Methodological Appendix to Chapter 3
We created the Reputation Institute in 1999 to study corporate reputations. The annual consumer polls that we’ve sponsored since then have been conducted with our market research partner Harris Interactive and other affiliated companies. Harris relies on a 7 million person panel of voluntary participants to identify and poll representative samples of consumers in the United States. The opinions of those sampled are weighted to be representative of what the general public thinks in each country. In Europe, we relied on telephone sampling to carry out the studies.
Projects unfolded in two stages in each country: a nominations phase and a rating phase. In the nominations phase, we began by asking two basic questions:
Of all the companies that you’re familiar with or that you might have heard about, which TWO—in your opinion—stand out as having the BEST reputation overall?
Of all the companies that you’re familiar with or that you might have heard about, which TWO—in your opinion—stand out as having the WORST reputation overall?
Polls of the general public were conducted in 12 countries and three regions of the world between fall 2000 and spring 2001. Over 20,000 people participated in these initial polls. From their nominations, we extracted a list of the companies that were “most visible” to the public in each country—for better or worse. These companies are the ones whose reputations were measured in detail in the second phase of research—the rating phase.
Detailed studies sponsored by Harris Interactive and the Reputation Institute using the RQ have been featured annually in the Wall Street Journal since 1999. Here we discuss the results of parallel studies completed in 2001 in the United States, Australia, Denmark, Italy, and the Netherlands. The projects involved close collaboration with the market research firms of Harris Interactive, Blauw Research, and AMR Interactive, as well as close working relationships with the Corporate Communications Center at Erasmus University, with SDA-Bocconi in Italy, and with Copenhagen Business School and Interplay in Denmark.
As interesting as they are, nominations are not accurate depictions of the reputations of companies. To fully examine corporate reputations requires more detailed measurement—hence the launch in 2000 of a worldwide project to measure corporate reputations accurately—we call it the rating phase. To carry out the rating phase, we needed a reliable instrument to measure the reputations of the nominated companies. The instrument we used to rate those companies was the Reputation Quotient (the RQ), a measure developed by Charles Fombrun and Harris Interactive that was designed to overcome weaknesses in existing instruments and that we have proposed as a useful benchmark for reputation measurement.
More than 30,000 people participated in the rating phase across the five countries. We relied on these RQ interviews to obtain accurate ratings of the most visible companies in each country. The ratings enabled us to calculate reputation scores for these companies and set the stage for us to explore the underlying reasons why they are thought to be tops by the public.
Sampling Process in the United States: A total of 10,038 U.S. respondents were polled in seven separate polls conducted between April and August 2001: Of these, 5,975 were done online and 4,063 by telephone. An identical nomination phase was conducted during summer 2000 with 5,661 respondents, of which 4,651 were interviewed online and 1,010 by telephone. The online respondents were randomly selected from a large online panel that Harris Interactive created to carry out online research. At the time, the Harris online panel included over 7 million voluntary participants. To ensure that the study was not biased to online respondents, a separate set of telephone interviews was conducted with a representative sample of the general public. The nominations for best and worst corporate reputation were tallied and summed, and the 60 most visible corporate reputations in the United States were identified.
A total of 21,630 consumers then rated the reputations of the 60 most visible companies using the RQ instrument. All interviewing was conducted online in October 2001, and all companies were rated by respondents who indicated that they were at least somewhat familiar with that company. Respondents were asked to rate up to two randomly chosen companies with which they were familiar. Respondents who were familiar with more than two companies were randomly assigned to rate two companies. After the first rating was completed, they were given the option to rate their second company, and interviews lasted an average of 22 minutes.
Sampling Process in Europe: Sampling was done on a country-by-country basis. Various research firms were used to obtain phone nominations from representative samples of 750 to 1,000 people in each country in fall 2000. Commercial brands, purely financial holdings, and subsidiaries were excluded. The nominations enabled creating lists of the 20 to 30 companies with the most visible corporate reputations in each country. RQ interviews in each country were conducted by telephone, between February and April 2002, and companies were rated by respondents who had indicated that they were “very familiar” with, “somewhat familiar” with, or “had heard the name of” that company.
Sampling Process in Australia: A similar process was used to sample the general public in Australia for both the nominations phase and the rating phase. The Australian sample was national in scope, and stratified to represent the population of the States and Territories, and then split between urban and rural populations. Data collection was conducted by AMR Interactive, based in Sydney, Australia.
The comparable sampling and rating methodologies used in the United States, Europe, and Australia enable us to compare the relative RQ scores given to companies across the regions.