Performance Assessment for California Teachers;
 2010. Read the article. 

Coming to a Credential Program Near You- National Exit Exams for Teachers

By Ann Berlak.  June ,2010.


Near the end of his Teachers College speech, Arne Duncan  named the particular exam he had in mind—PACT (Performance Assessment for California Teachers). Few listening to Duncan’s speech had heard of PACT, but it has been a fixture in California for several years, and many teacher educators and teacher credential candidates in California are, unfortunately, quite familiar with it. This article was written in the hope that the rest of the nation might learn from and not replicate our experience.

The Lesser of Two Evils?

In 1998 the California legislature passed a law requiring teacher credential candidates to pass a state-approved high-stakes exit exam. Many credential programs in California, including San Francisco State University (where I teach in the College of Education), chose PACT, devised by members of the education faculties at Stanford University, the University of California-Berkeley, and several other institutions. The alternative to PACT was CalTPA, a system designed by the state in consultation with the Educational Testing Service. Our College of Education elected to use PACT because it purported to be qualitative, not quantitative, and to assess “authentic” teaching performance in “real-world” contexts.

PACT’s creators and chief advocates claim that PACT provides the crucial link in the chain of evidence connecting the classroom performance of individuals at the end of their credential programs to achievement of their pupils in their first year of teaching, as measured by standardized tests.1 In academic lingo this is called “predictive validity.” The assumption, then, is that PACT will identify how well various credential programs promote teaching practices and learning outcomes valued by federal and state authorities.

Like all other states, California already had an elaborate set of state-mandated entrance and exit assessments for teacher education programs. In California this included the CBEST (a test of basic literacy), CSET (a standardized test of content knowledge), RICA (a test of knowledge about teaching reading), student teaching supervision, and GPA requirements. Our programs were also assessed by an increasingly prescriptive NCATE, the national accrediting body that imposes an entirely separate and largely redundant assessment system. PACT adds another layer to what is already a complex system of assessment. Each additional test throws up yet another hurdle to becoming a teacher—hurdles that disproportionately affect people of color and those for whom English is not their first language.

Near the end of his Teachers College speech, he named the particular exam he had in mind—PACT (Performance Assessment for California Teachers). Few listening to Duncan’s speech had heard of PACT, but it has been a fixture in California for several years, and many teacher educators and teacher credential candidates in California are, unfortunately, quite familiar with it. This article was written in the hope that the rest of the nation might learn from and not replicate our experience.


            SB 2042,  passed by the California Legislature in in 2000, required a major revision of teacher preparation in California based upon a new  set of  state standards and a set of teacher performance expectations (TPEs) .   The universities have responded by revising their programs. In 2042  The legislature  created a system where the state must continually train new teachers to replace those driven out by inadequate working conditions.   One element of 2042 required  the development of high stakes  performance assessment of California teachers (TPA) based upon the teacher performance expectations (TPE)   to be developed by the Commission on Teacher Credentialing.  As of 2009, three options have been approved;  1) TPA, 2) Performance Assessment for California Teachers (PACT), 3) Fresno Assessment for Student Teachers.   The following is an analysis of PACT.



Assessing PACT,

A state-mandated Teacher Performance Assessment for identifying highly qualified teachers.

Ann Berlak, 2008.



The US is in crisis.  The economy is in shambles, the degree of economic inequality is greater than at any time since the years preceding the Great Depression, and it is unclear to what extent the destruction of the environment can be undone. Over the past decade politicians have promulgated a regime of high stakes centralized standardized testing of K-12 students for the purpose of holding schools accountable for doing their part to address the social, political, and economic problems we face.  It is, however, now clear to most of us-- teachers, teacher educators, citizens and policy makers --that standardized testing is itself part of the problem, and certainly not a solution.  Like the economic policies of the past eight years, K-12 standardized testing is now widely recognized as a failed policy.

It is therefore stunning that state and national politicians and policy makers are presently mandating an additional system of high stakes testing in further pursuit of the improvement of K-12 schools.  They are requiring teaching credential programs to administer standardized high stakes tests to credential candidates.  They seem to believe that this will produce a greater number of “highly effective” teachers, and do so cost effectively.1  At a time when the budgets of virtually every social program across the country are being cut to the bone, and the Governor of California is asking public universities for still more budget cuts, politicians and policy makers are imposing upon teacher education programs a new, expensive and time-consuming assessment system of no proven value. And no one seems to know where the money to pay for it will come from.

In California each credential program has the choice between participating in the state-sponsored assessment system designed in consultation with Educational Testing Service,  CalTPA, or the Performance Assessment of California Teachers or PACT,.2  This paper will explore how PACT, the assessment system being implemented in about half the credential programs in California, including the University where I teach, affects teacher credentialing programs, the preparation of credential candidates, and, ultimately, K-12 schools. An assessment of PACT is vital not only to Californians, but to teacher educators and citizens across the country, since Linda Darling-Hammond of Stanford University, presently one of the most prominent teacher educators in the US, and the lead researcher in the development of PACT, is putting PACT forward as a national model for teacher assessment.

What PACT is and what it claims to do

 In 1998 the California Legislature mandated that credentialing programs institute a Teacher Performance Assessment that uses state-approved rating scales aligned with and derived from the California academic content standards and Teacher Performance Expectations (TPE’s) that were already well established criteria for teacher certification in the state. This was an effort to establish criteria for the Highly Qualified Teacher required under No Child Left Behind.  Since then  credentialing institutions have been overhauling their programs to bring them into compliance with the mandate.3 The result has been an unprecedented degree of state control over public education and a substantial transformation of credentialing programs in California.4

PACT was initiated by a consortium of schools of education at all the Universities of California (or UCs), Mills College and Stanford University, joined early on by three California State Universities or CSUs. (The UCs are not to be confused with CSUs which are of lower status, have fewer resources, heavier workloads and lower salaries.) The intent of the consortium was to create an alternative to the system constructed by the state and ETS. There have been two major published studies assessing the effects and value of PACT, a 2006 study by Raymond Pecheone,  PACT’s project director, and Ruth Chung, a member of the project’s research and program staffs,  the other by Chung, published in 2008.  These studies, conducted by PACT insiders, and funded by the University of California Office of the President and the Hewlett and Morgan Family Foundations, purport to document the validity, reliability, promise and usefulness of PACT.

The assessments are required by law to be performance-based, that is, they are expected to assess “evidence from teachers’ actual teaching practice” in “real world” contexts; they are to be, in the lingo, authentic assessments.5  PACT assesses two types of “performance,” the Teaching Event (TE) and several embedded signature assignments (ESAs). The Teaching Event is a fifteen-minute video-taped lesson taught by the candidate during the final semester of the certification program, accompanied by a portfolio of approximately fifty pages which includes multiple sources of data related to the TE: teaching plans, teaching artifacts, student work samples and written reflections and commentaries.  The guidelines specifying the elements of the portfolio are laid out in great detail in a 47 page Student Handbook. At many institutions, credential candidates enter the data on the TE (including the video) into a computer program called Taskstream.  The signature assignments are course assignments in specified content areas graded by the teacher of the course using state-approved criteria.

Using another Taskstream program, anonymous trained scorers, who are “calibrated” yearly to assure scorer consistency, assess the data students have submitted according to a series of PACT-established rubrics; the Taskstream program then transforms the assigned scores on each of the rubrics into a single number that is taken to represent the effectiveness of the individual credential candidate, and that number is sent to the state. If the submitted number is not below the state-determined cut-score, the candidate will have fulfilled the TPA criteria for earning a credential. 6

Its advocates believe that using PACT will improve teacher quality through its gate-keeping function, by directly improving the teaching of credential candidates, and by providing answers to many questions that have long plagued educational research:  how to make comparisons of effectiveness among candidates, establish the relative effectiveness of different teacher credential programs, refine programs, and improve the preparation of beginning teachers. 7  PACT advocates intend for PACT eventually to provide the missing link in the chain of evidence between classroom performance of credential candidates during their student teaching (as measured by PACT) and the achievement of the pupils they teach during their first year, as measured by standardized tests that remain the gold standard of school learning. 8

Though many teacher educators who have participated in piloting PACT privately question PACT’s value and catalogue its limitations, only a few discuss their concerns openly, and virtually nothing has been published that offers a systematic critique of Pecheone and Chung’s defense of PACT.  One reason for this reticence is made clear by an e-mail forwarded from a high-ranking official of the California State University August 1, 2008 to many CSU teacher educators.  The gist of the letter is that though the CSU expected the state to recognize the importance of appropriate funding to support the State-mandated Teacher Performance Assessments no funding will be forthcoming; nevertheless the law has gone into effect and campuses must comply. The message was clear: any student who completes a program at an institution not using PACT or another TPA will not be granted his or her credential.

Funding for teacher education, as for all public programs designed to serve the general welfare, continues its downward spiral, resulting in a student body that is increasingly and disproportionately white, and in larger class sizes, higher ratios of student teachers to supervisors, decreased numbers of supervisor visits per credential candidate, the elimination of course offerings, hikes in tuition and fees, and even proposals to close down entire teacher education programs. One Department Chair has resorted to asking Office Max to donate office supplies. Though reduced funding is diminishing teacher education programs, the impetus to implement the TPA continues unabated, at a scoring cost of approximately $400 per candidate.  There are a number of additional and significant costs including faculty training time and full and part time administrative cost as well. 9  The authors of the legislation likely had no idea of the consequences of mandating the TPA without funding it: the TPA legislation snuck in under the radar as part of an omnibus education bill and the issue of funding was not debated nor even raised10.

There seem to be three possibilities for funding PACT: drawing upon existing instructional resources at the expense of other identifiable and/or unidentifiable elements of programs (Chico State’s Department of Education has been allocated $100,000 for implementing PACT in 2009 and no one seems to know what part of the University budget the money is coming from); requiring faculty to do assessments as an overload, without compensation, and in contradiction of the Union contract (as also seems to be the case at Chico State);  and asking credential candidates to pay the cost of administering the tests themselves, adding an additional fee to their already skyrocketing tuition and burden of educational debt. As is the case with K-12 schools, the assumption seems to be that teacher education can do ever more with even less.

Though many teacher educators comply with PACT because they see no alternative, others do so because they genuinely believe PACT is on balance good and useful. Below I evaluate this later view.  I first identify and examine some basic assumptions about research in the social sciences that are fundamental to PACT. Next, I consider whether PACT is a valid and reliable way to evaluate teacher effectiveness. I then ask whether increasing state and federal control over teacher education through the use of PACT is likely to produce more effective teachers and contribute to reducing inequalities of opportunity and wealth.  Finally I explore the question of what is to be done.


Is PACT a valid and reliable measure of teacher performance?

Is PACT good science?

The traditional view of science: Science is science is science:  

Most of us--parents, legislators, policy makers and teacher educators-- have, from an early age, internalized the deep beliefs about science that inhabit our culture.  These include the beliefs that the only forms of investigation that can legitimately be called science are those that emulate the natural or physical sciences and that the methods used for studying the physical world are equally appropriate for studying human institutions and experiences. We have also come to take for granted that, as in the physical sciences, good educational research produces “findings” that are empirical, verifiable, objective and generalizable and are in no way affected by our preconceptions, beliefs, hopes, and biases.  Most of us have also come to identify assessment with measurement: the best science produces quantifiable evidence, and to expect that research findings can be applied to practice.11 

This view of science is recycled by politicians, bureaucrats and policy makers who have enshrined it in terms such as “scientific reading,” “evidence-based,” “data-driven,”  “outcomes-based,” “value-added,” and “scientific rigor” which are stitched into the logic of educational policy so seamlessly that this is now for the most part imperceptible. 12  Politicians reproduce it when they write standardized high stakes testing into legislation, 13 and include in the Federal Higher Education Act the requirement  that credential programs be evaluated based on graduates’ performance on licensing tests.  The National Council for the Accreditation of Teacher Education recycles this view of science by requiring that programs provide “objective” evidence of outcomes as they respond to each of the accreditation standards. The U.S. Department of Education keeps this view alive by funding the narrowest forms of research and by “web-scrubbing,” for example deleting research from ERIC digests that it considers unsupportive of the Department of Education’s view of research. 14  Given the pervasiveness of these assumptions it is not surprising that we accept this view of research mindlessly despite contrary views a few of us might have encountered in our formal training.

An alternative view of science  

For the past forty plus years (particularly since Thomas Kuhn’s The Structure of Scientific Revolutions) challenges to these “orthodox” or traditional assumptions have appeared in both the “hard” and the social sciences, including educational research. From these alternative viewpoints the concepts of interpretation, discovery, illumination, clarification, values, politics, human volition, and non-hierarchical investigative partnerships are central, and the pursuit of hard truth or objectivity is replaced with acknowledgment of uncertainty and the understanding that we inevitably perceive “data” or “evidence” through lenses that are shaped by what we have learned to see and value. This view of science is collaborative rather than imperious, and modest rather than megalomaniac.  It rejects the notion that standardized tests, rubrics and measurements are untouched by human judgment, standing outside history, apart from the flux of everyday events, able to yield precise, quantifiable readings of human relationships and institutions.  From this perspective science is a human endeavor marked by uncertainty and controversy.

PACT’s romance with science The traditional view of science is deeply embedded in PACT (and in CalTPA) and in a number of claims Darling-Hammond and Pecheone and Chung make about PACT, for example, the claim that at some time in the future PACT scores will be able to predict who will become a good teacher, that aggregated mean scores for candidates within a program are a useful basis for internal and external reviews of programs, course development, and accreditation,15 and that establishing inter-rater reliability will enable scorers to identify teacher effectiveness objectively. Nor is there, in the research and commentary in support of PACT, even a hint that the “traditional” conceptions of social scientific research are contested.

Is PACT a valid assessment of the highly qualified teacher? 16

PACT is touted as an evidence-based, valid and reliable assessment instrument for enforcing professional standards and assessing teacher effectiveness. First, let’s consider its validity.  An assessment instrument is valid to the extent that it accurately and fairly assesses the concept, idea, performance activity or belief it claims to assess.  In the case of PACT this means that the mean of the scores assigned to a candidate by a trained rater represents the degree of the candidate’s future effectiveness as a teacher.

How is effective teaching conceptualized or defined?  In order to decide if this number accurately or validly represents (and predicts) effective teaching we first need a common idea of what effective teaching is. According to Linda Darling-Hammond such agreement is not problematic. She writes, “(T)he question  (of what teachers ought to know and be able to do to be considered effective) is easier to address than it once was because… (there are) now performance-based standards developed during the past decade… for beginning teacher licensing that have been adapted or adopted in more than thirty states and reflect a consensual, research- grounded view of what teachers know and should be able to do.” 17

The claim that there is a consensual view of effective teaching raises some thorny issues. First, defining quality always involves value judgments.  Conceptions of effective teachers are inseparable from and as highly politicized as conceptions of the good society.  It seems obvious to me that disagreements about the qualities of a good teacher like disagreements over what makes a good society, do and should abound. Thus, assessing effective teaching is not, as Darling-Hammond claims, an objective, “research-grounded” empirical process.

I will give only two of many possible examples of bias in the PACT definition of good teaching.  One is suggested by Pecheone and Chung’s musings about why candidates teaching in “suburban schools” get higher PACT scores than those in “urban” or “inner-city” schools. Pecheone and Chung’s hypothesis is that the differences in scores can be attributed to credential candidates in urban settings experiencing more constraints on their teaching decisions related to district-mandated curricula.18 Pecheone and Chung do not entertain an alternative possibility--that the source of these difference in scores might be that the PACT conception of the effective teacher is a generic one that excludes qualities of effective teaching that may be specific to teachers of non-dominant (non- “suburban”, non- “inner city”) racial and linguistic groups and social classes.  The PACT conception of an effective teacher may be weighted towards a conception of an effective suburban teacher.

Consider also how the requirement that credential candidates demonstrate their evaluation skills by evaluating student learning during the one-week period of the TE affects the concept of effective teaching.  Evaluating candidates’ assessment skills within this time frame in effect requires candidates to assess relatively simple skills such as spelling or adding fractions that can be taught and assessed within a week.  However, in the time frame provided by the TE it is not possible to ascertain how well the candidate assesses more complex and long term goals, such as the development of mutual respect between Asian and Latino students, or the growth of student’s sense of agency. Though such learning may be highly valued in some conceptions of effective teaching, it cannot be recognized within the PACT.  Note also that empathy, enthusiasm, fairness, respect, commitment, and other so-called dispositions have no place in PACT’s conception of the effective teacher. 19

Highly effective teachers and social justice. That anyone would contest the idea of effective teaching implicit in the PACT rubrics may initially seem puzzling, particularly to people who have a liberal-mainstream view of effective teaching. This is because the PACT conception accords with their common sense. Who would for example quarrel with the State’s requirement that credential candidates be able to

“List the content standards that are the target of student learning  (list the complete text of the relevant parts of each standard):  (TPE 1)”


as part of their documentation of the Teaching Event. However, implicit in the State Content Standards is a particular viewpoint that not everyone endorses about what is most important for children to learn. For example, it is of great concern to some that there is no mention of racism, poverty, sexism, or income inequalities in the state content standards, nor any alternative to the view that US foreign policy might be motivated by reasons other than humanitarianism. 20 TPE 1 in effect defines good teaching as teaching uncritically what the state mandates. Thus the PACT Teaching Event assessment corrals teacher educators and their students into a singular and narrow conception of good teaching and the good society to which the promotion of social justice and critical thinking about the larger purposes of schooling are  peripheral.

PACT is particularly troubling to those who believe that central purposes of schooling should include promoting a vigorous, more egalitarian and sane democracy and preparing children to lead rich and generous lives, not preparation for jobs that are for the most part dull, boring, meaningless and poorly paid, nor improving the competitiveness of US workers in the global economy.21 Educators who want schools to promote social justice consider it essential that teachers look critically at the standards they are expected to promote. 22

 Credential programs committed to social justice can always add to the TPEs. as the Department of Bilingual/Multicultural Education at California State University Sacramento did.  Finding the state-mandated Teacher Performance Expectations that underlie PACT too limited, the faculty voted to add the following to the State’s 13 TPEs:


TPE 14. Candidate exhibits a commitment to democratic practices, intercultural communication, and civic participation, including rejection of all forms of discrimination based upon race, ethnicity, social class, gender, gender identity, age, ability, sexual orientation, and political beliefs, among others.


TPE 15. Candidate develops an educational environment that fosters personal and professional integrity, civility, respect and support for students from diverse cultural and ethnic groups.


Most if not all of those who constructed PACT are, I am sure, very well-intentioned, and would likely would not oppose these additions.  However, it is axiomatic that what is not required by high stakes tests will become marginalized, and what is required will overwhelm other considerations. A single standard or definition of effective teaching promulgated state- and nation-wide is likely to exclude the interests of those on the margins and to marginalize the cultivation of creativity in both teachers and their students.  Perhaps this can be illuminated by imagining what criteria for assessing teacher effectiveness would be central to an alliance of teachers and poor parents who are primarily concerned with their impotence within the political order and/or whose primary and overriding goal is to reduce race and class inequalities in schools.

Is PACT a valid assessment of performance?  PACT claims that it assesses teachers’ actual teaching performance in “real world” contexts, rather than teachers’ knowledge about or dispositions towards teaching. It’s not quite so simple as that.  Judgments about the quality of a piece of art-work or a player’s performance in a particular foot-ball game are examples of assessments where no inferences are necessarily made about underlying traits or skills that the products or observed behavior represent. (“It’s a wonderful painting,” not necessarily, “she’s a talented artist.”) In contrast, we are expected to make inferences from the TE about candidates’ ongoing qualities or competencies such as generalized planning or assessing skills.23

There are a number of reasons to question the validity of making inferences about candidates’ skills by observing a fifteen-minute video-tape of the TE, and then rating it  according to the PACT protocol. I will suggest just a few.   First, the fifteen- minute video-taped Teaching Event is highly contrived.  Since the children who are videoed must have received written permission from their parents, they know this is not a “normal” lesson. The children may have been prompted to be on their best behavior or promised rewards for behaving well; the video may be shot when the children most difficult to manage are out of the room. In at least one case normal seating arrangements were changed to make sure students who had not been given permission were at the periphery.  In addition some credential candidates may be highly anxious about being video-taped, particularly if they know the tape will be assessed by a stranger, to whose questions and concerns they will be unable to respond.  It would not be surprising if credential candidates chose simple relatively foolproof lessons for their teaching event.  If so the assessment would say little about the capacity of the candidates to teach arguably more important complex concepts and ideas.

Though Pecheone and Chung claim that PACT encourages and captures a unified and integrated learning sequence, it is expected that “(T)he focus of the PACT assessments (be).. on candidates’ application of subject-specific pedagogical knowledge that research finds to be associated with successful teaching.”24 I think what this means is that each lesson must apply theory to practice.  It is no news that practitioners do not normally alter their practice on the basis of research findings, and that there has long been a vigorous debate about the relation of theory to teaching practice. This requirement to teach a lesson that is “research based” likely has the effect of diverting the candidate from an intuitive and creative performance, and suggests a shocking lack of attention to the last twenty years of utilization research—research on the viability of ‘applying theory to practice’--- that indicates that top-down, research and development models of the 1950’s and 1960’s didn’t work. 25

How valid is assessment using rubrics? 26 Rubrics are sets of criteria intended to enable an “objective” determination of the quality of performances or responses. To evaluate the role the rubrics play in the PACT assessments consider as an example one of its eleven rubrics for assessing the “knowledge base” of elementary literacy.  The concept knowledge base is divided into four domains: planning, teaching, assessing and reflection.   Each of these four domains is assessed on a minimum of two rubrics.  Figure One is an example of one of the rubrics for planning.






EH1:  How do the plans structure student learning of developmentally appropriate[1] analytic reasoning skills in history or social science?  (TPEs 1,4,9)

Level 1

Level 2

Level 3

Level 4

The standards, learning objectives, learning tasks, and assessments either have no central focus or a one-dimensional focus (e.g., solely on facts or planning activities that do not engage students in the use of analytic reasoning skills).

The standards, learning objectives, learning tasks, and assessments have an overall focus that is primarily one-dimensional (e.g., learning facts or a singular interpretation of a topic in history/social science).

·               The focus provides students an opportunity to use facts and concepts to make interpretations or judgments about a topic in history or social science.

Learning tasks or the set of assessment tasks focus on multiple dimensions of history-social science learning through clear connections among facts, concepts, interpretations, and judgments about a topic in history or social science.

·               A progression of learning tasks and assessments is planned to build understanding of the central focus of the learning segment.

Both learning tasks and the set of assessment tasks focus on multiple dimensions of history-social science learning through clear connections among facts, concepts, interpretations, and judgments about  a topic in history or social science.

·               A progression of learning tasks and assessments guides students to build deep understandings of the central focus of the learning segment.


After an assessor rates the Teaching Event (the video and porfolio) on each of the four rubrics that taken together comprise “planning,” using a four-point or four-level scale exemplified in the rubric above, the Taskstream program computes a mean score for planning.  Scorers follow the same process for each of the remaining domains and then one mean of the scores of each of the domains is calculated and sent to the State as a measure of the degree of the candidate’s teaching effectiveness.

In spite of claims to the contrary each rubric by no means assesses a unitary concept or process.  When calculating a score on the rubric above, for example, the rater must assign a single number that represents the extent to which planning “establishes a balanced instructional focus.” As is clear in the above rubric, to assign a three on that rubric requires the rater to make inferences about at least three separate skills: the ability to “focus on multiple dimensions of history social science learning,” to “make clear connections between concepts,” and to plan a “progression” of the concepts.  Scorers can go back again and again over the fifteen minute TE and portfolio and make an individual decision about how to weigh each of these three skills when calculating the single number (and then rate all the other rubrics as well), though, given the enormous complexity of and time involved in scoring even this one rubric, it seems unlikely they will do so.  However, the candidates, for their part, must teach a fifteen-minute lesson from which they know a scorer is supposed to infer all the underlying traits or capacities that are, according to the rubrics, components of “establishing a balanced instructional focus,” as well as all the other capacities that are components of all the other rubrics as well. These pre-established PACT rubrics will certainly leave no time for teacher educators to teach qualities other than those the creators of PACT have defined as essential elements of teacher effectiveness. The socially acceptable terminology for preparing students to score well on the PACT rubrics is curricular alignment, a euphemism for ‘teaching to the test.’ If credential candidates are given the rubrics ahead of time and their futures are dependant on their scores, who among them would not focus to the highest degree possible on fulfilling the criteria by which they will be scored?

Like other forms of assessment, rubrics require the raters (and “ratees”) to pay attention to the limited number of teaching qualities the creators of the assessment consider most important. Though the use of rubrics takes for granted that the mean score of the rater-assigned numbers for each rubric is a valid (and reliable) measure of how good the teaching is, in fact the score only identifies the candidates’ degree of compliance to the criteria in the rubric, criteria that have been established by particular human beings who, like the rest of us, have preferences for and biases about the qualities of effective teaching.  Many critics of using rubrics to assess writing or other forms of creative or artistic expression share the view that high scores based upon assessments using rubrics do not mean that the product or process is necessarily good. Their argument is that rubric assessments cannot take into consideration aspects of the overall effect of the process that are achieved by means not anticipated in the scoring criteria. Rubric assessments do not, in effect, give points for creativity.

Pecheone & Chung’s research indicates a strong level of agreement between scorer’s assessments of the TE and holistic judgments by faculty and supervisors, and claim that this data confirms the validity of the TE scoring.27  They do not, however, address the question of what value the TE scoring process adds. There is no evidence that this time consuming, unwieldy and invalid assessment process is superior to the clinical supervision and evaluation of candidates that was in place before the advent of PACT.

  How reliable are PACT scores?  For scores to be considered reliable, raters must learn to assign candidates quite similar scores—in the case of PACT no more than a one points difference on a four-point scale. To achieve this raters must undergo a “calibration” process where they are all supposed to learn to make the same inferences from their common observations.  For example, for the rubric above, scorers must come to agreement about whether or not the credential candidate has “built a deep understanding of the central literacy focus of the learning segment” (scoring a four) or simply “built an understanding” (scored as a three). When I was trained as a PACT scorer I and the one other member of the group who is a specialist in second language acquisition gave a score of one to the teacher in the video part of the TE with respect to a rubric assessing the teaching of second language learners.  All the others assigned a 3 or 4. It was the end of the day; everyone was impatient to finish up; there was no time left to discuss our differences, and the training required that we reach consensus within one point.  The two of us agreed to change our scores to three.

What beliefs support the assumption that an assessment is more reliable if it is done by a stranger during a two to three hour period rather than by someone (such as a student teaching supervisor) who gets to observe the credential candidate and the candidate’s schooling context over a period of time? The stranger-assessor must assess tasks such as : 


Briefly describe the following:

a.     Type of school/program in which you teach, (e.g.,  elementary/middle school, themed magnet, or charter school)

b.     Kind of class you are teaching (e.g., third grade self-contained, sixth grade core English/social science) and organization of subject in school (e.g., departmentalized, interdisciplinary teams)

c.     Degree of ability grouping or tracking, if any

Here the assessor must determine, among other things, if the credential candidate is able to identify tracking or ability grouping.  Whether or not candidates can do so is important since tracking, frequently a site of institutional racism and classism, may initially be invisible to them.  To accurately assess candidates’ responses to this question, stranger-assessors would need to observe candidates’ classrooms and schools themselves. This is only one example of one element of the PACT assessment that initially appears quite simple but actually requires a highly discerning observer who spends extended time observing and dialoguing with candidates in order to make valid judgments about their knowledge. It is likely that not all assessors, who are paid by the “head’ if they are paid at all, will take the time required to make thoughtful and nuanced assessments of candidates’ performances on every rubric; some are likely to earn an average of twenty dollars an hour (when the time for training and calibrating is factored in) as “efficiently” as possible.

The most useful assessments of credential candidates’ performances are created through dialogues between the candidate and an experienced professional who build a common understanding of the observed lesson together as they listen to and learn from one another. There is no place for such dialogues in the scoring of the TE. There is a space on Taskstream for scorers to write short explanations for the scores they assign on a given rubric, but if a candidate believes the scorer has made an inferences that completely misses the mark or is unwarranted he or she has no recourse. Perhaps this matters little if the candidate gets a passing score.  However, much thought, time and money will still  have been spent that could have been put to so much better use.

Intuitively it may seem fair for everyone to be judged by the same criteria, and for assessors to be trained to apply the criteria as similarly as possible. However, rubric scoring always requires interpretation.  Irreducible, irresolvable and utterly necessary, interpretation is constitutive of science itself. 28 At what cost do we strive to create scorers who achieve high levels of agreement with one another by making the same interpretations of the same phenomena?  Can we be certain that reliable PACT scores are not the product of collective tunnel vision?


PACT in Practice

PACT and the chain of evidence: the research value of PACT. 

Pecheone and Chung describe the chain of evidence they hope to establish as follows:

(A) longitudinal study of the predictive validity of the PACT assessment is currently being planned.. Credentialed teachers who have completed the PACT will be followed into their 1st years of teaching to examine…the relationship of their scores on the PACT assessment with their students’ achievement gains.  If it can be shown that performance (on PACT)… significantly correlates with student learning, …then PACT would have a compelling evidence base to support (its) .. use to credential prospective teachers. 29


Realizing this dream of building a chain of evidence faces many obstacles. We should not be mystified by Darling-Hammond’s claim that the use of  “appropriate statistical controls” including  “multivariate multi-level analyses”26 allows us to rule out “extraneous” factors, such as social class.  “Measurement of a teacher’s success in promoting learning is so fraught with psychometric problems and the limits of causal models so great given the preponderance of interaction effects that no current assessment system can do the job” 30

Pecheone and Chung’s reference to student learning as the end point in the chain of evidence is shorthand (in spite of references to multiple measures) for student achievement as measured by students’ standardized test scores. The search for a chain of evidence that links two sets of standardized test scores can only result in a massive bureaucratic system seeking control over what people learn in schools, a system that represses or ignores both dissent and diversity and fosters docile and compliant teachers and citizens.

 Assessing PACT claims about its effects on learning to teach

Pecheone and Chung and Darling-Hammond promise many positive outcomes from using PACT, among them much stronger skills for a wider range of credential candidates 31 To date, however, their research seems to focus on candidates’ self-reports and subjective perceptions that they both liked the performance assessments and found them useful. 32 The two PACT studies suggest there were a significant number of credential candidates who viewed PACT as an interruption of rather than an asset to their learning. This seems to be true of many students in our program as well.

When evaluating the significance of such self reports it is important to know student’s points of comparison to the PACT-driven programs.  Perhaps the students might have preferred the credential program offered prior to the introduction of PACT, of which they had no knowledge.  An analogy would be to automobile workers who worked in teams to build entire automobiles, until the option to work in teams disappeared with the introduction of the assembly line.   It took just two generations for workers to lose memory of the experience of building an automobile together, and to come to accept the assembly line as “the way it is.” Like the assembly line workers, prospective teachers may be content with PACT because they have had no opportunity to experience or envision an alternative.

Darling-Hammond claims that PACT will reduce the achievement gap by “replacing the decentralized non-system of US education with a more systematic professional policy.” 33 This is not the place to rehearse the many and diverse explanations of the so-called achievement gap.  The alternative ways of conceptualizing the vastly unequal outcomes of schooling—“the education debt,” 34“the opportunity gap,” for example--- suggest that our understanding of the origins of the gap and the related notions of how to address it range widely.  I see no reason to believe that this long-standing problem will be addressed by a centralized professional program that has PACT rubrics at its center.

 How PACT and the accountability movement impacted our program: a personal reflection

What I and many of my colleagues noticed from the time we began, hurriedly and without released time or any other form of support, to revise our curriculum to bring it into compliance with the new state laws mandating Teacher Performance Assessments was increasing student dissatisfaction with our program and loss of faculty morale. A number of experienced faculty took early retirement.  The piloting of PACT four years later accelerated these processes.  It also promoted a reduction of time spent by faculty supervising student teachers, a focus in teaching seminars on technical issues surrounding PACT rather than on deeper issues facing schooling and society, and a  significant redirection of professional development towards the implementation of PACT. Once PACT was introduced almost every faculty meeting focused on how we could satisfy the PACT requirements, how to integrate PACT into existing course components, and how to incorporate the scoring of the PACT TE into the contracted responsibilities of supervisors and cooperating teachers.

Given that that the advent of the TPA coincided with the most intense period of implementation of NCLB and scripted curriculum, it goes without saying that PACT was not the only source of discouragement.  But PACT’s focus on content standards, that are the first consideration in every lesson plan for PACT,  went hand in hand with the scripted curriculum mandated by No Child Left Behind. Presently there is some evidence that in some school districts in the Bay Area the pressure to follow scripted curricula is at last subsiding; it is indeed ironic that, in spite of this shift, the pressure for credential candidates to focus on the TPA continues with forceful and unwavering political and administrative support.

Credential candidates’ preparation for the TE is a very labor-intensive process that begins during the first semester and dominates the workload of the final semester, a time when students formerly focused their energies on student teaching. Presently in at least some PACT institutions, cooperating teachers, course instructors, and supervisors have reduced their expectations to make way for PACT, and the time student teachers have to prepare creative and challenging teaching projects has declined.


Hierarchy and power

The introduction of Teacher Performance Assessments in California signified a definitive transfer of power over teacher education from university-based teacher education to the state and federal governments and their agents. Credential programs were told that each institution could construct its own Teacher Performance Assessment, though the TPA had, of course, to be approved by the State.  On the face of it the opportunity for each credential program to write its own TPA appears egalitarian.  However, the results were exceedingly hierarchical.  Only the foundation-supported and well-connected UCs had the resources to construct and implement an alternative.  CSU Dominguez Hills wrote an alternative in pursuit of “an elegant and rich assessment system” but after three years decided to join PACT “in order to ameliorate the costs associated with an independently designed system” 35  It takes a lot of time and money to construct a valid and reliable instrument that satisfies state regulations.  Only the most prestigious and wealthiest institutions had the resources to do so. 36  In the conclusion of their paper  Pecheone and Chung wrote:

“Despite legitimate concerns about the costs of implementing performance-based assessments, the alternative is to judge teacher competence solely on the basis of standardized multiple-choice tests of content and/or pedagogical knowledge.” 37


The assumption that PACT and multiple-choice tests are the only two alternatives will be dealt with below.  Here I want to note that the issue of cost, here passed off as of minor concern, is a central issue at the State Universities which are vastly under-resourced compared to UCs..

My hunch is that the writers and creators of PACT and their students engaged in  discussions and debates surrounding  PACT’s construction and that these experiences contributed to some of the positive effects they claim for PACT.  However, the experience of the creators of any system is bound to be quite different from the experience of the system-users who, in this case, found themselves at the bottom of the pecking order, implementing a pre-fabricated, already “packaged” assessment system.  Though it is claimed that the new reforms are now owned and sustained by the profession,38 this is only true if one defines “the profession” very narrowly, excluding the many teacher educators and credential candidates who have had no input or buy-in into PACT. There is no better way to destroy morale and to promote mindless implementation than to impose conformity and obedience to dictates from above.

 PACT was the creation of a well-funded elite and its deputies from the beginning. It depended heavily upon the financial support of the Carnegie Foundation and the Office of the President of Stanford.  Pecheone and Chung make the role of funding sources quite clear:  “The cost of developing PACT could not have been accomplished without… financial support from the university of California office of the President and corporate funded foundations.” 39 The important task of creating the rubrics that define and proclaim the qualities of highly qualified teachers was given to selected faculty and doctoral students, primarily from the high status UCs and private universities, and their appointees who held like-minded views of scientific research and of the qualities of highly effective teachers with foundation-funders.  If the views of those in leadership positions or those who did the work of constructing and defending the details of PACT had diverged too much from those of their funders or failed to see and understand science and scientific research the way the funders did, they would not have been given the opportunity to create and research PACT in the first place, or would soon have been dismissed.

The assumption that PACT’s authors have a particular entitlement to assess teacher education programs is evident in Darling-Hammond’s statement that our group (the creators and developers of PACT) “can play a special role in developing leaders who have sophisticated knowledge of teaching and are prepared not only to practice effectively in the classroom but also to take into account the bigger picture of schools and schooling.”40

The testing frenzy, including PACT and other forms of TPA, is a diversion from confronting the realities of inequality and schooling.  This frenzy has been likened to a political spectacle:  “Pure theater with no other purpose than to look like something positive is happening, whereas it is not.”41 Although the call is to ensure highly qualified teacher, there is no evidence to suggest that credentialed teachers as a group are not highly qualified, 42 and there certainly are other more productive roads to improving schools than TPAs----paying teachers more, providing better facilities, lowering class size, and insuring that highly qualified teachers end up teaching in predominantly poor, Latino, and Black schools. The massive campaign to hold teacher educators accountable is in part a way to name, analyze, and normalize the disproportionate educational failures of poor and Black and Brown children in a “scientific” enough way, ultimately, to gain assent, and minimize resistance, to privatization.  If the intent of the TPA were really to improve teaching and schooling then at the very least provision would have been made to provide feedback from the TE assessment to credential candidates that would help them improve their teaching.  However, there is no meaningful provision for such feedback. The story of PACT is just a permutation of an old story, nothing new. 


Conclusion: What is to be Done?

Larry Cuban of Stanford University “characterized the momentum building in the late 1980s for national tests and curriculum as a train rushing down a track.  He asked whether scholars should accommodate to what appeared by then to be almost a political reality—by helping to build a better track for the train… or whether… to use their expertise to try publicly to slow down the train by speaking out to lay and professional audiences. ” 43 As Marilyn Cochran-Smith writes, the train Cuban described is now streaking through a deep dark tunnel that instead of light at its end may well have a concrete wall.44  PACT may be seen as a well-intentioned but ultimately flawed attempt to build a better track.

To use another analogy, the teacher performance assessments are a house of cards.  Each of us could suggest policies that would be more likely to improve the schools, narrow the opportunity gap, and address issues of inequality more productively than teacher performance assessments. There are at least two reasons teacher educators are simply going along with the mandates.  One is the deep hold that PACT and similarly flawed assessment policies have upon many teacher educators’ minds. As a nation and a profession we have come to accept unquestioningly a view of science that ultimately maintains hierarchies of power.   As a beginning we need to re-think our conceptions of assessment and science.

The second reason teacher educators are not resisting the TPA is that fear of challenging authority is deeply embedded in most of us, who have become accustomed to being at the bottom of the academic heap. Challenging state mandated TPA will require a major change in the degree of respect accorded to teacher educators.  This will only follow a major change in how teacher educators view ourselves.  Included in the 2008 reauthorization of the Higher Education Act is the mandate that the Department of Education must toughen standards for teacher education programs. In the same document is another mandate: that the federal government is barred from issuing regulations governing higher education by measuring (non-teacher credential) student learning outcomes.  This is the clearest expression I have seen of the low status and regard accorded teacher educators.  Some teacher educators view the TPA as direct attack by the state and federal governments on our integrity and academic freedom, and a devaluation of our wealth of cumulative knowledge and experience, but most of us view it as our just deserts. We have not learned to stand up as professionals and proclaim that the emperor has no clothes.

I see speaking out as an obligation for all who intend to prepare teachers to promote a just and democratic society. How can we as teachers and scholars speak out against policies if we believe they are seriously flawed in logic and evidence?  The Teacher Education Caucus of the California Faculty Association (CFA) has been organizing opposition to the TPA.  The CFA is filing a system wide work-load grievance to address the increased workloads that result from the unfunded TPA mandate.  At several campuses Statutory Grievances have been filed challenging PACT in terms of its assault on academic freedom.  So far we have received only limited support. Nothing will happen unless more teacher educators join in.  At the very least we need to act as role models for succeeding generations of teachers.



Thanks particularly to Roberta Ahlquist, Harold Berlak and Duane Campbell for their excellent feedback.



Berlak, Ann (2003) Who’s in  charge here? Teacher education and 2042.  In Teacher Education Quarterly, Vol 30, No.1 Winter 31-41


Berlak, Harold (2000) Cultural politics, the science of assessment and democratic renewal of public education.  In Ann Filer (ed.) Assessment: social practice and social product.  London: Falmer press.


Berlak, Harold et. al. (1992) Towards a new science of educational testing and assessment. Albany, N.Y.: SUNY Press


Bracey, Gerald (2003) Schools should not prepare students for the world of work.  In Bracy, On the Death of Childhood. Portsmouth: Heinemann, 30-44.


Berliner, David & and Bruce Biddle (1995) The manufactured crisis.  New Yorrk: HarrperCollins


Berliner, David (2005) The near impossibility of testing for teacher quality. Journal of Teacher Education, Vol. 56, No. 3, May/June 205-213


Chung, Ruth (2008) Beyond Assessment: Performance Assessments.  Teacher Education.  Teacher Education Quarterly, winter, 7-28


Cochran-Smith (2006) Evidence, efficacy and effectiveness: Introduction to the double issue.  Journal of Teacher Education Volume 57, # 1, January/February. 3-5


Cochran-Smith, Marilyn (2005a).  Studying teacher education. What we know and need to know. Journal of Teacher Education Vol. 57, No.4, September/October 301-306


Cochran-Smith, Marilyn (2005b) 2005 Presidential address: The new teacher education for better or for worse.  Educational Researcher, Vol. 34, No. 7 October. 3-18


Cochran-Smith, Marilyn (2005c) Teacher education and the outcomes trap.  Journal of Teacher Education, Vol. 56, No. 5.  November/December.  411-417.


Darling-Hammond, Linda  (2006)  Assessing teacher education: The usefulness of multiple measures for assessing program outcomes. Journal of Teacher Education, Vol. 57, No. 2 March/April. 120-138


Darling-Hammond, Linda (1994) National standards and assessment.  American Journal of Education. 478-510


Delandshere, Ginette & Anthony Petrosky (1998) Assessment of complex performances: Limitations of key measurement Assumptions.  Educational Researcher, Vol. 26, No. 2, 14-24


Eisner, Elliot W. (1999) The uses and limits of performance assessment Phi Delta Kappan. May.  658-60


Keller, Evelyn Fox, (1985) Reflections on gender and science.  New Haven: Yale University Press


Kohn, Alfie (2006) The trouble with rubrics.  English Journal, Vol. 95,  No. 4. March, 12-14


Lather, Patti (2003) This IS your father’s paradigm: Government intrusion and the case of qualitative research in education.  Guba Lecture, American Educational Research Association, Chicago, April


Lather, Patti (2005) Foulcauldian scientificity: Rethinking the research, policy and practice nexus.  American Educational Studies Association, Nov. 2005


Ladson-Billings, Gloria. (2006) From the achievement gap to the education debt: Understanding achievement in U.S. schools Educational Researcher Vol. 35, No.7. October. 2-13


Mabry, Linda (1999) Writing to the rubric.  Phi Delta Kappan. Vol. 80.9, 673-79


Messick, Samuel (1994) The interplay of evidence and consequences in the validation of performance assessments.  Educational Researcher. Vol. 23, #2, March.  13-23


Nelson, Thomas. (2003) Editor’s introduction: In response to increasing state and national control over the teacher education profession. Teacher Education Quarterly.  Vol. 30, No. 1, Winter . 19-31


Pecheone, Raymond l. & Ruth R. Chung, (2006) Evidence in teacher education.  The performance assessment for California Teachers (PACT).  Journal of Teacher Education.  Vol. 57, No. 1 January/February , 22-36


Russell, Sharon (2006) Reforming urban teacher education: SB 2042 implementation five years later.  Issues in Teacher Education, Vol.15, #1, Spring. 27-51


Sandy, Mary (2006).  Timing is everything: Building state policy on teacher credentialing in an era of multiple, competing and rapid educational reforms.  Issues in Teacher Education, Vol 15, No.1, Spring.  7-19


St. Clair, Ralf (2005) Similarities and superunknowns: An essay on the challenges of educational research.  In Harvard Educational research Vol. 75, Number four. Winter 435-454


Sleeter, Christine (2003) Reform and control: An analysis of SB 2042. Teacher Education Quarterly.  Vol. 30, NO. 1, Winter . 19-31


Sylvester, Paula M., Deborah Summers, & Edward F. Williams (2006), Costs and benefits of accountability: A case study of Credential Candidates’ performance assessments. Issues in Teacher Education, Vol. 15, #1, Spring. 21-35.


Villegas, Ana Maria (2007) Dispositions in Teacher Education.  Journal of teacher education, Vol. 58, No.5, Novembr/December 370-380.


Wineberg, Mona S., Evidence in Teacher preparation.  Establishing a framework for accountability.  Journal of Teacher Education, Vol. 57 No. 1, January/February 51-64


Wolf, Dennie & Nancy Pistone (1991). Taking full measure: rethinking assessment through the arts.  New York: College entrance examination board.



1 See Darling-Hammond, 2006

2. See Nelson, 2003, Berlak, A., 2003, Sandy, 2006

[1]   Both the content and skills that are the focus of the learning segment should be appropriate for the grade level taught (K-2; 3-5; 6-8).  Candidates and scorers should consult the “Historical and Social Sciences Analysis Skills” on pages 1 and 21 of the History-Social Science Content Standards for California Public Schools.


3 This is in addition to existing requirements including passing the CSET, a state-administered test of general knowledge, maintaining a B or better GPA in their credential courses, completing student teaching successfully, and passing the state required RICA--a test of knowledge about teaching reading. Credential candidates must pay for taking (and re-taking if necessary) each of these tests.  See Russell, 2006, A. Berlak, 2003, Sandy, 2006,  Chung, 2008 for Teacher Performance Assessment history  in California.

4 Sandy, 2006, 700

5 Pecheone & Chung,2006, 23.The pencil and paper aspects of PACT and CalTPA, including the signature assignments are considered performance based assessments. Performance refers not only to performance in the classroom but also performance of the sort of work teachers do outside the classroom.  See Darling-Hammond, 2006(a), 13. and 2006(b), 123.

6 The information in this paragraph is drawn from Pecheone and Chung, 2006, 24-25. To earn a credential students must fulfill a number of other requirements, including coursework, student teaching, and several additional pencil and paper tests. See footnote #3

7 Pecheone and Chung,2006, 24

8 Pecheone and Chung, 2006, 32.  Pecheone and Chung do not specify how the “achievement gains” (33) will be assessed.  However, there is no conceivable alternative in a study of this complexity to the used of standardized test scores.

9 The greatest cost of PACT is paying for the assessments.  However, at our university money was also spent to fund a full time position to organize the TPA, and additional space to store portfolios was required.  The time required for curriculum revision increased workload without reimbursement.

10 SB2042, passed in 1998 which established teacher performance assessments stated explicitly that the implementation of teacher performance assessment is subject to the availability of funds. SB 1209 (2006) which mandated the TPA was unfunded.

11 See Delanshere (1998) for an excellent analysis of the respect for quantification and Keller, 1985, for an analysis of ideas about the relation of theory to practice.

12  Cochran-Smith, 2005, 4

13  This quotation and the claims in the preceding paragraph can be found in Lather, 2003, 1, 2

14 See Lather 2003,6.  I depend heavily upon Lather’s analysis in this and the following paragraphs.

15Pecheone and Chung, 2006, 25

16 The following critique of validity in Pecheone and Chung, 2006, holds for Chung,2008, as well since it is based in the same view of scientific research.

17 Darling-Hammond, 2006b, 123

18 Pecheone and Chung, 2006, 29

19 See Villegas , 2007, for a persuasive argument regarding the assessment of dispositions.

20 . See Sleeter ,2003. for an excellent  critique of the state content standards.

21 See Bracey,  2004

22.. Cochran-Smith, 2003,11,13; Cochran -Smith,2005c, 415; Eisner, 1999. Sleeter, 2003,8

23. Messick, 1994, 14,16


24 Pecheone & Chung, 2006, 23.

25 Lather, 2002, 2

26 The discussion of rubrics is drawn from Mabry, 1999; Eisner, 1999 and Kohn, 2006

27 . Pecheone & Chung, 2006, 31

28 Lather, 2003

29 Pecheone and Chung, 2006, 33;  See also Darling-Hammond, 2006, 132.

30 Berliner, 2005, 208.

31 Pecheone and Chung, 2006, 24

32. See Pecheone & Chung, 2006 & Chung, 2008.

33.Darling-Hammond, 2006a, 18

34 See Ladson-Billings, 2006

35 Russell, 2006

36 Darling-Hammond, 2006, conveys the process by which PACT was created over a period time by a collaborative group of people who were supported in ongoing dialogue.  The difference between implementing a process that was created by those who implemented and revised it and those who had no say in creating it cannot be overstated.

 37 Pecheone and Chung, 2006,33

38 See Sandy, 2006

39 Pecheone and Chung 2006, 33

40 Darling-Hammond, 2006, 122

41 See Berliner,2005,  for a persuasive argument supporting the notion of testing as political spectacle, 205

42 Berliner, David & and Bruce Biddle (1995)

45 Cochran-Smith, 2005b, 14

44 Ibid





28 Lather, 2003

29 Pecheone and Chung, 2006, 33;  See also Darling-Hammond, 2006, 132.

30 Berliner, 2005, 208.

31 Pecheone and Chung, 2006, 24

32. See Pecheone & Chung, 2006 & Chung, 2008.

33.Darling-Hammond, 2006a, 18

34 See Ladson-Billings, 2006

35 Russell, 2006

36 Darling-Hammond, 2006, conveys the process by which PACT was created over a period time by a collaborative group of people who were supported in ongoing dialogue.  The difference between implementing a process that was created by those who implemented and revised it and those who had no say in creating it cannot be overstated.

 37 Pecheone and Chung, 2006,33

38 See Sandy, 2006

39 Pecheone and Chung 2006, 33

40 Darling-Hammond, 2006, 122

41 See Berliner,2005,  for a persuasive argument supporting the notion of testing as political spectacle, 205

42 Berliner, David & and Bruce Biddle (1995)

43 See Wineberg, . 2006, 53

44 Cochran-Smith, 2005b, 14

45 Ibid

<a rel="license" href="http://creativecommons.org/licenses/by-nc-nd/3.0/us/"><img alt="Creative Commons License" style="border-width:0" src="http://i.creativecommons.org/l/by-nc-nd/3.0/us/88x31.png" /></a><br /><span xmlns:dc="http://purl.org/dc/elements/1.1/" href="http://purl.org/dc/dcmitype/Text" property="dc:title" rel="dc:type">Assessing PACT</span> by <a xmlns:cc="http://creativecommons.org/ns#" href="http://sites.google.com/site/assessingpact" property="cc:attributionName" rel="cc:attributionURL">Ann Berlak </a> is licensed under a <a rel="license" href="http://creativecommons.org/licenses/by-nc-nd/3.0/us/">Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License</a>.<br />Based on a work at <a xmlns:dc="http://purl.org/dc/elements/1.1/" href="http://sites.google.com/site/assessingpact" rel="dc:source">sites.google.com</a>.



















Subpages (1): Assessing PACT II