ED Response to OMB Comments

Memorandum United States Department of Education

Institute of Education Sciences

National Center for Education Statistics

DATE: September 16, 2010

TO: Shelly Martinez, OMB

FROM: Stephen Cornman, NCES

THROUGH: Kashka Kubzdela, NCES

RE: Response to 09-15-10 and 09-16-10 OMB Passback for TCS 2010-2013 (OMB# 1850-new v.1)

Supporting Statement, Part A

A3 – Here it says that “NCES never sees the statewide teacher ID,” but in A10, it says that data will be maintained on NCES servers in the coming year.  Please clarify.

When the Census Bureau collects the data from the State Educational Agencies (SEA’s) they create an NCES teacher ID for each teacher.  NCES utilizes only the NCES Teacher ID and does not ever see any state identifiers that were transmitted to the Census Bureau.  Although Census has a crosswalk between the state teacher ID’s and the NCES Teacher ID, NCES does not ever see that crosswalk.

After consultation with Marilyn Seastrom, we have removed the language in A-10 that “data will be maintained on NCES servers in the coming year.”

A4 – instead of “anonymous,” we think you mean “de-identified.”

We deleted the word anonymous.

A10 – Please clarify whether all of the data collected from the States will be protected under ESRA or whether it is only certain variables.  If partial, which is quite atypical but not unprecedented in the statistical system, we will need to work carefully with NCES to be sure that this is fully communicated to respondents.

The Education Sciences Reform Act (ESRA) applies to all TCS variables in all States.  (We consulted with Marilyn Seastrom on this issue.)

Supporting Statement, Part B

Introductory section – In the final sentence, what does it mean that “the designation” may be changed?

This sentence has been changed to “Items reported as missing or not applicable are confirmed with SEAs and the responses may be changed accordingly.”

Questionnaire

Item 17.  We recommend changing “guarantee” to a word like “assurance.”

We have adopted this recommendation.

Instruction Manual

Introduction.  The last line of the first paragraph seems to have a typo “Title P.L….”  Also, we prefer to see the standard NCES citation for confidentiality rather than this one.  Also, the order of the sentences in the final paragraph should be reversed to provide the broad confidentiality assurance before getting into issues of researcher access since those are governed by the confidentiality protections in the statute.

We have inserted the correct statutory authority for both NCES and Census to conduct this data collection.  We have inserted the standard NCES citation for confidentiality under ESRA.

Background.  Why is section this so lengthy and policy oriented?  Particularly given the burden associated with this collection, we recommend trimming substantially to about one paragraph.  We also are concerned that the second paragraph seems to be equating teacher quality with years of experience and training.

We have trimmed this introduction to one paragraph with language that has been used in the recent R&D reeport and approved by IES.

General Information on the Website.  We recommend changing the sentence about the Census Bureau editing responses (which may concern or provide perverse incentive to provide high quality data) to something like “The Census Bureau is the data collector for NCES.”

We have also adopted this recommendation.

9-16-2010 Second Passback

One follow up on the confidentiality.  If we are offering to keep all data submitted confidential, then why ask about whether the states keep the teacher variables confidential?  How will NCES use responses to those questions?

Through these questions NCES will be able to identify those states that make these data publicly available. NCES will use this information for two purposes. For the current year’s data, since those data were collected under a pledge of confidentiality, NCES will need to take additional precautions in preparing a public use file in order to guard against disclosures in the NCES data compared to the publicly available records. In subsequent years, NCES will be able to segment the data collection and not extend a pledge of confidentiality to those states that place this information in the public domain.