June 5, 2002
Dr. Joseph F. Johnson, Director
Compensatory Education Programs
U.S. Department of Education
400 Maryland Avenue, SW, Room 3 W 230
Washington, DC 20202-6132
RE: PROPOSED TITLE I REGULATIONS ON STANDARDS AND ASSESSMENTS
Dear Dr. Johnson:
Attached are comments submitted by the Center for Law and Education, the Citizens’ Commission on Civil Rights, the National Council of La Raza on the proposed Title I regulations
on standards and assessments, as published in the Federal Register on May 6, 2002. As you will note, we are quite concerned about what we believe are troubling departures from the Act’s requirements in the proposed regulations’ treatment of norm-referenced testing and state/local assessment, as well as a range of other issues.
We appreciate this opportunity comment and would be pleased to discuss constructive approaches to these issues with the Department.
Sincerely,
Paul Weckstein, Co Director
Center for Law and Education
1875 Connecticut Avenue, NW, Suite 510
Washington, DC 20002
(202) 986-3000
on behalf of:
Center for Law and Education
Citizens’ Commission on Civil Rights
National Council of La Raza
PROPOSED TITLE I REGULATIONS ON STANDARDS AND ASSESSMENTS
submitted by:
Center for Law and Education
Citizens’ Commission on Civil Rights
National Council of La Raza
Preliminaries
Timing – problems with release of proposed regulations
We believe there is much that can be gained from the Department’s convening meetings, such as those scheduled on the proposed regulations, where people have a chance to engage in dialogue and discuss issues regarding interpretation and implementation of the Act. We encourage the Department to continue convening such meetings. The potential value of these particular meetings, however, was impaired by the lack of sufficient notice of, and opportunity to review and reflect on, the proposed regulations in advance. The proposed regulations appeared on the Department’s web site on Friday, May 3rd, one work day before the first of the regional hearings, in Cincinnati. Indeed, most interested individuals and groups did not even have this one day, if they received no notice that the web site had been changed on that day and that proposed regulations were then available. Instead, participants at the meetings typically saw them for the first time in the packets they received at the meetings – which is not an effective basis for thoughtful and informed reaction at the meetings to the complexities of the regulations.
Additionally, outreach was limited, particularly in drawing in parents. While state education department personnel and other education officials were informed and had the wherewithal, for example, to fly in from other states in the region, the need to ensure that parents and their representatives are aware, have sufficient background information and assistance to participate in a full and informed manner, and have the means to attend all require more attention. Who is, and who is not, present and active in the discussions determines the perspectives that the Department hears.
Public and parent involvement in State standards and assessment planning
Greater public and parent involvement in setting State standards and in planning for State assessments, particularly in light of additional temptations to weaken standards and assessments.
The more rigorous approach to determining adequate yearly progress (AYP) and the new or accelerated consequences for failing to make adequately yearly progress naturally pose a temptation to avoid the AYP triggers by watering down the State’s academic content standards, lowering the achievement standards that define proficient and advanced levels of mastery of the content standards, using lower-level assessments of those standards, and/or lowering the assessment cut-off scores that are adopted for proficiency. Indeed, we are already learning about state-level discussions to transform vast numbers of currently non-proficient students to “proficiency” by suddenly treating as proficient what was previously defined as basic and well below what the State previously thought was proficient (In addition to the potential for this occurring through revision of existing standards and assessments, there are multiple new opportunities for this arising – many States are still be in the process of continuing to develop their initial, final assessment systems for reading/language arts and mathematics; will be developing new assessments to meet the new grades-three-through-eight requirements, and will be developing both standards and assessments for science.)
In the end, the main quality control for avoiding watered-down standards and proficiency levels upon which we must rely is the democratic and deliberative process by which they are set within a State. The law provides, in Sec. 1111(b)(1)(F), that nothing prohibits a State from “revising, consistent with this section, any standard adopted under this part before or after the date of enactment of the No Child Left Behind Act of 2001” – which would be true even without the explicit provision. For this process (as well as the development of new standards and assessments noted above) to serve the Act, rather than result in counter-productive weakening of standards and assessments, it is essential that the Department make sure that the requirements for development of standards and assessments are properly understood and implemented. There are two aspects of this – the procedural requirements for ensuring adequate participation in the development of standards and assessments and the substantive requirements for what the standards and assessments must look like.
In terms of procedure, the Act requires that the State plan be developed in consultation with various parties, including parents. (We emphasize parents here not only because we represent them but also because their unalloyed interests in securing a high-quality education for their children is usually less mixed with the temptations noted above and felt by the educators and education officials who experience the fear and burden of being held responsible for lack of adequate progress.) The Department needs to provide more help in understanding this “consultation” in a robust way, if the standards and other components of state planning are truly going to meet the requirements of, and fulfill the promise of, the Act. “Consultation” far too often turns out to be pro forma, and all the more so when officials are feeling pressed to submit documents to meet the necessary deadlines. Ensuring real consultation requires attention to the details of, for example:
∙ Whether the consultation is iterative, at each stage of the process (rather than one-shot);
∙ Whether there is an unswerving focus throughout the process, in addressing each aspect (the content standards, setting the achievement levels for proficiency, addressing the requirements for challenge, higher-order skills, etc.), on the broad questions of what students should emerge from school knowing and being able to do, in relation to successful participation in a range of adult roles in a democracy?
∙ Whether participants are given full and accessible information and assistance (e.g., regarding standards developed b other states, academic discipline groups, NAEP, etc.; relevant research in such areas as developmental stages of learning, successes and problems in standard-setting, the knowledge and skills involved in significant achievement in a wide range of adult roles and endeavors; and the process and criteria required by Title I);
∙ Whether the full range of program beneficiaries (parents and students) and their advocates
(including civil rights organizations), and of program providers are consulted and given
adequate weight (including all relevant sub-populations; with outreach in a manner and
inform sufficient to reach them and help them understand the relevant issues; and with
sufficient time to develop thoughtful input at each stage).
Also, the timing of the submission of consolidated State applications, which the Department is requiring by June 12th, is further creating pressures to short-circuit real involvement, including involvement in the States’ decisions and plans concerning standards and assessments. In one state, for example, the draft of the consolidated state plan is being released just one day before public hearings are held on the plan, and just three days before the deadline for written comments. This is hardly conducive to high-quality and thoughtful planning and deliberation or to ensuring that the public and parents are able to play a full and informed role in shaping the plan and in participating in critical decisions about State standards and assessments.
Additionally, in relation to these issues, a Department strategy for assuring real and fully participatory planning over a longer period of time, past the June 12th deadline, in the State plan issues (on standards and assessments, as well as other key issues) also becomes important.
§200.1 State responsibilities for developing challenging academic standards
(a)(1) and (a)(3). Subjects to be taught
We are concerned about interpretation of these requirements designed to ensure that disadvantaged students are not subject to narrowed, lower expectations about what they should know and be able to do, and the continued potential for erroneously assuming, in practice, that as long as they can read and do math, it is not important that they learn what other students are learning in other areas, such as history. More attention is needed on the Act’s provisions for addressing subjects other than mathematics and reading/language arts (and, beginning in 2005-06, science). States are free not to adopt standards in other subjects, but addressing two related topics, discussed below, is necessary to fulfill the Act’s basic purpose in ensuring that disadvantaged students are not subject to lower expectations about what they should know and be able to do. Title I does not prescribe what it is that disadvantaged students should be taught. Title I does, however, insist that they be taught the same knowledge and skills as other students. It is simply not acceptable to say, for example, that as long as they can read and do math, it is not important that they learn what other students are learning in other areas, such as history. That is precisely the narrowing of expectations and opportunities that Title I is designed to overcome.
1. Other subjects for which the State has standards
Sec. 1111(b)(1)(C) provides that “The State shall have such academic standards for all public elementary school and secondary school children, including children served under this part, in subjects determined by the State, but including at least mathematics, reading or language arts, and (beginning in the 2005-2006 school year) science, which shall include the same knowledge, skills, and levels of achievement expected of all children.” Thus, it is the State which determines whether there will be standards in other subjects. What is not permitted, however, is for the State to adopt such standards in other subjects and then not use them for Title I purposes. Once the State has articulated, through its standards, what all students should know and be able to do, then it cannot provide an educational program to students participating in Title I which is aimed at mastering only some of that knowledge and skills. If the State has standards, it must use them for Title I purposes. This is a distinction often not understood and in need of clarification. Instead, it is sometimes assumed that a State can adopt standards for all children in other subjects but then can pick and choose whether they will apply to disadvantaged students under Title I.
The need here – to ensure that the full range of such academic standards as a State chooses to adopt are applied – is reinforced by Section 1111(b)(1)(E), discussed immediately below. Since that provision requires the State to have a strategy for ensuring that students are taught the same knowledge and skills as other students in subjects where the State has no standards, it would make no sense to say that the State is under no obligation to ensure such teaching in subjects where the State has standards.
Using for Title I purposes the full range of standards that the State has chosen to develop for all students, and applying them to the education of the students served by Title I, requires States, districts, and schools to address them adequately in fulfilling their various standards-related obligations in order to ensure that students served by Title I are taught the same knowledge and skills in those standards. See, for example, the school-level provisions (in Sec. 1114 and 1115) under which students must be provided with enriched curricula aligned with the standards; effective instructional methods for learning them; and effective and timely help when a student is having difficulty learning them; and the State’s obligations to describe [in Sec. 1111(b)(8)] how it will assist schools and LEAs to, for example: develop the capacity to comply with each of such local requirements, develop or identify high-quality effective curricula aligned with such standards, provide additional assistance to individual students needing help in achieving the standards, and address other factors appropriate to provide students an opportunity to achieve the knowledge and skills described in those standards.
2. Subjects for which the State has no standards
Section 1111(b)(1)(E) provides “For the subjects in which students will be served under this part, but for which a State is not required by subparagraphs (A), (B), and (C) to develop, and has not otherwise developed, such academic standards, the State plan shall describe a strategy for ensuring that students are taught the same knowledge and skills in such subjects and held to the same expectations as are all children.” For the reasons noted above, this provision is important, but it is not well understood and implemented.
First, while successful strategies under this provision may vary among the States, the Department needs to help States understand that such strategies must indeed be successful in ensuring this result, including some further explanation of what ensuring that result means. At a minimum, this includes each of the following, without which there cannot be an effective strategy which ensures this result.
(a) Because it is schools, not the SEA, which does the actual teaching the result cannot be ensured by the State unless the State ensures that schools and districts understand this obligation – so the strategy must address how the State will ensure that understanding. This means the entire school community, including the teachers, the parents, and the students themselves, so that everyone understands what students should be learning.
(b) A successful strategy must ensure that schools and districts get the help they need in developing curriculum and instruction which meets this requirement.
(c) Given the explicit equity focus of this provision – on ensuring that students served by Title I are taught the same knowledge and skills in these other subjects and held to the same expectations as are all children – the strategy needs to include a focus on instructional grouping practices, to ensure that the curriculum in all such groups are equally enriched and effective in addressing the full range of such knowledge and skills. Disadvantaged students are often placed in lower tracks which fail to teach the same range of challenging knowledge and skills.
(d) Because it is the State which must ensure that students are so taught, no strategy can be successful in ensuring this result unless it includes an effective and specific State system for monitoring local practice and for taking effective action to correct any failures of implementation.
Second, Department explanation is needed in order to avoid a disabling misinterpretation of the first part of this provision – “the subjects in which students will be served under this part.” This can only be understood as the full range of academic subjects taught to the students participating in the Title I program – either those specifically identified in targeted assistance schools or all students in schoolwide programs. We are long past the day when Title I was about a separate add-on to student’s academic program. In both targeted assistance schools and schoolwide programs, Title I is designed to assist in addressing the entire core of the academic program of the students served. (As is evident from the parallel nature of the provisions in sections 1114 and 1115, the distinction in the two types of programs is in which students are served, not in deviating from the focus on the entire academic program of those who are served.)
§200.2 State responsibilities for assessment
(a)(1), (b)(1), and throughout. State “system” of assessments
In order to make the regulations consistent with the Act, it is necessary to change references to “assessment system” and “system of assessments” to “assessments,” particularly in §200.2(b)(1).
Throughout this section, the proposed regulations refer to States establishing a “system” of assessments. The Act, however, does not use this language and refers only to the State assessments. While in some ways the concept of a system of assessments is fine, in understanding how it all fits together into a coherent system, the use of the term also seems designed to justify the decision to allow States to depart from using the same assessments for all children, something which is not permissible under the Act. This problem becomes clearest in paragraph (b)(1) of the proposed regulations.
Sec. 1111(b)(3)(C) of the Act specifically requires: “Such assessments shall – (i) be the same academic assessments used to measure the achievement of all children.” This is critical to the basic aim of the Act, in ensuring that expectations for the level of education are not lower in some areas of the State than in others. By changing this language here only to require the same assessment system, and then later [in proposed §200.3(b)] permitting the use of different local assessments, the regulation becomes inconsistent with the Act in allowing such differences. See below for further discussion.
(a)(2) Assessments in “other” subjects.
The proposed regulations say that the State “may” also measure achievement in other subjects in which the State has adopted standards. This tracks Sec. 1111(b)(C)(viii) of the Act, which provides that the assessments shall, at the discretion of the State, measure students’ proficiency in other subjects in which the State has adopted challenging content and achievement standards. Nevertheless, in order to ensure proper interpretation, this provision must be read in conjunction with other provisions of the Act, and with the comments under §200.1(a), above, concerning subjects for standards: While the State is not obligated to develop standards in other subjects, (a) if the State has done so, there is an obligation to use and apply them to Title I students, and (b) there is an obligation through some mechanism to ensure effectively that Title I students are taught the same knowledge and skills, and not subjected to lower expectations in the subjects in which they are taught beyond those for which standards are required by the Act. Thus, States need to understand that, regardless of how they use their discretion under paragraph (viii) here, they must still meet those obligations in some effective way (a task which will be considerably more challenging, to say the least, without effective assessment).
(b)(7) Multiple measures
This paragraph, like the Act [Sec. 1111(b)(3)(C)(vi)], requires that the assessments involve multiple up-to-date measures of student academic achievement, including measures that assess higher-order thinking skills and understanding.
1. The Department should make it clear that multiple measures means multiple ways of
measuring or assessing the same proficiencies,
in order to help assure the validity of the
determination that students are or are not proficient. Given misunderstanding of this provision in
current law in ways that have limited its meaning, additional language is warranted and needed to
get across its meaning, particularly in light of the more serious consequences in the new Act that
will flow from these determinations of student proficiency. Further, it will help address a major
concern about heightened accountability – that volatile and spurious changes in assessment
results will trigger findings that a school is in need of improvement; multiple methods for
assessing the same proficiencies helps reduce this volatility and increasing the capacity of the
assessment system to meet the requirements for validity and reliability. As also discussed above,
it is also critical to ensuring the validity and reliability of the assessments in a related sense – by
providing a range of ways that the full range of students, with a variety of experiences and
learning styles, can demonstrate their proficiency on the same skills and knowledge.
2. While some parties may propose that the way to deal with these issues is to indicate that States have flexibility, if they choose, to adopt additional measures beyond traditional standardized tests, that approach would be misleading and insufficient in ways that will result in improper implementation of the Act’s requirements. Whether to use multiple measures which together provide valid information about the full range of knowledge and skills that together constitute proficiency on the State’s challenging standards is not an option – it is a core requirement.
3. Related to this, the Department should take steps to ensure that people not be confused about the important distinction between the “multiple measures” required here and either the additional academic assessment measures permitted under Sec. 1111(b)(4) of the Act or the additional academic indicators permitted under §200.2(c) of the proposed regulations [Sec. 1111(b)(2)(C)(vi) and (vii) and (D) of the Act]. These latter provisions apply only to measures and indicators which do not meet the requirements for valid and reliable assessment of the proficiencies in the State standards, and neither of these additional sets of measures and indicators may reduce the number of schools which are subject to the improvement and accountability provisions of section 1116 of the Act. In contrast, those multiple measures of academic performance which actually contribute to validity and reliability of the judgments about whether students have demonstrated proficiency in the skills and knowledge in the State standards should thus not be viewed as additional measures. Rather, they must be viewed as part of the core measures which indeed will and should be the primary indicators of student progress and the basis for review of school and LEA performance, particularly given the obligation to have multiple measures.
200.3 Designing State Academic Assessment Systems
(a)(1)(i) Address the depth and breadth of the State’s standards
The draft rules, prior to negotiated rule-making, stated that the assessments should “Fully address the depth and breadth of the State's academic content standards” (emphasis added). “Fully” has been dropped from the proposed rules. It should be restored. Deleting “fully” drains the provision of much of its meaning. Virtually any assessment could be argued to “address” the depth and breadth of the standards in some fashion or to some extent. The question is to what extent, and the right answer should indeed be fully. There has been a tendency not to fully or adequately address the question of alignment of assessments with the standards, and weakness of the regulations here will be exploited. The question is whether the full breadth and depth of the standards are captured by the assessments. In the negotiations, “fully” was dropped out of a concern that it would be interpreted in too rigorous a manner that is never feasible even with a rich set of multiple measures. We believe that this fear is unwarranted, and that the Department’s intent in its original draft was correct and clear.
(a)(2) Use of norm-referenced tests
The provisions permitting the state assessment system to include norm-referenced tests
(indeed, if the State chooses, to include only norm-referenced tests, provided that the tests are
augmented with additional items as necessary to measure accurately the breadth and depth
of the
state’s standards) is contrary to the Act and will be very damaging to effective implementation of
the Act.
Standardized tests developed to “yield national norms,” comparing students against the performance of other students, cannot do the assessment job required by this Act – determining which students have or have not become proficient in the particular skills and knowledge identified in a State’s academic standards. The problems with trying to use norm-referenced tests to do so are legion:
• Such tests are not sufficiently “aligned” with the range of challenging skills and knowledge in a particular State’s standards.
• They typically focus on lower-level, more basic skills and are not very good at assessing higher-order skills that the States’ standards are required to address.
• Even if they were to “cover” some of the same knowledge and skills as are in a State’s standards, they are not constructed to validly differentiate which students are proficient in those particular standards and those who are not.
• Indeed, norm-referenced tests are constructed – and norm-referenced test items are selected – in order to produce a normal curve distribution. Even if, in a particular set of skills, the large majority of students in the norming group have a similar level of attainment, the test is designed to spread them out across that curve. And items that don’t help do so are tossed out, in order to produce that curve. This is quite contrary to assessments that are truly well designed to assess students in terms of the criteria of proficiency in the skills and knowledge in a set of standards – in terms of which students are rarely distributed on such a curve.
The Department has argued that its regulation addresses these concerns, because it requires that the assessments, whether norm-referenced or not, must still meet the Act’s requirements for validity, alignment with standards, multiple measures of proficiency, etc. This is more than a bit like saying we’re allowing squares to serve as circles because we’re still requiring that they be round. If the requirements were in fact to be rigorously applied, norm-referenced tests, while allowed in principle, would be rejected in fact – that is, “We’ll accept any square that can prove it’s a circle.” But in reality, that is not what will happen.. The regulations, by blessing norm-referenced tests for this purpose for which they are not designed, will in practice result in many such tests being approved (with the addition of a few additional items that won’t and can’t cure the systemic defects of this approach) – that is, we’ll end up with approval of squares so long as they have some rounded corners. There will be enormous pressure to do so, because adopting off-the-shelf national norm-referenced tests – and indeed continuing to use the ones most commonly already used in many states – is far easier than developing the rich set of multiple measures of proficiency, in relation to a particular State’s standards, that the Act requires. This pressure will increase because of the new Act’s requirements for development of additional assessments in additional grades and subjects, and because of the tighter accountability system that makes it natural to turn to tests that might not be so rigorous in order to avoid having “too many” schools be deemed in need of improvement.
(c) Inclusion of local assessments in the “state” system
The issues here are complex, and we believe that there is a very important local role in any assessment process that will be rich enough and valid enough to serve the purposes of this Act. However, the proposed regulations are not adequate to address this issue. Indeed, they instead take an approach which even on their face are directly contrary to the Act.
Section 1111(b)(3)(C)(i) of the Act explicitly requires that the assessments at issue here “shall” “be the same academic assessments used to measure the achievement of all children” in the State. The regulations first ignore this clear requirement of the Act by eliminating it from the regulations altogether. As noted earlier, under proposed §200.2(b)(1), instead of following the Act and saying that the same “assessments” must be used to measure achievement of all students, the proposed regulations instead say that it must be the same “assessment system.” And then, here, the “system” is defined to allow the use of different assessments for different students.
There are important reasons for this statutory requirement, that has been ignored, mandating the same assessments. First, without it, we are returned to a non-uniform state “system” in which very unequal expectations for the depth and quality of what a child will learn prevail depending upon what school, in what kind of locale, the child attends. And localities would be selecting the standards and assessments under which their own schools would be subject to the now much tougher accountability triggers for adequate yearly progress, so the likelihood of efforts to use any local selection of assessments as a way for lower-achieving districts to adopt, subtly or otherwise, easier assessment measures is by no means abstract.
Second, another important reason for this emphasis is that leaving the process to the local level ignores the reality of what it takes to meet the Act’s requirements for accurate and inclusive measures of student proficiency. For example, it takes significant resources to validate carefully the assessments to ensure that they are accurately measuring the full range of skills and knowledge on any one set of standards; it takes significant resources to develop accurate and rich multiple measures of proficiency in those skills and knowledge. Similarly, what is “practicable” in terms of assessing the proficiency of limited English proficient students on standards looks very different if the State’s capacity is applied to the task rather than asking each district to address that task separately for each of its language groups. And finally, in the face of more limited resources, many districts, rather than adopting richer multiple measures of proficiency, are more likely to feel little alternative than turning to off-the-shelf tests – producing the exact opposite result desired by those advocating for local assessments.
The regulations seek to guard against these concerns by highlighting various requirements that must be met by any system using local assessments – both the same requirements (of validity, etc.) that apply to assessments generally and additional requirements, including that the local assessments (i) be equivalent in content coverage, difficulty, and quality to one another and to state assessments, where they exist; and (ii) have comparable validity and reliability with respect to the various groups of students. But, aside from the basic problem that the statute’s requirement to use the same assessments is being ignored, this too is a bit like the round squares. While one reading would be that in fact, no set of varying local assessments will be able to pass muster and meet these criteria of equivalence and comparability (criteria that would be critical if such a system were to accomplish basic purposes of the Act), a more likely scenario will be tremendous pressure to avoid a null set and to approve some systems without being firm on insistence about those criteria.
There is, we believe, a different path – one that does not create these statutory, policy, and practical problems – for addressing the important local role in a uniform system in which the same assessments are used for all students. The Act calls for a rich system of multiple measures, which must serve not only for determining a school’s progress under the State accountability system but also, under the provisions of the Act, to improve teaching of advanced skills and provide ongoing useful information for teachers, parents, students, and building administrators about the full range of skills and knowledge in the standards, including higher-order thinking and understanding. As indicated in the discussion of multiple measures above, such multiple measures cannot be seen as a perhaps allowable add-on to the state system of assessment – they must be part of the core measures. If this is to be done in a way which improves teaching of advanced skills and does not result in counter–productive approaches which narrow teaching, then multiple measures can and should properly include a range of measures for students to fully demonstrate what they know and can do, including for example extended written work, demonstrations, portfolios, etc. As some States are now starting to show, there are ways to create a statewide uniform system of assessment in which teachers and local staff can become an integral part.
For both these ends to be met, there must be a very careful approach to ensuring both the needed uniformity and the needed richness. It means, for example, very rigorous validity analysis taking the criteria in the Act very seriously as applied to more open-ended forms of assessment. (In this sense, the proposed focus on equivalence and comparability could be drawn on, not for approving different assessments, but rather for rigorous and continuous review of the equivalence, comparability, validity, etc. of the way the students’ achievement is evaluated with these statewide, uniform measures.) And it means very extensive and well-designed approaches to training and technical assistance for both State and local personnel.
§200.4 State Law Exception
This section [and the underlying provision of the Act, Sec. 1111(b)(5)] carves out a narrow exception to deal with a State that lacks sufficient authority under State law to adopt standards and aligned assessments applicable to all students in the State’s public schools. Such a State is given the choice of either adopting Statewide standards and assessments that will apply only to students served by Title I (rather than all students in the State), or adopting and implementing State policies to ensure that each local district adopts standards and aligned assessments that meet all the Title I criteria and that will apply to all students in those district.
First, it is worth noting that the wording of this exception in the Act reinforces the interpretation of the basic rule to which the exception is created – namely that, outside of this exception, the State must adopt standards and assessments on a Statewide basis. This further highlights our point above that §200.3(c) above is contrary to the Act. Indeed, it is interesting to note that the proposed criteria here for local assessments – where clearly permitted under the Act as a State law exception, are essentially the same that the proposed regulations apply to local assessments under §200.3(c), where the Act does not permit such exceptions.
Second, in utilizing this exception, the Department needs to require, under §200.4(a), strong evidence of lack of State authority.
Third, where a State does lack such authority, it is essential that the Department ensure
that all of the criteria for standards and assessments – including those discussed in the comments
above – are fully met. As noted in the discussion of the general rule immediately above, this is a
significant challenge (both for combating and eliminating unequal expectations and for
complying with the other requirements for standards and assessments), and it must not be
slighted.
Moreover, in fully implementing it, attention needs to be paid not only to the
substantive criteria and the rigor of the State reviews (for which we support the criteria in the
regulations), but also the required processes at the local level. Since key parts of standards and
assessment development would be carried out at the local level, the requirements in section 1118
for parent involvement must be fully and vigorously implemented in making those local
decisions. This is an area, not sufficiently addressed by the proposal, where strong regulation or
policy guidance is needed.
Finally, as a technical matter, we believe subsection (b) should refer to “paragraph (a)(2),” rather than to “paragraph (a).” The criteria discussed in subsection (b) apply only to those States which choose to address the exception by having each LEA adopt standards and assessments applicable to all students served by the LEA – i.e., the option in (a)(2). If instead a State addresses the exception by the choosing the option in (a)(1) – adopting Statewide standards and assessments but limiting their applicability to students served by Title I – then the criteria in subsection (b) are inapplicable.
§200.6 Inclusion of All Students
(a)(2) Alternate assessments for students with disabilities
First, the last sentence, concerning the subjects to be assessed, needs to be changed. Instead of saying “in at least reading/language arts, mathematics, and, beginning in the 2007-08 school year, science,” the provision should read: “in the same subject areas in which all other students are assessed.” Under the parallel language in §200.2(a)(1) about reading, math, and (starting in 2007-08) science, a State may be assessing students, for example, in history or, prior to 2007-08, in science. If the State has chosen to do so for other students, the alternate assessments for students with disabilities must (as a matter of non-discrimination under the civil rights laws, as well as under Title I) assess the same areas.
Second, we are concerned that the provision does not sufficiently link alternate assessments to the State’s standards (and to the criteria for valid and reliable assessment), as necessary to comply with both disability law and Title I. Alternate assessments need to encompass those instances where a student with disabilities cannot be validly assessed with the regular State assessments, even with accommodations, but can be meaningfully assessed in relation to the same standards, but with a different, alternative form of assessment – i.e., s/he capable of demonstrating some level of performance in regard to some or all the knowledge and skills in the State’s content standards, but requires a different form of assessment to determine that level of performance. While many students with disabilities can be assessed, with accommodations, using the same assessments, of those who cannot, many can, and therefore must be, assessed with different instruments (for instance more performance-based or open-ended instruments) for validly measuring the same knowledge and skills – even if, in particular schools, such students are, initially at least, disproportionately failing to reach proficient levels. It is critical that we not have lower expectations for what such students are taught. There should only be the narrowest exception – for the small number of students who are so severely cognitively impaired that any meaningful measure of their performance in relation to the State’s standards is impossible. Paragraph (2), however, is written as if to assume that once the assessments used for other students are not appropriate for a particular student, then the standards are not appropriate either. (It requires only that the assessments “yield results” in the required subject areas – not validly and reliable measure, or yield results of, the student’s skills and knowledge in relation to the State’s standards.) This must be changed – by using language similar to that in paragraphs (a)(i) and (ii) to clarify that (with the narrowest exception noted above) the alternative assessments must measure their academic achievement relative to the State’s standards for the grades in which the student is enrolled, consistent with Section 200.1(b)(2) and (3) and (c). This is necessary both (i) to ensure that (again, with the narrowest of exceptions) the alternate assessment measure performance in relation to the state standard, and (ii) that the alternate assessments be valid and reliable, etc.
Third, we note and question the omission of reference to Section 504 in paragraph (2), in contrast to paragraph (1). Limiting alternate assessment only to students covered by IDEA, as opposed to students covered only under Section 504, might make sense if alternate assessments were only for students with the severest cognitive impairments for whom performance in relation to knowledge and skills in the State standards was not meaningful. As noted above, however, those students should be only a small minority, and alternate assessment is also for students who simply need a different method of assessment of the same knowledge and skills. Once that is properly clarified, we are concerned that it is not proper to exclude all 504-classified students from alternate assessment.
(d) Homeless Children
We support the additional language that has been added to clarify that homeless children must be included in the State’s assessment, reporting, and accountability systems. And it is true, as the paragraph also says, that the State is not required to report results for homeless children as a separate disaggregated category. Having said that they do not have to be reported separately, however, additional clarification is needed to indicate that the required disaggregated category of economically disadvantaged students includes homeless students. First, meeting the McKinney-Vento Act definition of homelessness should be enough to be counted as economically disadvantaged, without requiring further evidence of participation in free or reduced lunch programs – to which homeless students often face barriers. Second, having said in the regulations how these students are not to be counted (as a separate category), it would mislead many people not to point out, equally prominently in the regulations, how they are to be counted (as economically disadvantaged students). It is, of course, true (and should be noted in the regulations) that certain students who are homeless may also be students with disabilities or students with limited English proficiency, and in those cases should be counted in those categories as well, but that is equally true for other economically disadvantaged students – they may meet other criteria as well by reason of other characteristics. But the point here is that, by reason of their homelessness under McKinney-Vento alone, they should be counted as economically disadvantaged.
§200.7 Disaggregation of Data
(a) Statistically reliable information
Additional clarification is needed in order to deal with what happens with data that may be statistically reliable for one purpose even though it may not be for other purposes. (It is the inferences made with the data – the use of the data – that may or may not be proper because of the size of the numbers.) For example, the numbers of students within a particular demographic category may be too small to yield statistically reliable information for purposes of comparing this year’s 5th graders with last year’s 5th graders – i.e., for purposes of helping to determine whether the school is making adequate yearly progress, it is not statistically reliable and cannot be used for that purpose. At the same time, because that same data is a report on all students within that category, rather than a sample, it is statistically reliable for purposes of reporting on that category and should be reported as such (provided it does constitute personally identifiable information).
(b) Personally identifiable information
We are quite troubled by one interpretation of “personally identifiable” articulated by
Department staff during negotiated rule-making. First, it was said that reporting that all students
of a particular category were at a particular achievement level (e.g., “proficient,” “basic,” or
“below basic”) would be personally identifiable and thus improper, because it tells us specific
information about each individual student in that group – with which we strongly agree. But then
it was said that reporting that, of a known group of 8 students, 5 were in one category and 3 were
in another was not personally identifiable – for example, if 5 were “basic” and 3 were “below
basic” – because we do not know what the performance level is of any one student. However,
this is equivalent to reporting that every one of the 8 students is not “proficient” – i.e., has not
achieved the goal that has been set for all students. This should clearly be viewed as improper
reporting of individually identifiable information.
The characterization should not hinge on the
form by which that information is revealed – information that reveals that Johnny is either at a
“basic” or “below basic” level is the same as revealing that Johnny is not proficient.