Section B. Description of Statistical Methodology
B.1. Respondent Universe
The StLA survey is a universe survey of StLAs in the 50 states and DC (51 total). The chief officer of the StLA designates a staff person as the “Key Holder” with primary responsibility for completion of the survey, although additional staff may assist in its completion. In most states, the key holder is the Public Libraries Survey state data coordinator.
B.2. Statistical Methodology
Not applicable, as this is a universe survey.
B.3. Methods for Maximizing Response Rate
The official request for data collection is mailed to the chief officers of StLAs and stresses the importance of their participation in the survey, as partners in the process. The key holders receive a separate letter that recognizes their primary responsibility for completion of the survey and encourages their response.
The web survey was designed to minimize response burden, to enable timely data submissions of high quality, and to require minimal or no edit follow-up for data problems. A user guide and on-line tutorial is included in the web survey application. A 'Help' function quickly links data items to their definitions. An edit check tool alerts the respondent to questionable data during the data entry process (via on-screen "edit check warnings") and through an edit check report that can be viewed on-screen or printed. These tools ease the completion of the survey and enable the review and submission of the data in a timely fashion. In addition, the survey is transmitted with pre-entered prior-year data for items that are not expected to change annually—most of Parts A through E and M, or about 65 percent of the survey items. The respondent is instructed to review the pre-entered data and update any information that has changed from the previous year. (All other data cells are left blank for the respondent to fill in, not update.)
As the survey due date approaches, the Census Bureau contacts nonrespondents to encourage their completion of the survey. IMLS staff also contact the state library agencies, if necessary. IMLS Library Statistics Working Group (LSWG) members are available as mentors and may provide technical assistance to states in reporting their data. Census conducts edit follow-up of the data submissions and tries to obtain any data not reported in the original submission, in order to maximize response rate.
The John Lorenz Award (a certificate signed by the Director of IMLS) is presented to Chief Officers and Key Holders who submitted their data a month before the due date.
B.4. Tests of Procedures and Methods
The web-based reporting system is updated and maintained by a contractor. Census serves as primary administrator of the data collection. IMLS, the Census Bureau and the contractor share responsibility for testing the product prior to its official release.
The LSWG provides guidance in survey design and content, and makes recommendations for improvements to the survey. The web application’s operation is reviewed on an ongoing basis by Census, and improvements are made as needed. Evaluations of the reliability and validity of the data are conducted periodically. The edit specifications are evaluated as part of every survey cycle.
B.5. Reviewing Statisticians
Suzanne Dorinksi, a mathematics statistician for the Census Bureau, reviews the data and is consulted on technical aspects of the data collection. In addition to Ms. Dorinksi’s review, the survey statistician at IMLS will conduct a final data review prior to release of the data and report.
| File Type | application/msword |
| File Title | Section B |
| Author | Barbara Smith |
| Last Modified By | KMILLER |
| File Modified | 2010-08-31 |
| File Created | 2010-03-10 |