Research Reports

You are here: Census.govSubjects A to ZResearch Reports Sorted by Year › Abstract of RSM2013/22
Skip top of page navigation

2005 Census Internet Prototype Applications: Usability and Accessibility Testing

Elizabeth D. Murphy, Susan Ciochetto, Kathleen T. Ashenfelter, Lawrence A. Malakhoff, David Coon, and Nika Smith


2005 National Census Test, Internet, usability, accessibility, person-based form, topic-based form.


We performed usability and accessibility testing of the Internet form for the 2005 National Census Test (NCT) in May, 2005. The purpose of usability testing was to identify any aspects of the user-interface design that might detract from the success and satisfaction of Internet respondents during the 2005 NCT. The purpose of accessibility testing, with automated software and test participants, was to identify any violations of the Federal accessibility regulations (Section 508) and detect user problems with their understanding of labels in the Internet form. Twenty-two participants were recruited for usability testing and evenly divided between the topic-based (i.e., all question topics –age, sex, race, etc.- for each person of the household before moving on) and the person-based (i.e., all questions asked for each person in a household before moving on) forms, by a process roughly resembling alternating. Most of the accessibility testing used automated tools, but we did recruit two Census employees who used a screen-reader and screen magnification, respectively, to access the Internet form. Both of these Census employees evaluated the topic-based form. The screen reader user also evaluated the person-based form.

All usability test participants completed their forms and exceeded the goal of 80 percent completion. Data-entry accuracy was high. Data-entry efficiency ranged from 70 percent for the person-based version to 90.5 percent overall, when adjusted for outliers. Navigational effectiveness was high. However, 82 percent of topic-based participants committed field-changing errors; and 70 percent of person-based participants had some difficulty moving between screens. Ninety-one percent of participants completed the form efficiently. Accessibility testing with two Census Bureau employees confirmed the Section 508 issues identified by the automated accessibility testing. Although both users with visual disabilities completed their forms, they often faced usability obstacles in the process.


Elizabeth D. Murphy, Susan Ciochetto, Kathleen T. Ashenfelter, Lawrence A. Malakhoff, David Coon, and Nika Smith. (2013). 2005 Census Internet Prototype Applications: Usability and Accessibility Testing. Center for Survey Measurement Study Series (Survey Methodology #2013-22). U.S. Census Bureau. Available online at <>.

Source: U.S. Census Bureau, Research and Methodology Directorate, Center for Survey Measurement

Published online: November 6, 2013
Last revised: November 6, 2013


Source: U.S. Census Bureau | Research and Methodology Directorate | Center for Survey Measurement | (301) 763-3215 (or |   Last Revised: November 06, 2013