Measuring Educational Gain in the National Reporting System for Adult Education

AGENCY:

Office of Vocational and Adult Education, Department of Education.

ACTION:

Final regulations.

SUMMARY:

The Secretary establishes procedures for determining the suitability of tests for use in the National Reporting System for Adult Education (NRS). These final regulations also include procedures that States and local eligible providers must follow when using suitable tests for NRS reporting.

DATES:

These regulations are effective February 13, 2008.

The incorporation by reference of certain publications listed in the rule is approved by the Director of the Federal Register as of February 13, 2008. However, affected parties do not have to comply with the information collection requirements in §§ 462.10, 462.11, 462.12, 462.13, and 462.14 until the Department of Education publishes in the Federal Register the control number assigned by the Office of Management and Budget (OMB) to these information collection requirements. Publication of the control number notifies the public that OMB has approved these information collection requirements under the Paperwork Reduction Act of 1995.

FOR FURTHER INFORMATION CONTACT:

Mike Dean, U.S. Department of Education, 400 Maryland Avenue, SW., Room 11152, Potomac Center Plaza, Washington, DC 20202-7240. Telephone: (202) 245-7828 or via Internet: Mike.Dean@ed.gov.

If you use a telecommunications device for the deaf (TDD), call the Federal Relay Service (FRS), toll free, at 1-800-877-8339.

Individuals with disabilities can obtain this document in an alternative format (e.g., Braille, large print, audiotape, or computer diskette) on request to the contact person listed in the preceding paragraph.

SUPPLEMENTARY INFORMATION:

These final regulations further the Department's implementation of section 212 of the Adult Education and Family Literacy Act (Act), 20 U.S.C. 9201 et seq., which establishes a system to assess the effectiveness of eligible agencies in achieving continuous improvement of adult education and literacy activities.

On October 18, 2006, the Secretary published a notice of proposed rulemaking (NPRM) for 34 CFR part 462 in the Federal Register (71 FR 61580). In the preamble to the NPRM, the Secretary discussed on pages 61581 and 61582 the significant proposed regulations. As a result of public comment, these final regulations contain several significant changes from the NPRM. While we fully explain these changes in the Analysis of Comments and Changes section elsewhere in these regulations, they are summarized as follows:

• Rather than immediately establishing, in § 462.4, a deadline for State and local eligible providers to stop using tests that are currently listed in the Implementation Guidelines: Measures and Methods for the National Reporting System for Adult Education (Guidelines), the Secretary will announce a deadline in a notice published in the Federal Register after reviewing the first group of tests submitted under these regulations.
• On April 14, 2008, the Secretary will provide test publishers the first opportunity to submit tests for review under these final regulations. In subsequent years, in accordance with § 462.10(b), test publishers must submit applications to the Secretary by October 1 of each year.
• We have revised several sections of the regulations to distinguish between (1) traditional tests, which use items that have been generated before the test is administered, and (2) computerized tests, which use an algorithm to select test items while the test is being administered. The changes affect §§ 462.3(b) regarding the definition of test, 462.11 regarding the information that must be included in a test publisher's application, 462.12 and 462.13 regarding the Secretary's review of tests, and 462.41 regarding the administration of tests.
• Section 462.12(e) has been revised to clarify that test publishers can request that the Secretary reconsider a decision to revoke a determination that a test is suitable before the Secretary makes a final determination about the test's suitability for measuring educational gain for the NRS.

Through these final regulations, we formalize the process for the review and approval of tests for use in the NRS. We believe that the uniform process in these regulations will facilitate test publishers' submissions of tests to the Department for review and will help strengthen the integrity of the NRS as a critical tool for measuring State performance on accountability measures. This process also will provide a means for examining tests that are currently approved for use in the NRS, but that have not been updated recently and, therefore, need to be reassessed for their continuing validity.

Analysis of Comments and Changes

In response to the Secretary's invitation in the NPRM, 13 parties submitted comments on the proposed regulations. An analysis of the comments and of the changes in the regulations since publication of the NPRM follows.

We group and discuss issues under the sections of the regulations to which they pertain, with the appropriate sections of the regulations referenced in parentheses. Generally, we do not address technical and minor changes—and suggested changes the law does not authorize the Secretary to make.

General Comment

Comments: A commenter stated that, because each State uses its own curriculum frameworks, the validity of a particular test may vary to the extent that the test aligns with a State's curricula. The commenter, therefore, stated that the Department could not approve a test for use in all States without evaluating its validity for each State that uses it.

Discussion: We agree that not all States can use any single test. States are expected to select a suitable test or tests that best align with their particular curricula. If a State's curriculum is not aligned with an existing test, the State will need to develop its own test aligned with the State curriculum and submit the test to the Department for review under these final regulations.

Changes: None.

Definitions

Adult Education (§ 462.3)

Comments: A commenter stated that the proposed regulations incorrectly defined adult education. The commenter noted that the regulations refer to students “who are not enrolled in secondary school” while the Act refers to students “who are not enrolled or required to be enrolled in secondary school under State law.” The commenter recommended using the definition in the Act.

Discussion: Section 462.3(a) indicates that certain terms used in the regulations, including adult education, are defined in section 203 of the Act. The language the commenter quotes is from the definition of adult education population in § 462.3(b), which is not defined in the Act. Nevertheless, we agree that the two definitions should be consistent.

Changes: We have modified the definition of adult education population to include individuals who are not required to be enrolled in secondary school under State law in order to make it consistent with the definition of adult education in the Act.

Content domains and skill areas (§ 462.3)

Comments: One commenter stated that the term skill areas should be used consistently throughout the regulations, instead of the regulations using this term interchangeably with the terms content domain and content specifications.

Discussion: In drafting the proposed regulations, we used the terms content domain and content specifications in the sections of the regulations applicable to test publishers because those are terms of art in the test publishing industry. Likewise, the term NRS skill areas is used in the sections of the regulations that are applicable to States and local eligible recipients because this is a term of art in the adult education program. Although we used the term content specifications in the proposed regulations, we did not include it as a defined term. We think it is appropriate to do so in the final regulations because the term has the same meaning as the terms content domains and NRS skill areas.

Changes: We have modified the defined term, content domains or NRS skill areas, in proposed § 462.3 to also include the term content specifications.

Test publisher (§ 462.3)

Comments: A few commenters expressed concern that the definition of test publisher might be too restrictive and could prevent the review of some tests. Commenters recommended expanding the definition to include universities; adult education programs; other entities that possess sufficient expertise and capacity to develop, document, and defend assessments; entities in the process of copyrighting a test; and entities holding an unregistered copyright to a test. One commenter agreed that the Secretary should review only tests from test publishers owning a registered copyright.

Discussion: The proposed regulations did not prohibit universities, adult education programs, or other legitimate test developers from submitting a test for review. We explained in the preamble to the NPRM that entities submitting tests for the Secretary's review must be knowledgeable about the test, be able to respond to technical questions the Secretary raises during the review process, and have the legal right to submit the test for review. With regard to the recommendation to have the Secretary approve other entities who can submit a test for review, it would be inappropriate and counter-productive for the Secretary to determine the suitability of a test submitted for review without the permission of the rightful owner of a registered copyright of the test or the entity licensed by the copyright holder to sell or distribute the test.

Changes: None.

June 30, 2008, deadline for transitioning to suitable tests (§ 462.4)

Comments: Several commenters expressed concern that States might not have adequate time by the June 30, 2008, deadline to change assessment instruments, particularly if the Secretary determines that a test is no longer suitable for use in the NRS. The commenters stated that States need time to rewrite assessment policies, select replacement tests, retrain personnel, purchase materials, modify complex data systems, and, possibly, hire special contractors to assist with modifying those data systems. A different commenter stated that it might take two years to implement a change in State assessment instruments. Commenters recommended that the regulations permit a State to negotiate a practical transition timeline with the Secretary.

Commenters also recommended that, because transitioning from unsuitable tests places a burden on States' financial resources and professional development capabilities, the regulations should be deferred until the amount of funds available for State leadership becomes 15 percent of the Federal allocation. One commenter indicated that local programs often pay the significant cost of purchasing assessment instruments and that replacing an entire assessment system in a single budget year could devastate a local budget.

Discussion: Proposed § 462.4 would have permitted States and local eligible providers to continue to measure educational gain using a test that was identified in the Guidelines until June 30, 2008. However, we specifically asked for comments on whether this deadline would provide sufficient time for States and local eligible recipients to make the transition to suitable tests because we recognized that changing tests significantly affects a State's accountability system. Our intention in proposing the June 30, 2008, deadline was to ensure that States stop using unsuitable tests on a date certain but still provide enough time for (1) the Secretary to complete one review of tests and (2) States and local eligible recipients to transition from unsuitable tests to suitable tests. We also intended to impose a deadline that would result in the efficient removal of unsuitable tests from use in the NRS. Once the Secretary determines that a test is unsuitable for use in the NRS, permitting States to continue using it for long periods of time would be inconsistent with the Secretary's intent to improve data quality.

While we understand the desire to defer implementation of the regulations because of cost factors and timing constraints, improving the quality of State accountability systems and the data reported by the NRS is of immediate importance and should not be unduly delayed. Adult Education and Family Literacy Programs, like other Federal programs, must report on progress made, achievements, and overall program effectiveness using valid and reliable measures of performance. The regulations are designed to improve the reliability and validity of data used to report the educational gains of students, and thereby improve the reliability and validity of data on overall program effectiveness.

In light of the commenters’ concerns, and to accommodate States’ needs to make system revisions, provide training, and acquire tests, we will not specify a date in these regulations by which States and local eligible providers must cease using unsuitable tests; instead, we have provided for the Secretary to announce this deadline in a notice published in the Federal Register.

Changes: We have revised § 462.4 to provide that the Secretary will announce, through a notice in the Federal Register, a deadline by which States and local eligible providers must stop using tests that are currently listed in the Guidelines and that the Secretary has determined not to be suitable for use in the NRS under these final regulations.

Deadline for submitting tests for review by the Secretary (§ 462.10(b))

Comments: One commenter agreed that the regulations should provide an annual deadline for test publishers to submit tests to the Secretary for review. Other commenters requested clarification on when the review cycle begins and ends. Another commenter asked if the first opportunity to submit tests would be in 2007 or in 2008. Yet another commenter suggested that the date for submission of tests be no sooner than two months and no later than four months after the effective date of the final regulations.

Discussion: We are establishing April 14, 2008 as the first date by which test publishers must submit tests for review under these regulations. In subsequent years, test publishers must submit applications to the Secretary by October 1 of each year. However, because we cannot predict the number of tests that will be submitted for review nor the amount of time it will take to review the tests, it is not possible to predict how long the process will take from year to year. We, therefore, do not think it is appropriate to establish a date on which we will announce the results of the Secretary's review. We will publish the list of suitable tests well before the program year in which they might be used.

Changes: None.

Content of an application—General (§ 462.11)

Comments: A commenter responded positively to the regulations’ specific delineation of what an application for test review must include. Another commenter asked whether test publishers must use a form in addition to submitting the information outlined in proposed § 462.11(b) through (j). Another commenter stated that it may be too constraining to require test publishers to arrange application information in the order established by proposed § 462.11(b) through (j).

Discussion: To facilitate the review process, the regulations in § 462.11 describe the specific requirements for the contents of an application. A test publisher is not required to submit any form or information except as required in § 462.11. We believe that organizing the information in the application in the order presented in § 462.11(b) through (j) will help to ensure that information about a test is easily available to and reviewable by the educational testing and assessment experts who will review the tests; however, to provide test publishers with some flexibility in organizing their applications, we will permit them to include in their applications a table of contents that identifies the location of the information requested in § 462.11(b) through (j).

Changes: We have revised § 462.11(a)(3)(ii) to permit test publishers to include a table of contents in their applications as an alternative to presenting information in the application in the order described in § 462.11(b) through (j).

Content of an application—Involvement of the adult education population (§ 462.11)

Comments: A commenter stated that proposed § 462.11 would generally require a test publisher to demonstrate that adult educators have been involved in a test's development and maintenance, and that some publishers would not meet that requirement easily. The commenter also stated that compliance with the regulations would require customized tests developed specifically for use in adult education, which would increase the cost and exclude some quality assessments.

Discussion: The regulations do not require a test publisher to demonstrate that adult educators have been involved in a test's development and maintenance. We realize that tests developed for other populations might not be suitable for use in the NRS because they were not developed with the adult education population in mind and do not readily measure the educational functioning levels used in the NRS. The regulations are clear that the Secretary reviews tests to determine their suitability for use in the NRS. For instance, § 462.13(a) indicates that, in order for the Secretary to consider a test suitable for use in the NRS, the test must measure the NRS educational functioning levels of members of the adult education population. Accordingly, § 462.11(c)(1)(ii) requires information that demonstrates the extent to which the adult education population was used to develop and evaluate a test, which is appropriate because the tests will be used with that population.

Changes: None.

Content of an application—Motivation of examinees (§ 462.11(c)(1)(iii))

Comments: Two commenters were concerned that test publishers would have to include information in the application on the motivation of examinees used in the development of a test. One commenter indicated that “there is no generally accepted method for identifying and classifying the degree and level of motivation of examinees.” The commenter stated that a test publisher could make some assumptions about motivation, but indicated that the assumptions would be subjective and not scientifically valid. The second commenter requested clarification of the expectation that examinees would be motivated.

Discussion: The regulations only require test publishers to provide in their applications information on the steps, if any, taken to ensure that examinees were motivated while responding to the test. The regulations do not require test publishers to take steps to ensure that examinees were motivated while responding to the test. Further, if a test publisher were to take such steps, the test publisher would not be required to use any particular methodology for doing so.

Changes: None.

Content of an application—Item development (§ 462.11(c))

Comments: A commenter noted that the proposed regulations did not require test publishers to include in their applications information on item selection or form development for the test under review.

Discussion: The commenter's observation is correct and calls attention to the need for the regulations to require test publishers to include this information in their applications and for the regulations to clarify the distinction between traditional tests, which use items that have been generated before the test is administered, and those that use a computerized algorithm to select test items while the test is being administered.

Changes: We added a new paragraph (3) to § 462.11(c) to require test publishers to describe in their applications the procedures used to assign items (1) to forms, for tests that are constructed prior to being administered to examinees, or (2) to examinees, for adaptive tests in which items are selected in real time.

Content of an application—Maintenance: history of test use (§ 462.11(d)(4))

Comments: A few commenters recommended that the regulations require test publishers to include in their applications additional information on the history of the test's use.

Discussion: The regulations require test publishers to provide documentation of how a test is maintained, including a history of the test's use. We are particularly interested in information on how many times the test forms have been administered. This information is useful in gauging how much the test forms have been exposed and the likelihood of test items being compromised.

Changes: We have revised § 462.11(d)(4) to clarify that information submitted in the application regarding the history of a test's use must include information on the number of times the test has been administered.

Content of an application—Maintenance (§ 462.11(d)(5))

Comments: A commenter recommended that the regulations require test publishers to include in their applications the procedures used for computerized adaptive tests to select subsets of items for administration, determine the starting point and termination conditions, score the tests, and control item exposure.

Discussion: We agree that requiring test publishers to provide the recommended information will help experts to better assess the suitability of computerized adaptive tests for use in the NRS.

Changes: We added a new paragraph (5) to § 462.11(d) to require test publishers to include in their applications for computerized adaptive tests the information recommended by the commenter.

Content of an application—Match of content to the NRS educational functioning levels (content validity) (§ 462.11(e)(2) and (4))

Comments: A few commenters asked if proposed § 462.11(e)(2) and (4) were requesting the same information, and sought clarification regarding the difference between the paragraphs.

Discussion: The paragraphs are requesting the same information.

Changes: We have removed § 462.11(e)(4) to eliminate the duplicate information requirement and renumbered the remaining paragraphs.

Content of an application—Procedures for matching scores to NRS educational functioning levels (§ 462.11(f)(2))

Comments: A commenter stated that requiring the judgments of subject-matter experts to translate an examinee's performance to the examinee's standing with respect to the NRS educational functioning levels might not prove fruitful and would substantially increase the cost of test development. The commenter stated that determination of score ranges and their fit to the existing NRS levels can be made based on an analysis of skills being assessed and an intimate knowledge of the assessment tools being used.

Discussion: Section 462.11(f) does not require the use of subject matter experts. It requires test publishers to document the procedure they use to translate the performance of an examinee to the examinee's standing with respect to the NRS educational functioning levels. A test publisher can choose the procedure it thinks is best. However, if a test publisher chooses to use judgment-based procedures to translate performance, the regulations require the publisher to provide information on that procedure, including information concerning the subject matter experts the test publisher used. Requiring this information is consistent with accepted professional test development and standard-setting procedures in the 1999 edition of the Standards for Educational and Psychological Testing and will help test publishers demonstrate the suitability of their tests for measuring educational gain for the NRS.

Test scores are only useful in this context if they can accurately classify individuals according to NRS levels. Therefore, it is necessary for test publishers to demonstrate how the range of test scores map onto the NRS levels and do so in a reliable and valid fashion. In the test development process, developers need to show that the range of test scores produced on their tests covers the range of skills depicted in the NRS levels, and more importantly, shows which range of scores corresponds to a specific NRS level.

Changes: None.

Content of an application—Reliability (§ 462.11(g))

Comments: A commenter noted that in discussing reliability the proposed regulations used the phrase “the correlation between raw or number correct scores.” The commenter noted that this phraseology is not applicable to tests that use an adaptive structure or a multi-parameter item response theory model. The commenter stated that, in such situations, the particular items answered correctly, not the number of items answered correctly, determine the score.

Discussion: We agree with the commenter that the phrase in the regulations is not applicable to computerized adaptive tests.

Changes: We revised § 462.11(g)(1) to require that, in the case of computerized adaptive tests, test publishers document in their applications the correlation between raw (or scale) scores across alternate administrations of the test.

Comments: With regard to proposed § 462.11(g)(2), a commenter suggested that information about the number of individuals classified into NRS levels would only provide useful data if the information were submitted after the Department approved a test's scores-to-NRS-levels crosswalk. The commenter stated that requiring this information prior to test approval could produce information that is not meaningful. Another commenter responded positively to the requirement for “inclusion of information about decision/classification consistency.”

Discussion: We do not agree that the Department should approve the rules a test publisher uses to transform the scores of a test into estimates of examinees’ NRS educational functioning levels prior to the test publisher providing evidence that the transformation rules result in reliable, i.e., consistent, educational functioning level classifications. We believe that, when an application is submitted, a test publisher should be able to provide documentation of the degree of consistency in performance across different forms of the test, particularly regarding which examinees are classified into the same NRS educational functioning levels across different forms of the test. By demonstrating that a test can consistently classify individuals into the same NRS educational functioning levels across different forms of the test, the test publisher assures the Department that assessments of educational gain are the result of instruction and other interventions to improve literacy, not measurement error. Without this demonstration of classification consistency, reports of educational gain are uninterpretable. This information is very important to determinations about the suitability of a test and whether the test measures the NRS educational functioning levels as required in § 462.13(a).

Changes: None.

Content of an application—Construct validity (§ 462.11(h))

Comments: A commenter expressed concern that proposed § 462.11(h) would have required the results of several studies on the adult education population in connection with other tests designed to assess educational gain, which can be useful and meaningful, but also time-consuming and expensive. The commenter indicated that imposing this requirement after, not before, test approval would permit test publishers to collaborate and conduct the studies in a more cost-effective manner. Further, the commenter stated that the requirement could exclude some qualified assessments. The commenter recommended that the regulations be rewritten so that (1) these studies would only be required of tests that have been approved and (2) a five-year period could be provided for conducting the studies.

Discussion: We do not agree that a test should be approved for use in the NRS prior to the test publisher providing documentation of the appropriateness of a test for measuring educational gain for the NRS, i.e., documentation that the test measures what it is intended to measure. Section 462.11(h) is consistent with the 1999 edition of the Standards for Educational and Psychological Testing, which stresses the importance of gathering evidence of the test's construct validity. The Secretary cannot determine whether a test is suitable for use in the NRS without having evidence of the test's construct validity.

Changes: None.

Content of an application—Construct validity (§ 462.11(h)(1))

Comments: A commenter expressed concern that proposed § 462.11(h)(1) would have required test publishers to document the appropriateness of a given test for measuring educational gain in the NRS, including the correlation between the NRS test results and the results of other tests that assess educational gain in the same adult education population. The commenter stated that this comparison could lead to faulty conclusions if the other test is not an accurate measure of the construct.

Discussion: We believe that it is appropriate to look at test correlations as one criterion for evaluating construct validity. Generally, a test should correlate with other tests known to measure the same construct, and it should not correlate (or have a very low correlation) with tests known to measure different constructs. This latter relationship depends upon the nature of the comparison constructs. To the extent that the two constructs are theoretically related, a correlation that approximates their theoretical relationship is expected.

Changes: None.

Content of an application—Construct validity (§ 462.11(h)(2))

Comments: Two commenters stated that proposed § 462.11(h)(2) should be reconsidered because “hours of instruction” is not a variable that correlates highly with test scores.

Discussion: Proposed § 462.11(h)(2) would have required test publishers to document that a test measures what it is intended to measure, including the extent to which the raw or scale scores are related to other relevant variables, such as hours of instruction or other important process or outcome variables. While we are aware of data establishing the relationship between Adult Basic Education (ABE) test scores and hours of instruction, the reference to “hours of instruction” in § 462.11(h)(2) was only intended to provide an example of a possibly relevant variable.

Changes: We have revised § 462.11(h)(2) to clarify that “hours of instruction” is an example of possibly relevant variables.

Content of an application—Other information (§ 462.11(i)(1))

Comments: A commenter requested clarification of the phrase “an analysis of the effects of time on performance” used in proposed § 462.11(i)(1). The commenter thought the phrase meant “the effects of the time it takes to administer the test or for a student to complete it.”

Discussion: The commenter is correct. Section 462.11(i)(1) requires an application to include a description of the manner in which test administration time was determined and an analysis of the speededness of the test. The term “speededness” as used in § 462.11(i)(1) refers to the effects on test performance that result from the time it takes to administer or to complete the test.

Changes: We have revised § 462.11(i)(1) to clarify that we require both information on the manner in which test administration time was determined and an analysis of the speededness of the test.

Content of an application—Other information (§ 462.11(i)(5))

Comments: A commenter requested clarification of the term “retesting procedures” used in proposed § 462.11(i)(5). The commenter asked if the term referred to “pre- and post-testing.”

Discussion: The term “retesting” as used in § 462.11(i)(5) refers to the re-administration of a test that might be necessary because of problems in the original administration (e.g., the test taker becomes ill during the test and cannot finish, there are external interruptions during testing, or there are administration errors).

Changes: We have revised § 462.11(i)(5) to clarify that “retesting” refers to the re-administration of a test that might be necessary because of problems in the original administration such as, the test taker becomes ill during the test and cannot finish, there are external interruptions during testing, or there are administration errors.

Content of an application—Other information (§ 462.11(i)(6))

Comments: A commenter noted that proposed § 462.11(i)(6) would require test publishers to provide such other evidence as the Secretary determines is necessary to establish the test's compliance with the criteria and requirements in proposed § 462.13. The commenter requested clarification regarding what that evidence would be.

Discussion: While § 462.11 includes the information we anticipate the Secretary will need to determine whether a test is suitable for use in the NRS, we recognize that the Secretary can require a test publisher to provide additional information to establish a test's compliance with the criteria and requirements in § 462.13. Section 462.11(i)(6) merely alerts test publishers to this possibility.

Changes: None.

Content of an application—Previous tests (§ 462.11(j))

Comments: A commenter asked if a test publisher submitting an application for a test that is currently approved for use in the NRS would have to provide the information in proposed § 462.11(b) through (i) or would have to provide only documentation of periodic review of the content and specifications of the test as specified in proposed § 462.11(j)(1).

Discussion: As indicated in § 462.11(a)(1), a test publisher must include with its application information listed in paragraphs (b) through (i) as well as the applicable information in paragraph (j). All applications must, therefore, include the information in § 462.11(b) through (i).

Changes: None.

Comments: A commenter requested clarification of the term “periodic review” used in proposed § 462.11(j)(1) and (2) with regard to the documentation that must be submitted for a previous test to ensure that it continues to reflect NRS educational functioning levels. The commenter asked if annual or bi-annual reviews constitute a “periodic review,” and suggested that the regulations use the term “current review” and specify the intervals for the reviews.

The commenter also suggested that the regulations address the possibility that the NRS educational functioning levels might change. The commenter stated that realigning tests to revised NRS educational functioning levels would cause a burden for test publishers, and that test publishers need at least one year advance notice of any proposed changes in the NRS educational functioning levels.

Discussion: Section 462.11(j) requires test publishers to provide specific information about currently used tests to ensure that the tests continue to reflect NRS educational functioning levels. The shorter the period of time between reviews of a test, the more relevant the results would be in determining the test's content validity with regard to NRS educational functioning levels. However, we are reluctant to specify a time-frame for reviews because we do not want to create an additional burden for test publishers by requiring reviews more frequently than a test publisher typically would perform.

With regard to providing advance notice to test publishers about changes in the NRS educational functioning levels, in the past we have customarily provided notice to all concerned parties, including test publishers, well in advance of any changes to the NRS educational functioning levels by posting notices on the Internet at http://www.nrsweb.org. We intend to continue that practice and, because the educational functioning levels are included in § 462.44 of these final regulations, will publish a notice of proposed rulemaking in the Federal Register requesting comment on any proposed changes to § 462.44.

Changes: None.

Comments: A commenter asked if the phrase “previous tests used in the NRS” in proposed § 462.11(j)(1) referred only to those tests previously listed in the NRS Implementation Guidelines.

Discussion: The “previous tests” discussed in § 462.11(j)(1) are tests that were listed in the Guidelines and used to measure educational gain in the NRS before the effective date of these final regulations.

Changes: None.

Comments: A commenter requested clarification of the requirement in proposed § 462.11(j)(3) for test publishers to submit “new data” for tests that have not changed in the seven years since the Secretary determined the tests were suitable for use in the NRS.

Discussion: The intent of § 462.11(j) is to ensure that a test publisher provides specific information about a test that has been in use for years, but that has not changed in the seven years since the Secretary determined the test was suitable for use in the NRS. In this circumstance, the regulations require test publishers to provide new, i.e., updated, data that support the validity of the test.

Changes: We have revised § 462.11(j)(3) to clarify that test publishers must provide updated data to support test validity.

Computerized tests (§§ 462.3, 462.11(c)(3), 462.12(a)(2)(iii), 462.13(e), and 462.41(c)(3))

Comments: A commenter acknowledged that the proposed regulations attempted to distinguish between traditional paper-based tests and computerized tests. However, the commenter recommended that the regulations be modified to clarify that the term “parallel forms” refers to “paper-based tests or computerized tests that do not involve an item selection algorithm, such as computerized adaptive tests, multistage adaptive tests, or linear on-the-fly tests.” The commenter suggested specifically changing the proposed regulatory language from “and have two or more secure, parallel, equated forms” to “and have two or more secure, parallel forms if the test is produced before it is administered to a learner (e.g., paper-based tests). If the test uses a computerized algorithm for administering items in real time, the size of the item pool and the method of item selection must ensure negligible overlap in items across pre- and post-test.” Another commenter also emphasized the importance of recognizing computerized tests.

Discussion: As we have discussed elsewhere in this notice, we agree that the regulations should clarify the distinction between (1) traditional tests, which use items that are generated before the test is administered, and (2) computerized tests, which use an algorithm to select test items while the test is being administered.

Changes: We have revised §§ 462.3 (the definition of test), 462.11(c)(3), 462.12(a)(2)(iii), 462.13(e), and 462.41(c)(3) to provide the clarification recommended by the commenters.

Tests that measure some, but not all, educational functioning levels (§§ 462.12(a)(2)(iv) and 462.13(b))

Comments: A commenter requested clarification on whether the Secretary will review tests that currently measure only some, but not all, educational functioning levels.

Discussion: The regulations provide for the Secretary to review tests that measure some, but not all, educational functioning levels. Sections 462.12(a)(2)(iv) and 462.13(b) indicate that the Secretary reviews and determines the suitability of a test if an application includes a test that samples one or more of the major content domains of the NRS educational functioning levels of Adult Basic Education (ABE), Adult Secondary Education (ASE), or English-As-A-Second Language (ESL) with sufficient numbers of questions to represent adequately the domain or domains. Further, § 462.12(b)(2) provides flexibility for the Secretary to determine that a test or a sub-test is suitable for use in the NRS if it measures the content domain of some, but not all, of the NRS educational functioning levels.

Changes: None.

Procedures the Secretary uses to review the suitability of tests (§ 462.12(b))

Comments: A commenter asked if the results of the Secretary's review of a test would be posted for public review. Another commenter suggested that the regulations should provide that test publishers be notified at least 30 days before the Secretary notifies States and local eligible providers that a test is unsuitable for use in the NRS. With regard to tests that the Secretary determines are unsuitable, a different commenter recommended that the regulations provide a time-frame by which the Secretary would review any additional information submitted by the test publisher and make a final determination.

Discussion: In accordance with § 462.12(c)(2), the Secretary will annually notify the public through a notice published in the Federal Register and posted on the Internet at http://www.nrsweb.org of tests that are suitable for use in the NRS. Under § 462.12(e)(5), the Secretary will follow the same procedure to notify the public when a test that was previously determined suitable is determined to be unsuitable. However, the Secretary will not post for public review the Secretary's determination regarding a test that has not been previously reviewed and that the Secretary determines to be unsuitable. We do not believe this is necessary because the Secretary's determination regarding these tests will not have an impact on the public, States, or local eligible providers.

Proposed § 462.12(d) provided that a test publisher would have 30 days to request that the Secretary reconsider the decision that a test is unsuitable. Therefore, it will be after this 30-day period that the Secretary will notify the public of a decision regarding a test.

Because it is impossible to anticipate the complexities that might be involved in decisions regarding a test that was initially determined to be unsuitable, we do not believe it would be appropriate to limit the time the Secretary takes to review additional information provided by a test publisher and make a final determination regarding the suitability of a test. However, we intend to conduct the review as expeditiously as possible.

Changes: None.

Publishing the list of suitable tests (§ 462.12(c))

Comments: A commenter suggested that if tests must be submitted for review by October 1 of each year, the test publisher and adult education programs should be informed of which tests are approved no later than February 1 of the following year.

Discussion: Within a reasonable time-frame, the Secretary will review tests, notify test publishers of the results of the review, and provide States and local eligible providers with a list of suitable tests. However, because we cannot predict the number of tests that will be submitted for review nor the amount of time it will take to review the tests, it is not possible to predict how long the process will take from year to year. We, therefore, do not think it is appropriate to establish a date by which we will announce the results of the Secretary's review. We will publish the list of suitable tests well before the program year in which they can be used.

Changes: None.

Revocation of determination that a test is suitable (§ 462.12(e))

Comments: A commenter suggested that test publishers be notified of the Secretary's decision to revoke a determination that a test is suitable before the Secretary notifies the general public. The commenter stated that test publishers should be given sufficient time to address the Secretary's concerns before the Secretary revokes the determination that the test is suitable. This would provide a process comparable to the process proposed for a test that the Secretary determines is unsuitable.

Discussion: We agree that test publishers should have an opportunity to address the Secretary's decision to revoke a determination that a test is suitable before that determination becomes a final decision.

Changes: We have revised § 462.12(e) to give test publishers an opportunity to request that the Secretary reconsider a decision to revoke a determination that a test is unsuitable.

Comments: A commenter stated that proposed § 462.14(b) was unclear concerning the reasons a test's seven-year approval status might be revoked.

Discussion: Section 462.12(e), not § 462.14(b), establishes the reasons for which the Secretary can revoke the determination regarding the suitability of a test. In the proposed regulation, we stated that the Secretary can revoke the determination if the Secretary determines that the information submitted as a basis for the Secretary's review of the test was inaccurate. In proposed § 462.14(b), however, we implied that the Secretary also could revoke a determination regarding the suitability of a test if the test is substantially revised—for example, by changing its structure, number of items, content specifications, item types or sub-tests.

The proposed regulations were, therefore, unclear as to whether the Secretary could revoke a determination about the suitability of a test if the test had been substantially revised. We intended that revision of a test would be a valid reason for revoking a determination about test suitability.

Changes: We have revised § 462.12(e) to clarify that substantial changes to a test is one of the reasons the Secretary can revoke a determination regarding the suitability of a test.

Criteria and requirements for determining test suitability (§ 462.13(c)(1), (d), and (f)(3) (proposed (g)(3))

Comments: A commenter noted that The Standards for Educational and Psychological Testing referenced in § 462.13(c) provides specific guidelines for the technical documentation, including evidence of validity that should be made available to interested parties. The commenter recommended clarifying that publishers should provide the Secretary with information regarding test development and validity.

Discussion: In proposed § 462.13(c)(1), we indicated that, in order for a test to be determined suitable for use in the NRS, the test would have to meet all applicable and feasible standards for test construction provided in The Standards for Educational and Psychological Testing. We did not intend this language to imply that the Secretary would review only information that is related to test development. We agree with the commenter that validity is an important factor in test evaluation and make it clear in § 462.11(e) through (g), that test publishers must include in their applications information on content validity, reliability, and construct validity. However, we think adding the reference to “validity” in § 462.13(c)(1) will reinforce the importance of test publishers providing information on validity.

Changes: We have revised § 462.13(c)(1) to require a test to meet all applicable and feasible standards for test construction and validity that are provided in the Standards for Educational and Psychological Testing.

Comments: A commenter noted that many factors influence the appropriate time between testing and retesting, including intensity of instruction, frequency and length of class meetings, and class size. The commenter wanted to “decrease the emphasis on a single protocol for post-testing” and suggested that, when discussing the time between test-taking, the regulations use the term “publisher's recommendations” instead of the term “publisher's guidelines” that was used in proposed §§ 462.13(d) and 462.40(c)(3)(ii).

Discussion: We do not agree that the term “publisher's guidelines” emphasizes the use of a single protocol when determining when to administer a post-test. State and local eligible providers, like the commenter, are aware that many factors influence the appropriate time between pre-testing and post-testing. These factors are taken into consideration by test administrators at the local level. However, because tests differ, test administrators rely on the guidelines, developed during test construction and validation and provided by test publishers, to help ensure that a sufficient amount of time has passed before a post-test is given in order to optimize the measurement of educational gains.

Changes: None.

Comments: A commenter requested clarification of § 462.13(f)(3) (proposed § 462.13(g)(3)), which requires the publisher of a test modified for an individual with a disability to recommend educational functioning levels based on the previous performance of test-takers who are members of the adult education population of interest to the NRS in order for the Secretary to consider a test suitable for use.

Discussion: In order for the Secretary to consider a test that has been modified for individuals with disabilities suitable for use in the NRS, test publishers must (1) demonstrate that adult education students with disabilities were included in the pilot or field test referred to in § 462.11(c)(1); (2) match scores to the NRS educational functioning levels based on the information obtained from adult education students with the disability who participated in the pilot or field test and for whom the test has been adapted; and (3) provide in the application, as required in § 462.11(f), documentation of the adequacy of the procedure used to translate the performance of adult education students with the disability for whom the test has been modified to an estimate of the examinees' standing with respect to the NRS educational functional levels.

Changes: In response to the comment, we changed § 462.13(f)(3) to clarify that the Secretary considers a test modified for individuals with disabilities suitable for use in the NRS if test publishers (1) recommend educational functioning levels based on the information obtained from adult education students who participated in the pilot or field test and who have the disability for which the test has been adapted and (2) provide documentation of the adequacy of the procedure used to translate the performance of adult education students with the disability for whom the test has been modified to an estimate of the examinees' standing with respect to the NRS educational functional levels.

Subpart D—General (§§ 462.40-462.44)

Comments: Several commenters noted that the Department currently provides non-regulatory guidance to States on the NRS. The commenters asked why the Department is issuing regulations when most States comply with and are successfully implementing the non-regulatory guidance and the guidance provides each State with needed flexibility. The commenters stated that “as long as States are complying with guidelines and meeting performance standards, we understand that the Federal role is to limit regulations and allow the States to continually increase their capabilities to improve program services.” The commenters, therefore, recommended that subpart D be removed from the final regulations and that the Secretary provide technical assistance and resources to the few States that are having difficulty creating effective assessment and data procedures. Additionally, one of these commenters stated that while guidance is valued, regulations sometimes seem arbitrary, intrusive, and unnecessary.

Discussion: It is the Department's policy to regulate only when essential to promote quality, and then in the most flexible, most equitable, and least burdensome way possible. We believe these regulations comply with that policy. While the regulations in subpart D are legally binding, they largely codify the guidance provided in the Guidelines and the “State Assessment Policy Guidance.” Although States have made significant progress, there remains variability in the quality of State processes and procedures in the collection and reporting of data on student assessment and performance. These regulations, like the Guidelines, technical assistance activities, and other efforts the Department has supported to improve data quality, provide a significant tool to create a standard level of quality among all States and thereby strengthen the integrity of the NRS as a critical tool for measuring State performance and the impact of adult education.

Changes: None.

Meaning of the term “placement” (§§ 462.40(c)(4), (c)(6) and (c)(10), 462.41(c)(3) (proposed), 462.42(a), (b)(2), and (c)(1), (d)(1), and (d)(2), and 462.43(a)(2) and (b))

Comments: A few commenters stated that using the term “placement” in the regulations to mean the “assignment of a student to the appropriate NRS educational functioning level” is confusing because the traditional meaning of placement is “placing a student in a particular class or other instructional offering.” Commenters noted that tests designed for placement are different from those used to measure educational gain.

Discussion: The regulations, like the Guidelines currently used by States, use the terms “place” and “placement” as terms of art to refer to the placement of students in NRS educational functioning levels in order to measure and then report on educational gain in the NRS. It is only within this context that the terms are used, and no other meaning should be inferred.

Changes: None.

Placing students (§ 462.42(d)(1) and (2))

Comments: A commenter stated that placing a student in an NRS educational functioning level using the lowest test score could result in programs providing instruction in skill areas that are not most relevant to the student. Further, the commenter stated that programs (1) should not place students based on one test and (2) should be allowed to use multiple placement tools and other pertinent information, such as student goals, to determine placement. Another commenter stated that all scores have some measurement error and that instituting a policy of using the lowest test score might introduce systematic, rather than random, measurement errors. This commenter also indicated that the policy of using the lowest test score could encourage programs to teach students only in an area where gain is to be expected. The commenter stated that a better policy would be either to focus on the learner's primary subject area at pre-test, or evaluate gain based on a composite score across areas.

Discussion: With regard to placing a student using the lowest test score when the student is tested in multiple skill areas, § 462.42(d)(1) and (2) are consistent with the policy in the Guidelines, which indicates: “States should provide to local programs the criteria for placing students at each educational functioning level, using test scores from the initial assessment. Not all of the skill areas described in the [functioning] level descriptors need to be used to place students, but the skill used should be the areas most relevant to the students' needs and the program's curriculum. If multiple skill areas are assessed and the student has differing abilities in each area, however, NRS policy requires that the program place the student according to the lowest skill area.” The Department's policy ensures that States use a standardized approach when reporting educational gain, which ensures comparability of data. These regulations use the term “placement” as a term of art to refer to the placement of students in NRS educational functioning levels in order to measure and then report on educational gain in the NRS. Placement of a student in an educational functioning level for NRS purposes does not affect placement in an instructional program. States can use a variety of tools when devising an instructional program for students. States and local eligible providers know that using the lowest test score is a convention used for reporting purposes and it is not intended to encourage programs to teach students only in skill areas in which students scored the lowest on tests or only in skill areas in which gain is expected. The policy also is not intended to restrict the number or type of assessments used to identify a student's needs and to customize an instructional program to meet those needs. In fact, programs are expected to teach students in multiple areas depending on the students' needs and the programs' curricula.

We do not agree with the commenter that using the lowest test score might introduce systematic, rather than random, measurement error. Using the lowest test score in order to operationalize educational gain in a consistent manner across States for reporting purposes is error free. If tests have met the standards established in these regulations, they should be able to assess skills at any level with minimum error.

Changes: None.

Measuring educational gain (§ 462.43)

Comments: A commenter suggested the Department measure significant gain as determined by test developers. The commenter thought this approach would be the most accurate measure of gain. The commenter opposed the approach in proposed § 462.43, stating that it would only capture learning gain when a learner completes an educational functioning level by crossing artificial cut points regardless of the starting level. Another commenter stated that the approach for measuring gain in § 462.43 is “too coarse, is likely to capture only extreme gain, and will miss very significant educational gains that occur within an EFL [educational functioning level].” The commenter suggested that the Department consider using other options for demonstrating educational gain such as “achieving a gain score that is beyond chance expectations (using a scale score, which is more reliable than a proficiency classification).” Another commenter stated that “awarding only one educational gain to students completing more than one educational functioning level to a high degree misrepresents and under reports the effectiveness and accomplishment of both adult teacher and learner.” This commenter also stated that the current NRS policy of reporting an educational gain for obtaining a General Educational Development (GED) diploma by learners in ASE II, but not to those in ABE or ASE I, misrepresents and under-reports accomplishments. A different commenter suggested that attainment of a GED diploma should be recognized as an educational gain.

Discussion: The approach for measuring educational gain in § 462.43 is well established and accepted by the field. Over a two-year period, the Department consulted with States and convened a statutorily mandated panel in order to develop a performance accountability system for the adult education program. During that time, consensus was reached on defining the performance measures, including how to measure educational gain. States and other participating entities agreed that the approach in § 462.43 is the most effective means of obtaining accurate, reliable, and valid information on student performance.

Changes: None.

High Advanced ESL (§ 462.44)

Comments: Some commenters opposed proposed § 462.44 because it, like the Guidelines, eliminated the “High Advanced ESL” literacy level from the Functioning Level Table. The commenters noted that the change means that State and local agencies will no longer be able to report educational gain for adult English language learners above the “Advanced ESL” literacy level or Student Performance Level (SPL) 6. The commenters stated that, based on communications with the Department's Office of Vocational and Adult Education (OVAE), the change also means that programs can no longer provide “High Advanced ESL” instruction in adult education programs. The commenters indicated that OVAE has suggested that students who would have received “High Advanced ESL” instruction should now complete their preparation by moving from ESL classes to either adult basic education or adult secondary education classes. The commenters expressed their belief that OVAE's suggestion is pedagogically problematic and, more importantly, would prevent programs from adequately addressing the learning needs of ESL learners. Further, the commenters opposed stopping ESL instruction at the “Advanced ESL” literacy level or SPL 6 because they believe students at that level do not have language and literacy skills that are sufficient to transition successfully to postsecondary education or to meet the demands of the workplace—two of the purposes for adult education that are cited in the Act. Commenters noted that the regulations do not place this same restriction on native English speakers and recommended that the “High Advanced ESL” literacy level be restored to the Functioning Level Table. Finally, the commenters recommended that the regulations provide States with flexibility and not limit program services.

Discussion: Commenters are correct in that proposed § 462.44 would codify a change in the Guidelines by eliminating the “High Advanced ESL” literacy level from the Functioning Level Table. The change was made in 2006 as part of the information collection request for the Guidelines. However, before the change was made, (1) OVAE consulted with Adult Education State Directors and (2) the Office of Management and Budget (OMB) provided interested Federal agencies and the public, including States and local eligible providers, an opportunity to comment on the information collection request, which included the removal of the “High Advanced ESL” literacy level from the Functioning Level Table. We received no negative comments regarding our intent to eliminate the “High Advanced ESL” literacy level, and therefore, changed the Functioning Level Table in the Guidelines and in the proposed regulations.

The change in the Functioning Level Table does not mean that adult education programs can no longer provide services to “High Advanced ESL” students. These regulations are consistent with the Act, which authorizes services below the postsecondary level, and do not change who can be served by adult education programs. The educational functioning levels largely serve to classify students for reporting purposes and should not be viewed as the standard for establishing the type of instruction that can be provided. Placement of a student in an educational functioning level for NRS purposes does not affect placement in an instructional program. A student who scores above an SPL 6 might still be served in adult education programs if he or she has an educational need below the postsecondary level. States have the flexibility to determine how programs are structured, how students are placed in programs, and how instruction is delivered.

Changes: None.

Student Performance Levels (SPL) (§ 462.44)

Comments: A commenter requested clarification of the relationship between the NRS literacy levels and SPLs in the Functioning Level Table. The commenter indicated that no exact definition of SPL levels is provided and, therefore, their relationship to the NRS levels is unclear.

Discussion: The SPLs were developed by the Center for Applied Linguistics to provide a standard description of adult refugees' abilities at a range of levels and a common standard for ESL level descriptions for use by programs nationwide. They are nationally recognized in the adult ESL education community and represent a standard metric for identifying skill levels of adult ESL students in general language ability, listening comprehension, and oral communication. The Functioning Level Table provides educational functioning level descriptors for students at literacy levels in ABE, ASE, and ESL. The literacy levels are divided into equivalent grade levels for ABE and ASE and into SPLs for ESL. The descriptors illustrate the types of skills students functioning at a given level are likely to have. The descriptors do not provide a complete or comprehensive delineation of all the skills at a particular level, but instead, provide examples to guide assessment and instruction.

Changes: None.

Executive Order 12866

We have reviewed these final regulations in accordance with Executive Order 12866. Under the terms of the order we have assessed the potential costs and benefits of this regulatory action.

The potential costs associated with the final regulations are those resulting from statutory requirements and those we have determined to be necessary for administering this program effectively and efficiently.

In assessing the potential costs and benefits—both quantitative and qualitative—of these final regulations, we have determined that the benefits of the regulations justify the costs.

We have also determined that this regulatory action does not unduly interfere with State, local, and tribal governments in the exercise of their governmental functions.

We discussed the potential costs and benefits of these final regulations in the preamble to the NPRM under the headings Significant Proposed Regulations (pages 61581 and 61582), Regulatory Flexibility Act Certification (pages 61582 and 61583), and Paperwork Reduction Act of 1995 (page 61583).

Paperwork Reduction Act of 1995

The Paperwork Reduction Act of 1995 does not require you to respond to a collection of information unless it displays a valid OMB control number. With the exception of §§ 462.10 through 462.14, we display the valid OMB control number assigned to the collection of information in these final regulations at the end of the affected sections of the regulations.

Intergovernmental Review

These regulations are subject to Executive Order 12372 and the regulations in 34 CFR part 79. One of the objectives of the Executive order is to foster an intergovernmental partnership and a strengthened federalism by relying on processes developed by State and local governments for coordination and review of proposed Federal financial assistance.

In accordance with the order, we intend this document to provide early notification of the Department's specific plans and actions for this program.

Assessment of Educational Impact

In the NPRM we requested comments on whether the proposed regulations would require transmission of information that any other agency or authority of the United States gathers or makes available.

Based on the response to the NPRM and on our review, we have determined that these final regulations do not require transmission of information that any other agency or authority of the United States gathers or makes available.

Electronic Access to This Document

You can view this document, as well as all other Department of Education documents published in the Federal Register, in text or Adobe Portable Document Format (PDF) on the Internet at the following site: http://www.ed.gov/news/fedregister.

To use PDF you must have Adobe Acrobat Reader, which is available free at this site. If you have questions about using PDF, call the U.S. Government Printing Office (GPO), toll free, at 1-888-293-6498; or in the Washington, DC, area at (202) 512-1530.

Note:

The official version of this document is the document published in the Federal Register. Free Internet access to the official edition of the Federal Register and the Code of Federal Regulations is available on GPO Access at: http://www.gpoaccess.gov/nara/index.html.

(Catalog of Federal Domestic Assistance Number does not apply.)

List of Subjects in 34 CFR Part 462

• Administrative practice
• Adult education
• Grants program—education
• Incorporation by reference, and Reporting and recordkeeping requirements

For the reasons discussed in the preamble, the Secretary amends title 34 of the Code of Federal Regulations by adding a new part 462 to read as follows:

PART 462—MEASURING EDUCATIONAL GAIN IN THE NATIONAL REPORTING SYSTEM FOR ADULT EDUCATION

Authority: 20 U.S.C. 9212, unless otherwise noted.

Subpart A—General

Subpart B—What Process Does the Secretary Use To Review the Suitability of Tests for Use in the NRS?

Subpart C—[Reserved]

Subpart D—What Requirements Must States and Local Eligible Providers Follow When Measuring Educational Gain?