Chapter 1 Honolulu Survey 2000 and Its Objectives|
1.1 Backdrop of Honolulu Survey
In the tradition of Socrates, a young Chikio Hayashi at the Institute of Statistical Mathematics, wondered who Japanese are as Japan regained its independence status following the signing of Peace Treaty in San Francisco in 1951. He gathered data on the characterization of what constituted Japanese people「日本国民性の研究」[A Study of Japanese National Character] long before the advent of 日本人論」[Theory of Japanese People] in the 1970s. From this body of the literature, he developed a questionnaire designed to the tap the essence of Japanese culture. He used a random sampling of 3,000 or more Japanese adults to conducted a nation-wide survey every five years since 1953.
As the decade of 1970s began, Dr. Hayashi and his colleagues wondered again what happens to Japanese values or Japanese-ness once the Japanese emigrate abroad. Japan by that time acquired enough trade surplus generated as a result of the Vietnam War among other events to allow scholars to conduct research abroad. In 1971, Yasumasa Kuroda at the University of Hawaii joined Hayashi's team to begin the study of Japanese Americans in Honolulu. We conducted our first random sample survey of Japanese Americans living in urban Honolulu based upon the list of registered voters (Suzuki, 1972, The Research Committee on the Study of Honolulu Residents, 1980, 1984, 1986, 1990). Since then, we realized that Japanese Americans do not live alone in the island. They live among other ethnics in Honolulu where there is no ethnic group that constitutes a majority. The Japanese Americans constitute a plurality among ethnic groups in Hawaii as a whole and in Honolulu as well. They continue to account for more than one-third of the voters in Honolulu followed by Caucasians with about one-third of the population. The courageous and wise young generation of the 1940s and 1950s largely under the leadership of Hawaii's Democratic Party especially after 1954 brought about the beginning to the end of segregated housing along other forms of ethnic/racial discrimination in Hawaii. Although certain forms of discrimination continue to exist, much of discriminatory practices were eliminated by the 1980s. Japanese Americans experienced difficulties in finding their homes in desirable places as late as the 1970s.
Chikio Hayashi developed a chain-link approach or theoretical framework to study various countries. He calls it 「連鎖的国際比較調査方法」. His approach, for example, is that Japanese Americans maintain some values they share with Japanese while they also have American values the Japanese do not have and values they have in common with other local ethnic groups that make the rest of Hawaii's population. Japanese Americans constitute a link to Americans as a whole who in turns relates to Europeans.
His chain-link continued to expand its territories covered beyond Hawaii. We conducted Hawaii surveys in 1972, 1978, 1983, 1988 and 2000. He and his teams conducted surveys in U.S. mainland (1988), France (1987), Italy (1987), Germany (1987) and England (1987) as well as Japanese Brazilians. They also collected some data in China, the Philippines and other parts of the world as well. In earlier surveys in Hawaii, we used to have a Japanese version of the questionnaire administered among some Japanese Americans who preferred Japanese to English. We discontinued this option after a couple of surveys, but Hayashi observed that the traditional translation of the questionnaire did not seem to elicit equivalent responses from our respondents. In other words, the questionnaire may be translated in a valid and reliable fashion, but that did not constitute a sufficient condition to assume that responses we receive from respondents will be functionally equivalent and reliable. His team initiated bilingual surveys to measure the validity and reliability of translation among Tsukuba University students and foreigners staying in Tokyo.
The Theory of Rashomonesque Yamazakura 羅生門・山桜論
Subsequently, Yasumasa Kuroda of the University of Hawaii and Tatsuz? Suzuki of the Institute of Statistical Mathematics in Tokyo received a couple of grants from the Toyota Foundation to conducted bilingual surveys of Arabic, English and Japanese in the United States, Egypt and Jordan. He and his team conducted a follow-up study of Arabic and English languages through bilingual focused-interviews in Amman and Cairo. The most important finding from these studies is that whenever there is a neutral or middle-response category in a questionnaire Japanese language users are most likely to select the middle-response category while Arabic language users are least likely to do so. English speakers are found between the two extremes but definitely closer to Arabic than to Japanese. Nationality played less significant roles than language in use. Language structures our observational sense as well as our responses at least as far as this dimension is concerned. Arabic and American language speakers feel no resistance in choosing between two alternative such as yes and no. Japanese language speakers' proclivity is not to choose but to adjust to a particular situation, which minimizes conflicts with others. From the cross-language surveys and others, Hayashi and Kuroda (1997) developed a theory of Rashomonesque Yamazakura consisting of two key elements in Japanese culture; diffuse concept of self and multiple (relativistic) view of the reality. Their theory assumes that language and religion constitutes the basic foundation of a culture, shaping the way we respond to a situation. They found their longitudinal survey data as well as qualitative historical data such as poems from the 7th century on support the theory of Rashomonesque Yamazakura.
Self, Others, Interpersonal Relationship, Organization and Adjustment
In 1988, S. Frank Miyamoto of the University of Washington, Stephen Fugita of Santa Clara University of California and Tetsuden Kashima of the University of Washington joined Hayashi's team to study Japanese Americans in the West Coast. Miyamoto proposed. They brought their theoretical concerns to be incorporated into their surveys in the West Coast. Miyamoto's longstanding theoretical focus originating in the 1930s of strong organizational and interpersonal relationship in predominantly the Issei-community was one. In addition, his more recent work on the Nisei's interpersonal style characterized by sensitivity to others in the tradition of George Herbert Mead was incorporated to our questionnaire in Hawaii. Kashima brought (Kashima 1977) his concern for how Buddhists modified their organizations to meet the needs of Japanese American community in the West Coast. Fugita too is interested in Japanese American organizations that continue to thrive despite their successful assimilation into the mainstream of American life. Their largely sociological and social psychological concerns were added to the growing list of items to be included in our survey of 2000.
We wish to note that the theory of Rashomonesque Yamazakura focused on the concept of self while the West Coast concerns added another dimension that focuses more on self and others. There are similarities in our theoretical concerns. The Japanese self characterized by its diffuse features means that Japanese are sensitive to others and others' views of them. Hence, group becomes most salient unit in Japanese society. Kashima's findings of changing nature of Buddhist organizations and Fugita's discovery of the success of Japanese Americans' assimilation in America all also represent what we say is an integral part of being a Japanese who tends to face a situation with an adjustment in mind, not a choosing between alternatives.
We attributed the nature of Japanese culture to language and religion, while the three West Coast sociologists seem to imply that Japanese Americans' tendency to maintain persistent and strong ethnic organizations are derived from their immigrant backgrounds in Japan. In a way, what they propose is that even if the Japanese language is lost among younger generations of Japanese Americans, they will remain Japanese Americans and that this trend may be reinforced by their ethnic organizational membership.
It is within these conceptual and theoretical concerns within which the original question of what makes up a Japanese mind or culture is continued to be studied through our longitudinal and comparative surveys in a chain-link fashion. Our surveys have covered all three continents since its inception in 1953 to 2000. We continued to expand our search for the original question of what is meant to be Japanese to include how they are similar to and different from their emigrant siblings and their posterities to others in American and African continents and in Europe. This survey report may mark the beginning of an end of our search for our questions as far as our generation of researchers are concerned.
What constitutes the nature of Japanese American culture, so-called the "local culture," and multi-ethnic culture of Hawaii at large? How are all these groups different from and similar to others we have surveyed in the past? To what extent and in what direction are they changing when compared longitudinally with our past data? How are Honolulu's Japanese Americans similar to and different from their counterparts in the West Coast? How are they different from the Japanese in Japan? Are Honolulu's Japanese Americans closer to the Japanese in Japan than those of the West Coast? It is these questions and other related-questions that we have attempted to answer in this report.
1.3 Research Site and the Universe
The city and county of Honolulu covers the entire island of Oahu, most populated island of all Hawaiian islands. Oahu's population accounts for three-fourths of the total State population of 1.1or 2(?) million people. Approximately one-half of the population lives in the area of Honolulu we designated as the area for the universe of our inquiry, Kokohead in the east to Diamond Head side of Middle Street in the west. It is most urbanized part of the island. We find East Honolulu to be largely represented by well-to-do families of Caucasian, Chinese and Japanese origins and by less-well-to-do indigenous Hawaiians, Filipinos and other more recent immigrants and their posterities in the west. The universe includes registered voters of both very rich and poor in addition to the middle class, but does not include rural voters.
Secured condominiums were rare when we first started our surveys in 1972. We observed the advent of gated-communities staffed by security guard in the 1980s. The number of them has increased with time. In the 1990s, we saw the rise of the use of cellular phones, caller identification, unlisted telephone subscribers and fast-food restaurants. Young people no longer eat dinner at home, making it very difficult to get hold of them to secure an interviewing. The recession in the 1990s in Hawaii proved to be severe blow to Hawaii economy, resulting in the drastic cut in the State spending and the rise of unemployment while the mainland enjoyed its unprecedented prosperity. For example, the State cut the University of Hawaii's budget by one-third of the budget in the early 1990s. Many moved to the mainland to seek a better economic opportunity. It was not until the summer of 2000 when signs of economy recovery started to appear. What do these mean to the 2000 survey?
We first used the list of registered voters to define the universe of our survey and continued to do so this time as well. One problem has been that we used Representative District to define the population at the outset. Unfortunately, boundaries of the Districts do not remain the same over the years due to periodic reapportionments. We wanted to maintain the same area to be covered in our population, resulting in defining the population to be all registered voters who reside in areas from Koko Head in the east and Diamond or East side of Middle Street in the west.
We purchased a CD containing a list of registered voters of the State of Hawaii (N=431,428). We sent the copy to Fumi Hayashi of our research team at the Toyo Eiwa Jyoshi Gakuin University. Hayashi was in charge of sampling and sent us back a list of sample respondents on a diskette. She selected 2,101 sample respondents out of 178,428 registered voters in the universe (Representative District 15-30) by using systematic random sampling.
We also developed another sample of about 3,000 respondents for possible uses if we could not get about 500 respondents out of the original sample. We ended up using this sample list as well as we succeeded in getting only 449 valid responds out of 2,101 possible sample respondents. In consultation with Tokyo, we decided to select only voters with recognizable Japanese names. We recognized and selected 215 Japanese names out of the first 692 names that appeared in the list. We secured 51 interviews out of 215 possible sample respondents. In the survey of Japanese Americans, Yasumasa Kuroda conducted for the US Department of Health and Welfare in 1985, he found approximately 16 percent of the Japanese Americans had names that are not recognizable to be Japanese origin in Honolulu. It is safe to assume we missed nearly 20 percent of the Japanese Americans in the process. We should also report that we found five who turned out to be non-Japanese among 51 respondents we succeeded in interviewing. We discovered one Filipino, one Hawaiian and three Caucasians. The last to be interviewed was a Caucasian woman physician married to a Japanese who kept postponing her interview due to emergencies for several times before our interviewer secured an interview with this busy physician. We decided to over-sample Japanese Americans in order for us to have a sufficient number of Japanese Americans with whom to compare Japanese Americans on the West Coast.
We followed the same method of conducing the survey as we have done in the past.
1. Letter of Introduction: Yasumasa Kuroda using his University stationary wrote a letter of introduction and request for an interview to respondents. A copy of the letter is attached in the Appendix 1-A. The contents of his letter varied somewhat from the beginning to the end with little variation. It essentially introduces what the survey is all about and how important it is for us to seek the cooperation of all potential respondents. Kuroda left his office telephone number and e-mail address for any response respondents may wish to make. He did receive calls as well as e-mail messages, largely to ask that their names be deleted. But there are some who kindly let us him know that they moved or give telephone number to call etc. One busy physician sent him a message to keep his interviewer try again to contact the physician's office. This is the first survey where e-mail messages are exchanged with potential respondents. He mailed a limited number of letters once a week or so to make sure that not too many days elapses between the time respondents receive their letter and the time of our interviewers to contact them.
2. Interviewers Hired and Trained: We hired and trained more than two-dozen interviewers but only about a half of them turned out to our interviewers in the sense that they produced any results. Most of those who worked for at least a few months did well. We verified each survey completed through either calling or checking on respondent's year of birth. We had information on the date of birth for most of the respondents from the list of registered voters. One of them returned only one completed survey, which we had to discard. Another one did only three completed interviews. Most of them quit before they start. An instruction for interviewers is attached in the Appendix 1-B.
3. Interviewing: We instructed our interviewers to practice interviewing with their friends or family members until they felt they are comfortable. Whenever possible, they made phone calls in advance to make an appointment. Valid phone numbers are available only a few times out of ten potential respondents. We instructed interviewers to make at least three attempts to locate their respondents. We told them to make three attempts at different times and different day of the week preferably in late afternoon and early evening and weekends. After two attempts, we instructed them to leave a message to call them back. And if they think they can get cooperation of any respondent who was not home three times. We encourage them to visit any number of times until they can get the respondent.
4. Verification: Our interviewers turned-in their completed questionnaire once a week. We then verified through either telephoning or checking the year of birth between what we find in the completed survey and the data contained in the voter list. We had a few cases of interviewing wrong people. One case involved father and son who had the same name but different middle initials. We are happy to report that the result of verification produced no falsified surveys at all. We did not just verify a sample of the respondents but we verified every single survey completed.
5. Success Rate for Completed Surveys:
There are several reasons for the low rate of success. First, a recent court ruling changed the interpretation of election law, making it impossible to delete names from the list of registered voters for those who did not vote in two consecutive time as done in the past. Now, a voter must fail to vote over two election years, which could amounts to four years. One father wrote me a letter stating that his daughter died several years ago and he reported her death to the City Clerks office. Despite his request to remove his daughter's name from the list, he keeps getting election notices from the State.
Second, Hawaii's economy suffered from a lengthy recession in the 1990s, prompting a number of residents to leave the island for a better economic opportunity in the mainland.
Third, an increased number of gated communities and secured condominium hampered our interviewers from making the initial contact.
Fourth, a decreasing number of telephone numbers are published and more people use cellular telephones than ever before.
Fifth, the largest murderer case in Hawaii involving a Japanese American employee at the local Xerox and other violent crimes occurring during the duration of survey did not help us at all in securing cooperation of some respondents. People are cautious in opening the door to strangers when our interviewers call on them. One of our female interviewers told me that an elderly Japanese American lady in Kaimuki pleaded our young and attractive Caucasian interviewer to stop being an interviewer. She told her it is too dangerous for young and pretty women like her to be going from house to house.
We found we succeeded in securing 449 interviews while 279 potential respondents refused to take part in the survey after being contacted. We secured 52 interviews out of the second sample consisting of 215 potential respondents. What follows is a breakdown of the results of our efforts to interview 2,101 potential respondents. Figures appearing following the first set of figures represent the result of the second sampling of Japanese Americans.
・ Pau (Completed): 449 (21%), 51 (24%)
・ Refused: 279 (13%), 34 (16%)
・ Moved: 550 (26%), 25 (12%)
・ Not home: 425 (20%), 62 (29%)
・ Not home/Left Message: 155 (7%), 14 (6%)
・ Contact later: 32 (2%), 9 (4%)
・ No English: 31 (2%), 1 (0%)
・ Ill/in hospital: 24 (1%), 5 (2%)
・ Physical: 7 (0%), 4 (2%)
・ Deceased: 19 (1%) 0 (0%)
・ No access: 68 (3%), 9 (4%)
・ Can't locate: 44 (2%), 1 (0%)
・ Out-of-town: 14 (1%), 1 (0%)
・ West of Middle Street: 4 (0%), 0 (0%)
・ Total N = 2101, N = 215
These figures indicate that Japanese Americans move less often and constitute more stable than other ethnics. They tend to stay home less often when our interviewers attempt to make contact. However, their rate of cooperation as well as refusal is not significantly different from that of the public at large.
We added an extra set of questions for Japanese Americans to respond as evidenced in the Appendix 1-C.
1.6 The Questionnaire
It is from theoretical concerns described in Chapter 1 that went into the construction of our questionnaire for Hawaii survey 2000. In addition to Japanese national character study items we took from the following sources: CREDOC [Centre de Rescherche pour l'etude et L'observation des Conditions de vie] in Paris, NORC [National Opinion Research Center] distributed by the Roper Public Opinion Research Center, items we developed for the Honolulu survey and items generated by Frank Miyamoto, Tetsuden Kashima and Stephen Fugita for the West Coast survey. A copy of the questionnaire is attached in the Appendix 1-D.
We exchanged e-mail messages, faxes and telephone calls to develop our questionnaire in the summer of 1999 with our team members on the West Coast and Tokyo. As with the case in the past surveys, we always have more questions we wish to include than we have space for them. We had to limit the length of our questionnaire to a reasonable length in order to maintain the validity of responses we receive from our respondents. It takes about 35 to 40 minutes to complete the questionnaire for most people. There were cases where it took close to three hours to complete the survey.
1.7 The Coding and Data Cleaning
We made a few decisions on coding.
First, we decided to take the first of two responses whenever a respondent gave two answers to a question. We have instructed our interviewers to ask for one answer for each question unless otherwise so instructed (e.g., Q45). Despite our efforts to the contrary, some respondents insisted on giving us two answers and some interviewers allowed two responses. However, we found no more than a dozen of such nature.
Second, we instructed our interviewers to simply record the answer to the question of one's main occupation (Q72), rather than for them to interpret and categorize respondents into one of the given categories. Despite our efforts, some interviewers kept classifying respondents into categories without specifying the nature of the main occupation. We have decided to change some of my interviewers' classifications whenever we felt the classification was inappropriate. For example, a high school graduate is unlikely to be a professional (attorney, architect, scientist, engineer, professor, teacher, etc.) even when our interviewer classified him to be such. we reclassified him to the position of clerk instead of professional. Consequently, there are some discrepancies between what is in some of the completed questionnaire and the data. These discrepancies are intentional rather than errors.
Third, a question occurred to me as we started to see one organization in questions 80-92 classified as one organization type such the United Japanese Society being classified as a cultural organization and sometimes that of a community. We thought of unifying the classification system, however, We found that it is an impossible task since there are organizations with which we are not familiar. We decided to allow respondents to decide and interviewers to record whatever respondents said.
We completed the initial coding and data entry in Honolulu. In August of 2000 Kuroda sent the data as an attachment to his e-mail message to Fumi Hayashi. Ry?z? Yoshino and his research team members at the Institute took the completed questionnaire to Tokyo with them on their way home from the Honolulu Conference held at the end of August of 2000.
CREDOC [Centre de Rescherche pour l'etude et L'observation des Conditions de vie], Enquete, 1979-1980: Sur Le Situations et Perceptions Relatives aux Conditions de vie et a la Qualite de la vie des Francais. Paris: CREDOC, (Juillet, 1980).
Fugita, Stephen S. and David J. O'Brian. Japanese American Ethnicity: The Persistence of Community. Seattle: University of Washington Press, 1991.
Hayashi, Chikio, Fumi Hayashi, Yasumasa Kuroda, Tatsuzo Suzuki, and L. Lebart. "Comparative Study of Quality of Life and Multidimensional Data Analysis: Japan, France and Hawaii." Fourth International Symposium: Data Analysis and Informatics, Versailles, France, 1985, 573-585.
Hayashi, Chikio and Yasumasa Kuroda. Japanese Culture in Comparative Perspective. New York: Praeger, 1997.
Kashima, Tetsuden. Buddhism in America: The Social Organization of an Ethnic Religious Institution. Englewood Cliffs, NJ: Greenwood Press, 1977.
Kuroda, Yasumasa. "Public Opinion and Cultural Values" in Michael Haas ed. Multicultural Hawai'i. New York: Garland Publishing, Inc., (1998), 131-146.
Kuroda, Yasumasa, Chikio Hayashi, and Tatsuzo Suzuki. "A Cross-National Analysis of the Japanese Character among Japanese Americans in Honolulu," Ethnicity. Vol. 5 (March 1978), 41-59.
Kuroda, Yasumasa, Chikio Hayashi and Tatsuzo Suzuki. "The Role of Language in Cross-National Surveys: American and Japanese Respondents." Applied Stochastic Model and Data Analysis. Vol. 2, No. 2 (1986), 43-59.
Kuroda, Yasumasa and Tatsuzo Suzuki. "Language and Attitudes: A Study in Arabic, English, and Japanese on the Role of Language in Cross-cultural Thinking." in Donald M. Topping, Doris C. Crowell, and Victor N. Kobayashi. eds. Thinking Across Cultures. Hillsdale, New Jersey: Lawrence Erlbaum Associates, Publishers, (1989), 147-161.
Kuroda, Yasumasa and Tatsuzo Suzuki. "A Comparative Attitudinal Analysis of Rationalism: Arab, American and Japanese Students." in Urbanism in Islam: The Proceedings of the International Conference on Urbanism in Islam. Vol. 3 (1989), 65-96.
Kuroda, Yasumasa and Tatsuzo Suzuki. "Arab Students and English: The Role of Implicit Cultures." Behaviormetrika. No.29, (1991), 23-44.
Kuroda, Yasumasa and Tatsuzo Suzuki. "A Comparative Analysis of the Arab Culture: Arabic, English and Japanese Languages and Values." Behaviormetrika, No. 30, (1991), 35-53.
Kuroda, Yasumasa and Tatsuzo Suzuki. "Tahalil Muqarin Thaqafa Al-Arabiye: Al-Lughat Wa Al-Queem Al-Arabiye Wa Al-Ankelizidyeh Wa Al-Yabaniye" Al Mustaqbal Al Arabi (The Arab Future), No. 163 (September 1992), 14-31.
Miyamoto, S. Frank. Social Solidarity Among the Japanese in Seattle. Seattle: University of Washington Press, 1981.
NORC [National Research Center]. National Data Program for the Social Sciences: Cumulative Codebook for the 1927-1977 General Survey. Distributed by Roper Public Opinion Research Center, Yale University, New Haven, Connecticut, 1977.
Suzuki, Tatsuzo et al., "A Study of Japanese-Americans in Honolulu, Hawaii" Annals of the Institute of Statistical Mathematics, Supplement 7, 1972, 60 pp.
The Research Committee on the Study of Honolulu Residents. Honolulu Residents and Their Attitudes in Multi-Ethnic Perspective: Toward a Theory of the American National Character. Monograph 1, Tokyo: The Institute of Statistical Mathematics, 1980. Distributed by the University of Hawaii Press.
The Research Committee on the Study of Honolulu Residents. Japanese Americans in Comparative Perspective. Monograph Series, Monograph 2, Tokyo: The Institute of Statistical Mathematics, 1984. Distributed by the University of Hawaii Press.
The Research Committee on the Study of Honolulu Residents. The Third Attitudinal Survey of Honolulu Residents, 1983. Monograph 3, Tokyo: The Institute of Statistical Mathematics, 1986. Distributed by the University of Hawaii Press.
The Research Committee on the Study of Honolulu Residents. The Fourth Attitudinal Survey of Honolulu Residents, 1988.Monograph 4, Tokyo: The Institute of Statistical Mathematics, 1990. Distributed by the University of Hawaii Press.
1 One should not confuse the census population figures with those of the registered voters since former include any resident who happen to be living in the island at the time of census regardless of citizenship or voter registration. Because of a relatively large number of military personnel and their family members in the island and the fact that Hawaii has one of the highest income tax rates, many residents are not registered vote in Hawaii.
2 Japanese American faculty members including Yasumasa Kuroda along with other locals were found to be about 8 percent below the average salary of Caucasian colleagues of comparable positions in 1995. Women faculty turned to be below 3 percent of men of comparable positions. The University administration made efforts to adjust this inequity but not to the satisfaction of all parties in view of serious financial constraints suffered by the State of Hawaii.
3 The traditional method is to translated the original questionnaire into a second language and have someone other than the one who translated the original questionnaire translates the second language version back into the original language. If the two versions of the questionnaire in the original language more or less are the same, we considered the translation valid.
4 Yasumasa Kuroda's reasoning for including Arabic was that we are concerned with written language. Ancient Greeks from whom stem modern world civilization including English language obtained their written language and the concept of money from Syro-palestinians. Kuroda felt that he could do more meaningful comparison of the data if he had three different languages than two.
5 On this finding and other related findings, see Hayashi and Kuroda (1987) and the references therein.
6 Rashomon in this instance is derived from Akira Kurosawa's movie that won the first Japanese movie to win Cannu Movie Festival award in 1951. Kurosawa's theme is that there may be no single absolute truth or reality, but rather the world is relativistic in the Sophist tradition of "Man is the measure of all things."
7 We found the need for verification necessary after discovering some falsified surveys. It takes a lot of time and efforts to verify by phone and mail. Fortunately, when we use the list of registered voters, we can also obtain their date of birth, which helps us verify. A limited number of phone calls to check on each interviewer are necessary to find out how well she or he performs as an interviewer. In so doing, not just whether or not she or he went to interview but also how long she or he took complete an interview. We have had cases in the past where an interviewer asked only a few questions and filled out the rest of questionnaire on his own. This is similar to what the Japanese calls "kiseru."