ALS 2010 2012 Supp Stat Part B revised

A.Collection of information employing statistical methods

1. Respondent Universe

The respondent universe for the 2010 and 2012 ALS will be taken from the 2009 IPEDS database of at least a 2-year, postsecondary, degree granting institutions that are eligible for Title IV funding, and have libraries.

Table 1. Postsecondary, degree granting institutions that are eligible for Title IV funding, and are expected to have libraries, by sector: 2008.

	2-year institutions	4-year institutions	Total
Public	966	610	1,576
Private
Private non-profit	82	1,430	1,512
Private for-profit	386	353	739
Total	1,434	2,393	3,827

SOURCE: U.S. Department of Education, National Center for Education Statistics, Academic Libraries Survey, 2008.

2. Statistical Procedures for Collecting Information

ALS is a census of academic libraries, and does not involve sampling.

3. Procedures for Collection of Information

For the 2010 and 2012 rounds of ALS, NCES will continue to use a web-based collection. Opening and closing dates of the survey will be broadcast over the ALS web site. For respondents that are unable to report via the web, ALS staff will enter the data for the respondent, produce an edit report, and resolve potential errors. When an institution submits the data, responses will be reviewed for consistency and validity. The data will be edited for internal consistency and prior-year comparisons will be performed, based on detailed standards provided by NCES. Respondents will be contacted to resolve potential errors. Pre-tabulation data checks will be performed, based on detailed specifications provided by NCES, to ensure data quality prior to tabulation.

4. Methods for Maximizing Response Rates

As indicated previously, ALS is an electronic survey that was redesigned in FY 2000 to collect data over the Internet via a web-based reporting system. Respondents enter their data into the web survey and edit the data, assisted by interactive error warnings and error reports built into the web application. When they have completed the survey, they lock the data to indicate it is final and ready for access by NCES. The web application is designed as a user friendly and efficient tool for reporting data on a timely basis and with as few errors as possible. The web reporting system includes a User Guide and a Help function that explain the use of the web application, the survey instructions and definitions, and a directory of library representatives and contact persons as a resource in responding to the survey. As the survey due date approaches, the Census Bureau will contact nonresponding institutions to encourage their participation in the survey.

NCES also attempts to maximize response by reducing response burden. The survey universe is limited to only postsecondary, degree granting institutions that are eligible for Title IV funding, and have libraries. NCES also coordinates library professionals within the states who assist in the nonresponse follow-up process by contacting the institutions within their state that have not responded to the ALS questionnaire.

The ALS survey is recognized and supported by the American Library Association (ALA). As part of the biennial data collection process, a letter from the NCES Commissioner is mailed to the chief academic officers of the institutions and the library directors describing the survey and requesting their participation in the survey.

The Census Bureau conducted a study of the 2008 ALS data to identify potential nonresponse cases. This information will be used to assist us in nonresponse follow-up for the 2010 survey.

5. Tests of Procedures and Methods

The data collection procedures and data items described in this submission have been thoroughly tested. Most of the data elements requested have already been collected in previous IPEDS surveys since 1988. Other items have been successfully used in surveys conducted by library associations.

Data quality is a constant concern and NCES continues to assess and evaluate the appropriateness and effectiveness of the ALS questionnaire items. One approach NCES uses is to assess relevant data from the Academic Libraries Survey from different years to evaluate the consistency and reliability of reported data. The interrelationships among surveys and the relationship of items over time were used, during the most recent administration of ALS, to develop the automated tests for editing each data submission. Edit checks currently help to identify potential problems and provide opportunities to correct them early in the data collection process. By implementing a web-based data collection effort that requires error resolution and correction prior to data submission, NCES anticipates producing improved data quality in a more timely fashion.

The ALS survey data are analyzed by NCES. The evaluation includes a review of the data for reliability and validity; a review of the data items, definitions, and instructions; the comparability of data from state to state, and recommendations for improvements to the survey. The ALS Technical Working Group is actively involved in survey design and content, and makes recommendations for improvements to the survey.

The ALS survey system is maintained and updated by an NCES contractor. The contractor is responsible for testing the product, with assistance from Census and NCES staff. The web-based data collection method was tested in a successful pilot collection of the 1999 IPEDS data and in the full-scale collection of the 2000 ALS.

6. Reviewing Statisticians

The ALS survey is conducted by NCES. The U.S. Census Bureau is the data collection and processing agent for the survey, through an interagency agreement with NCES. At NCES, Marilyn McMillen Seastrom, Chief Statistician, and Neil Russell, Mathematical Statistician, are consulted on technical aspects of the data collection. Terri Craig, Chief of the Statistical Methods Branch at the Census Bureau, is consulted on issues concerning imputation and data editing.

Tai Phan of NCES is the program director for ALS.

2