In fact, NASA’s EOSDIS has never received a CSI score lower than 74 since the first EOSDIS ACSI survey was conducted in 2004. As with all previous surveys spanning more than 15 years, NASA’s EOSDIS outscored the federal government, which received an aggregate 2020 CSI score of 68. In addition, CSI scores for EOSDIS Distributed Active Archive Centers (DAACs) ranged from 71 to 83, with an average aggregate CSI of 78.
The results from the annual ACSI survey provide something of a blueprint that the DAACs and NASA’s Earth Science Data and Information System (ESDIS) Project (which manages EOSDIS science operations, including data archival and distribution) can use to tailor their products and services so they better align with and serve user needs.
The ACSI survey is administered by the CFI Group, an independent organization contracted by the federal government to assess user satisfaction with products and services at numerous federal entities. Along with conducting the annual NASA EOSDIS survey, the CFI Group also conducts surveys for the National Weather Service, the General Services Administration, and the U.S. Department of Education, among others.
The ACSI model used by the CFI Group to conduct the annual survey is a set of causal equations linking customer expectations, perceived quality, and perceived value to customer satisfaction, which is reflected numerically in the CSI score. Satisfaction, in turn, is further linked to a customer’s likelihood to recommend products and services and their willingness to use products and services in the future.
One expected result of high customer satisfaction with services is user trust and loyalty. This is accounted for in the ACSI algorithm and reflected by a number indicating the likelihood of a respondent to recommend the evaluated products and services to others coupled with the likelihood of a respondent to use the services in the future.
Scores pertaining to respondents’ “likelihood to recommend EOSDIS products and services” (87 out of 100) and “likelihood to use EOSDIS services in the future” (88 out of 100) were one point higher than in the 2019 survey (although one point lower than in the 2018 survey). These consistent rankings suggest that, over the years, EOSDIS has consistently provided products and services its users value.
ESDIS, which coordinates and facilitates the annual ACSI survey, opted to use a “long-form” version of survey questionnaire in 2020. One benefit of this longer survey are the more detailed and in-depth questions allowing for a more complete survey of EOSDIS products and services. In all, 9,178 completed surveys were returned in the 2020 vs. 6,337 completed surveys in 2019. This higher response rates lends further credence that the 2020 survey more accurately reflects user satisfaction. Starting with its 2019 survey, the ESDIS Project began alternating short-form surveys with more detailed longer surveys as a way to maintain continuity with questions from longer surveys (e.g., the 2020 survey was a long-form survey; the 2021 survey will use the short-form, and so on.).
The 2020 survey was conducted online between September 21 and October 23. The CFI Group sent email invitations to individuals who used EOSDIS data and/or products. They received 9,178 completed surveys, a significant increase from last year’s 6,337, representing a 1.2% response rate and, according to the CFI Group, a large enough sample size for calculating a statistically valid CSI score.
Outside of three survey questions required by the CFI Group asking respondents to rate their overall expectations, perceived quality, and perceived value of EOSDIS services and products on a 1 to 10 scale (which are used to calculate the CSI score), the remaining questions on the EOSDIS survey asked respondents to evaluate their experience with the specific DAAC or DAACs from which they receive data and products (respondents were allowed to skip questions pertaining to DAACs they did not consult with or contact). The survey also allowed respondents to provide open-ended comments, providing EOSDIS with direct and valuable insights regarding respondents’ likes, dislikes, and thoughts on suggested improvements.
In the following summary tables, total percentages may not equal 100% due to survey questions allowing for multiple responses; all non-percentage values are out of 100. (Click on any image for a larger view.)