ACSI Methodology Paper

American Customer Satisfaction Index
Methodology Report

March 2005

Methodology

Market Strategies, Inc. Major Corporate Sponsor

American Customer
Satisfaction Index (ACSI)
Methodology Report

March 2005

 Copyright 2005, The Regents of the University of Michigan

PREFACE
ACSI is designed, conducted, and analyzed by the National Quality Research Center
(NQRC), Stephen M. Ross School of Business at the University of Michigan. The ACSI
technical staff includes:
•

Professor Claes Fornell, Donald C. Cook Professor of Business Administration, and
Director, National Quality Research Center

•

David VanAmburg, Managing Director, ACSI

•

Forrest Morgeson, Research Associate

•

Barbara Everitt Bryant, Ph.D., Research Scientist

•

Professor Eugene W. Anderson, Associate Dean for Degree Programs, Professor of
Marketing

•

Assistant Professor Fred Feinberg

•

Professor Michael Johnson, Dr. Maynard Phelps Professor of Business
Administration

•

Assistant Professor Carolyn Yoon

•

Julie M. Trombly, Editor and Graphic Production

For questions on research or an interpretation of this report, contact:
National Quality Research Center
Stephen M. Ross School of Business
University of Michigan
Ann Arbor, MI 48109-1234
Telephone: 734-763-9767; Fax: 734-763-9768
Web: www.theacsi.org

iii

Quarterly update reports on customer satisfaction are available at an annual subscription
rate of $129 ($99 to ASQ members) and are delivered electronically. To order ACSI
reports, contact:
ASQ
600 N. Plankinton Avenue
P.O. Box 3005
Milwaukee, WI 53201-3005
Telephone: 800-248-1946; Fax: 414-272-1734
Web: www.asq.org

iv

CONTENTS
PREFACE ...................................................................................................................................... iii
I. INTRODUCTION .......................................................................................................................1
A. HISTORY ..............................................................................................................................2
B. METHODOLOGY IN BRIEF ...............................................................................................4
II. PURPOSE, ECONOMETRIC MODELING, AND INDEX PROPERTIES.............................7
A. PURPOSE OF MEASUREMENT.........................................................................................7
B. ECONOMETRIC MODELING.............................................................................................8
1. MODEL AND METHODOLOGY.................................................................................8
2. THE CUSTOMER SATISFACTION INDEX (ACSI) ................................................11
3. ACSI ANTECEDENTS (DRIVERS OR CAUSES OF SATISFACTION).................11
4. ACSI CONSEQUENCES (OUTCOMES OF SATISFACTION)................................13
5. MANIFEST VARIABLES (QUESTIONS) USED IN THE MODEL.........................13
C. INDEX PROPERTIES.........................................................................................................17
1. PRECISION..................................................................................................................17
2. VALIDITY ...................................................................................................................17
3. RELIABILITY..............................................................................................................18
4. PREDICTIVE POWER AND FINANCIAL IMPLICATIONS OF ACSI...................18
5. COVERAGE.................................................................................................................20
6. SIMPLICITY ................................................................................................................21
7. DIAGNOSTICS............................................................................................................21
8. COMPARABILITY......................................................................................................22
D. THE ACSI EQUATIONS....................................................................................................23
E. THE ACSI FORMULA........................................................................................................25
III. SELECTION OF ECONOMIC SECTORS, INDUSTRIES, COMPANIES, AND
GOVERNMENT SERVICES...................................................................................................27

v

IV. QUESTIONNAIRES..............................................................................................................29
A. SCREENING QUESTIONNAIRES....................................................................................29
B. BRAND/COMPANY IDENTIFICATION..........................................................................30
C. QUESTIONNAIRE..............................................................................................................31
V. HOUSEHOLD SURVEY SAMPLE........................................................................................33
A. HOUSEHOLD SAMPLE AND TELEPHONE NUMBER SELECTION..........................33
1. SELECTION OF HOUSEHOLD NUMBERS FOR SCREENING.............................33
2. HOUSEHOLD SAMPLE SIZE ....................................................................................35
3. CALLBACKS AND HOUSEHOLD SUBSTITUTION ..............................................35
B. SELECTION OF DESIGNATED RESPONDENT WITHIN HOUSEHOLD....................35
C. SCREENING HOUSEHOLD RESPONDENTS TO IDENTIFY
QUALIFIED CUSTOMERS .................................................................................................36
D. INTERVIEW RESULTS .....................................................................................................36
1. RESPONSE AND COOPERATION RATES ..............................................................36
2. REFUSAL CONVERSIONS........................................................................................37
E. PROFILE OF INTERVIEWED CUSTOMERS ..................................................................37
VI. DATA COLLECTION VIA TELEPHONE ...........................................................................39
VII. E-BUSINESS AND E-COMMERCE SAMPLES AND DATA COLLECTION
VIA INTERNET .......................................................................................................................43
A. SAMPLE SELECTION .......................................................................................................43
B. SCREENING TO QUALIFIED CUSTOMERS..................................................................43
C. DATA COLLECTION VIA INTERNET ............................................................................44
REFERENCES ..............................................................................................................................45

vi

APPENDIX A: COMPANIES AND GOVERNMENT SERVICES EVALUATED BY
CUSTOMERS IN ACSI ............................................................................................................49
APPENDIX B: ACSI INDUSTRY DEFINITIONS AND CUSTOMER IDENTIFICATION....69
APPENDIX C: EXAMPLE SCREENING QUESTIONNAIRES AND
BRAND/COMPANY IDENTIFICATION ...............................................................................79
APPENDIX D: CUSTOMER SATISFACTION MEASUREMENT QUESTIONNAIRES—
PRIVATE SECTOR AND GOVERNMENT SERVICES........................................................91
APPENDIX E: RESPONSE AND COOPERATION RATES...................................................107
APPENDIX F: SAMPLE PROFILE ..........................................................................................113

vii

Introduction

I. INTRODUCTION
The American Customer Satisfaction Index (ACSI) is a uniform, national, cross-industry
measure of satisfaction with the quality of goods and services available to household
consumers in the United States. Established in 1994, ACSI is both a trend measure and a
benchmark for companies to compare themselves with others in their own or other
industries. It provides measures by which a large number of federal agencies, and two local
government services, can compare user satisfaction with the quality of their services over
time, and with services provided in the private sector.
Research shows that ACSI is predictive of corporate performance, growth in the Gross
Domestic Product (GDP), and changes in consumer spending (Fornell, 2001; commentaries
posted quarterly on the Web site: www.theacsi.org).
Produced by a consortium of the Stephen M. Ross School of Business at the University
of Michigan, ASQ,1 and CFI Group, ACSI is funded by multiple sources. These are:
(1) annual fees from corporate subscribers, (2) ACSI-related research conducted for
corporate subscribers, (3) licensing of the ACSI model, (4) Market Strategies, Inc., a major
corporate sponsor, and (5) sponsorship of the measurement and analyses of e-business and
e-commerce industries by Foresee Results.
The National Quality Research Center (NQRC) at Michigan’s Ross School of Business is
the research and production center for the index, analyses of data, and report writing. ASQ
distributes published reports and news releases. CFI Group provides software design and
assistance, as well as marketing consulting.
The ACSI statistical models, discussed in Chapter II, provide customer satisfaction
indices (on 0 to 100 scales) and indices of antecedents (drivers or causes) and outcomes of
satisfaction with the products and services of specific companies, government agencies, and
industries for ten economic sectors that market to U.S. household consumers. The sectors
are broadly representative of the national economy: (1) Utilities, (2) Manufacturing/
Nondurable Goods, (3) Manufacturing/Durable Goods, (4) Retail Trade, (5) Transportation
and Warehousing, (6) Information, (7) Finance and Insurance, (8) Health Care and Social
Assistance, (9) Accommodation and Food Services, and (10) Public Administration. ACSI
also measures e-business and e-commerce, but the companies within these fall under the
other economic sectors.

1

At the time of ACSI’s establishment, this organization was called the American Society for Quality Control.

1

The American Customer Satisfaction Index

ACSI Methodology Report
Each company or government service, industry, and sector is measured annually. The
national index is updated quarterly, on a rolling basis, with new data for one or more
measured sectors replacing data from the prior year.

A. HISTORY
Fifteen years ago, ASQ saw the need for a national measure of quality. In 1990-91, the
organization commissioned National Economic Research Associates (NERA) to determine
whether a national, cross-company, cross-industry measure existed or could be developed.
NERA examined 60 different approaches to measuring quality. There was no standard
definition of quality and indices for different categories of products and services in use in
the United States were not comparable. It was not possible to assign values to the separate
measures of quality in order to aggregate these into a national index. A key failing was the
inability to obtain quality measures that translated to customer-perceived value.
NERA concluded that a comprehensive assessment of quality required a mechanism that
assigned values to dimensions of quality that influence customer behavior and that any
design that did not reflect the customer’s voice and notion of value would not meet the goal
of a national quality index. NERA recommended adaptation of the Swedish Customer
Satisfaction Barometer to the United States.
According to the NERA report (National Economic Research Associates, 1991), the
Swedish Barometer, established in 1989, used an econometric model designed by Claes
Fornell and colleagues at the University of Michigan’s NQRC that (1) was the most
comprehensive effort to date to measure product and service quality; (2) illustrated
the feasibility of using the survey approach to assess quality on a broad scale; and
(3) recognized the necessity of relating measures of quality to customer behavior
(Fornell, 1992).
With funding from ASQ and individual corporations, NQRC conducted an extensive
design, development, and pretest phase in 1993. In the following year, the baseline
American Customer Satisfaction Index was produced covering 7 sectors of the economy,
30 industries, and 180 companies. The baseline ACSI study demonstrated that household
screening to identify qualified customers, interviews of these customers with the NQRCdesigned survey questionnaire, and econometric modeling could be used on a large scale to
produce comparable indices across a wide variety of companies and industries in the United
States.
As of 2005, ACSI covers 41 industries (including those in e-business and e-commerce)
and over 200 companies and federal or local government services. ACSI measures
2

ACSI: The National Measure of Customer Satisfaction

Introduction
satisfaction with companies that produce 43% of GDP and the specific products and services
of these companies that account for 36%. The first ten years provided both point-in-time
and trend measures of satisfaction based on approximately 650,000 interviews.
ACSI measures ten economic sectors in the North American Industry Classification
System (NAICS) that produce products and services sold directly to household customers.
These sectors are: (1) Utilities, (2) Manufacturing/Nondurable Goods, (3) Manufacturing/
Durable Goods, (4) Retail Trade, (5) Transportation and Warehousing, (6) Information,
(7) Finance and Insurance, (8) Health Care and Social Assistance, (9) Accommodation and
Food Services, and (10) Public Administration.2 The sectors included in ACSI produce 66%
of the GDP.
Not included in ACSI are Agriculture/Forestry/Fishing and Hunting, Mining,
Construction, Wholesale Trade, Real Estate/Rental/Leasing, Professional/Scientific/
Technical Services, Management of Companies and Enterprises, Administrative
Support/Waste Management and Remediation (although satisfaction with solid waste
disposal provided by local governments is measured in ACSI), Educational Services
(although satisfaction with some educational services provided by the federal government
is measured), Arts/Entertainment/Recreation, and Other Services.
Within each sector, satisfaction is measured with large companies in representative
industries. Within each industry, 2 to 30 companies are selected (although in most
industries the number is 4 to 8 companies). The companies are chosen to be those with the
largest U.S. market shares in each industry, whether or not the company is a domestic or a
non-U.S. company. It is the customers of these companies who are identified by interview
screening and then interviewed about their satisfaction with the specific company.
Each company, industry, and sector is measured annually. The national ACSI score is
updated quarterly, on a rolling basis, with new data for one or more measured sectors
replacing data from the prior year.

2

While the NAICS groups nondurable goods and durable goods together in a single manufacturing sector, ACSI measures
these as two separate sectors.

3

The American Customer Satisfaction Index

ACSI Methodology Report
Table 1: Data Collection and Sector Update Schedule
Sector

Data Collection Period

Utilities, Transportation &
Warehousing, Information,
Health Care & Social Assistance,
Accommodation & Food Services

ACSI Release of
Results

January to March

May

April to June

August

Manufacturing/Nondurable Goods

July to September

November

Retail Trade, Finance & Insurance,
E-Commerce

October to December

February

Public Administration

Throughout the year

December

Manufacturing/Durable Goods,
E-Business

The companies whose satisfaction is measured in ACSI as of early 2005 are listed in
Appendix A, along with their revenues and Fortune 1000 ranks.

B. METHODOLOGY IN BRIEF
The ACSI methodology is distinguished from other measures of quality by four
significant characteristics:
1. ACSI has a uniform, customer-based definition of quality: “customer satisfaction
with the quality of goods and services purchased and used.”
2. ACSI treats satisfaction with quality as a cumulative experience, rather than a mostrecent-transaction experience.
3. ACSI uses a cause-and-effect model that measures satisfaction quantitatively as the
result of survey-measured input of customer expectations, perceptions of quality, and
perceptions of value (i.e., quality for cost).
4. The ACSI model links satisfaction quantitatively with customer-survey-measured
outcomes: complaints (a negative outcome) and customer loyalty (a positive
outcome). Customer loyalty is derived from measures of customer retention and
price tolerance.
ACSI uses an empirically tested, cause-and-effect model. It is a multi-equation, latent
variable, econometric model that produces four levels of composite index measures. These
are: (1) a national customer satisfaction index; (2) indices for 10 sectors of the economy;

4

ACSI: The National Measure of Customer Satisfaction

Introduction
(3) indices for 41 industries; and (4) indices for over 200 major companies and federal or
local government services, including indices for an “all others” category in each industry.
Input to the econometric modeling comes from surveys conducted on a computerassisted-telephone-interviewing (CATI) system. Customers are selected randomly from
national and regional probability samples of continental U.S households. Random-digit-dial
(RDD) selection of households includes those with both listed and unlisted numbers.
Selection of a respondent within the household based on the individual with the most recent
birthday provides a representative distribution of respondents by age, gender, and other
characteristics. For e-business and e-commerce companies, selecting user samples,
screening respondents, and interviewing customers are all done on the Internet.
To be eligible for interview, either by telephone or online, a prospective respondent must
qualify as the purchaser of specific products or services within defined time periods. These
vary from three years for the purchase of major durables, to “in the last month” for
frequently purchased consumer goods and services, to currently having utility services,
insurance policies, or bank accounts in one’s own name. Thus the definition of “customer”
in the American Customer Satisfaction Index is an individual chosen randomly from a large
universe of potential buyers who qualifies by recent experience as a purchaser/user of
products or services of specific companies or agencies that supply household consumers in
the continental United States.
The process of qualifying respondents as customers of specific products and services and
thus eligible for interview is described in Chapter V, Household Survey Sample.
For federal agencies with high incidence of use by the public (for example, National Park
Service visitors or Internal Revenue Service tax filers), respondent selection and customer
screening are done in the same manner as for private sector companies (that is, from RDD
samples of U.S. households). For agencies for which users/customers represent a
specialized service (for example, Head Start parents, patent and trademark applicants, or
beneficiaries of entitlement programs), the procedure differs. Each agency supplies the
sampling frame of the customer population for a particular time period (or a large random
sample of that population) to NQRC, which then draws a random sample from that frame.
Completed survey interviews are input to the econometric model described in Chapter II,
Purpose, Econometric Modeling, and Index Properties, which computes the indices at
company or government service, industry, sector, and national levels.

5

The American Customer Satisfaction Index

Purpose, Modeling, and Properties

II. PURPOSE, ECONOMETRIC MODELING, AND
INDEX PROPERTIES
This chapter discusses the purpose of the American Customer Satisfaction Index; the
ACSI model; what the desirable properties of the index should be; and the extent to which
ACSI can be said to have these properties. ACSI aims to contribute toward more accurate
and comprehensive information on economic output as a measure of national customer
satisfaction and as a long-term indicator of economic returns for the nation and for
individual companies.

A. PURPOSE OF MEASUREMENT
ACSI provides a perspective for understanding the economy and national, industry, and
company competitiveness. The perspective is that of the customer’s experience with the
quality received from goods and services available in the U.S. marketplace.
Much attention has been focused on productivity as the key to competitiveness for
companies and nations. Ideally, productivity should reflect not only efficiency in
production, but also how quality and service are incorporated into market prices. In
practice, however, productivity measures often fall short, particularly in the service sector
where the value of improved or reduced quality is not easily captured.
Measurement of productivity relates to measurement of price changes. As noted by
several economists (for example, Gordon, 1990), the measurement of prices would be
straightforward if there were a single, generally accepted index of economic and social well
being that would tell how much better or worse off consumers are each year. Without good
measurement of both price and qualityand how these change over timeassessment of
productivity is extremely difficult.
As an economic indicator, ACSI suggests a context within which to interpret both price
and productivity changes. An objective of ACSI is to help with this interpretation by
capturing the elusive character of a product (attributes, price, market fit) from the user
perspective.
Economic data attempt to capture a myriad of transactions between buyers and sellers in
many types of markets. ACSI measures the overall satisfaction of buyers in household
consumer markets, but it is not limited to transactions per se. Rather, it is the subjective
evaluations of the goods and services acquired and consumed in the United States that are
measured. In the final analysis, all human decision making is subjective. It is the
7

The American Customer Satisfaction Index

ACSI Methodology Report
customer’s evaluationnot engineering standardsthat ultimately affects the demand
curve. Measurement of that evaluation, however, is not necessarily subjective.
Customer satisfaction also incorporates price, how well the companies have chosen their
markets, and the resulting degree of fit between the nature of demand and the nature of
supply. Further, customer satisfaction, in contrast to quality, assumes actual consumption
experience. Since most products and services are repeat purchases, customer satisfaction
has a large effect on demand.

B. ECONOMETRIC MODELING
1. Model and Methodology
The ACSI methodology has four basic properties:
1. ACSI uses an econometric model with measures of an index of satisfaction (ACSI)
and related indices for latent variables or constructs (boxes in Figures 1, 1a, and 2)
that are general enough to be comparable across companies, industries, and sectors.
These measures come from manifest variables (survey questions) that are inputs to
the model. The latent variables and their relationships apply to public services and
competitive product markets alike.
2. ACSI is embedded in a system of cause-and-effect relationships. This serves to
validate the index from a nomological standpoint. Nomological validity, a form of
construct validity, is the degree to which a construct behaves as it should behave
within a system of related constructs called a nomological net (Bagozzi, 1980;
Cronbach and Meehl, 1955). If the model predictions are supported, then the validity
of ACSI is supported.
3. Consistent with its definition, satisfaction is measured as a latent variable (central box
in Figures 1, 1a, and 2) using multiple manifest variables (questions). Any one
concrete measure of satisfaction, such as a single survey question, is at best a proxy
for latent satisfaction (Simon, 1974). Instead, ACSI uses several proxies that reflect
overall consumption experience. These proxies are combined into an index on a 0 to
100 scale to operationalize satisfaction.

8

ACSI: The National Measure of Customer Satisfaction

Purpose, Modeling, and Properties
4. A primary objective is to estimate the effect of ACSI on customer loyalty, a construct
of universal importance in the evaluation of current and future business performance.
Figure 1 shows the ACSI model used for the private sector, with an expanded version
shown in Figure 1a. Figure 2 represents the model used for government services
(applicable to nonprofits).
Figure 1. ACSI Model: Private Sector
Reliability

Customization

Overall

Complaint Behavior

PERCEIVED
QUALITY

Price
Given
Quality

CUSTOMER
COMPLAINTS

CUSTOMER
SATISFACTION
(ACSI)

PERCEIVED
VALUE

Quality
Given
Price

CUSTOMER
EXPECTATIONS

Reliability

Customization

Satisfaction

Comparison
Confirm/
Disconfirm With Ideal
Expectations

CUSTOMER
LOYALTY

Repurchase
Likelihood

Overall

Price Tolerance
(Reservation Price)

Expansion of the Model to Measure Product Quality and Service Quality as
Inputs to Perceived Overall Quality
In some industries, particularly those in the Manufacturing/Durable Goods and Retail
Trade sectors, the product and the service required to maintain it are provided over different
time periods. That is, there is an initial purchase followed by a period of maintenance. The
service provider may not be the manufacturer. For retailing, products are manufactured by
one company, but delivered by another. For those industries, ACSI uses the expanded
model shown in Figure 1a. Customer ratings of product and service quality are often
statistically significantly different, with product quality rated higher than service quality.

9

The American Customer Satisfaction Index

ACSI Methodology Report
Figure 1a. Expanded ACSI Model to Measure Product Quality and Service Quality
as Inputs to Perceived Overall Quality
Reliability
Customization
Overall

PERCEIVED
PRODUCT
QUALITY

Reliability
Customization
Overall

PERCEIVED
SERVICE
QUALITY
Complaint Behavior

PERCEIVED
OVERALL QUALITY

Price
Given
Quality

CUSTOMER
COMPLAINTS

CUSTOMER
SATISFACTION
(ACSI)

PERCEIVED
VALUE

Quality
Given
Price

CUSTOMER
EXPECTATIONS

Reliability

Customization

Satisfaction

Comparison
Confirm/
Disconfirm With Ideal
Expectations

CUSTOMER
LOYALTY

Repurchase
Likelihood

Overall

Price Tolerance
(Reservation Price)

The Model for Government Services and Nonprofit Organizations
As researchers expanded ACSI by adding customer segments of a large number of
federal government agencies, parts of the model structure needed to change. Repurchase
likelihood and price tolerance are not relevant for most agencies as outcome measures.
Likewise, perceived value in terms of price/quality relationships is not a driver as there is
usually no direct charge (or very nominal charge) for tax-supported organizations.
Instead, the desired outcome of customer satisfaction for a majority of government
agencies is user trust, for which the indicators are: (1) the degree to which the
user/customer would recommend the agency’s services to others (advocacy); and (2) the
extent to which the user has confidence in relying on the agency in the future (confidence).
The antecedents, or drivers, of satisfaction vary within each agency as each provides
different services.
Figure 2 shows the ACSI model for government services. Major latent variables (indices) in

10

ACSI: The National Measure of Customer Satisfaction

Purpose, Modeling, and Properties
the model remain unchanged: customer expectations, perceived quality, customer
satisfaction (ACSI), and customer complaints.
Figure 2. ACSI Model: Government Services and Nonprofit Organizations
Q1

ACTIVITY 1
Complaint Behavior

Q2

Q3

ACTIVITY 2

CUSTOMER
COMPLAINTS

PERCEIVED
QUALITY

Q4

CUSTOMER
SATISFACTION
(ACSI)

Q5

ACTIVITY 3
Q6

CUSTOMER
EXPECTATIONS

Satisfaction

Comparison
to Ideal

USER TRUST

Confirm/
Disconfirm
Expectations

Advocacy

Confidence

2. The Customer Satisfaction Index (ACSI)
To estimate the customer satisfaction index (ACSI), the model software weights the three
manifest variables (questions) that comprise satisfaction. Customers’ responses for a
company or government service are aggregated to produce its ACSI. Thus, the estimate is
specific to each company or government service. The ACSI score is affected by all latent
and manifest variables in the system as described later in this chapter (see Section D, “The
ACSI Equations”).

3. ACSI Antecedents (Drivers or Causes of Satisfaction)
Customer satisfaction (ACSI) has three antecedents in the model: perceived quality,
perceived value, and customer expectations. Perceived quality should have a direct, positive
effect on satisfaction. As a general psychological phenomenon, satisfaction is primarily a
function of a customer’s quality experience with a product or service (Churchill and
Surprenant, 1982; Fornell, 1992; Tse and Wilton, 1988; Westbrook and Reilly, 1983).
11

The American Customer Satisfaction Index

ACSI Methodology Report
Quality experts (Deming, 1981; Juran and Gryna, 1988) delineate two primary components
of this quality experience: (1) the degree to which a product or service provides key
customer requirements (customization), and (2) how reliably these requirements are
delivered (reliability). The greater the perceived quality, the greater the level of customer
satisfaction (ACSI). This prediction is consistent with a growing number of studies in
marketing and consumer research literature (Yi, 1991). As described previously, the drivers
of satisfaction for government are tailored to the activities or services an agency provides to
its customers.
The second determinant of customer satisfaction (ACSI) is perceived value, or the
perceived level of product or service quality relative to the price paid. Adding perceived
value incorporates price information into the model and increases the comparability of
results across companies, industries, and sectors. Quality per dollar, or value, is a common
denominator that consumers use to compare brands and categories alike (Johnson, 1984).
Using value perceptions to measure performance also controls for differences in income and
budget constraints across respondents (Hauser and Shugan, 1983; Lancaster, 1971), which
allows comparisons of very high-priced and very low-priced products and services. As with
perceived quality, the basic prediction is that as value increases, customer satisfaction
improves. The differential impact of perceived quality and perceived value in the model
provides important diagnostic information. As the impact of value increases relative to
quality, price becomes a more important determinant of satisfaction.
The third determinant of customer satisfaction is the level of quality that customers
expect to receive. These expectations should positively affect customer satisfaction because
they serve as an anchor in the evaluation process (Oliver, 1980; Van Raaij, 1989).
Expectations capture all of the customer’s prior knowledge about and consumption
experience with a company’s products or services. Because each of these sources of
information forecasts a company’s ability to provide a positive customer experience, they
should all have a positive effect on satisfaction. Customer expectations provide the anchor
that is adjusted or updated in light of a customer’s more recent purchase and consumption
experience or what he or she has heard about the product or service.
Finally, customer expectations should be positively related to both perceived quality and
perceived value. These relationships capture a customer’s ability to learn from experience
and predict quality and value (Howard, 1977). The size of these predictive relationships
should vary with customer experience, as well as with factors such as the level of
observation (individual customers versus markets), the nature of the information (price
versus performance), and environmental changes (Johnson, Anderson, and Fornell, 1995).

12

ACSI: The National Measure of Customer Satisfaction

Purpose, Modeling, and Properties

4. ACSI Consequences (Outcomes of Satisfaction)
Following Hirschman’s (1970) exit-voice theory, the immediate consequences of
increased customer satisfaction are decreased customer complaints and increased customer
loyalty (Fornell and Wernerfelt, 1987 and 1988). When dissatisfied, customers have the
option of exiting (going to a competitor) or voicing their dissatisfaction in an attempt to
receive retribution. Thus, an increase in satisfaction should decrease the incidence of
complaints. Increased satisfaction should also increase customer loyalty. Customer loyalty
is the ultimate dependent variable in the model because of its value as a proxy for customer
retention and its effect on profitability.
The final relationship in the model is the effect of customer complaints on customer
loyalty. The direction and size of this relationship measures, in large part, the effectiveness
of a company’s complaint-handling system (Fornell, 1992). When the relationship is
positive, then a company is successful in turning complaining customers into loyal
customers. When negative, complaining customers are predisposed to defect.
In the government model, as explained earlier, user trust replaces customer loyalty.

5. Manifest Variables (Questions) Used in the Model
Tables 2 and 3 identify the manifest variables (questions) from the ACSI survey that are
used in the model estimation, as well as which questionnaire items operationalize each latent
variable (boxes in model diagrams). These are linked by numbers to the actual questions
shown in Appendix D: Customer Satisfaction Measurement Questionnaires— Private Sector
and Government Services.
Most questions are asked on a 1 to 10 rating scale, running from low to high. The
surveys used as input to the model are described in Chapter IV, Questionnaires; Chapter V,
Household Survey Sample; Chapter VI, Data Collection Via Telephone; and Chapter VII,
E-Business and E-Commerce Samples and Data Collection Via Internet.

13

The American Customer Satisfaction Index

ACSI Methodology Report
Table 2. Survey Questions Used in the ACSI Private Sector Model
Question
Number*

Manifest Variable
(Question) Description

1

Overall expectation of quality (pre-purchase)

2

Expectation regarding customization, or how
well the product and service fits the customer’s
personal requirements (pre-purchase)

3

Expectation regarding reliability, or how often
things would go wrong (pre-purchase)

4P

Overall evaluation of quality experience with
product (post-purchase)

5P

Evaluation of customization experience, or how
well the product fits the customer’s personal
requirements (post-purchase)

6P

Evaluation of reliability experience, or how often
things have gone wrong with product (postpurchase)

4S**

Overall evaluation of quality experience with
service (post-purchase)

5S**

Evaluation of customization experience, or how
well the service fits the customer’s personal
requirements (post-purchase)

6S**

Evaluation of reliability experience, or how often
things have gone wrong with service (postpurchase)

9

Rating of price given quality

10

Rating of quality given price

11

Overall satisfaction

12

Expectancy disconfirmation (performance that
falls short of or exceeds expectations)

13

Performance versus the customer’s ideal
product and service in the category

14

Has the customer complained to the company
within specified time period

15

Repurchase likelihood rating

16

Price tolerance (increase) given repurchase

17

Price tolerance (decrease) to induce repurchase

Latent Variables
(Indices)
Customer Expectations

Perceived Product Quality

Perceived Service Quality

Perceived Value
Customer Satisfaction (ACSI)

Customer Complaints
Customer Loyalty

*Questionnaires are shown in Appendix D. Question numbers not shown are not used in modeling.
**Used only in expanded model in Figure 1a.

14

ACSI: The National Measure of Customer Satisfaction

Purpose, Modeling, and Properties
Table 3. Survey Questions Used in the ACSI Model for
Government Services and Nonprofit Organizations
Question
Number*
1

Manifest Variable
(Question) Description

Latent Variables
(Indices)

Overall expectation of quality of services
(pre-use)
Activity 1 (Defined by agency)

2

Rating of aspect of Activity 1

3

Rating of aspect of Activity 1

Construct (Driver) of Perceived Quality

Activity 2 (Defined by agency)
4

Rating of aspect of Activity 2

5

Rating of aspect of Activity 2

Customer Expectations

Construct (Driver) of Perceived Quality

Activity 3 (Defined by agency)

Construct (Driver) of Perceived Quality

6

Rating of aspect of Activity 3

7

Rating of aspect of Activity 3

10

Overall evaluation of quality experience
with services (post-use)

11

Overall satisfaction

12

Expectancy disconfirmation (performance
that falls short of or exceeds expectations)

13

Performance versus the user’s ideal
service in the category

14

Has the user complained to the agency
within specified time period

15

Willingness to recommend agency’s
services

16

Confidence agency will do a good job in
the future

Perceived Quality
Customer Satisfaction (ACSI)

Customer Complaints
User Trust

*Question numbers not shown are not used in modeling.

Customer expectations are measured first by asking customers to think back and
remember the level of quality they expected based on their knowledge and experience with a
product or service (that is, “will” expectations as opposed to “should” expectations). Three
expectation measures are collected: (1) overall expectations, (2) expectations regarding
customization, and (3) expectations regarding reliability. For practical reasons, only overall
expectations are asked for the government services model.
Customers then rate their recent experience with the product or service using three
measures of perceived quality: (1) overall perceived quality, (2) perceived customization,
15

The American Customer Satisfaction Index

ACSI Methodology Report
and (3) perceived reliability. For some industries, these three questions are repeated—once
for the measure of product quality and a second time for the measure of service quality.
Next, two questions tap perceived value: (1) a rating of quality received relative to price
paid and (2) a rating of price paid relative to quality received.
As several authors have argued, there is no single standard for evaluating customer
satisfaction. Instead, satisfaction should be reflected in a variety of comparison standards
(Cadotte, Woodruff, and Jenkins, 1987; Johnson and Fornell, 1991; Woodruff, Cadotte, and
Jenkins, 1983). Customer satisfaction (ACSI) is operationalized using three manifest
variables: (1) an overall rating of satisfaction with a product or service to date, (2) the
degree to which performance falls short of or exceeds expectations (expectancy
confirmation/disconfirmation), and (3) a rating of performance relative to the customer’s
ideal product or service in the category. These manifest variables are the same as those
developed for the Swedish Customer Satisfaction Barometer (Fornell, 1992) to measure
satisfaction as a latent variable. Each manifest variable (question) is a qualitatively different
benchmark or evaluation that customers use throughout their purchase and consumption
experience. As a latent variable, ACSI extracts shared variance or that portion of each
rating that is common to all three manifest variables. Thus, satisfaction is not confounded
by either performance or expectations. Only the psychological distance between
performance and expectations, and between performance and the customer’s ideal point, are
used to measure satisfaction. All ACSI models use these same three manifest variables to
measure satisfaction.
Customer complaints are measured by whether or not a customer has indicated that he or
she has complained within the same time period that qualifies the potential respondent as a
customer.
Finally, two variables measure customer loyalty for private sector companies. The first is
repurchase likelihood and the second is price tolerance (reservation price). The latter is
constructed from two survey questions: (1) the degree to which a company could raise its
price(s) as a percentage before the customer would definitely choose not to buy from that
company the next time (given that the customer has indicated that he or she is likely to
repurchase); and (2) the degree to which a company would have to lower its price(s) as a
percentage before the customer would definitely choose to buy from that company the next
time (given that the customer has indicated that he or she is unlikely to repurchase).
For government, the consequence of satisfaction is typically user trust, most often
measured by (1) would the user recommend the agency’s services to others (advocacy) and

16

ACSI: The National Measure of Customer Satisfaction

Purpose, Modeling, and Properties
(2) does the user have confidence that the agency will do a good job in the future
(confidence).

C. INDEX PROPERTIES
In ACSI, customers are asked to evaluate products and services that they have purchased
and used. A straightforward summary of what customers say in their responses to the
questions described as manifest variables in Tables 2 and 3 (for example, means and
percentage distributions) may have certain simplistic appeal, but such an approach will fall
short on any other criterion. For the index to be useful, it must meet criteria related to its
objectives. If ACSI is to contribute to more accurate and comprehensive measurement of
economic output, predict economic returns, provide useful information for economic policy,
and become an indicator of economic health, it must satisfy certain properties in
measurement. These are: (1) precision, (2) validity, (3) reliability, (4) predictive power,
(5) coverage, (6) simplicity, (7) diagnostics, and (8) comparability.

1. Precision
Precision refers to the certainty about the recorded value of ACSI. Very high standards
for precision are set, particularly at the national level, in order to detect changes from one
period to the next. ACSI results show that the 90% confidence interval (on a 0 to 100 scale)
for the national index is plus or minus 0.2 points. For each of the measured sectors, it is an
average plus or minus 0.5 points. For industries, the confidence intervals are an average
plus or minus 1.0 points for manufacturing industries and 1.1 points for private sector
service industries. For the typical company, these are an average plus or minus 2.0 points
for manufacturing companies and 2.5 points for service companies and government services.
This level of precision is obtained as a result of great care in data collection, careful
manifest variable specification, and latent variable modeling. According to ACSI research,
latent variable modeling (using weighted averages of multiple questions to produce the
indices shown in the boxes in the models in Figures 1, 1a, and 2) produces an average
improvement of 22% in precision over use of responses from a single question.

2. Validity
Validity refers to the ability of the individual measures to represent the underlying latent
variable (index) customer satisfaction (ACSI) and to relate effects and consequences in an
expected manner. Discriminant validity, or the degree to which measured latent variables
differ from other measured latent variables, also is evidenced. For example, there is not

17

The American Customer Satisfaction Index

ACSI Methodology Report
only an important conceptual distinction between perceived quality and customer
satisfaction, but there is an empirical distinction as well. That is, the covariance between the
three manifest variables (questions) measuring ACSI is higher than the covariances between
ACSI and any other latent variable in the system.
The nomological validity of the ACSI model can be examined by: (1) latent variable
covariance accounted for and (2) explained variance (R2). On average, the structural model
accounts for 94% of the latent variable covariance structure. The average R2 of the
customer satisfaction equation in the model is .75. In addition, all coefficients relating the
latent variables (indices) of the model have the expected sign. All but a few are statistically
significant.
In measures of customer satisfaction, there are several threats to validity. The most
serious of these is the skewness of the frequency distributions as respondents tend to
disproportionately use the high scores on a scale. Skewness is addressed by using a fairly
high number of scale categories (1−10) and by using a multiple-indicator approach (Fornell,
1992 and 1995). It is an established fact that validity typically increases with the use of
more categories (Andrews, 1984), and it is particularly so when the respondent has good
knowledge about the subject matter and when the distribution of responses is highly skewed.
An index of satisfaction is much preferred over a categorization of respondents as either
“satisfied” or “dissatisfied.” Satisfaction is a matter of degree—it is not a binary concept.
If measured as binary, precision is low, validity is suspect, and predictive power is poor.

3. Reliability
Reliability of a measure is determined by its signal-to-noise ratio. That is, the extent to
which the variation of the measure is due to the “true” underlying phenomenon versus
random effects. High reliability is evident if a measure is stable over time or equivalent
with identical measures (Fornell, 1992). Signal-to-noise in the items that make up the index
(in terms of variances) is about 4 to 1.

4. Predictive Power and Financial Implications of ACSI
An important aspect of ACSI is its ability to predict economic returns. The model,
of which ACSI is a part, uses two proxies for economic returns as criterion variables:
(1) customer retention (estimated from a nonlinear transformation of a measure of
repurchase likelihood) and (2) price tolerance (reservation price). The manifest variables
included in the index are weighted in such a way that the proxies and ACSI are maximally
correlated (subject to certain constraints). Unless such weighting is done, the index is more

18

ACSI: The National Measure of Customer Satisfaction

Purpose, Modeling, and Properties
likely to include matters that may be satisfying to the customer, but for which he or she is
not willing to pay.
The empirical evidence for predictive power is available from both the Swedish data and
ACSI data. Using data from the Swedish Customer Satisfaction Barometer (SCSB), a
1-point increase in SCSB each year over five years yields, on the average, a 6.6% increase
in current return-on-investment (Anderson, Fornell, and Lehmann, 1994). Of the companies
traded on the Stockholm Stock Market Exchange, it is also evident that changes in SCSB
have been predictive of stock returns. Similar findings exist for ACSI (Fornell, 2001).
A basic tenet underlying ACSI is that satisfied customers represent a real, albeit
intangible, economic asset to a company. By definition, an economic asset generates future
income streams to the owner of that asset. Therefore, if customer satisfaction is indeed an
economic asset, it should be possible to use ACSI for prediction of company financial
results. It is, of course, of considerable importance that the financial consequences of ACSI
are specified and documented. If it can be shown that ACSI is related to financial returns,
then the index has value.
Faculty, doctoral students, and consultants have done considerable research on the
linkage between ACSI and economic returns, analyzing both accounting and stock market
returns from measured companies. The pattern from all of these studies suggests a
statistically strong and positive relationship. Specifically:
•

There is a positive and significant relationship between ACSI and accounting
return-on-assets.

•

There is a positive and significant relationship between ACSI and the market value of
common equity. When controlling for accounting book values of total assets and
liabilities, a five-unit gain (on the 0 to 100 scale used for ACSI) is associated with an
average of 15% increase in market value. There are also significant and positive
relationships between ACSI and market-to-book values and price/earnings ratios.
There is a negative relationship between ACSI and risk measures, implying that
companies with high loyalty and customer satisfaction have less variability and
stronger financial positions.

•

There is a positive and significant relationship between ACSI and the long-term
adjusted financial performance of companies. Tobin’s q is generally accepted as the
best measure of long-term performance. It is defined as the ratio of a company’s
present value of expected cash flows to the replacement costs of its assets.

19

The American Customer Satisfaction Index

ACSI Methodology Report
Controlling for other factors, ACSI has a significant relationship to Tobin’s q
(Mazvancheryl, Anderson, and Fornell, 2004).
•

At the more macro level, ACSI predicts S&P’s 500 corporate earnings well. Recent
findings suggest a similar predictive power for consumer spending.

The ACSI scores of approximately 130 publicly traded companies display a statistically
positive relationship with the traditional performance measures used by companies and
security analysts (that is, return-on-assets, return-on-equity, price-earnings ratio, and the
market-to-book ratio). In addition, companies with higher ACSI scores display stock price
returns above the market-adjusted average (Fornell, Mithas, Morgeson, and Krishnan,
2005). ACSI also is positively correlated with “market value added.” This evidence
indicates that the ACSI methodology produces a useful measure for customer satisfaction
that is forward-looking and relevant to a company’s economic performance.

5. Coverage
ACSI covers a substantial portion of the U.S. economy. In terms of revenues, it is 36%
of GDP. Total revenues of the measured companies account for 43% of GDP, but ACSI
measures only products and services sold to household consumers in the domestic market.
The economic sectors and industries covered are discussed in Chapter III. Within each
industry, the number of companies measured varies from 2 to 30. (See Appendix A:
Companies and Government Services Evaluated by Customers in ACSI.)
The national index and the indices for each industry and sector are reflective of the total
value (quality times sales) of products and services provided by companies at each
respective level of aggregation. Relative revenues are used to determine each company’s
contribution to its respective industry index. In turn, relative revenues by each industry are
used to determine each industry’s contribution to its respective sector index. To calculate
the national index, the percentage contributions of each sector to GDP are used to topweight the sector indices. Mathematically, this is defined as:
F

Index for Industry i in Sector s at time t = Iist = ∑
f

S fist I fist

∑

F
f

S fist

I S I
ist ist

Index for Sector s at time t = Ist = ∑
i

where

20

ACSI: The National Measure of Customer Satisfaction

∑S
I

i

ist

Purpose, Modeling, and Properties

S fist = Sales by firm f , industry i , sector s at time t
I fist = Index for firm f , industry i , sector s at time t
and
F

Sist = ∑ S fist = Total Sales for Industry i at time t
f

I

Sst = ∑ Sist = Total Sales for Sector s at time t
i

ACSI is updated on a quarterly basis. For each quarter, new indices are estimated for one
or more sectors with total replacement of all data annually. The national index is comprised
of the most recent estimate for each sector

National Index at time t = It =

T

S

∑∑

t =T −3

s

T

Sst Ist

∑∑S

t =T −3

S

s

st

where I st = 0 for all t in which the index for a sector is not estimated, and I st = I st for all
quarters in which an index is estimated. In this way, the national index represents company,
industry, and sector indices for the prior year.

6. Simplicity
Given the complexity of model estimation, ACSI maintains reasonable simplicity. It is
calibrated on a 0 to 100 scale. Whereas the absolute values of ACSI are of interest, much of
the index’s value, as with most other economic indicators, is found in changes over time,
which can be expressed as percentages.

7. Diagnostics
As evident from the model specification in Figures 1, 1a, and 2, the ACSI methodology
estimates the relationships between customer satisfaction and its causes as seen by the
customer: customer expectations, perceived quality, and perceived value. Also estimated
are the relationships between ACSI, customer loyalty [as measured by customer retention
and price tolerance (reservation price)], and customer complaints. ACSI generates
information about levels of satisfaction, expectations, and so forth, as well as the
21

The American Customer Satisfaction Index

ACSI Methodology Report

antecedents (causes) and consequences (outcomes) of satisfaction. For example, it is
possible to estimate the impact of product and service reliability, the effect of increased
customization (fitness for use), the role of expectations, and the expected economic returnon-investment in customer satisfaction as the result of improved customer retention.
There are, however, certain limitations with respect to diagnostics. Cassell (1993)
reports on the advantages of the approach chosen in ACSI to measure satisfaction at the
corporate rather than the leading-brand level in terms of unbiased estimates. This refers to
the manifest and latent variables, but not necessarily to the coefficients that relate the index
to causes and consequences. For the coefficients to be unbiased and consistent, the standard
statistical assumptions apply. Further, it is not realistic to believe that all diagnostics are
equally useful. For companies selling many different products to different markets, it is
obvious that any set of diagnostics that do not relate to specific brands are of limited value.

8. Comparability
A fundamental question is whether or not it is possible to compare the satisfaction levels
(ACSIs) of different customers, companies, industries, and sectors. As evident from welfare
economics, any interpersonal comparison of utility is complicated. In ACSI, as in any
measurement that uses surveys as input, it is not certain that the questionnaire scales have
the same meaning to each respondent. The ACSI methodology addresses this difficulty by
treating customer satisfaction as a latent (nonobservable) construct at a higher level of
abstraction where there is a basis for comparing things that are fundamentally different
(Johnson and Fornell, 1991). This is not to suggest that such comparisons are without error,
but that the error is reasonably small relative to what is gained. ACSI meets a reasonable
criterion of comparability (across individuals, companies, and so forth). “Objective” factors
such as degree of industry concentration and heterogeneity in demand and supply can
account for the variation in the index. Results show that much of this variation across
industries and sectors is, in fact, accounted for by variables of industrial organization.
A further discussion of ACSI and results in its 1994 baseline year is given in Fornell,
Johnson, Anderson, Cha, and Bryant (1996). A ten-year history of ACSI is described in
“The American Customer Satisfaction Index at Ten Years: ACSI 1994-2004, A Summary of
Findings: Implications for the Economy, Stock Returns and Management” (Fornell,
VanAmburg, Morgeson, Anderson, Bryant, and Johnson, 2005). Scores for companies and
government services across all years of measurement are updated quarterly on the Web site:
www.theacsi.org.

22

ACSI: The National Measure of Customer Satisfaction

Purpose, Modeling, and Properties

D. THE ACSI EQUATIONS
The formal expression of the model depicted in Figure 1 can be written as a series of
equations estimated by partial least squares (PLS). The systematic part of the predictor
relationships is the conditional expectation of predictands for given values of predictors.
The general equation is thus specified as stochastic:
E[η |η , ξ ] = Βη + Γξ

where η′ = (η1, η2, ..., ηm) and ξ′ = (ξ1, ξ2, ..., ξn) are vectors of unobserved endogenous
and exogenous variables, respectively; B (m × m) is a matrix of coefficient parameters for η;
and Γ (m × n) is a matrix of coefficient parameters for ξ. This implies that
E[ηζ′] = Ε[ξζ′] = Ε[ζ] = 0, where ζ = η − Ε[η|η, ξ].
The equation that relates the latent variables in the model shown in Figure 1 is:
η 1   0
η   β
 2   21
η 3  =  β 31
η   0
4
  
η
 5  0

0 0 η 1  γ 11 
ζ1 





ζ 
γ 21
0 0 0 0 η2
   
 2
β 32 0 0 0 η 3  + γ 31  ξ +  ζ 3 
ζ 
0 β 43 0 0 η 4   0 
4
   
 
ζ
0 β 53 β 54 0 η 5   0 
 5
0

0

where
ξ = Customer Expectations
η1 = Perceived Quality
η2 = Perceived Value
η3 = Customer Satisfaction (ACSI)
η4 = Customer Complaints
η5 = Customer Loyalty

23

The American Customer Satisfaction Index

ACSI Methodology Report

The general equations for relating the latent variables to empirical variables are:

y = Λyη + ε
x = Λxξ + δ

where y′ = (y1, y2, ..., yp) and x′ = (x1, x2, ..., xq) are the measured endogenous and
exogenous variables, respectively. Λy (p × m) and Λx (q × n) are the corresponding
regression coefficient matrices. By implication from PLS estimation (Fornell and
Bookstein, 1982), we have E[ε] = E[δ] = E[ηε′] = E[ξδ′]= 0. The corresponding equations
in the model are:

 x1   λ11  δ 1 
     
 x 2  =  λ21 ξ + δ 2 
 x 3   λ31  δ 3 

and

 y1 
y 
 2
 y3 
y 
 4
 y5 
 
 y6  =
 y7 
 
 y8 
y 
 9
 y10 
y 
 11 

 λ 11
λ
 21
λ 31

 0
 0

 0
 0

 0
 0

 0
 0


0
0
0

λ 12
λ 22
0
0
0
0
0
0

0
0
0
0
0

λ 13
λ 23
λ 33
0
0
0

0
0
0
0
0
0
0
0

λ 14
0
0


 ε1 

ε 

 2

ε3 

 
 η 1   ε 4 
 η 2   ε 5 
   
 η 3  +  ε 6 
 η 4   ε 7 
   
 η 5   ε 8 

ε 

 9
λ 15 
ε 10 

ε 
λ 25 
 11 
0
0
0
0
0
0
0
0
0

24

ACSI: The National Measure of Customer Satisfaction

Purpose, Modeling, and Properties

where:
x1 = Customer Expectations About Overall Quality
x2 = Customer Expectations About Reliability
x3 = Customer Expectations About Customization
y1 = Overall Quality
y2 = Reliability
y3 = Customization
y4 = Price Given Quality
y5 = Quality Given Price
y6 = Overall Satisfaction
y7 = Confirmation of Expectations
y8 = Distance to Ideal Product (Service)
y9 = Formal or Informal Complaint Behavior
y10 = Repurchase Intention
y11 = Price Tolerance (Reservation Price)

E. THE ACSI FORMULA
The general form of ACSI is as follows:

ACSI =

E[ξ ] − Min[ξ ]
× 100
Max[ξ ] − Min[ξ ]

where ξ is the latent variable for customer satisfaction (ACSI), and E . , Min . and Max .
denote the expected, the minimum, and the maximum value of the variable, respectively.
The minimum and the maximum values are determined by those of the corresponding
manifest variables
n

Min[ ξ ] = ∑ wi Min[ xi ]
i =1

and
n

Max[ ξ ] = ∑ wi Max[ xi ]
i =1

25

The American Customer Satisfaction Index

ACSI Methodology Report

where xi’s are the manifest variables of the latent customer satisfaction, wi’s are the weights,
and n is the number of measurement variables. In calculating ACSI, unstandardized weights
must be used if unstandardized measurement variables are used.
In ACSI, there are three indicators for customer satisfaction that range from 1 to 10.
Then the calculation is simplified to:
3

ACSI =

3

∑ wi x i − ∑ wi
i =1

i =1

3

9∑ w i
i =1

where wi’s are the unstandardized weights.

26

ACSI: The National Measure of Customer Satisfaction

× 100

Selection of Sectors, Industries & Companies

III. SELECTION OF ECONOMIC SECTORS,
INDUSTRIES, COMPANIES, AND
GOVERNMENT SERVICES
The selection of sectors, industries, companies, and government services is premised on
obtaining a representation of the U.S. economy that provides goods and services to
households by measuring companies with total sales that represent a significant proportion
of the GDP. In addition to U.S.-based companies that produce the nation’s GDP, ACSI
includes goods and services produced by foreign companies with major U.S. market shares.
For reasons of both efficiency and precision, analysis is at the aggregate company level,
rather than at the product or brand level (Cassel, 1993). The ACSI score for an individual
company thus reflects the proportional mix of its product and service offerings. For
government, analysis is at the level of certain services provided to specific segments of the
population.
ACSI includes ten sectors of the economy that sell to household consumers:
(1) Utilities, (2) Manufacturing/Nondurable Goods, (3) Manufacturing/Durable Goods,
(4) Retail Trade, (5) Transportation and Warehousing, (6) Information, (7) Finance and
Insurance, (8) Health Care and Social Assistance, (9) Accommodation and Food Services,
and (10) Public Administration. Additionally, ACSI measures two online groups of
industries whose companies are included in the prior sectors: e-business and e-commerce.
Government spending accounts for approximately 11% of GDP. ACSI includes federal
and local, but not state government. Local government is represented in ACSI by two
services that reach most of the public: solid waste disposal and police. Figure 3 shows the
sectors and industries measured in ACSI.

27

The American Customer Satisfaction Index

Figure 3

The ACSI: National Economy, Sectors, and Industries*

Manufacturing/
Nondurable Goods
9.4%
Utilities
1.8%
- Food
manufacturing

Manufacturing/
Durable Goods
11.1%

- Pet food
- Soft drinks
- Energy
utilities

- Breweries
- Cigarettes
- Apparel
- Athletic
shoes
- Personal
care &
cleaning
products

- Personal
computers
- Cellular
telephones
- Electronics
(TV/VCR/DVD)
- Major
appliances

Retail Trade
5.8%

Information
5.3%

Health Care &
Social Assistance
6.9%

Public
Administration
11.4%

Accommodation &
Food Services
3.7%
- Solid waste

Transportation &
Warehousing
3.0%
- Supermarkets

- Newspapers

- Gasoline
stations

- Motion
pictures

disposal
- Police

- Department
& discount
stores
- Specialty
retail stores

- Broadcasting
TV news

- Federal
agencies

- Airlines
- U.S. Postal
Service
- Express
delivery

- Fixed line
telephone service

Finance &
Insurance
7.3%

- Hospitals

- Banks

- Hotels

- Life insurance

- Limited-service
restaurants

- Wireless
telephone
service

- Health
insurance

- Cable &
satellite TV

- Property &
casualty
insurance

- Automobiles
& light
vehicles

*The percentage shown is the contribution of each sector to the GDP. The 10 measured sectors produce 65.7% of GDP.
**E-Business / E-Commerce is a portion of other sectors.

E-Business/
E-Commerce**

- News &
information
- Portals
- Search engines
- Retail
- Auctions
- Brokerage
- Travel

Questionnaires

IV. QUESTIONNAIRES
A. SCREENING QUESTIONNAIRES
There are multiple versions of questionnaires to screen respondents (for information
about selection of respondents, see Chapter V, Household Survey Sample). Different
versions of screening questionnaires are employed because product usage is determined
specifically; relevant time periods for purchase/usage vary; and all brand names must be
linked to parent companies. Table 4 shows the relevant time periods for qualifying
customers.
Table 4. Time Periods for Purchase and Use to Qualify as Customer
(All purchases must be new, not used)
Currently
• Have active checking account, savings account, or bank loan in own name.
• Have life, homeowner’s, automobile, property/casualty, or health insurance in
own name.
• Have garbage and trash collection (solid waste disposal) provided by local
government (asked only of those living in metropolitan areas).
• Have telephone service in home or cellular service in own name.
• Have cable or satellite TV service in home.
• Have electric or natural gas service in home.
Within past week
• Read a newspaper that was subscribed to or purchased.
Within past month
• Purchased and consumed food (five categories).
• Purchased and smoked cigarettes (asked only if 21 or older).
• Purchased and consumed beer (asked only if 21 or older).
• Purchased and consumed soft drinks or pop.
• Own a dog or cat and purchased pet food.
• Watched a national TV news program.
Within past three months
• Purchased food from a fast food or pizza (limited-service) restaurant to eat in the
restaurant, for carry out, or ordered for home delivery.
• Purchased gasoline for own automobile.
• Shopped for groceries at a supermarket or other store.
• Purchased personal care or cleaning products (five categories).

29

The American Customer Satisfaction Index

ACSI Methodology Report
Within past six months
• Shopped for merchandise at a department or discount store or specialty retail store.
• Used a parcel delivery, overnight, or two-day mail delivery service for sending a
letter, a document, or a package either from home or work.
• Visited a U.S. Post Office to buy stamps, pick up mail, or use any of the counter
services.
Within past year
• Flew as a passenger on a scheduled airline.
• Stayed overnight at a hotel for business or pleasure.
• Went to a motion picture theater, rented or purchased a videotape or DVD, or
watched a movie on pay-per-view.
• Purchased apparel (multiple categories) or athletic shoes.
• Qualified as a user/customer of federal services. (Use of specific services within the
past year is the criterion for qualifying as the user/customer for most federal
services. A few have longer or shorter time periods.)
Within past two years
• Purchased a cellular phone.
Within past three years
• Purchased a personal computer for home.
• Purchased a television or videocassette recorder/DVD player.
• Purchased a major appliance such as washer, dryer, stove, refrigerator, or
dishwasher.
• Used hospital services as inpatient or outpatient.
• Had contact with local police (asked only of those living in metropolitan areas).
More than six months, but within three years
• Purchased or (personally) leased an automobile, van, or light truck.

For examples of the screening questionnaires for the major appliance industry and for the
supermarket industry, see Appendix C: Example Screening Questionnaires and
Brand/Company Identification.

B. BRAND/COMPANY IDENTIFICATION
Because customers often respond with a brand name rather than a company name when
asked about the purchase of goods or services, all of the brands produced by the companies
measured in ACSI are programmed into the computer-assisted-telephone-interviewing
(CATI) system. For many industries, customers may also respond with a subsidiary name
rather than a parent company name (as with brand names, the CATI system links all
subsidiary names to parent companies). For a description of how brand and subsidiary
names are linked with company names, see Chapter VI, Data Collection via Telephone.

30

ACSI: The National Measure of Customer Satisfaction

Questionnaires

Highlighting and entering the brand in the CATI system automatically designates the
interview by company name, although the brand name appears on the interview screen for
all questions asked of the respondent. Examples of the brand and subsidiary lists for the
major appliance industry and the supermarket industry are shown in Appendix C.
If the respondent answers with a brand name for a company not measured, the response is
coded under the “all others” category. The respondent is not administered the complete
questionnaire for the particular product, but only items used in the ACSI score.

C. QUESTIONNAIRE
The satisfaction questionnaire items are shown in Chapter II, Tables 2 and 3. These
items are the product of multiyear empirical testing in the United States. The private sector
questionnaire has had few modifications since the first wave of measurement in 1994. The
examples used in questionnaires are tailored to be relevant to each industry, but the basic
question format is constant across all industries. Questions for federal government are
customized for each measured service. The generic questionnaires for interviews for the
private sector and government services are shown in Appendix D.
The number of substantive questions used in the ACSI models are as follows: 15 for
Figure 1 (private sector model), 18 for Figure 1a (expanded model), and 14 for Figure 2
(government services model). Respondents are also asked six demographic questions and,
on occasion, optional questions that are not used for modeling. Identification of the
respondent’s geographic location is coded from the sample. The substantive contents of the
questions are as follows:

31

The American Customer Satisfaction Index

ACSI Methodology Report
Questions Used for Private Sector Models (Figures 1 and 1a)

•
•

•
•
•
•
•
•
•

Three questions on customer’s expectations before purchase/use of
products/services.
Three questions on customer’s perceived quality based on actual experience with
products/services (overall quality, customization, reliability). (These are
expanded to six questions for the model in Figure 1a.)
Two questions on perceived valueprice given quality and quality given price.
One question on overall satisfaction with products/services.
One question on products/services exceeding or falling short of expectations.
One question on comparison with ideal products/services.
One question on complaints.
One question on repurchase likelihood.
Two questions on price tolerance.

ACSI is comparable across industries, as the questions are the same for each. Examples
that the interviewers provide to illustrate quality, customization, and reliability are
customized by industry to make these questions relevant to the specific product or service.
Questions Used for Government Services and
Nonprofit Organization Model (Figure 2)

•
•

•
•
•
•
•
•

One question on user’s expectations before use of services.
Six questions rating three activities. The agency identifies activities it provides
that are inputs to perception of quality (for example, customer service,
information, services provided).
One question on user’s perceived quality based on actual experience with
services.
One question on overall satisfaction with services.
One question on services exceeding or falling short of expectations.
One question on comparison with ideal services.
One question on complaints.
Two questions on user trust.

32

ACSI: The National Measure of Customer Satisfaction

Household Survey Sample

V. HOUSEHOLD SURVEY SAMPLE
A. HOUSEHOLD SAMPLE AND TELEPHONE NUMBER
SELECTION
This chapter discusses the random selection of adults and identification of customers of
specific products and services in households for private sector companies and for
government services with high incidence of use. For other government services, sample
selection is random from population frames of users (for example, program beneficiaries or
applicants).

1. Selection of Household Numbers for Screening
The universe from which the sample is drawn for private sector companies and highincidence government services is households with telephones in the continental United
States. Numbers to be dialed are selected using Genesys CSS, the sample design and
generation system offered by Marketing Systems Group (Marketing Systems Group, 2004).
Interviewing is done quarterly, with data collected for one or more of the ten measured
economic sectors each quarter. Regional samples are selected and screened for customers of
companies that have regional markets (for example, energy utilities and cable television
service).
Creation of the Random-Digit Database

A modified Epsem (equal probability of selection method) is used to select numbers to be
dialed. Replicate samples are screened successively to maintain national (or regional)
representation for companies with both high and low incidence of purchase. The same RDD
method is used for national replicates and for the targeted regional RDD samples. RDD
includes listed and unlisted telephones.
Samples are generated using a database of “working blocks,” prescreened to remove
identifiable nonresidential numbers. A block (also known as a 100-bank or a bank) is a set
of 100 contiguous numbers identified by the first two digits of the last four digits of a
telephone number. For example, in the telephone number 255-4200, “42” is the block, and
4200-4299 the numbers in it. A block is termed to be working if one or more listed
telephone numbers are found in that block.
Each exchange is assigned to a single county. Nationally, about 72% of all exchanges
appear to fall totally within single-county boundaries. For those overlapping county and/or
33

The American Customer Satisfaction Index

ACSI Methodology Report

state lines, the exchanges are assigned to the county of plurality or the county with the
highest number of listed residents within the exchange. This assignment prevents
overrepresentation of these exchanges.
Sample Stratification

Samples are generated using stratified sampling procedures. A separate sample is
selected from the sampling units in each stratum. The database used has been stratified
by county.
Prior to sample selection, the sample is allocated proportionally across all strata in the
defined geography using several frame adjustment options. The sampling frame determines
the way a sample is distributed across geography at the county level.
The sample is distributed by county in proportion to the total active blocks (with one
or more listed numbers) in the exchanges assigned to that county. Rather than being an
estimate of target population, all frame units are represented with equal probability across
counties. Active blocks in each exchange are counted with each database update. The
number of active blocks in an exchange is multiplied by 100 (the number of possible
ten-digit telephone numbers in a block) to calculate the total possible phone numbers.
Sample is allocated to each county in proportion to its share of these possible ten-digit
telephone numbers.
Sample Selection

After the sample has been allocated, sample selection is made. Samples of random
numbers are systematically selected with equal probability across all eligible blocks. All
blocks within a county are organized in ascending order by area code, exchange, and block
number.
Once the quota has been allocated to all counties in the frame, a sampling interval is
calculated for each county by summing all the eligible blocks in the county and dividing that
sum by the number of sampling points assigned to the county. From a random start between
zero and the sampling interval, blocks are systematically selected from each county. Once a
block has been selected, a two-digit random number in the range of 00 to 99 is appended to
the exchange and block to form a ten-digit telephone number.
Telephone numbers selected for use are marked on the database to protect against reuse
in the same calendar year.

34

ACSI: The National Measure of Customer Satisfaction

Household Survey Sample
Dual-Sample Frame for Manufacturing/Durable Goods

Currently, a dual-sample frame is used for the Manufacturing/Durable Goods sector.
This is because of the enormous number of households that need to be screened to identify
and interview purchasers (within the last three years) of companies with low market shares
in the automobile and light vehicle industry. The dual frame consists of both RDD national
samples, as used for the other sectors of the economy, and purchased lists of automotive
vehicle owners.3 The vehicle owners also are eligible to be screened for purchase and use of
other products and services measured in ACSI.
Throughout interviewing for the Manufacturing/Durable Goods sector, the sample from
each frame is monitored for demographic characteristics to ensure that similar types of
customers are interviewed in each frame.

2. Household Sample Size
There is no fixed sample size for initial screening of households. There is a quota of 250
customer interviews per company and 260 per federal government service. For only a few
companies and services with low incidence, fewer than 250 interviews may be completed.

3. Callbacks and Household Substitution
If the household is not reached on the initial call, three callbacks are made on different
days at different time periods. If no contact is made after this total of four attempts, a
substitute number is selected and the process is repeated. All telephone interviewing is
conducted from 5 p.m. to 9 p.m. local time, Monday through Friday; 9 a.m. to 9 p.m.
Saturday; and 10 a.m. to 9 p.m. on Sunday. Scheduled callbacks can be completed from
8 a.m. to 9 p.m. local time, upon request from the respondent.

B. SELECTION OF DESIGNATED RESPONDENT WITHIN
HOUSEHOLD
Eligible respondents for screening are defined as adults age 18 to 84, except that for the
brewery (beer) and cigarette industries, only those 21 and older are eligible. The adult to be
interviewed is selected randomly by asking for the individual age 18 to 84 with the most
recent birthday. No substitution of respondent within the household is allowed. If that
respondent is unreachable or unwilling, a new household is dialed.

3

Lists of licensed vehicle registrations from the 43 states from which such information is available (plus the District of
Columbia) are purchased from R.L. Polk (Southfield, Michigan).

35

The American Customer Satisfaction Index

ACSI Methodology Report

C. SCREENING HOUSEHOLD RESPONDENTS TO IDENTIFY
QUALIFIED CUSTOMERS
The designated respondent is qualified as a customer from questions about the purchase
of products and services within specified time periods for up to ten industries. No single
respondent is interviewed about more than three companies (none of which are competing in
the same industry or product category).
Once a respondent qualifies as the customer of three measured companies, screening is
discontinued. If the respondent qualifies for one or two out of ten, he or she is interviewed
about the products or services of these companies. If the respondent does not qualify as a
customer for three measured companies, but has purchased goods or services from another
company within the industries for which he or she was screened, the respondent is asked
three satisfaction questions about that company. These interviews are aggregated within
each industry as “all others.” Qualification as a customer is based on purchase/usage during
the time periods shown in Chapter IV, Table 4, and the questions shown in Appendix B:
ACSI Industry Definitions and Customer Identification.
There are multiple screening questionnaires to cover the measured industries. If more
than ten industries are measured during the quarterly field interviewing period, ten are
selected to start. As the quota of interviews for an industry is filled, the screening
questionnaire for that industry is dropped and one for another industry is added. Once the
number of companies for which the respondent qualifies is determined, the interviewer then
proceeds with the customer satisfaction questionnaire for each of these.
Standard errors and confidence intervals for the companies are based on the completed
sample size. Because all latent constructs used in the ACSI model are weighted averages of
multiple questions, the final measures have more precision than single questions have with
these sample sizes (see Chapter II, Section C-1, “Precision”).

D. INTERVIEW RESULTS
1. Response and Cooperation Rates
In 2003, ACSI telephone interviewers completed 51,297 private sector company
interviews with 36,952 customer respondents (an average of 1.4 interviews per respondent)
plus 10,140 interviews with users of federal government services.
To standardize the reporting of random-digit-dial telephone survey results, the American
Association of Public Opinion Research (AAPOR) has produced Standard Definitions:

36

ACSI: The National Measure of Customer Satisfaction

Household Survey Sample

Final Dispositions of Case Codes and Outcome Rates for Surveys (American Association of
Public Opinion Research, 2004).
The ACSI methodology is unusual among surveys for several reasons. First, the design
calls for four attempts (initial call and three callbacks at different times of day and different
days) on each sampled phone number rather than an unlimited number of calls. Secondly, a
randomly selected adult (age 18 to 84) in the household is then screened to determine if
he/she has been the customer of specific companies within specific time periods. Thirdly,
interviews for any company or government service are discontinued once a quota of 250 to
260 interviews is completed. Call results, therefore, do not exactly fit AAPOR categories.
Appendix E: Response and Cooperation Rates reports the disposition of each of the
sampled telephone numbers that produced the interviews.
Using AAPOR standards, the Cooperation Rate is 94.2% and the Response Rate (RR3) is
22.7%. The calculations of these are shown in Appendix E.

2. Refusal Conversions
If the person who answers at a household refuses for anyone in the household to be
interviewed, or if an identified qualified customer refuses, an attempt is made by a more
experienced interviewer on a different day to contact the household or customer to obtain
cooperation. Refusal conversions account for 8% to 10% of completed interviews.

E. PROFILE OF INTERVIEWED CUSTOMERS
Appendix F gives a demographic profile of interviewed customers for private sector
companies. All are between the ages of 18 and 84, or 21 and 84 if interviewed about
breweries or cigarettes. The sample, screened to purchasing adults in telephone
households,4 has expected differences from all adults ages 18 to 84 in the direction of having
higher socioeconomic characteristics of education and income and less representation of
non-whites. The higher proportions of females (62% compared to 51% in the population) is
partially attributable to the mix of companies measured in ACSI. Many of these produce
household and food products purchased disproportionately by women.
4

In 2000, 99% of owner-occupied households and 95% of renter-occupied households had telephone service. By 2003,
according to the Bureau of Labor Statistics, cellular-only households had grown to approximately 4.5%. Interviews are not
allowed on cellular numbers.

37

The American Customer Satisfaction Index

Data Collection Via Telephone

VI. DATA COLLECTION VIA TELEPHONE
The National Quality Research Center, Ross School of Business, works closely with its
telephone sampling and interviewing sources to fulfill the unusual interviewing
requirements called for by the ACSI sampling design described in the previous chapter. The
interviewing requirements are unusual because of the need to screen households to identify
customers of about 200 companies and the users of specific government-provided services,
and to maintain nationally (or regionally) representative samples of the customers of each of
these. Both industries and companies have greatly varying incidence of usage in U.S.
households. This incidence ranges from 100% use of the U.S. Postal Service to a far lower
percentage of adults who have stayed within the past year at one of several hotel chains with
low household penetration and market share.
Customers for some federal government services are in such low incidence that it is not
feasible to find them by screening households (for example applicants for certain grants,
benefits, patents, or trademarks). For these, the federal agency provides the population
frame in the form of a list of the total population or a very large random sample of it. From
this frame, a random sample for interview is selected.
Most respondents in ACSI are identified by telephone screening. This screening is made
more complex because many customers respond with brand names or subsidiary/
divisional names of companies rather than with parent company names. A further
complexity is that a qualified respondent can be a customer for a product or service from
more than one industry, and is thus eligible for interview about more than one company (up
to a maximum of three).
To accommodate these requirements, the computer-assisted-telephone-interviewing
(CATI) system and the database created during interviewing are programmed and designed
to have the following capabilities:
1. Assign multiple screening questionnaires and multiple company interview questionnaires
to each designated respondent.
•

A respondent is administered only a portion of possible screening questionnaires
(maximum ten). As interviews for each industry are completed, the screening
questionnaire for another industry is added.

•

A respondent is eligible for interviews in a maximum of three industries, but only for
one company within any industry. He or she may not be interviewed more than once
for the same company if it is measured in another industry.

39

The American Customer Satisfaction Index

ACSI Methodology Report

•

The quota of company interviews for some industries in the Manufacturing/
Nondurable Goods sector can be filled by multiple products from multiple screenings.
For example, the Colgate-Palmolive Company has products in several categories in
the personal care and cleaning products industry and Sara Lee Corporation has
products in various categories of the food manufacturing industry.

•

A respondent who does not qualify as a customer of one of the measured companies,
but has purchased or used the products or services for which screened from another
company, can be administered the three ACSI satisfaction questions about the “other
company”—if this respondent has not already been interviewed for a maximum of
three companies.

•

Demographic questions are asked of a responding customer only once, but can be
linked to each company interview that the customer completes.

2. Fill interview quotas for low-incidence companies.
•

Screen respondents for use of products or services of industries with varying
incidence of use.

•

Identify respondent purchases from specific companies with varying market shares.

•

Within industries, if a respondent is a customer for products or services from more
than one company, select the company with the lowest market share as the subject of
the interview.

3. Associate product, brand, and subsidiary names with parent companies.
•

A specialized database includes all identifiable brand names and subsidiaries of each
measured company. This database is incorporated in the CATI program for ACSI.
Names associated with the companies in each industry come up on the screen for
interviewer use (see examples for the major appliance industry and for the
supermarket industry in Appendix C: Example Screening Questionnaires and
Brand/Company Identification). Over 5,000 brand and subsidiary names are in the
database, which is updated every quarter. The brand database was initialized using
the Brands and Their Companies database of Gale Research, Inc. (Southfield,
Michigan). It is updated at the start of each interviewing period by checking the Web
pages of measured companies and by using sources such as business media reports on
acquisitions and mergers and company annual reports.

40

ACSI: The National Measure of Customer Satisfaction

Data Collection Via Telephone

•

The program allows the customer being interviewed to identify to the interviewer
which product or brand he or she purchased, and the brand database associates the
brand with the parent company.

•

The program inserts the name of the brand or subsidiary identified by the responding
customer into the company interview questionnaire so that the linkage is transparent
to the customer who is then interviewed using the specific brand or subsidiary name.

4. Monitor representativeness of sample for each industry throughout interviewing period.
•

Demographics are tracked and reviewed weekly throughout the field interviewing
period. Because both list and random-digit-dial (RDD) samples are used for the
Manufacturing/Durable Goods sector, demographics of each sample are tracked
separately and in combination. Interviewing is switched over to RDD only if the list
sample shows beyond expected deviations from the demographics of the RDD
sample.

Interviews for the baseline 1994 American Customer Satisfaction Index were conducted
simultaneously for all sectors, industries, and companies between May 10 and July 22, 1994.
Since then, interviews have been conducted quarterly for one or more sectors per quarter.
The field interviewing schedule is shown in Table 1 in Chapter I, Introduction. The
University of Michigan competitively bids sampling and interviewing work periodically.5

5

Pretest interviewing in 1993 and sampling, interviewing, data collection, and consulting from 1997 to 1999 were provided
by Wirthlin Worldwide (McClean, Virginia), now part of Harris Interactive. For the years 1994 to 1996 and 2000 to 2004,
ACSI sampling, interviewing, data collection, and consulting have been provided by Market Strategies, Inc. (Livonia,
Michigan).

41

The American Customer Satisfaction Index

E-Business & E-Commerce Data Collection

VII. E-BUSINESS AND E-COMMERCE SAMPLES
AND DATA COLLECTION VIA INTERNET
Since 2000, the American Customer Satisfaction Index has been expanded to add
measurement of customer satisfaction with companies that sell products and services on the
Internet. These companies are in three e-business industries (news and information, portals,
and search engines) and in four e-commerce industries (retail, auctions, brokerage, and
travel). Internet interviewing is appropriate for obtaining customer evaluations for these
Internet-based companies.

A. SAMPLE SELECTION
Samples are drawn from the Survey Spot Panel developed and maintained by Survey
Sampling International (Survey Sampling International, 2004). Survey Spot is an online
panel with 1,329,409 active members 18 years or older in the United States. Respondents
give explicit permission to receive invitations for surveys and receive incentives via a
monthly cash drawing system that rewards respondents for participation in surveys,
regardless of whether they qualify for participation in a particular survey.
For ACSI, a series of random samples are drawn from the panel so as to match the age
and gender demographics of the population based on the 2000 census. The selected sample
persons are then invited by e-mail to visit the MSInteractive Web site and complete the
survey. Invitations are sent out in three waves for each interview period, and reminder
e-mails are sent to nonresponders.

B. SCREENING TO QUALIFIED CUSTOMERS
During the screening process for the e-business and e-commerce interviews, individuals
are informed that they are not eligible to participate unless they are between the ages of 18
and 84 and are U.S. citizens. To qualify as a customer for any of the industries, the
respondent has to have used one of the measured companies or an “other” company in the
industry.

43

The American Customer Satisfaction Index

ACSI Methodology Report

C. DATA COLLECTION VIA INTERNET
For e-business and e-commerce, the standard private sector ACSI questionnaire shown in
Appendix D is reformatted for screen display and ease of respondent use. Customers who
qualify on the basis of screening proceed through a series of screens on the Internet and
input responses to survey questions on their computers.
In 2004, 7,348 Internet interviews were conducted with 6,150 customer respondents (an
average of 1.2). Internet interviewing results are reported in Appendix E: Response and
Cooperation Rates following the tabulation of telephone interviews.

44

ACSI: The National Measure of Customer Satisfaction

References

REFERENCES
American Association of Public Opinion Research (2004). Standard Definitions: Final
Depositions of Case Codes and Outcome Rates for Surveys.
Anderson, Eugene W., Claes Fornell, and Donald R. Lehmann (1994). “Customer
Satisfaction, Market Share, and Profitability: Findings from Sweden.” Journal of
Marketing, 58 (July), 53-56.
Andrews, Frank M. (1984). “Construct Validity and Error Components of Survey
Measures: A Structural Modeling Approach.” Public Opinion Quarterly, 48, 409-442.
Bagozzi, Richard P. (1980). Causal Models in Marketing. New York: John Wiley & Sons.
Cadotte, Ernest R., Robert B. Woodruff, and Roger L. Jenkins (1987). “Expectations and
Norms in Models of Consumer Satisfaction.” Journal of Marketing Research, 24 (August),
305-314.
Cassell, Claes (1993). “Comparison of the Flagship-Product and the Company Methods.”
Unpublished paper, Stockholm School of Economics, Stockholm, Sweden.
Churchill, Gilbert A., and Carol Suprenant (1982). “An Investigation Into the Determinants
of Customer Satisfaction.” Journal of Marketing Research, 19 (November), 491-504.
Cronbach, Lee J., and Paul E. Meehl (1955). “Construct Validity in Psychological Tests.”
Psychological Bulletin, 52 (4), 281-302.
Deming, W. Edwards (1981). Management of Statistical Techniques for Quality and
Productivity. New York: New York University, Graduate School of Business.
Fornell, Claes (2001). “The Science of Satisfaction.” Harvard Business Review, (March),
120-124.
 (1995). “The Quality of Economic Output: Empirical Generalizations About Its
Distribution and Relationship to Market Share.” Marketing Science, 14 (Summer), 3,
203-211.
 (1992). “A National Customer Satisfaction Barometer: The Swedish Experience.”
Journal of Marketing, 56 (January), 6-21.
 and Fred L. Bookstein (1982). “Two Structural Equation Models: LISREL and
PLS Applied to Consumer Exit-Voice Theory.” Journal of Marketing Research
(November), 440-452.

45

The American Customer Satisfaction Index

ACSI Methodology Report

, Michael D. Johnson, Eugene W. Anderson, Jaesung Cha, and Barbara Everitt
Bryant (1996). “The American Customer Satisfaction Index: Nature, Purpose, and
Findings.” Journal of Marketing, 60 (October), 7-18.
 , Sunil Mithas, Forrest Morgeson, and M. S. Krishnan (2005). “Customer
Satisfaction and Stock Prices: High Returns, Low Risk.” Working paper, National
Quality Research Center, Stephen M. Ross School of Business, University of Michigan,
Ann Arbor, MI.
, David VanAmburg, Forrest Morgeson, Eugene W. Anderson, Barbara Everitt
Bryant, and Michael D. Johnson (2005). “The American Customer Satisfaction Index at
Ten Years: ACSI 1994-2004, A Summary of Findings: Implications for the Economy, Stock
Returns and Management.” Working paper, National Quality Research Center, Stephen M.
Ross School of Business, University of Michigan, Ann Arbor, MI.
 and Birger Wernerfelt (1988). “A Model for Consumer Complaint Management.”
Marketing Science, 7 (Summer), 271-286.
 and  (1987). “Defensive Marketing Strategy by Customer Complaint
Management.” Journal of Marketing Research, 24 (November), 337-346.
Gordon, Robert J. (1990). The Measurement of Durable Goods Prices. Chicago, IL:
University of Chicago Press.
Hauser, John R., and Steven M. Shugan (1983). “Defensive Marketing Strategies.”
Marketing Science, 2 (4), 319-360.
Hirschman, Albert O. (1970). Exit, Voice, and Loyalty— Responses to Decline in Firms,
Organizations, and States. Cambridge, MA: Harvard University Press.
Howard, John A. (1977). Consumer Behavior: Application of Theory. New York:
McGraw-Hill, Inc.
Johnson, Michael D. (1984). “Consumer Choice Strategies for Comparing Noncomparable
Alternatives.” Journal of Consumer Research, 11 (December), 741-753.
, Eugene W. Anderson, and Claes Fornell (1995). “Rational and Adaptive
Performance Expectations in a Customer Satisfaction Framework.” Journal of Consumer
Research, 21 (March), 128-140.
 and Claes Fornell (1991). “A Framework for Comparing Customer Satisfaction
Across Individuals and Product Categories.” Journal of Economic Psychology, 12 (2),
267-286.

46

ACSI: The National Measure of Customer Satisfaction

References

Juran, Joseph M., and Frank M. Gryna (1988). Juran’s Quality Control Handbook,
4th Edition. New York: McGraw-Hill, Inc.
Lancaster, Kelvin (1971). Consumer Demand: A New Approach. New York: Columbia
University Press.
Marketing Systems Group (2004). Genesys CSS, 565 Virginia Drive, Fort Washington, PA
19034.
Mazvancheryl, Sanal, Eugene W. Anderson, and Claes Fornell (2004). “Customer
Satisfaction and Shareholder Value: The Association Between ACSI and Tobin’s q.”
Working paper, National Quality Research Center, Stephen M. Ross School of Business,
University of Michigan, Ann Arbor, MI.
National Economic Research Associates (1991). Developing a National Quality Index: A
Preliminary Study of Feasibility. Washington, D.C.
Oliver, Richard L. (1980). “A Cognitive Model of the Antecedents and Consequences of
Satisfaction Decisions.” Journal of Marketing Research, 17 (November), 460-469.
Simon, John L. (1974). “Interpersonal Welfare Comparisons Can Be Made and Used for
Redistribution Decisions.” Kyklos, 27, 63-98.
Survey Sampling International (2004). “Survey Spot Panel,” Fairfield, CT.
Tse, David K., and Peter C. Wilton (1988). “Models of Consumer Satisfaction Formation:
An Extension.” Journal of Marketing Research, 25 (May), 204-212.
Van Raaij, W. Fred (1989). “Economic News, Expectations, and Macro-Economic
Behavior.” Journal of Economic Psychology, 10 (December), 473-493.
Westbrook, Robert A., and Michael D. Reilly (1983). “Value-Percept Disparity: An
Alternative to the Disconfirmation of Expectations Theory of Consumer Satisfaction.” In
Advances in Consumer Research, Vol. 10, Richard P. Bagozzi and Alice M. Tybout (eds.).
Ann Arbor, MI: Association for Consumer Research, 256-261.
Woodruff, Robert B., Ernest R. Cadotte, and Roger L. Jenkins (1983). “Modeling
Consumer Satisfaction Processes Using Experience-Based Norms.” Journal of Marketing
Research, 20 (August), 296-304.
Yi, Youjae (1991). “A Critical Review of Customer Satisfaction.” In Review of Marketing
1990, Valerie Zeithaml (ed.). Chicago, IL: American Marketing Association, 68-123.

47

The American Customer Satisfaction Index

APPENDIX A:
COMPANIES AND GOVERNMENT
SERVICES EVALUATED BY
CUSTOMERS IN ACSI

49

COMPANIES AND GOVERNMENT SERVICES IN ACSI (2005)
Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

UTILITIES SECTOR
Energy Utilities
Allegheny Energy, Inc.
Ameren Corporation
American Electric Power Company, Inc.
CenterPoint Energy, Inc.
Cinergy Corp.
CMS Energy Corporation
Consolidated Edison, Inc.
Dominion Resources, Inc.
DTE Energy Company
Duke Energy Corporation
Edison International
Energy East Corporation
Entergy Corporation
Exelon Corporation
FirstEnergy Corp.
FPL Group, Inc.
KeySpan Corporation
National Grid Transco plc
NiSource Inc.
Northeast Utilities
Pepco Holdings, Inc.
PG&E Corporation
PPL Corporation
Progress Energy, Inc.
Public Service Enterprise Group Incorporated
Reliant Energy, Inc.
Sempra Energy
Southern Company
TXU Corp.
Xcel Energy Inc.
Total Energy Utilities
TOTAL UTILITIES SECTOR

2,742
4,593
15,441
9,772
4,416
6,017
9,827
12,078
7,062
23,483
12,156
4,919
9,195
15,812
12,318
9,630
6,915
6,247
6,401
6,065
7,271
11,221
5,587
9,027
11,340
11,707
7,887
11,251
11,325
7,939
279,644
279,644

6

(Residential revenue)
806
1,240
1,127
1,780
490
1,486
3,862
5,677
2,097
6,857
4,765
1,845
2,685
6,122
4,262
4,430
3,617
3,785
1,741
1,759
2,545
2,962
1,067
1,363
2,994
2,576
3,581
3,623
2,492
4,390
88,026
88,026

605
382
132
201
393
307
198
164
273
75
163
354
217
126
158
205
274
NL
292
304
270
179
324
224
176
166
257
178
177
254
0
0

The asterisk under “Total Revenues” means that the company’s total revenue is shown under a larger category
for that company.

51

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

MANUFACTURING/NONDURABLE GOODS SECTOR
Food Manufacturing (baked goods, cereal,
confectionery products, canned & packaged
fresh foods, meat & cheese)
Campbell Soup Company
ConAgra Foods, Inc.
Dole Food Company, Inc.
General Mills, Inc.
H.J. Heinz Company
Hershey Foods Corporation
Kellogg Company
Kraft Foods Inc.
Mars, Incorporated
Nestlé S.A.
Quaker Foods (PepsiCo, Inc.)
Sara Lee Corporation
Tyson Foods, Inc.
Total Food Manufacturing

6,678
22,053
4,608
10,506
9,328
4,173
8,812
*
17,000
*
18,291
24,459
125,908

52

5,626
19,848
1,898
10,987
3,842
3,338
5,264
21,907
12,240
1,592
1,500
5,233
21,909
115,184

39,788

39,788

280
84
380
186
213
415
233
15
NL
Non-US
63
104
72

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

Pet Food
Del Monte Foods Company
Hill's Pet Nutrition, Inc. (Colgate-Palmolive
Company)
The Iams Company (The Procter & Gamble
Company)
Mars, Incorporated
Nestlé Purina PetCare Company (Nestlé S.A.)
Total Pet Food
Soft Drinks
Cadbury Schweppes plc
The Coca-Cola Company
PepsiCo, Inc.
Total Soft Drinks
Breweries
Adolph Coors Company
Anheuser-Busch Companies, Inc.
Miller Brewing Company (SABMiller plc)
Total Breweries

2,171
*

412
1,316

670
196

*

2,000

28

*
*
2,171

1,700
1,592
7,020

39,585
40,473
80,058

2,903
19,292
16,835
39,030

4,000
14,147

Cigarettes
Philip Morris USA Inc. (Altria Group, Inc.)
Reynolds American Inc.
Total Cigarettes

53

18,147

2,410
11,182
3,465
17,057

69,704
5,267
74,971

17,001
2,633
19,634

*
0

8,538

NL
Non-US

Non-US
91+112+989
62+190+498

8,538

8,295
8,295

430
142
Non-US

15
341
0

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

Apparel (casual clothes, jeans & sportswear,
underwear & hosiery)
Jones Apparel Group, Inc.
Levi Strauss & Co.
Liz Claiborne, Inc.
Sara Lee Corporation
VF Corporation
Total Apparel

4,375
4,091
4,241
*
5,207
17,914

4,375
2,496
3,305
4,311
5,207
19,694

Athletic Shoes
NIKE, Inc.
Reebok International Ltd.
Total Athletic Shoes

10,697
3,485
14,182

3,020
1,591
4,611

4,171
9,903

57,451

3,354
2,872
873
18,876
6,143
32,118

390,802

254,348

Personal Care & Cleaning Products
(cleaners, detergent, shampoo, soap,
toothpaste)
The Clorox Company
Colgate-Palmolive Company
The Dial Corporation (Henkel KGaA)
The Procter & Gamble Company
Unilever
Total Personal Care & Cleaning Products

43,377

TOTAL MANUFACTURING/NONDURABLE
GOODS SECTOR

54

399
422
407
104
343
0

184
477
0

11,844
53,674
65,518
122,139

416
196
Non-US
28
Non-US

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

MANUFACTURING/DURABLE GOODS
SECTOR
Personal Computers
Apple Computer, Inc.
Dell Inc.
Gateway, Inc.
Hewlett-Packard Company
Total Personal Computers

6,207
41,444
3,402
73,061
124,114

Cellular Telephones
Kyocera Corporation
Motorola, Inc.
Nokia Corporation
Samsung Electronics Co., Ltd.
Total Cellular Telephones

27,058

27,058

Electronics (TV/VCR/DVD)
TV, VCR, DVD aggregated
Total Electronics (TV/VCR/DVD)
Major Appliances (washer, dryer, stove,
refrigerator, dishwasher)
General Electric Company
Maytag Corporation
Whirlpool Corporation
Total Major Appliances

134,187
4,792
12,176
151,155

55

3,338
26,524
3,402
13,223
46,487

301
31
484
11
0

1,052
2,606
2,656
1,183
7,497

37,031
7,563
55,526

9,670
9,670

0
0

4,562
4,538
7,875
16,975

10,932

Non-US
61
Non-US
Non-US

5
361
161
0

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

Automobiles & Light Vehicles
Bayerische Motoren Werke AG (BMW)
DaimlerChrysler AG
Chrysler
Dodge
Jeep
Mercedes-Benz
Ford Motor Company
Ford
Lincoln, Mercury
Volvo
General Motors Corporation
Buick
Cadillac
Chevrolet
GMC
Oldsmobile
Pontiac
Saturn Corporation
Honda Motor Co., Ltd.
Hyundai Motor Company
Mazda Motor Corporation
Nissan Motor Co., Ltd.
Toyota Motor Corporation
Volkswagen AG
Total Automobiles & Light Vehicles

52,122
171,529
*
*
*
*

360,191

10,494
82,186
*
*
*
*
98,698
*
*
*
142,857
*
*
*
*
*
*
*
31,040
6,546
7,355
21,576
51,382
21,705
473,839

662,518

554,468

717,062

164,496
*
*
*
195,695
*
*
*
*
*
*
*

TOTAL MANUFACTURING/DURABLE
GOODS SECTOR

56

Non-US
Non-US

4

3

67,479
20,900
19,730
56,905
163,637
109,234
661,536

Non-US
Non-US
Non-US
Non-US
Non-US
Non-US

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

RETAIL TRADE SECTOR
Supermarkets
Albertson's, Inc.
The Kroger Co.
Publix Super Markets, Inc.
Safeway Inc.
SUPERVALU INC.
Winn-Dixie Stores, Inc.
Wal-Mart Stores, Inc.
Total Supermarkets

34,436
53,791
16,848
35,553
20,210
12,168
*
173,006

34,436
53,791
16,848
35,553
20,210
12,168
20,430
193,436

0

Gasoline Stations
Gasoline stations (aggregated)
Total Gasoline Stations

559,236
559,236

731,806
731,806

434,299
434,299

Department & Discount Stores
Army and Air Force Exchange Service (AAFES)
Dillard's, Inc.
Federated Department Stores, Inc.
J.C. Penney Corporation, Inc.
Kmart Corporation
Kohl's Corporation
The May Department Stores Company
Sears, Roebuck and Co.
Target Corporation
Wal-Mart Stores, Inc
Total Department & Discount Stores

6,500
7,864
15,264
32,923
26,032
10,282
16,518
41,124
44,988
258,681
460,176

6,500
7,864
15,264
32,923
26,032
10,282
16,518
41,124
44,988
153,790
355,285

57

38
19
117
37
99
162
1

NL
258
134
43
67
189
147
32
23
1
0

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

Specialty Retail Stores
Best Buy Co., Inc.
Circuit City Stores, Inc.
Costco Wholesale Corporation
The Home Depot, Inc.
Lowe's Companies, Inc.
SAM'S CLUB (Wal-Mart Stores, Inc.)
Total Specialty Retail Stores
TOTAL RETAIL TRADE SECTOR

58

22,673
9,954
42,546
64,816
31,263
*
171,252

22,673
9,954
42,546
64,816
31,263
34,357
205,609

78
195
29
13
50
1
0

1,363,670

1,486,136

434,299

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

TRANSPORTATION & WAREHOUSING
SECTOR
Airlines
American Airlines (AMR Corporation)
Continental Airlines, Inc.
Delta Air Lines, Inc.
Northwest Airlines Corporation
Southwest Airlines Co.
United Airlines (UAL Corporation)
US Airways Group, Inc.
Total Airlines

17,440
8,870
13,303
9,510
5,937
13,724
6,846
75,630

17,440
8,870
13,303
9,510
5,937
13,724
6,846
75,630

0

U.S. Postal Service7
U.S. Postal Service
Total U.S. Postal Service

58,126
58,126

58,126
58,126

0
0

Express Delivery
FedEx Corporation
United Parcel Service, Inc.
U.S. Postal Service—Express & Priority Mail
Total Express Delivery

22,487
33,485
5,634
61,606

15,588
25,000
5,634
46,222

0
0

195,362

179,978

0

TOTAL TRANSPORTATION &
WAREHOUSING SECTOR

7

110
231
150
207
310
145
276

This measures USPS monopoly mail and counter services only. The USPS’s Express Mail and Priority Mail
services are included in the competitive Express Delivery industry.

59

82
42
NL

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

INFORMATION SECTOR
Newspapers
Newspapers [aggregate of Advance
Publications, Inc. (Time Warner Inc.).; Dow
Jones & Company, Inc.; Gannett Company,
Inc.; Knight-Ridder, Inc.; The New York Times
Company; The Times Mirror Company; Tribune
Company]
Total Newspapers
Motion Pictures (motion pictures, video,
DVD)
Motion pictures (motion pictures, video, DVD)
(aggregated nationally)
Total Motion Pictures
Broadcasting TV News
Broadcasting TV News (aggregate of ABC,
CBS, CNN, Fox, NBC)
Total Broadcasting TV News
Fixed Line Telephone Service
AT&T Corp.
BellSouth Corporation
MCI, Inc.
Qwest Communications International Inc.
SBC Communications Inc.
Sprint FON Group (Sprint Corporation)
Verizon Communications Inc.
Total Fixed Line Telephone Service

60

22,806

22,806

0

22,806

22,806

0

34,000

34,000

34,000

34,000

0

38,470

38,470

0

38,470

38,470

0

34,529
22,635
27,315
14,936
40,843
26,917
67,752
234,927

9,484
16,957
6,375
4,230
28,842
7,923
23,510
97,321

40
80
NL
136
33
65
12
0

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

Wireless Telephone Service
Cingular Wireless LLC (BellSouth Corporation,
SBC Communications Inc.)
Nextel Communications, Inc.
Sprint PCS Group (Sprint Corporation)
T-Mobile USA, Inc. (Deutsche Telekom AG)
Verizon Wireless (Cellco Partnership)
Total Wireless Telephone Service
Cable & Satellite TV
Charter Communications, Inc.
Comcast Corporation
Cox Communications, Inc.
The DIRECTV Group, Inc.
EchoStar Communications Corporation
Time Warner Cable
Total Cable & Satellite TV

32,178

NL

67,752
78,572

10,820
12,690
8,358
22,489
86,535

183
65
Non-US
12

4,819
21,263
5,759
10,121
5,551
43,877
91,390

4,819
21,263
5,759
10,121
5,551
16,129
63,642

0

500,165

342,774

70,200

10,820
*

TOTAL INFORMATION SECTOR

61

70,200
70,200

358
89
318
NL
327
27

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

FINANCE & INSURANCE SECTOR
Banks
Bank of America Corporation
J.P. Morgan Chase & Co.
Wachovia Corporation
Wells Fargo & Company
Total Banks

62,427
65,817
24,474
31,800
184,518

62,427
65,817
24,474
31,800
184,518

24+140
26+88
73
49
0

Life Insurance
MetLife, Inc.
New York Life Insurance Company
The Northwestern Mutual Life Insurance
Company
Prudential Financial, Inc.
Total Life Insurance

36,261
25,700
17,060

36,261
25,700
17,060

36
70
115

27,907
106,928

27,907
106,928

57

Health Insurance
Aetna Inc.
Blue Cross and Blue Shield Association
UnitedHealth Group Incorporated
Total Health Insurance

17,976
182,700
28,823
229,499

17,976
182,700
28,823
229,499

108
NL
54

47
NL
14
165
18

0

Property & Casualty Insurance
The Allstate Corporation
Farmers Group, Inc. (Zurich Financial Services)
GEICO (Berkshire Hathaway Inc.)
The Progressive Corporation
State Farm Mutual Automobile Insurance
Company
Total Property & Casualty Insurance

32,149
3,281
63,859
11,892
56,065

32,149
3,281
7,784
11,892
56,065

167,246

111,171

0

TOTAL FINANCE & INSURANCE SECTOR

688,191

632,116

0

62

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

HEALTH CARE & SOCIAL ASSISTANCE
SECTOR
Hospitals
Hospitals (aggregated nationally)
Total Hospitals
TOTAL HEALTH CARE & SOCIAL
ASSISTANCE SECTOR

63

451,220
451,220

451,220
451,220

0
0

451,220

451,220

0

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

ACCOMMODATION & FOOD SERVICES
SECTOR
Hotels
Global Hyatt Corporation
Hilton Hotels Corporation
Holiday Inn (InterContinental Hotels Group
PLC)
Marriott International, Inc.
Ramada Franchise Systems Inc. (Cendant
Corporation)
Starwood Hotels & Resorts Worldwide, Inc.
Total Hotels

3,600
3,853

Limited-Service Restaurants
Burger King Corporation
Domino's, Inc.
Little Caesar Enterprises, Inc.
McDonald's Corporation
Papa John's International, Inc.
Wendy's International, Inc.
YUM! Brands, Inc.
KFC
Pizza Hut
Taco Bell
Total Limited-Service Restaurants
TOTAL ACCOMMODATION & FOOD
SERVICES SECTOR

64

3,600
3,082
5,210

6,192

9,198
18,192

7,358
1,820

3,801
38,644

2,802
23,872

125,839
1,333
435
17,141
917
3,149
8,380
*
*
*
157,194

4,000
1,247
435
6,943
416
2,191
*
1,576
1,802
1,802
20,412

0

195,838

44,284

6,192

NL
438
Non-US
216
106
441

6,192

NL
951
NL
114
NL
507
245

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

PUBLIC ADMINISTRATION SECTOR
Federal Agencies
Total Federal Government

19,855
19,855

19,855
19,855

Solid Waste Disposal
Police
Total Local Government

14,861
48,217
63,078

14,861
48,217
63,078

TOTAL PUBLIC ADMINISTRATION SECTOR

82,933

82,933

65

0

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

E-BUSINESS
News & Information
ABCNEWS.com (The Walt Disney Company)
CNN.com (Time Warner Inc.)
MSNBC.com (NBC, Microsoft Corporation)
NYTimes.com (The New York Times Company)
USATODAY.com (Gannett Co., Inc.)

27,061
*
*
3,227
6,711

383
295
344
60
102

60
27
5 & 46
499
278

Portals
America Online, Inc. (Time Warner Inc.)
MSN (Microsoft Corporation)
Yahoo! Inc.

43,877
32,187
1,625

6,982
1,855
1,625

27
46
813

Search Engines
AltaVista (Yahoo! Inc.)
Ask Jeeves, Inc.
Google Inc.

46
107
962

23
107
962

NL
NL
NL

TOTAL E-BUSINESS

115,803

12,738

66

0

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

Fortune
Rank by
Revenue
4/5/04

(millions $)
(Estimated)

E-COMMERCE
Retail
1-800-FLOWERS.COM, Inc.
Amazon.com, Inc.
barnesandnoble.com inc.
Buy.com Inc.

604
5,264
5,565
800

265
5,264
425
800

NL
342
325
NL

Auctions
eBay Inc.
priceline.com Incorporated
uBid, Inc. (Petters Group Worldwide, LLC)

2,165
864
225

2,165
864
225

673
NL
NL

Brokerage
The Charles Schwab Corporation
E*TRADE Financial Corp.

4,328
2,008

350
2,008

402
707

1,000
242
2,045

1,000
242
368

NL
NL
700

25,110

13,976

Travel
Expedia, Inc. (IAC/InterActiveCorp)
Orbitz, Inc. (Cendant Corporation)
Travelocity.com L.P. (Sabre Holdings
Corporation)
TOTAL E-COMMERCE

67

0

Companies/Government Services by
Sector & Industry

2003 Total
Revenues of
US
Companies
(millions $)6

2003
Revenues in
US Domestic
Market
Segments
Measured in
ACSI

Worldwide
Revenues of
Non-US
Companies
(millions $)

(millions $)
(Estimated)

TOTAL MEASURED U.S. COMPANIES AND
GOVERNMENT SERVICES (millions of $)
GROSS DOMESTIC PRODUCT (GDP)
(2nd Q 2004)

4,951,256

4,142,997

11,649,300

11,649,300

42.5%

35.6%

ACSI-MEASURED COMPANIES AS
% OF GDP
WORLDWIDE TOTAL REVENUES OF
NON-US COMPANIES MEASURED
IN ACSI (millions of $)

1,349,892

1,349,892

NL = Not listed in Fortune 1000.

SOURCES:
Annual reports of individual companies.
Hooversonline.com
Statistical Abstracts of the United States 2003.
Press clippings from The Wall Street Journal, The New York Times, Business Week.
“The Fortune 1000 Ranked Within Industries,” Fortune, April 5, 2004.
Hoover's Handbook of World Business 2004 (Hoover’s Business Press).
U.S. Trade and Industry Outlook 2004.
GDP from www.bea.gov

68

Fortune
Rank by
Revenue
4/5/04

APPENDIX B:
ACSI INDUSTRY DEFINITIONS AND
CUSTOMER IDENTIFICATION

69

ACSI INDUSTRY DEFINITIONS AND CUSTOMER IDENTIFICATION
Utilities Sector
Industry: Energy Utilities
Customer identified by these questions:
“Which company provides your electric service?”
“Do you have natural gas service?”
(IF YES) “Which company provides your natural gas service?”
Manufacturing/Nondurable Goods Sector
Industry: Food Manufacturing
Categories:
• baked goods (bread, cake, flour, cookies, crackers)
• cereal (cold or hot)
• confectionery products (chocolate, cocoa, chocolate candy)
• canned & packaged fresh foods (soup, canned vegetables or fruits, pickles, ketchup,
packaged salads, fresh vegetables or fruits with a brand name)
• meat & cheese (packaged cold meats, franks, sausage, poultry, cheese)
Customers identified by any of several questions:
“Have you purchased and consumed any baked goods, bread, cakes, flour, cookies, or
crackers in the last month?”
“Have you purchased and consumed cold or hot cereal in the last month?”
“Have you purchased and consumed chocolate, cocoa, or chocolate candy in the last
month?”
“Have you purchased and consumed any canned goods such as soup, vegetables,
fruits, pickles, ketchup, or packaged salads with a brand name in the last month?”
“Have you purchased and consumed any cold meats, franks, sausage, poultry, or
cheese in the last month?”

71

Industry: Pet Food
Customers identified by two-part question:
“Does your household have a cat or a dog?”
(IF YES) “Which brands of pet food have you purchased for your dog or cat to eat in the
last month?”
Industry: Soft Drinks
Customers identified by question:
“Have you purchased and consumed soft drinks or pop in the last month?”
Industry: Breweries
Customers identified by two-part question:
“Are you 21 years of age or older?”
(IF YES) “Have you purchased and consumed beer in the last month?”
Industry: Cigarettes
Customers identified by two-part question:
“Are you 21 years of age or older?”
“Have you purchased and smoked cigarettes in the last month?”
Industry: Apparel
Categories:
• casual clothes
• jeans & sportswear
• underwear & hosiery (underwear, pantyhose, hosiery, socks, tee-shirts, turtlenecks)
Customers identified by questions:
“Have you purchased casual clothes in the last year?”
“Have you purchased jeans or sportswear in the last year?”
“Have you purchased underwear, pantyhose, hosiery, socks, tee-shirts, or turtlenecks in
the last year?”
Industry: Athletic Shoes
Customers identified by question:
“Have you purchased athletic, running, or jogging shoes in the last year?”

72

Industry: Personal Care & Cleaning Products
Categories:
• cleaners (bleach, ammonia, cleansers, cleaning compounds)
• detergent
• shampoo
• soap
• toothpaste
Customers identified by any of several questions:
“Have you purchased bleach, ammonia, cleansers, or cleaning compounds in the last
three months?”
“Have you purchased detergent in the last three months?”
“Have you purchased shampoo in the last three months?”
“Have you purchased soap in the last three months?”
“Have you purchased toothpaste in the last three months?”
Manufacturing/Durable Goods Sector
Industry: Personal Computers
Customers identified by question:
“Have you purchased a new personal computer for your home in the past three years?”
Industry: Cellular Telephones
Customers identified by question:
“Have you purchased a new cellular phone in the past two years?”
Industry: Electronics (TV/VCR/DVD)
Customers identified by either of two questions:
“Have you purchased a new television within the last three years for personal use?”
“Have you purchased a new VCR or DVD player within the last three years for personal
use?”
(NOTE: TV/VCR/DVD combinations are recorded.)
Industry: Major Appliances (washer, dryer, stove, refrigerator, dishwasher)
Customers identified by question:
“Have you purchased a new major appliance such as a washer, dryer, stove,
refrigerator, or dishwasher in the past three years?”

73

Industry: Automobiles & Light Vehicles
Customers identified by question:
“Have you purchased a NEW automobile/van/light truck between six months and three
years ago, which you still own, or have you personally leased a NEW
automobile/van/light truck during that time period, even if you still don’t own it?”
Retail Trade Sector
Industry: Supermarkets
Customers identified by question:
“Have you shopped for groceries at a supermarket or another store that sells a wide
variety of goods including groceries in the past three months?”
Industry: Gasoline Stations
Customers identified by question:
“Have you purchased gasoline for your automobile in the past three months?”
Industry: Department & Discount Stores
Customers identified by question:
“Have you shopped at a department or discount store for merchandise NOT INCLUDING
GROCERIES in the past six months? By department or discount store, I mean a store
selling a wide variety of goods and arranged in several departments.”
Industry: Specialty Retail Stores
Customers identified by question:
“Within the past six months, have you shopped at a wholesale warehouse club, or a
store that specializes in selling only certain products or goods? Some examples of
specialty stores are those that mostly sell home improvement products, toys, electronics,
computer products, office products, pet supplies, clothes, books, music, and so forth.”
Transportation & Warehousing Sector
Industry: Airlines
Customers identified by question:
“Have you flown on a scheduled airline in the past year?”

74

Industry: U.S. Postal Service
Customers identified by question:
“Have you visited a U.S. Post Office in the past six months to buy stamps, pick up mail,
or use any of the counter services?”
Industry: Express Delivery
Customers identified by question:
“Have you used a parcel delivery, overnight or two-day mail delivery service for sending
a letter, a document, or a package from either your home or your place of work in the
past six months?”
Information Sector
Industry: Newspapers
Customers identified by question:
“Have you read a newspaper, which you purchased or subscribed to, in the last week?”
Industry: Motion Pictures (motion pictures, video, DVD)
Customers identified by question:
“Have you been to a motion picture theater or rented or purchased a video tape or DVD,
or watched a movie on pay-per-view in the past year?”
Industry: Broadcasting TV News
Customers identified by question:
“Have you watched a national TV news program in the last month?” (This DOES NOT
include shows like “Dateline,” “Primetime,” “60 Minutes,” or “20/20.” This DOES include
national network news on CNN, ABC, CBS, NBC, and Fox. Also includes “Good
Morning America” and “Today.”)
Industry: Fixed Line Telephone Service
Customers identified by questions:
“Which company provides your LONG DISTANCE telephone service?”
“Which company provides your LOCAL telephone service?”
Industry: Wireless Telephone Service
Customers identified by question:
“Do you currently have cellular phone service in your name?”

75

Industry: Cable & Satellite TV
Customers identified by question:
“Do you currently subscribe to cable television or satellite television service?”
Finance & Insurance Sector
Industry: Banks
Customers identified by question:
“Do you have an active checking account, savings account, or bank loan in your name at
a bank, not a credit union?”
Industry: Life Insurance
Customers identified by question:
“Do you have life insurance in your own name?”
Industry: Health Insurance
Customers identified by question:
“Do you have healthcare insurance?”
Industry: Property & Casualty Insurance
Customers identified by question:
“Do you have homeowner’s insurance, or automobile insurance, or other property or
casualty insurance in your own name?”
Health Care & Social Assistance Sector
Industry: Hospitals
Customers identified by question:
“Have you used the services of a hospital in the past three years?” (That includes a
parent that used the hospital services for a child.)

76

Accommodation & Food Services Sector
Industry: Hotels
Customers identified by question:
“Have you stayed overnight at a hotel when traveling for business or pleasure in the
United States in the past year?”
Industry: Limited-Service Restaurants
Customers identified by question:
“Have you purchased food from a fast food restaurant or pizza restaurant, to eat in the
restaurant, for carry out, or ordered for home delivery, in the last three months?”
Public Administration Sector
Solid Waste Disposal
Customers identified by the question:
“Does your local government provide your garbage and trash collection service?”
Police
Customers identified by question:
“Have you had any contact with your local police in the past three years—either asking
your police for information or help, being stopped for a traffic violation or some other
violation, or any other way in which you may have talked to police officers or the desk at
a police station?”
Federal Government
Customers are identified by qualifying questions specific to each federal service.
E-Business (online interviewing)
Customers for all industries are identified by the questions:
“Are you a resident of the United States?”
“Have you used a portal, news site, or search engine?”
Industry: News & Information
Customers identified by the question:
“Which of the following News and Information sites did you access in the past three
months?” (Check all that apply)

77

Industry: Portals
Customers identified by the question:
“Once you have connected to the Internet, which portals do you use most often to
search the Internet, find Web sites, etc.?” (Check all that apply)
Industry: Search Engines
Customers identified by the question:
“Which of the following search engines did you use in the past three months?” (Check
all that apply)
E-Commerce (online interviewing)
Customers for all industries are identified by the questions:
“Are you a resident of the United States?”
“Have you used a financial service, auction/reverse auction, retailer, or travel site?”
Industry: Retail
Customers identified by the question:
“From which retailers have you purchased merchandise on the Internet in the past three
months?” (Check all that apply)
Industry: Auctions
Customers identified by the question:
“From which auction and/or reverse auction sites on the Internet have you purchased
merchandise in the past three months?” (Check all that apply)
Industry: Brokerage
Customers identified by the question:
“From which financial services companies did you make a financial transaction (such as
buying stocks, bonds, or mutual funds), or receive financial consulting/planning
assistance on the Internet in the past three months?” (Note: this does not include online
bill paying or banking.) (Check all that apply)
Industry: Travel
Customers identified by question:
“From which Internet travel company did you make a reservation (for a flight, hotel room
or a rental car, or any vacation travel package) in the past six months?” (Check all that
apply)

78

APPENDIX C:
EXAMPLE SCREENING QUESTIONNAIRES
AND BRAND/COMPANY IDENTIFICATION

79

American Customer Satisfaction Index
Introduction
[ASK TO SPEAK TO PERSON RESIDING AT HOUSEHOLD WHO HAS HAD THE MOST RECENT
BIRTHDAY, BETWEEN 18 AND 84 YEARS OF AGE]
Hello, I’m _______________________________calling on behalf of the University of Michigan. We are
conducting research on how satisfied users are with services provided by federal government agencies and
private companies as part of the American Customer Satisfaction Index. You may have read something
about the American Customer Satisfaction Index in USA Today, The Wall Street Journal, or your local
newspaper. Your name will be confidential, and I will ask you only about products and services you have
recently purchased and used. Your participation is voluntary and you may stop at any time or skip any
question you do not wish to answer. Your opinions are important because you have been chosen randomly
to represent consumers across the United States.
____________________________________________________________________________

81

Example ACSI Screening Questionnaire
Major Appliances
Have you purchased a new major appliance such as a washer, dryer, stove, refrigerator, or dishwasher in
the past three years?
1 YES

(Continue)

2 NO

(R. does not qualify for this industry; go to next screener)

3 DON’T KNOW

(R. does not qualify for this industry; go to next screener)

4 REFUSED

(R. does not qualify for this industry; go to next screener)

Which brand did you purchase?
[INSERT CO./BRAND LIST]

(Continue)

OTHER (SPECIFY) __________

(Continue)

DON’T KNOW/REFUSED

(R. does not qualify for this industry; continue screening)

83

Example ACSI Company/Brand Database
Major Appliances
The following is a listing of the brands linked to measured companies in the computer-assistedtelephone-interviewing (CATI) system by the ACSI Company/Brand Database for this industry.
Accellis
Admiral
Advantium
Amana
Amana Distinctions
Atlantis
Avanti
Crosley
Dynasty
Equator
Estate
GarageWorks
GE
GE Advantium
GE Monogram
GE Monogram Advantium
GE Nautilus
GE Profile
GE Profile Advantium
GE Profile Arctica
GE Profile Harmony
GE Profile Prodigy
GE Profile Spacemaker
GE Profile Spectra
GE Profile Ultimate Advantium
GE Profile Wizard
GE Spacemaker
GE Spectra
GE Triton
GE Ultimate Advantium
Gemini
General Electric
Gladiator
Haier
Hotpoint
Inglis
Jade
Jenn-Air
Jenn-Air Designer
Jenn-Air Expressions
Jenn-Air Luxury
Jenn-Air ProStyle
Jetclean
KitchenAid
KitchenAid Architect
KitchenAid Briva
KitchenAid Ensemble

Maytag Corporation
Maytag Corporation
General Electric Company
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Whirlpool Corporation
Whirlpool Corporation
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
General Electric Company
Maytag Corporation
General Electric Company
Whirlpool Corporation
Maytag Corporation
General Electric Company
Whirlpool Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation

85

KitchenAid Pro Line
KitchenAid Superba
Magic Chef
Malber
Maytag
Maytag Accellis
Maytag Atlantis
Maytag Gemini
Maytag Jetclean
Maytag Neptune
Maytag Performa
Neptune
Norge
Performa
Roper
Speed Queen
Whirlpool
Whirlpool Calypso
Whirlpool Catalyst
Whirlpool Conquest
Whirlpool Duet
Whirlpool Durawash
Whirlpool GarageWorks
Whirlpool Gladiator
Whirlpool Gold
Whirlpool Polara
Whirlpool Power Clean
Whirlpool Powerscour
Whirlpool Resource Saver
Whirlpool Senseon
Whirlpool Ultimate Care II

Whirlpool Corporation
Whirlpool Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Maytag Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation
Whirlpool Corporation

86

Example ACSI Screening Questionnaire
Supermarkets
Have you shopped for groceries at a supermarket or another store that sells a wide variety of goods
including groceries in the past three months?
1 YES

(Continue)

2 NO

(R. does not qualify for this industry; go to next screener)

3 DON’T KNOW

(R. does not qualify for this industry; go to next screener)

4 REFUSED

(R. does not qualify for this industry; go to next screener)

Which supermarket or other store have you shopped for groceries at most frequently in the past three
months? (IF MORE THAN ONE, EMPHASIZE MOST FREQUENTLY)
[INSERT CO./BRAND LIST]

(R. qualifies for this industry; before going to questionnaire,
continue to screen for remaining groups)

OTHER (SPECIFY) __________

(R. does not qualify for this industry; continue screening)

DON’T KNOW/REFUSED

(R. does not qualify for this industry; continue screening)

87

Example ACSI Company/Brand Database
Supermarkets
The following is a listing of the store chains linked to measured companies in the CATI system by
the ACSI Company/Brand Database for this industry.
Acme Markets
Albertson’s
American Stores
Baker’s
Bell Markets
Bigg’s
Buttrey Food and Drug Stores
Cala Foods
Carr-Gottstein
Carrs
City Market
City Markets
Cub Foods
Dillon’s Food Stores
Dominick’s
Dominick’s Fresh Stores
Farm Fresh
Food4less
Foods Co.
Fred Meyer
Fresh Stores
Fry’s Food & Drug Stores
Fry’s Marketplace
Genuardi’s
Gerbes Supermarkets
Hilander
Hornbacher’s
Jay C
Jewel
Jewel-Osco
Kessel Food Markets
King Soopers
Kroger
Lucky Stores
Marketplace
Max Foods
Owen’s Market
Pavilions
Pay Less Super Markets
Pricerite
Publix
QFC (Quality Food Centers)
Quality Food Centers (QFC)
Ralphs
Randalls
Sack & Save
Safeway

Albertson’s, Inc.
Albertson’s, Inc.
Albertson’s, Inc.
The Kroger Co.
The Kroger Co.
SUPERVALU INC.
Albertson’s, Inc.
The Kroger Co.
Safeway Inc.
Safeway Inc.
The Kroger Co.
Winn-Dixie Stores, Inc.
SUPERVALU INC.
The Kroger Co.
Safeway Inc.
Safeway Inc.
SUPERVALU INC.
The Kroger Co.
The Kroger Co.
The Kroger Co.
Safeway Inc.
The Kroger Co.
The Kroger Co.
Safeway Inc.
The Kroger Co.
The Kroger Co.
SUPERVALU INC.
The Kroger Co.
Albertson’s, Inc.
Albertson’s, Inc.
The Kroger Co.
The Kroger Co.
The Kroger Co.
Albertson’s, Inc.
Winn-Dixie Stores, Inc.
Albertson’s, Inc.
The Kroger Co.
Safeway Inc.
The Kroger Co.
The Kroger Co.
Publix Super Markets, Inc.
The Kroger Co.
The Kroger Co.
The Kroger Co.
Safeway Inc.
Winn-Dixie Stores, Inc.
Safeway Inc.

89

Save Rite
Save-A-Lot
Scott’s Food
Seessel’s
Shaw’s
Shop ‘N Save
Shoppers Food Warehouse
Simon David
Smith’s Food & Drug Centers
Smitty’s
Star Markets
Super One Foods
Super Saver
Supervalu
Thriftway
Tom Thumb Food & Drug Stores
Vons
Wal-Mart
Wal-Mart Neighborhood Market
Wal-Mart Supercenter
Winn-Dixie
Winn-Dixie Marketplace

Winn-Dixie Stores, Inc.
SUPERVALU INC.
SUPERVALU INC.
Albertson’s, Inc.
Albertson’s, Inc.
SUPERVALU INC.
SUPERVALU INC.
Safeway Inc.
The Kroger Co.
Albertson’s, Inc.
Albertson’s, Inc.
Albertson’s, Inc.
Albertson’s, Inc.
SUPERVALU INC.
Winn-Dixie Stores, Inc.
Safeway Inc.
Safeway Inc.
Wal-Mart Stores, Inc.
Wal-Mart Stores, Inc.
Wal-Mart Stores, Inc.
Winn-Dixie Stores, Inc.
Winn-Dixie Stores, Inc.

90

APPENDIX D:
CUSTOMER SATISFACTION
MEASUREMENT QUESTIONNAIRES—
PRIVATE SECTOR AND
GOVERNMENT SERVICES

91

ACSI Questionnaire
For Private Sector Companies
(For input to models in Figures 1 and 1a)
NOTE: The questionnaire for most companies used for the model in Figure 1 is shown first; followed by a
section of six questions to be substituted for Q4, Q5, Q6 for the expanded model in Figure 1a in which
product quality and service quality are measured separately.
INTRODUCTION
[ASK TO SPEAK TO PERSON RESIDING AT HOUSEHOLD WHO HAS HAD THE MOST RECENT
BIRTHDAY, BETWEEN 18 AND 84 YEARS OF AGE]
Hello, I’m _______________________________calling on behalf of the University of Michigan. We are
conducting research on how satisfied users are with services provided by federal government agencies and
private companies as part of the American Customer Satisfaction Index. You may have read something
about the American Customer Satisfaction Index in USA Today, The Wall Street Journal, or your local
newspaper. Your name will be confidential, and I will ask you only about products and services you have
recently purchased and used. Your participation is voluntary and you may stop at any time or skip any
question you do not wish to answer. Your opinions are important because you have been chosen randomly
to represent consumers across the United States.
(INSERT SCREENER QUESTIONS APPROPRIATE TO PRODUCT/SERVICE)
In the questionnaire where the words [SCREENER INSERT] appear, the CATI (computer-assistedtelephone-interviewing) program inserts the name of the company, product, brand, service, store,
restaurant, hotel, etc., that the respondent has named in response to the screener questions. Within a
specific question, once the [SCREENER INSERT] has been mentioned, second or third mentions may be
more generic wordings for the product/service/outlet category.
(To begin/Next), think back to before you purchased your [SCREENER INSERT] and remember your
expectations about that particular [SCREENER INSERT]. I am going to ask you three questions about
your expectations. The first concerns your expectations of the overall quality of your [SCREENER
INSERT]; the other two questions consider your expectations of specific requirements of the
[SCREENER INSERT], and your expectations of potential problems with the [SCREENER INSERT].
Each time we will use a scale of 1 to 10, although the meaning of the scale will change slightly from
question to question.
Let’s begin:
Q1.

Before you purchased [SCREENER INSERT], you probably knew something about this particular
[SCREENER INSERT]. Now, think back and remember your expectations of the overall quality of
the [SCREENER INSERT]. Please give me a rating on a 10-point scale on which “1” means your
expectations were “not very high” and “10” means your expectations were “very high.”
How would you rate your expectations of the overall quality of [SCREENER INSERT]?
1 TO 10__________
11
12

Don’t know
Refused

93

(ROTATE Q2 AND Q3)
Q2.

(Again/At that same time), you probably thought about things you personally require from a
[SCREENER INSERT], such as [INSERT PRODUCT ATTRIBUTES]. Using a 10-point scale on
which “1” now means “not very well” and “10” means “very well,” how well did you expect your
[SCREENER INSERT] to meet your personal requirements?
1 TO 10__________
11
12

Q3.

Don’t know
Refused

(Again/At the same time), thinking about your expectations before you purchased (or your recent
experiences with) [SCREENER INSERT] . . . you probably thought about how often things could
go wrong with the [SCREENER INSERT] regarding such things as [INSERT PRODUCT
ATTRIBUTES]. Using a 10-point scale, on which “1” now means “very often” and “10” means
“not very often,” how often did you expect that things could go wrong with your [SCREENER
INSERT]?
1 TO 10__________
11
12

Don’t know
Refused

Next, I want you to think about your actual experience with your [SCREENER INSERT]. I am going to
ask you five questions, the first deals with your overall experience with [SCREENER INSERT]. The next
two questions deal with how well the [SCREENER INSERT] met your personal requirements, and how
often things go wrong with [SCREENER INSERT]. The other two questions are about specific
characteristics of the product or service…
Q4.

First, please consider all your experiences in the past [INSERT TIME PERIOD FROM
SCREENER] with your [SCREENER INSERT]. Using a 10-point scale, on which “1” means “not
very high” and “10” means “very high,” how would you rate the overall quality of your
[SCREENER INSERT]?
1 TO 10 __________
11
12

Don’t know
Refused

(ROTATE Q5 AND Q6)
Q5.

Now thinking about your personal requirements for a [SCREENER INSERT], such as [INSERT
PRODUCT ATTRIBUTES], please tell me how well your [SCREENER INSERT] has actually met
your requirements. Using a 10-point scale on which “1” now means “not very well” and “10”
means “very well,” how well has your [SCREENER INSERT] actually met your personal
requirements?
1 TO 10__________
11
12

Don’t know
Refused

94

Q6.

Now please think about how often things go wrong with the [SCREENER INSERT], regarding
such things as [INSERT PRODUCT ATTRIBUTES]. Using a 10-point scale on which “1” now
means “very often,” and “10” means “not very often,” how often have things actually gone wrong
with your [SCREENER INSERT]?
1 TO 10__________
11
12

Q7

Don’t know
Refused

(OPTIONAL QUESTION ABOUT CHARACTERISTICS OF PRODUCT/SERVICE)
1 TO 10 __________
11
12

Q8

Don’t know
Refused

(OPTIONAL QUESTION ABOUT CHARACTERISTICS OF PRODUCT/SERVICE)
1 TO 10 __________
11
12

Don’t know
Refused

Now I want you to consider the value of your [SCREENER INSERT] in terms of both (ROTATE: PRICE
AND QUALITY/QUALITY AND PRICE).
(ROTATE Q9 AND Q10)
Q9.

(FIRST/NEXT) Given the quality of your [SCREENER INSERT], how would you rate the price that
you paid (or prices that you pay) for [SCREENER INSERT]? Please use a 10-point scale on
which “1” means “very poor price given the quality” and “10” means “very good price given the
quality.”
1 TO 10 __________
11
12

Q10.

Don’t know
Refused

(FIRST/NEXT) Given the price that you paid (or prices that you pay at) for your [SCREENER
INSERT], how would you rate the quality of your [SCREENER INSERT]? Please use a 10-point
scale on which “1” means “very poor quality given the price” and “10” means “very good quality
given the price.”
1 TO 10__________
11
12

Don’t know
Refused

95

Satisfaction includes many things. Let’s move on and talk about your overall satisfaction with your
[SCREENER INSERT].
Q11

First, please consider all your experiences to date with your [SCREENER INSERT]. Using a 10point scale on which “1” means “very dissatisfied” and “10” means “very satisfied,” how satisfied
are you with your [SCREENER INSERT]?
1 TO 10__________
11
12

Q12.

Don’t know
Refused

Considering all of the expectations that we have discussed, to what extent has your [SCREENER
INSERT] fallen short of your expectations or exceeded your expectations? Using a 10-point
scale on which “1” now means “falls short of your expectations” and “10” means “exceeds your
expectations,” to what extent has your [SCREENER INSERT] fallen short of or exceeded your
expectations?
1 TO 10__________
11
12

Q13.

Don’t know
Refused

Forget your [SCREENER INSERT] for a moment. Now, I want you to imagine an ideal [INSERT
GENERIC NAME FOR SCREENER INSERT]. (PAUSE) How well do you think your
[SCREENER INSERT] compares with that ideal [INSERT GENERIC NAME FOR SCREENER
INSERT]? Please use a 10-point scale on which “1” means “not very close to the ideal,” and “10”
means “very close to the ideal.”
1 TO 10__________
11
12

Don’t know
Refused

Next, I want you to think about any communication you may have had with the company that produced
your [SCREENER INSERT] regarding complaints about your experience.
Q14.

Have you complained to the company about your [SCREENER INSERT] within the past [INSERT
SCREENER TIME PERIOD]?
1
2
3
4

Yes
No
Don’t know
Refused

96

{IF Q14 = 1, ASK Q14A; OTHERWISE GO TO Q15}
Q14A. How well, or poorly, was your most recent complaint handled? Using a 10-point scale on which
“1” means “handled very poorly” and “10” means “handled very well,” how would you rate the
handling of your complaint?
1 TO 10__________
11
12
Q15.

Don’t know
Refused

The next time you are going to [INSERT PURCHASE/USE/SHOP AT] a [SCREENER INSERT],
how likely is it that it will be a [SCREENER INSERT] again? Using a 10-point scale on which “1”
means “very unlikely” and “10” means “very likely,” how likely is it that it will be a [SCREENER
INSERT] again?
1 TO 10__________
11
12

Don’t know
Refused

{IF Q15 = 6-10, ASK Q16; OTHERWISE GO TO Q17}
Q16.

Let us imagine that [SCREENER INSERT] raises its prices. If other [COMPANIES/SUPPLIERS]
remain at the same prices, how much can [SCREENER INSERT] raise its price before you
definitely would not choose a(n) [SCREENER INSERT] the next time you purchase a [INSERT
GENERIC NAME FOR SCREENER INSERT ]?
Please provide your answer in percentages up to 25%
0 TO 25__________
26
101
102
103

26% or higher
Never would [PURCHASE/USE/SHOP AT] any other [SCREENER INSERT]
Don’t know
Refused

{IF Q15 = 1-5, ASK Q17; OTHERWISE GO TO QD1 CONTINUE/END}
Q17.

Let us now imagine that [SCREENER INSERT] lowers its prices. If other [COMPANIES/
SUPPLIERS] remain at the same prices, how much must [SCREENER INSERT] lower its price
before you would definitely choose a(n) [SCREENER INSERT] the next time you purchase a
[GENERIC NAME FOR SCREENER INSERT]?
Please provide your answer in percentages up to 25%
0 TO 25__________
26
101
102
103

26% or higher
Never would [PURCHASE/USE/SHOP AT] any other [SCREENER INSERT]
Don’t know
Refused

97

NOTE: FOR THE EXPANDED MODEL IN FIGURE 1A, SUBSTITUTE THIS INTRODUCTION AND
QUESTIONS 4P, 4S, 5P, 5S, 6P, 6S FOR QUESTIONS 4, 5, 6.
Up to this point I have asked you about your expectations prior to your recent experiences with [your]
[SCREENER INSERT]. Now I am going to ask you several questions about your ACTUAL
EXPERIENCES with [your] [SCREENER INSERT]. Some deal with your experience with the product itself.
Others are about your experience with service for that product.
Q4P.

First, please consider all your experiences in the last [INSERT TIME PERIOD FROM
SCREENER] with your [SCREENER INSERT]. Using a 10-point scale, on which “1” means “not
very high” and “10” means “very high,” how would you rate the overall quality of your
[SCREENER INSERT]?
1 TO 10 __________
11
12

Q4S.

Don’t know
Refused

Now please consider all your experiences in the last [INSERT TIME PERIOD FROM
SCREENER] with service for your [SCREENER INSERT]. Using a 10-point scale, on which “1”
means “not very high” and “10” means “very high,” how would you rate the overall quality of
service you have received for that [SCREENER INSERT]?
1 TO 10__________
11
12

Don’t know
Refused

(RANDOMIZE Q5 AND Q6 SERIES)
Q5P.

Now thinking about your personal requirements for a [SCREENER INSERT] such as [INSERT
PRODUCT ATTRIBUTES], please tell me how well your [SCREENER INSERT] has actually met
your requirements. Using a 10-point scale on which “1” now means “not very well” and “10”
means “very well,” how well has your [SCREENER INSERT] actually met your personal
requirements?
1 TO 10__________
11
12

Q5S.

Don’t know
Refused

Now thinking about your personal requirements for service for your [SCREENER INSERT], such
as [INSERT SERVICE ATTRIBUTES], please tell me how well service for your [SCREENER
INSERT] has actually met your personal requirements. Using a 10-point scale on which “1” now
means “not very well” and “10” means “very well,” how well has service for your [SCREENER
INSERT] actually met your personal requirements?
1 TO 10__________
11
12

Don’t know
Refused

98

Q6P.

Now please think about how often things go wrong with the [SCREENER INSERT], regarding
such things as [INSERT PRODUCT ATTRIBUTES]. Use a 10-point scale on which “1” now
means “very often,” and “10” means “not very often,” how often have things actually gone wrong
with your [SCREENER INSERT]?
1 TO 10__________
11
12

Q6S.

Don’t know
Refused

Now please think about how often things go wrong with the service for your [SCREENER
INSERT], regarding such things as [INSERT SERVICE ATTRIBUTES]. Using a 10-point scale
on which “1” now means “very often,” and “10” means “not very often,” how often have things
actually gone wrong with the service for your [SCREENER INSERT]?
1 TO 10__________
11
12

Don’t know
Refused

99

ACSI Questionnaire for
Government Services
(For input to model in Figure 2)
INTRODUCTION
Hello, I'm (NAME) calling on behalf of the University of Michigan. We are conducting research on how
satisfied users are with services provided by Federal government agencies and private companies as
part of the American Customer Satisfaction Index. You may have read something about the American
Customer Satisfaction Index in USA Today, The Wall Street Journal, or your local newspaper.
Today I want to ask you about services you may have received from the (AGENCY/PROGRAM). The
purpose of the research is to help this government agency improve its services to you and to people like
you. Your answers are voluntary, but your opinions are very important for this research. Your name will
be held completely confidential and never connected to your answers. This interview will take 8-10
minutes* and is authorized by Office of Management and Budget Control No. _________.
[ADD ANY NEEDED SCREENING QUESTIONS HERE]
Now, I am going to ask you some questions about the [AGENCY] [ACTIVITIES/SERVICES/PRODUCTS]
with which you have had experience.
Q1.

Before you used the [AGENCY/PROGRAM], you probably knew something about the [AGENCY/
PROGRAM]. Now think back and remember your expectations of the overall quality of the
[AGENCY/PROGRAM]. Please give me a rating on a 10-point scale on which “1” means your
expectations were “not very high” and “10” means your expectations were “very high.”
How would you rate your expectations of the overall quality of [AGENCY/PROGRAM]?
1 TO 10__________
11
12

Don’t know
Refused

Now, let’s think about _______
Q2.

[QUESTION RATING AN ASPECT OF ACTIVITY 1] Again, we will use a 10-point scale on
which…
1 TO 10 __________
11
12

*

Don’t know
Refused

Insert 15-17 minutes for RDD samples for which respondents can be interviewed about two agencies.

101

Q3.

[SECOND QUESTION RATING AN ASPECT OF ACTIVITY 1]
1 TO 10 __________
11
12

Don’t know
Refused

And next, considering ________
Q4.

[QUESTION RATING AN ASPECT OF ACTIVITY 2]
1 TO 10 __________
11
12

Q5.

Don’t know
Refused

[SECOND QUESTION RATING AN ASPECT OF ACTIVITY 2]
1 TO 10 __________
11
12

Don’t know
Refused

And thinking about_________
Q6.

[QUESTION RATING AN ASPECT OF ACTIVITY 3]
1 TO 10 __________
11
12

Q7.

Don’t know
Refused

[SECOND QUESTION RATING AN ASPECT OF ACTIVITY 3]
1 TO 10 __________
11
12

Don’t know
Refused

Q8.

[OPTIONAL QUESTION ABOUT SERVICES]

Q9.

[OPTIONAL QUESTION ABOUT SERVICES]

Q10.

Please consider all your experiences in the past two years with the [AGENCY/SERVICES/
PRODUCTS]. Using a 10-point scale, on which “1” means “not very high” and “10” means “very
high,” how would you rate the overall quality of the [AGENCY/SERVICES/PRODUCTS]?
1 TO 10__________
11
12

Don’t know
Refused

102

Satisfaction includes many things. Let’s move on and talk about your overall satisfaction with the
[AGENCY].
Q11.

First, please consider all your experiences to date with the [AGENCY/SERVICES/PRODUCTS].
Using a 10-point scale on which “1” means “very dissatisfied” and 10 means “very satisfied,” how
satisfied are you with the [AGENCY/SERVICES/PRODUCTS]?
1 TO 10__________
11
12

Q12.

Don’t know
Refused

Considering all of your expectations, to what extent has the [AGENCY/SERVICES/PRODUCTS]
fallen short of your expectations or exceeded your expectations? Using a 10-point scale on
which “1” now means “falls short of your expectations” and “10” means “exceeds your
expectations,” to what extent has the [AGENCY] fallen short of or exceeded your expectations?
1 TO 10__________
11
12

Q13.

Don’t know
Refused

Forget the [AGENCY] for a moment. Now, I want you to imagine an ideal [AGENCY/SERVICES/
PRODUCTS]. (PAUSE) How well do you think the [AGENCY] compares with that ideal
[AGENCY/SERVICES/PRODUCTS]? Please use a 10-point scale on which “1” means “not very
close to the ideal” and “10” means “very close to the ideal.”
1 TO 10__________
11
12

Don’t know
Refused

Next, I want you to think about any communication you may have had with the [AGENCY] regarding
complaints about your experience.
Q14.

Have you complained to the [AGENCY] in the past year?
1
2
3
4

Yes
No
Don’t know
Refused

{IF Q14 = 1, ASK Q14A – 14B; OTHERWISE GO TO Q15}
Q14A. How well, or poorly, was your most recent complaint handled? Using a 10-point scale on which
“1” means “handled very poorly” and “10” means “handled very well,” how would you rate the
handling of your complaint?
1 TO 10___________
11
12

Don’t know/not relevant/did not use
Refused

103

Q14B. How difficult or easy was it to make your most recent complaint? Using a 10-point scale on
which “1” means “very difficult” and “10” means “very easy,” how difficult or easy was it to make a
complaint?
1 TO 10___________
11
12
Q15.

Don’t know/not relevant/did not use
Refused

If asked, how willing would you be to say positive or good things about the job the [AGENCY] is
doing in [SPECIFIC SERVICE]? Using a 10-point scale on which “1” means “not at all willing”
and “10” means “very willing,” how willing would you be to say good things about the agency’s
handling of [SPECIFIC SERVICE]?
1 TO 10__________
11
12

Q16.

Don’t know
Refused

How confident are you that [AGENCY] will do a good job in the future of providing [SPECIFIC
SERVICE]? Using a 10-point scale on which “1” means “not at all confident” and “10” means
“very confident,” how confident are you that [AGENCY] will do a good job in the future providing
[SPECIFIC SERVICE]?
1 TO 10__________
11
12

Don’t know
Refused

104

ACSI Questionnaire
Demographics
Now, we need to ask a few demographic questions for the ACSI consumer profile...
QD1.

What is your age, please?
[RECORD NUMBER OF YEARS 18-84] _____

98 Don’t know
99 Refused
____________________________________________________________________________
QD2.

What is the highest level of formal education you completed? (READ CODES 1 - 5)
1 Less than high school
2 High school graduate
3 Some college or associate degree
4 College graduate
5 Post-Graduate
6 Don’t know
7 Refused

____________________________________________________________________________
QD3.

Are you of Hispanic, Latino or Spanish origin?

1 Yes
2 No
3 Don’t know
4 Refused
____________________________________________________________________________
QD4.

Do you consider your race(s) as:
(READ CODES 1 - 5; ACCEPT UP TO FIVE MENTIONS)

1 White
2 Black/African American
3 American Indian/Alaska Native
4 Asian
5 Native Hawaiian or other Pacific Islander
6 (DO NOT READ) Other race
7 Don’t know
8 Refused
______________________________________________________________________________

105

QD5.

What was your total annual family income in 2004? (READ CODES 1 - 7)

1 Under $20,000
2 $20,000 but less than $30,000
3 $30,000 but less than $40,000
4 $40,000 but less than $60,000
5 $60,000 but less than $80,000
6 $80,000 but less than $100,000
7 $100,000 or more
8 Don’t know
9 Refused
_______________________________________________________________________
[QD6] [RECORD GENDER BY OBSERVATION]
1 Male
2 Female
_______________________________________________________________________
[MOVE IN DMA FROM SAMPLE]: _____
________________________________________________________________________
[MOVE IN MSA CODE FROM SAMPLE]: _____
_________________________________________________________________________
[MOVE IN STATE FROM SAMPLE]: _____
_________________________________________________________________________
[MOVE IN MET STATUS CODE FROM SAMPLE]: _____
_________________________________________________________________________

106

APPENDIX E:
RESPONSE AND COOPERATION RATES

107

DISPOSITION OF SAMPLED TELEPHONE NUMBERS AND
RESPONSE AND COOPERATION RATES (2003 ACSI)
U

UNIVERSE OF SAMPLED TELEPHONE NUMBERS

592,415

Noneligible housing units
Disconnect/out of service
Business
Secondary line
Fax/modem
Number changed
Wrong phone number
Cellular phone
TOTAL NONELIGIBLE HOUSING UNITS

43,701
33,947
282
25,074
517
477
1,179
105,177

NE

Noneligible respondents
Noncustomer of any companies or government
services
Noneligible, age
Noneligible, non-English speaking or hard of hearing
Noneligible, other
TOTAL NONELIGIBLE RESPONDENTS

QF

Quota-filled, so respondent noneligible for
interview
Customer of quota-filled company
Scheduled for callback, but company quotas filled or
interview period ended
TOTAL QUOTA-FILLED

EU

ELIGIBLE UNIVERSE OF SAMPLED TELEPHONE
NUMBERS

109

8,496
5,038
11,879
642
26,055

4,487
55,580
60,067
401,116

I
P
I+P

Interviewed customers
Complete interview
Partial interview
TOTAL INTERVIEWS

R

Refusals
TOTAL QUALIFIED CUSTOMER REFUSAL

UE

Unknown eligibility/no contact with potential
household or customer—noninterview
Housing unit, unknown if eligible customer
respondent. Household refusal before screening for
eligible customer respondent
Unknown if a housing unit. No answer after repeated
calls (4), different times of day over multiple days
Answering machine for repeated calls (4) over multiday period
Busy or no answer, all calls (4)
TOTAL UNKNOWN ELIGIBILITY

36,952
199
37,151

2,067

143,996
160,059
35,927
22,016
361,998

Using the codes shown in the left column of the above table, survey cooperation and response
rates are calculated according to American Association of Opinion Research (AAPOR) reporting
standards.8
Based on completed (not partial) interviews, the cooperation rate for those screened and
identified as customers is:
COOPERATION RATE (AAPOR(2)) = (I)/(I+P)+R
COOP = 36,952/(37,151)+(2,067) = 94.2%
RESPONSE RATE = I/(I+P)+R+e(UE)
where e is the incidence of identified customers in all households able to be screened. The
numbers used in estimating this incidence are:
e = (I+P+R+QF)/(I+P+R+QF+NE)
e = (36,952+199+2,067+60,067)/(36,952+199+2,067+60,067+26,055) = 79.2%

8

American Association of Public Opinion Research (2004), Standard Definitions: Final Disposition of Case Codes and
Outcome Rates for Surveys.

110

However, knowing that the cooperation rate is 94.2%, rather than 100%, a response rate is now
calculated where both e and COOP rates are applied to the quota-filled cases and consider that
amount to be completed interviews. By applying the e and COOP rate, we have accounted for
the fact that some of the respondents would have refused or broken off the interview if we had
asked them to complete more of the interview. The response rate then is:
RESPONSE RATE (AAPOR RR(3)) = (I+COOP(QF))/(I+P+R+QF+NE+e(UE))
RR = 36,952+.94(60,067)/36,952+199+2,067+60,067+26,055+.79(361,998) = 22.7%

111

RESPONSE RATES FOR INTERNET INTERVIEWS (2004 ACSI)

E-Commerce Interview Results
20,553

Users who came to the site (9.8% of those sent e-mail invitations).

12,092

Respondents who were screened out because interviews for the
company from which they had purchased were already completed.

3,857

Qualified respondents.

3,680

Qualified respondents who completed the interview, including the
demographic questions (97%).

3,834

Completed company interviews (an average of 1.0 company
interviews per qualified respondent).

E-Business Interview Results
4,795

Users who came to the site (13.3% of those sent e-mail invitations).

2,147

Respondents who were screened out because interviews for the
company from which they had purchased were already completed.

2,596

Qualified respondents.

2,470

Qualified respondents who completed the interview, including the
demographic questions (95%).

3,514

Completed company interviews (an average of 1.4 company
interviews per qualified respondent).

112

APPENDIX F:
SAMPLE PROFILE

113

2003 ACSI CUSTOMER SAMPLE (AGE 18-84)
Demographic Characteristic

2000 Census
(%)

2003 ACSI
Customers
(%)

31
41
27

24
43
32

22
13
12
19
14
8
12
—

(2002)
12
11
12
18
14
11
14
10

Race
White
Black/African American
American Indian
Asian
Other

77
13
2
4
7

84
8
2
2
5

Ethnicity
% Hispanic/Latino

13

5

Gender
Male
Female

49
51

38
62

Educational attainment
Less than high school graduate
High school graduate
Some college
College graduate
Postgraduate

20
29
27
15
9

5
24
32
24
16

Age
18-34
35-54
55-84
Income
Under $20,000
$20,000-30,000
$30,000-40,000
$40,000-60,000
$60,000-80,000
$80,000-100,000
$100,000 or more
Refused/don’t know

115