American Customer Satisfaction Index - Methodology

Methodology

The ACSI uses two interrelated methods to measure customer satisfaction: customer interviewing and econometric modeling. Beginning with the interviewing, professional telephone interviewers working for a market research firm contracted by the ACSI and employing Computer Assisted Telephone Interviewing (CATI) technology collect data (in the form of survey responses) from randomly selected and screened customers of companies. The random-digit dial method of sampling is used to identify potential respondents, guaranteeing an accurate representation of the U.S. consumer population. In addition to these methods, the ACSI also collects a portion of its data using opt-in online interviewing and sampling from a representative Internet panel.

ACSI researchers analyze this data once collected with a structural equation model, which provides scores for the measured latent variable components (such as customer expectations, overall quality, perceived value, etc.), and the relationships (or "impacts") between these measured components. Most importantly, each measured company or organization receives a customer satisfaction index score (an "ACSI score") which reflects a weighted average of three satisfaction proxy questions. Each index score is on a 0-100 scale, and a company can (hypothetically) receive any score ranging from 0 to 100. In practice, over the history of the ACSI scores have tended to range from the low 50's to the high 80's. While slight differences between questionnaires administered to respondents across industries and sectors do exist, the three satisfaction questions used to create the ACSI score for each company are identical. Coupled with the standardized 0-100 index scale, these methods allow maximum comparability between companies and government agencies.

Using these methods, each year ACSI produces customer satisfaction scores for more than 225 companies, 45 industries, 10 economic sectors and the U.S. national economy overall. Measurement is done on a rolling basis. During each fiscal quarter, data is collected for particular sectors and industries and used to replace data collected 12 months earlier. This data is then weighted by company market share up to industry scores, by industry revenue to create sector scores, and by sector share of GDP to create the National ACSI score. The National ACSI score represents, albeit at a level of abstraction, the satisfaction of the "average American consumer." This broad perspective allows ACSI researchers to examine the impact of improving or declining satisfaction on macroeconomic performance. Once completed, all of these results are released to the public on a monthly basis through the ACSI website and a variety of media outlets.