Response to OMB Comments

memorandum

to: Shelly Wilkie Martinez, OMB

from: Monica Hill and Kathryn Chandler, NCES

subject: Responses to OMB Comments Received September 21, 2009

SSOCS 2010 & 2012 clearance, OMB# 1850-0761 v.4

date: September 25, 2009

CC: Kashka Kubzdela, NCES

The following memo details NCES responses to the OMB comments received 09/21/09.

1.   Please clarify whether NCES proposes to collect this information under CIPSEA (as suggested in SS A10) or not (as suggested in the affidavit and principal letter in Appendix C as well as the questionnaire).

• Because names and contact information for respondents is maintained, our understanding is that SSOCS cannot collect information under CIPSEA. Consistent with IES/NCES data collections that use and maintain respondent’s name and contact information, the following language appears in section A.10. We have added the highlighted language for clarification:

All information about individual respondents will be used only for statistical purposes and may not be disclosed, or used, in identifiable form for any other purpose except as required by law in compliance with the Education Sciences Reform Act of 2002 (ESRA 2002) Public Law 107-279, Title I, Part E, Section 183 which, except as amended by the Patriot Act of 2001, states that

No person may--

• use any individually identifiable information furnished under this title for any purpose other than a research, statistics, or evaluation purpose;

• make any publication whereby the data furnished by any particular person under this title can be identified; or

• permit anyone other than the individuals authorized by the Director to examine the individual reports.

• We understand the confusion in that authorizing legislation is included in section A.10, and we have therefore deleted the following sentence of this section:

  • The Census Bureau will collect the data for NCES by the authority of Public Law 107-279, Title I, Part C, Section 183 of the Education Sciences Reform Act of 2002 (20 USC 9573), which guarantees the protection of respondents’ data from disclosure in identifiable form, except as required by law.

• We do not see any specific mention to CIPSEA in section A.10. Could you clarify which statement refers to CIPSEA so that we can remove it?

• In addition, we have updated the confidentiality language in the questionnaire and in the principal, CSSO, and superintendent letters. In the questionnaire (supporting statement D) and the principal letters (supporting statement C) we have modified the language to read: Your answers may be used only for statistical purposes and may not be disclosed, or used, in identifiable form for any other purpose except as required by law [Education Sciences Reform Act of 2002 (ESRA 2002) Public Law 107-279, Section 183] and U.S. Code Title 20 Section 9573. Reports of the findings from the survey will not identify participating districts, schools, or staff. Individual responses will be combined with those from other participants to produce summary statistics and reports. In the CSSO and Superintendent letters (supporting statement C) we have modified the language to read: We recognize that some schools may not want to share information related to crime for fear of receiving negative attention. Please be assured that by federal mandate we are required to protect the identity of all schools included in our survey from public disclosure. The data we collect may be used only for statistical purposes and may not be disclosed, or used, in identifiable form for any other purpose except as required by law [Education Sciences Reform Act of 2002 (ESRA 2002) Public Law 107-279, Section 183] and U.S. Code Title 20 Section 9573. Reports of the findings from the survey will not identify participating districts, schools, or staff. Individual responses will be combined with those from other participants to produce summary statistics and reports.

2.   Did NCES test the likely effect on estimates of the changes in definitions of key terms, especially the new emphasis that boys can be victims of rape or sexual assault?

• NCES did not test likely effects on estimates of the changes in key terms. The definition changes for sexual harassment and sexual battery listed above were made in response to researchers’ requests to clarify and refine the definition and in order to bridge the NCES definition with the definition used by the Office of Civil Rights (see http://www.ed.gov/about/offices/list/ocr/docs/ocrshpam.html#_t1). It was NCES’ and the researchers’ concern that inconsistency between the two may cause confusion for school administrators in their efforts to comply with Title IX. NCES believes creating a more consistent definition in an effort to assist data users and administrators outweighs the potential change in estimates. NCES will footnote these estimates where appropriate to alert users to the change in definition.

• NCES will request that Census keep detailed records of follow up calls and outcomes for the variable C0310 (the numbers of rape or attempted rape) to determine whether over reporting is an issue for this variable. In addition, after data collection, NCES will analyze changes in this variable and will submit the results of that analysis to OMB within one month of receiving the final SSOCS:2010 data file.

3.   Can NCES clarify the reporting period in the instructions?  The survey instructions indicate: "Some questions refer to the 2009-10 school year. Please report for the school year to date."  If it is not specified, is it assumed that principals will report for both the 2008-2009 school year and the 2009-2010 school year? 

• The questions that do not refer to the specific timeframe are typically opinion items or skip patterns. Items 8, 9, 10, and 11 are part of the skip pattern that asks principals about the security staff reported in item 7, which specifies the 2009-10 school year. Items 13, 20, and 21 ask the principal for a best estimate or best of their knowledge estimate and therefore NCES has opted to not include the specific timeframe. Items 26-31 are demographic items which seek to capture a snapshot of the school characteristics at the time the principal is responding to the questionnaire.

• We do not intend for the principal to report on any item for the 2008-09 school year. The questionnaire will be sent to the principal in 2010, during the 2009-10 school year and it is therefore unclear how the absence of timeframe information for opinion or demographic items would cause confusion between the 2009-10 school year and the 2008-09 school year.

4.   Formatting issue: the answers to Question 22 subparts E and F need to be better aligned to the questions.

• We have modified the formatting so that the response boxes are better aligned to the questions. When the questionnaire is produced, we will ensure the alignment is correct.

5.   Is NCES confident that Question 23 has been adequately tested?  We are concerned about its clarity. Specifically:

• Yes, NCES conducted both pilot and on-site evaluations of this question before its’ inclusion in SSOCS 2004, and more recently, item 23 was vetted through cognitive lab in its’ current form. Past and more recent results indicated that respondents were generally able to follow directions and complete this question. Based on the most recent cognitive lab, NCES received a suggestion to emphasize the “total students involved in recorded offenses” column. NCES implemented this suggestion by adding the border around the column to differentiate the total number of students column from the disciplinary actions column.

a.   The relationship between column 1 and columns 2-5 seems unclear.

• Because multiple students can be involved in an incident, and a student may or may not receive disciplinary action for an offense, it is important to ask principals both about the students involved and the disciplinary actions taken. While column 1 and columns 2-5 are not dependent on one another, together they provide an important picture of the types of disciplinary actions taken for a selected offense, and whether or not any disciplinary action is taken for an offense. This item has been put through cognitive lab and administered during on-site visits and this relationship was not questioned by respondents.

• Because it is possible that principals will record not only the number of offenders in the “total students” column but also victims or bystanders, NCES will not imply in any release that the difference between the number of students and the number of disciplinary action implies wrongdoing (i.e., that a student should have been disciplined, but was not). In addition, NCES will consider clarification in future iterations of the survey (for example, column heading changes, or stressing that the students recorded in column one should have committed the offense), and will cog test this question in the future.

b. The rationale for changes noted that respondents double-counted columns 2 and 3. The change does not seem to obviously correct this concern. We can see how principals may regard the two choices as the same thing.

• Because the severity of disciplinary actions is subjective, the modification to the instructions emphasizes to principals to only count the most severe disciplinary action, and emphasizes that only one disciplinary action should be recorded for each offense. A definition for “specialized schools” is also provided on page 2.

c.   Is there a common understanding of what is the "most severe disciplinary action that was taken?" Is this just trying to avoid double-counting?

• The severity of disciplinary actions is subjective. The purpose of instructing respondents to count only the most severe action is not to eliminate double counting of the same disciplinary action, but rather to avoid double counting disciplinary actions for the same offense. That is, if a student is given both a detention and suspension for the same offense, only the action deemed more severe by the respondent should be recorded.

d. One general observation: this question seems very extensive for a school official who is supposed to work off estimates, rather than records. We are concerned that asking for such precise figures will give the responses the appearance that they are "counts" rather than estimates.

• NCES agrees that asking for such precise figures gives respondents the appearance that they should consult records and this is our intention. The instructions state that “it is not necessary to consult any records for items 5 and 27” only. NCES intends the rest of the questionnaire should be completed using records as a reference.

6.   We also found Question 24 confusing:

a.   Is the note at the end of subparts a and b really referring to 22 rather than 23? 

• The note at the end of subparts a and b should refer to item 23. This was a typo and has been corrected.

b. Why should the number "be greater than or equal to the sum of entries" in item 23?  We can think of examples where that may not be the case.

• The total number of removals for disciplinary actions cannot exceed the sum of removals for specified offenses. The total number of removals for disciplinary actions can be greater than the sum of removals for specified offenses, in that students may be removed for offenses not listed in item 23. Could you provide an example of a situation in which the number of removals for all disciplinary reasons could be less than the number of removals for certain disciplinary reasons?

• In the scenario where a student is transferred from the school (and has been removed from the school), the principal is instructed in item 23 to record only the most severe action, but in item 24 is instructed to include all removals and transfers. If a principal is confusing a removal and transfer, the student may be recorded in both items 23a and 23b. In future iterations of SSOCS, NCES will consider emphasizing the difference between removals and transfers and will cog lab this item.

4