
used the final column to convert the self- and teacher evaluations to
the 1-6 score that appeared on the report card. Sharon engaged students
in completing T-charts for each of the remaining skills, creating
similar instruments.
Participation in collaborative research led Sharon to transform
her methods of evaluation by (a) overtly negotiating the criteria for
student work and (b) including student judgments in her assessments of
student performance. In doing so she was guided by the goals of the
exemplary teachers but not by their methods. The actions taken by the
teacher-researchers in the other three stages of their framework for
teaching self-evaluation were similar. In each case they used the
Phase 1 findings in different ways, with every teacher ignoring some
results, modestly adapting others, and most frequently reconstructing
the intent of the exemplary teachers, typically avoiding literal
adoption of their methods.
Teacher-Researchers' Reactions to Data They Collected
The teacher-researchers thought the changes they made in
evaluating students enhanced student learning. Their evidence was mainly based
on classroom observations and marking of student work. The data were
anecdotal but internally persuasive. For example, teachers did not
collect student achievement data that could be compared to data from a
control group because they believed they did not need them. They
reported that as a result of teaching students how to evaluate their
work: (1) the quality of interaction in CL groups improved, (2) students
were able to articulate the criteria on which their work was judged,
(3) there were fewer complaints about the unfairness of evaluation,
and (4) student-teacher dialogues about the quality of student work
were more likely to focus on evidence than feelings. The most
important indicators of teacher success were perceived changes in student
self-regulation:
"I'm quite amazed that they are so knowledgeable about themselves.
children have taken responsibility for their own behavior they know
exactly where they need to improve" (Sharon, grade 1/2).
"The students have these forms in their possession, and they [are]
responsible for filling them out and keeping track of them the
ownership is on them" (Cheryl, grade 6).
"My cooperative learning lessons went up. I started doing more of
them because the students were more on task" (Michelle, grade 6).
"I think it really raised their level of striving. Now the students
are using and identifying those skills more frequently it has become
genuine" (Dianne, grade 7).
"They didn't take it as lightly ... they have bought into it.
They have been part of the making of something, so the pride in performance
is much greater" (Anne Marie, grade 12).
Teacher-Researchers' Reactions to Student Interviews
Students supported the changes their teachers made ("You get
better and better at stuff if you evaluate yourself" [grade 2]). They
believed self-evaluation was fairer because they had access to data not
visible to others ("You don't have to worry about someone missing
something you did well; you will know" [grade 6]). They liked having an
opportunity to state their case to the teacher, whereas before "a lot
of people [were] scared to go up to the teacher and say `I don't
think this is right'" (grade 6). The data meant more to them: "If the
teacher just takes notes, you don't really know what to improve on, but
if you're taking the notes, then you know what you're doing good on
and what you are not doing very good on" (grade 7). Students felt they
would be more likely to improve with self-evaluation ("It helps you
act better than you normally behave" [grade 6]), in part because the
data were recorded ("If you just think about it in your head you
usually forget it, but if it's written down, then you usually remember
it" [grade 1/2]).
Even students placed in the negative-attitude focus group were
mostly positive about the changes. But this group expressed concerns,
particularly about difficulty in assessing their work accurately ("When
I look at my own work, I can't find weaknesses" [grade 12]). Some
felt it was a waste of time ("We spend a lot of time filling out
evaluations when we could be doing work and stuff; I think that's sometimes
why people have a lot of homework at night" [grade 6]). Others felt
exploited ("We're doing the teachers' job, and the teachers are
getting paid to do this" [grade 7]). Students expressing less positive
attitudes toward evaluation on the initial survey were more likely to
develop misconceptions about new procedures. For example, in Cheryl's
grade 6 class the positive focus group appreciated the opportunity to
discuss discrepancies between their self-evaluations and the rating
given by their teacher. They found her willing to adjust her appraisal
if students presented a convincing argument. The negative focus
group in the same class thought it was unwise to challenge the teacher.
One student, in an analogy readily adopted by his peers, described the
teacher as a referee and described the consequences for players who
questioned the referee's ruling ("Don't bug me, kid; you're going out
of the game if you do" [grade 6]).
The student interview data influenced the teacher-researchers in
two ways. First, the interviews confirmed teacher perceptions that the
new approach to student evaluation had a positive effect on
students. Second, although the teacher-researchers knew that some students
continued to have negative feelings about evaluation, they were
surprised that some students harbored major misconceptions about
self-evaluation. None of the teacher-researchers had overtly addressed student
cognitions about evaluation in their action research.
Teacher-Researchers' Reactions to Student Survey Results
Table 1 displays the means, standard deviations, t tests, and
effect sizes (the difference between pre- and postmeans divided by the
pre-SD) for the attitude surveys. The table shows that students'
attitudes toward evaluation became more positive during the action research
of Sharon and Michelle, were unchanged in Dianne's and Anne Marie's,
and declined in Cheryl's.
TABLE 1. Changes in Students' Attitudes toward Evaluation,
by Teacher
Pre- Post
Teacher
Researcher Grade Mean SD Mean SD
Sharon 1/2 4.43 .419 4.73 .275
Michelle 6 4.02 .350 4.25 .383
Anne Marie 12 3.97 .405 4.04 .277
Dianne 7 3.75 .376 3.76 .382
Cheryl 6 4.21 .352 4.05 .535
Teacher
Researcher t test Effect Size
Sharon T(15) = -3.77(*) .72
Michelle T(26) = -2.63(*) .66
Anne Marie T(15) = -1.17 .17
Dianne T(27) = -.17 .03
Cheryl T(26) = -1.48(*) -.45
(*) p < .05.
These data were interpreted collectively. Sharon and Michelle were
pleased, but the teacher-researchers spent little time on the data
from these two classes because the results were what they had expected
for all classes. Initially, the team attributed the results for
Dianne and Anne Marie to a delay in testing. Both had completed their
self-evaluation activities 4 weeks prior to the second survey. Dianne
and Anne Marie, however, attributed the results to their own actions,
not to test timing. They concluded that the duration of treatment was
too short, that it took longer than they had anticipated to change
evaluation attitudes that had developed over students' entire school
careers. The teacher-researchers whose survey results showed no
significant effects resolved to increase their use of self-evaluation in the
future.
The results showing a decline in Cheryl's classroom were not fully
interpreted until the team had the transcripts of the student
interviews. The interviews indicated that although Cheryl and Michelle had
enacted similar procedures for combining self-evaluation with
appraisals by the teacher, misconceptions arose in Cheryl's class that were
not present in Michelle's. Several of Cheryl's students thought the
teacher's evaluation was a score of 1-5 that was assigned on a single
occasion. They were concerned that a student could be off-task at the
moment the teacher was observing ("At your worst possible time,
maybe she looks over and then for the rest of the time she doesn't ...
and you were doing really well, and you got like a 4 out of 5 [on your
self-evaluation] and she gives you 2 or something"). They did not
realize that the teacher selected five observational occasions, drawing
from each time slot during the CL activity, including start-up and
slow-down times during which off-task behavior might be more common.
Some students were also confused about how scores from the teacher were
combined with the scores students gave themselves. Some believed
that self-evaluation counted only when it matched the teacher's
assessment. These students thought self-evaluation was a waste of time
because they were either duplicating the teacher's work or their judgments
did not count. Cheryl also attributed the results of the surveys to
her own actions. She decided she needed to be more explicit in
describing self-evaluation procedures and to check students' understanding.
Cheryl's plans for acting on the survey data were similar to those of
the other teacher-researchers. They decided to give more attention
to student cognitions by identifying the benefits of self-evaluation
to students, responding to their negative feelings, and confronting
their misconceptions.
Discussion
This study confirms findings from previous research showing that
action research contributes to the knowledge base of teaching (e.g.,
Cochran-Smith & Lytle, 1993). In Phase 1, collective knowledge was
produced when the teacher-researchers assembled a four-step procedure
for teaching students how to evaluate their work that went beyond the
individual strategies of the exemplary teachers they interviewed. This
procedure provided the framework for generating self-evaluation
instruments (Rolheiser, 1996) and in-service activities (Ross, Rolheiser,
& Hogaboam-Gray, 1998a). In Phase 2, the teacher-researchers
contributed to collective knowledge by demonstrating that in some classrooms
teaching self-evaluation enhances student attitudes toward
evaluation. Evidence of the effects of action research on students is rare.
Most studies do not mention student effects, and those that do are
modest in their claims. For example, Calhoun and Allen (1994) found that
only 11 of 54 schools that engaged in action research provided
evidence of student effects. Phase 2 also enabled the teacher-researchers
to identify two factors, both under the control of teachers, that
mediated the effects of teaching self-evaluation. This finding influenced
subsequent in-service efforts by providing guidelines concerning the
duration of training required to change student attitudes and the
need to attend to student cognitions about evaluation.
Phase 1 influenced teacher-researchers' personal practical
knowledge by creating conditions for professional learning. Ross and Regan
(1993) found that the exchange of reports of professional experience
had a significant effect on teachers' professional learning by
creating conditions for constructive dissonance, synthesis, and
experimentation. Participation in action research in Phase 1 created these
conditions: Analyzing the practices of exemplary teachers enabled the
teacher-researchers to recognize deficiencies in their own practice, to do
so without lowering their confidence in their overall teaching
ability, and to devise a framework for self-improvement. Social comparison
was a key feature. The teacher-researchers selected an interview
sample of teachers they believed were exemplary in using an approach to
instruction (CL) that the teacher-researchers regarded as powerful.
Whether these teachers were exemplary in an absolute sense (and we
believe they were) is not important. What matters is that the
teacher-researchers believed they were. They also believed that Phase 1
increased their ability to conduct research. Phase 2 provided
teacher-researchers with personal practical knowledge about how to teach
self-evaluation in their grade and subject, information about the effects on
students of their first implementation of these methods, and direction
to guide their subsequent efforts.
The effect of participating in collaborative action research in
Phases 1 and 2 was to strengthen teacher-researchers' self-efficacy
beliefs. Teacher efficacy is the belief that teachers, individually and
collectively, will be able to bring about student learning. The
extensive research on teacher efficacy (reviewed in Ross, 1995c; Ross,
1998) demonstrates the generative power of teacher expectations.
Teachers who anticipate that they will be successful set higher goals for
themselves and their students, are more willing to engage in
instructional experiments, persist through obstacles to implementation, and
have higher student achievement. Action research provided
teacher-researchers with three kinds of information that have been linked in
previous research to enhanced teacher feelings of self-efficacy.
First and most important were mastery experiences. The data
collected by the teacher-researchers themselves, and to a lesser degree the
interviews, surveys, and field notes collected by the academics,
demonstrated to the teacher-researchers that they were able to improve
their evaluation of students. Repeated feedback on their effectiveness
created an upward cycle of expectations that enabled them to handle
disconfirming information and increase their aspirations. What
counted was not the absolute level of the teacher-researchers' performance
but how they interpreted their performance. It was not the objective
validity of the evidence but its internal credibility that mattered
(Bandura, 1986). Second, the interactions with other
teacher-researchers provided vicarious information about self-efficacy. Seeing
teachers like themselves being successful heightened teachers' expectations
about their ability to accomplish similar ends. Social comparisons
with the exemplary teachers also contributed to higher self-efficacy.
Third, persuasion by the academics and by their peers that the
teacher-researchers could enhance students' learning by designing student
self-evaluation procedures provided an initial impetus to self-efficacy
beliefs, one that was sustained when the predictions of others were
borne out by teacher-researchers' subsequent mastery experiences.
This study also supports a two-step approach to action research,
one in which teacher-researchers first learn how to study practice in
a particular domain by apprenticing with academics and then use the
results of the inquiry to design individual action research projects
to improve their teaching. Several aspects of this study, however,
constrain its broader application as a professional development
strategy. For example, the benefits of action research may be more accessible
to teachers who are confident and competent. Teachers with lower
self-efficacy are reluctant to engage in action research (Cousins &
Walker, 1995) because they anticipate being exposed as weaker teachers.
Reluctance may increase as the gap grows between teacher-researchers'
perceptions of their competence and that of the teachers they study.
Social comparison was a motivating factor in this study because the
teacher-researchers imagined they could do better than the exemplary
teachers they studied. For less capable teacher-researchers, social
comparisons may decrease motivation.
The benefits of collaborative action research may not accrue to
teachers who engage in action research independently of support from
academics (or other personnel with research training). The main
contribution of the academic researchers in this study was only partly
related to the training in research methods they provided. More important
was that by sharing decision making with teachers, the academics
communicated that the teacher-researchers were able to choose wisely.
Louis (1991) found that the strongest organizational influence on
teacher efficacy was receiving respect from relevant adults, in this case
academics who were viewed as leaders in offering practical strategies
for CL teaching through in-service training. In addition, teacher
efficacy is consistently associated with collaborative school cultures
and participation in school decision making (Ross, 1995c, 1998).
External resources the academics brought also provided the
teacher-researchers with release time to conduct action research and increased
teachers' accountability for follow-through. Equal ownership of research,
however, may have negative effects if it creates demands that
teachers cannot meet (Kreisberg, 1992), and collaboration with peers might
lower self-efficacy if teachers receive negative feedback from their
peers.
Action research continues to be recommended as a mechanism for
professional growth from preservice training to career end. We support
this approach and recommend a two-stage method in which
teacher-researchers begin by examining the practices of others before
experimenting with their own. Teachers should undertake such activities in
collaboration with supportive peers and professional researchers. We believe
that the action research methods developed in this study provide
greater opportunity than other methods for the enhancement of teacher
efficacy. Teachers' expectations about their professional abilities
have generative power. Action research processes are more likely to
contribute to teachers' expertise if the processes build confidence as
they simultaneously build knowledge and skill.
Note
An earlier version of this article was presented at the annual
meeting of the American Educational Research Association, New York,
April 1996. The research was funded by the Durham Board of Education, the
Social Sciences and Humanities Research Council of Canada, and the
Ontario Ministry of Education and Training. The views expressed in the
article are not necessarily those of the board, council, or
ministry. We wish to thank the teacher-researchers (Michelle Ferreira, Sharon
Hopkins, Cheryl Hoyle, Anne Marie Laginski, Dianne Serra), the
non-teaching members of the district partnership (Jim Craigen, Brian
Greenway, Don Real), and the 13 anonymous exemplary teachers who
participated in the project. Laurie Stevahn provided feedback on an earlier
draft. Correspondence should be sent to John A. Ross, Ontario Institute
for Studies in Education, Trent Valley Centre, Box 719, 150
O'Carroll Avenue, Peterborough, Ontario, Canada K9J 7A1.
References
Arter, J., Spandel, V., Culham, R., & Pollard, J. (1994, April).
The impact of training students to be self-assessors of writing. Paper
presented at the annual meeting of the American Educational Research
Association, New Orleans.
Bandura, A. (1986). Social foundations of thought and action: A
social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.
Bennett, B., Rolheiser, C., & Stevahn, L. (1991). Cooperative
learning: Where heart meets mind. Toronto: Educational Connections.
Buckmaster, L. (1994). A principal tutors four low achievers in a
third-grade mathematics classroom. Elementary School Journal, 95(1),
49-62.
Calhoun, E., & Allen, L. (1994, April). Results of school wide
action research in the league of professional schools. Paper presented
at the annual meeting of the American Educational Research
Association, New Orleans.
Caro-Bruce, C., & McCreadie, J. (1994). Establishing action
research in one school district. Elementary School Journal, 95(1), 33-40.
Cochran-Smith, M., & Lytle, S. (1990). Research on teaching and
teacher research: The issues that divide. Educational Researcher,
19(2), 2-11.
Cochran-Smith, M., & Lytle, S. (1993). Inside/outside: Teacher
research and knowledge. New York: Teachers College Press.
Cousins, J. B., & Earl, L. (1995). The case for participatory
evaluation: Theory, research, practice. In J. B. Cousins & L. Earl
(Eds.), Participatory evaluation in education: Studies in evaluation use
and organizational learning (pp. 3-18). London: Falmer.
Cousins, J., & Walker, C. (1995, June). Personal teacher efficacy
as a predictor of teachers' attitudes toward applied educational
research. Paper presented at the annual meeting of the Canadian
Association for the Study of Educational Administration, Montreal.
Darling-Hammond, L. (1994). Developing professional development
schools: Early lessons, challenges, and promises. In L. Darling-Hammond
(Ed.), Professional development schools: Schools for developing a
profession (pp. 1-27). New York: Teachers College Press.
Drass, K. (1986). TAP 1.0: Text analysis package. Mimeo. Dallas:
Northwestern University, Center for Urban Affairs and Policy Research.
Erickson, F. (1992). Ethnographic microanalysis of interaction. In
M. LeCompte, W. Millroy, & J. Preissle (Eds.), Handbook of
qualitative research in education (pp. 201-225). San Diego: Academic Press.
Fredericksen, J., & Collins, A. (1989). A systems approach to
educational testing. Educational Researcher, 18(9), 27-32.
Glass, G., McGaw, B., & Smith, M. (1981). Meta-analysis in social
research. Beverly Hills, CA: Sage.
Green, K., & Kvidahl, R. (1990, April). Research methods courses
and post-bachelor education: Effects on teachers' research use and
methods. Paper presented at the annual meeting of the American
Educational Research Association, Boston.
Henry, D. (1994). Whole language students with low self-direction:
A self-assessment tool. Charlottesville: University of Virginia.
(ERIC Document Reproduction Services No. 372 359)
Huberman, M. (1995). The many modes of participatory evaluation.
In J. B. Cousins & L. Earl (Eds.), Participatory evaluation in
education: Studies in evaluation use and organizational learning (pp.
103-111). London: Falmer.
Hughes, B., Sullivan, H., & Mosley, M. (1985). External
evaluation, task difficulty, and continuing motivation. Journal of Educational
Research, 78(4), 210-215.
Kester, V. M. (1994). Factors that affect African-American
students' bonding to middle school. Elementary School Journal, 95(1), 63-74.
Kreisberg, S. (1992). Transforming power: Domination, empowerment,
and education. Albany, NY: SUNY Press.
Louis, K.S. (1991, April). The effects of teacher quality of work
life in secondary schools on commitment and sense of efficacy. Paper
presented at the annual meeting of the American Educational Research
Association, Chicago.
Lytle, S., & Cochran-Smith, M. (1990). Learning from teacher
research: A working typology. Teachers College Record, 92(1), 83-103.
Mabry, L. (1992, April). Empirical evidence of validity and
reliability in measuring student achievement via alternative assessment
methods. Paper presented at the annual meeting of the American
Educational Research Association, San Francisco.
Merriam, S. (1990). Case study research in education: A
qualitative approach. San Francisco: Jossey-Bass.
Moss, P. (1992). Shifting conceptions of validity in educational
measurement: Implications for performance assessment. Review of
Educational Research, 62(3), 229-258.
Noffke, S. (1997). Professional, personal, and political
dimensions of action research. In M. Apple (Ed.), Review of research in
education (Vol. 22, pp. 305-343). Washington, DC: American Educational
Research Association.
Paris, S., Lawton, T., Turner, J., & Roth, J. (1991). A
developmental perspective on standardized achievement testing. Educational
Researcher, 20(5), 12-20, 40.
Paris, S., Turner, J., & Lawton, T. (1990, April). Students' views
of standardized achievement tests. Paper presented at the annual
meeting of the American Educational Research Association, Boston.
Rolheiser, C. (Ed.). (1996). Self-evaluation: Helping students get
better at it. Toronto: Visutronx.
Ross, J. (1995a). Assessment of group process and product. Final
report of Ministry of Education Transfer Grant Project. Peterborough:
Ontario Institute for Studies in Education Trent Valley Centre.
Ross, J. (1995b). Effects of feedback on student behavior in
cooperative learning groups in a grade 7 math class. Elementary School
Journal, 96(2), 125-143.
Ross, J. (1995c). Strategies for enhancing teachers' beliefs in
their effectiveness: Research on a school improvement hypothesis.
Teachers College Record, 97(2), 227-251.
Ross, J. (1998). The antecedents and consequences of teacher
efficacy. In J. Brophy (Ed.), Advances in research on teaching (Vol. 7,
pp. 49-74). Greenwich, CT: JAI.
Ross, J., & Regan, E. (1993). Sharing professional experience: Its
impact on professional development. Teaching and Teacher Education,
9(1), 91-106.
Ross, J., Rolheiser, C., & Hogaboam-Gray, A. (1998a). Skills
training versus action research in-service: Impact on student attitudes to
self-evaluation. Teaching and Teacher Education, 14(5), 463-477.
Ross, J., Rolheiser, C., & Hogaboam-Gray, A. (1998b). Student
evaluation in cooperative learning: Teacher cognitions. Teachers and
Teaching, 4(2), 299-316.
Sarason, I. (1987). Test anxiety, cognitive interference and
performance. In R. Snow & M. Farr (Eds.), Aptitude, learning, and
instruction: Vol. 3. Conative and affective process analyses (pp. 131-142).
Hillsdale, NJ: Erlbaum.
Schensul, J., & Schensul, S. (1992). Collaborative research:
Methods of inquiry for social change. In M. Lecompte, W. Millroy, & J.
Preissle (Eds.), Handbook of qualitative research in education (pp.
161-200). San Diego, CA: Academic Press.
Schunk, D. H. (1994, April). Goal and self-evaluative influences
during children's mathematical skill acquisition. Paper presented at
the annual meeting of the American Educational Research Association,
New Orleans.
Schunk, D.H. (1996). Goal and self-evaluative influences during
children's cognitive skill learning. American Educational Research
Journal, 33(2), 359-382.
Scott, C. (1994). Project-based science: Reflections of a middle
school teacher. Elementary School Journal, 95(1), 75-94.
Simms, M. (1994). A principal tutors four low achievers in a
third-grade mathematics classroom. Elementary School Journal, 95(1),
41-48.
Stipek, D., Recchia, S., & McClintic, S. (1992). Self-evaluation
in young children. Monographs of the Society for Research in Child
Development, 57(1), 1-84.
Stroeher, S.K. (1994). Sixteen kindergartners' gender-related
views of careers. Elementary School Journal, 95(1), 95-103.
Stuart, M. (1994). Effects of group grading on cooperation and
achievement in two fourth-grade math classes. Elementary School Journal,
95(1), 11-21.
Valli, L., Cooper, D., & Frankes, L. (1997). Professional
development schools and equity. In M. Apple (Ed.), Review of research in
education (Vol. 22, pp. 251-304). Washington, DC: American Educational
Research Association.
Wiggins, G. (1993). Assessing student performance: Exploring the
purpose and limits of testing. San Francisco: Jossey-Bass.
Appendix
THE EVOLUTION OF SELF-EVALUATION:
Giving Students Feedback On Their Self-Evaluations & Helping
Students Develop Action Plans
Michelle's Story
1
Michelle wanted to work on the project because she had used
student self-evaluation before but was unhappy with the results: "they
handed it in and that was it ... I put it away and it was forgotten."
2
For her involvement in the project Michelle selected goals for
teaching self-evaluation.
3
Michelle reviewed self-evaluation forms in cooperative learning
texts.
4
Michelle created recording forms (each giving a 1-5 score) for
each social skill, as well as s summary sheet. She checked the drafts
against her criteria and decided they were ready for use.
5
Michelle developed a T-chart with the students (what does it sound
like and look like) for two social skills.
6
Students evaluated their work (individually or as a group) on 4
occasions in each cycle.
7
She randomly selected 8 students to observe. Following the lesson,
Michelle shared the score (1-5) with the students observed.
8
Students graphed their evaluation data.
9
Students noted specific areas to work on the next time the social
skill was addressed.
10
Michelle reflected on the results. Students made better use of
self-evaluation data than they had in the past. They made plans and
stuck to them.
11
Student attitudes toward evaluation improved.
12
Michelle decided to continue using self-evaluation. She realized
that her forms needed to be modified to fit her new grade.
[ILLUSTRATION OMITTED]
John A. Ross is professor and Carol Rolheiser is associate chair
in the Department of Curriculum and Teaching, and Anne Hogaboam-Gray
is a research officer, at the Ontario Institute for Studies in
Education of the University of Toronto in Toronto, Ontario, Canada.