1.1

Research on working conditions and industrial relations within the Japanese automobile industry has hitherto usually been met with little understanding by managers and union officers. In a recent paper, Totsuka (1995) identifies, in addition to the 'front door' approach, four different research approaches which have subsequently evolved because of the need for circumvention when wanting to gather data on working conditions and industrial relations in Japan:

1. the 'front door' approach consisting in interviews with management (and union officers, in cases where there is a union), supplemented with tours of the factories and collection of documents intended for in-company use;
2. interviews with workers and supervisors, either in their homes or at some neutral ground,
3. research based on collaboration with union activists;
4. participant observation;
5. research based on cases brought to some kind of legal action (cf. Totsuka, 1995: pp. 109 - 110).

1.2

A review of some of the research projects conducted thus far on the Japanese-invested plants in North America reveals that these may be grouped into categories very similar to the types identified by Totsuka, or as combinations thereof (see review below). Although the history of research on the Japanese-invested plants in North America is of a relatively recent date there is thus evidence of an imaginative diversification of methods similar to research methodologies in use in Japan. The material presented in is paper is from a research which in fact utilised a combination of the first two approaches. This combination resulted in certain dilemmas related to issues of access, i.e. the relationship with the organisation's 'gatekeepers' (Argyris, 1969; Klein, 1976; Punch, 1994; Hammersley and Atkinson, 1995), and more generally to issues of research ethics when researching 'sensitive topics' (Lee, 1993; Renzetti and Lee, 1993). Overall the paper may be seen as an example of researching 'upwards' (Lee, 1993: pp. 8, 12), since the paper discusses the relationship between the researcher and some of the most powerful strata within a large corporation in spite of the fact that the field research itself focused on the day-to-day work within manufacturing operations.

1.3

The problems discussed within the paper should be of particular interest to researchers and students of Japanese factories in North America. It is, however, hoped for that the account may be of interest also to the general community of sociologists both because it is being based on a concrete case and because it may touch upon ethical issues of general relevance. In other words, although many 'social science textbooks on methodology usually provide an idealised conceptualisation of how social research ought to be designed and executed' (Shaffir and Stebbins, 1991, as cited in Waddington, 1994: p. 107), this account differs from these by being based on experience. Moreover, problems similar to the ones described in this paper might obviously be present also in connection with research within other types of organisations and other settings, and the paper might thus contribute as material for use in comparative studies regarding these issues.

1.4

The paper is divided into five sections. The first section explains the relevant terms from the literature on methodology and subsequently positions the case study in relation to other studies of Japanese-invested factories. Also discussed are some conceivable merits and demerits of the two first research strategies in terms of access and research ethics in connection with research on sensitive topics. The two subsequent sections contain the case study itself: firstly a description of the method used for finding and contacting informants without relying on company introduction, and secondly reflections on conduct in the field and relationship with the corporate 'gatekeepers'. The final section sums up the case in relation to issues of access, gatekeepers relationships and research ethics as these issues have been discussed in the social science research methodology literature, and it also contains some reflections regarding how research projects with constraints similar to this project could be organised in a better way in the future.

Issues of Access, Gatekeeper Relationships and Research Ethics within Sensitive Research on Japanese-Invested Factories

2.1

Issues of access have been discussed widely in the social science research methodology literature. The issue of access in return for certain adjustments of the research project itself is called 'conditional access' in Lee's terms (Lee, 1993). There are, according to Lee, three major kinds of conditional access: '(a) restrictions on the methodology used by the researcher; (b) the completion of a piece for the gatekeeper in return for access; and (c) the right of the gatekeeper to examine, modify or censor published material arising from the study' (1993: p. 125). We will see that the first kind of conditional access has been especially relevant in connection with the research which is the basis of this paper, while the third kind has been especially relevant in connection with the new field of research on Japanese-invested factories in general.

2.2

Access is often the main item for negotiation between the researcher and the 'gatekeeper(s)' of a field. Gatekeepers are in this context managers who are in a position which makes them interested in and able to influence or control research:

Whether or not they grant entry to the setting, gatekeepers will generally, and understandably, be concerned as to the picture of the organization or community that the ethnographer will paint, and they will have practical interests in seeing themselves and their colleagues presented in a favorable light. At least, they will wish to safeguard what they perceive as their legitimate interests. Gatekeepers may therefore attempt to exercise some degree of surveillance and control, either by blocking off certain lines of inquiry, or by shepherding the fieldworker in one direction or the other. (Hammersley and Atkinson, 1995: p. 66)

2.3

The relationship with the gatekeeper is seldom described in detail within the preceding literature on Japanese-invested factories, and neither is it often described in detail in monographs or articles from the field. This relationship is one of the main themes of this paper, and one of the conclusions is that the in this case somewhat problematic relationship sheds light also on the substance of the research project itself. Research topics and practical issues which might appear trivial at a first glance (such as interviews about the employment relations at large and the issue of whether it should be allowed for a researcher to contact an employee directly) were in this case sensitive topics, and the term 'sensitive' is thus in the case of this paper a wide term: i.e. not as a term pointing only to, for example, corporate secrets or issues pertaining to the personal or emotional lives of individuals, but pointing to all issues which may lead to sanctions or other consequences. The definition of 'sensitive research' in Lee's book is thus highly relevant, in other words 'research which potentially poses a substantial threat to those who are or have been involved in it' (Lee, 1993: p. 4).

2.4

The five approaches identified by Totsuka (1995)were, as we remember: (1) the 'front door' approach; (2) off-premises interviews according to qualitative methods; (3) collaboration with union activists; (4) participant observation; and (5) research based on cases brought to some kind of legal action (cf. Totsuka, 1995: pp. 109 - 110), and have been in use also in the research that has been conducted thus far on Japanese-invested factories. The early surveys by Shimada (1988), Suzuki (1991) and by Abo (1994a and b) are primarily depending on the first 'front door' category of data collecting. Within these studies there were sometimes employee interviews as well, albeit interviews that seem to have been by appointment with the company and on company premises.^[1] Adler (1993 and forthcoming), Kenney and Florida (1993) as well as Besser (1996) also utilised manager interviews and company documentation, but depended in addition on an extensive use of worker interviews.^[2] Babson (1995) combined worker interviews with access through union contacts, Graham (1994, 1995) as well as Parker and Slaughter (1988) conducted research according to the covert participant observation method in combination with interviewing, while Cole and Deskins (1988) conducted a research based on cases brought to legal action.

2.5

One may in general terms discuss the conceivable points of merit vs. demerit when it comes to the different approaches. I will examine only the first two in detail since these are the approaches relevant to the case study below. For the first, front door approach the greatest points of merit are obvious: after first having established contact there is a good chance that the researcher obtains detailed and trustworthy information impossible to obtain elsewhere, and in the cases of close relationship between corporation and researcher it will be easy to organise employee interviews. There may, however, be a great point of demerit linked to corporation~researcher closeness, since collaboration might entail restrictions or conditions in return for access. Especially the third kind of conditional access, censorship, is in the case of research on Japanese and overseas Japanese-invested factories occasionally evident in the form of concealing firm identity in return for access (see e.g. Suzuki, 1991; Abo, 1994a and b; as well as Williams et al, 1994).^[3] Another conceivable dubious point in connection with the front door approach is that organising interviews in conjunction with the corporation might result in biased respondents.

2.6

Following the second approach, which consists in independently finding, and establishing contact with, employees, may obviously alleviate the problem of potential bias. At the same time the method may result in another kind of problem: where the research topic is centred on achieving representativity it might become difficult to obtain a sufficient number of respondents without the aid of the corporation (or the union, if the firm is unionised). But especially in cases where sensitive topics are going to be discussed the independent approach entails points of merit, since the respondent may rest assured that the statements will remain confidential from his or her superiors. Another point of merit within the independent approach is that the method may circumvent the corporate gatekeepers altogether. With a completely independent approach circumventing the corporate gatekeepers the issue of access and gatekeeper relations ideally becomes one of researcher~respondent relations. The research might thus gain in strength when it comes to content and depth, and may in addition steer away from the problems of conditional access and gatekeeper influence. A completely independent approach means, however, that one must refrain from the general type information available from corporate (or union) gatekeepers, and one might even have to refrain from observing the premises where the work takes place.

2.7

Within this general context of pros and cons in connection with particular approaches, and within the specific context of an emerging research field focusing on Japanese-invested factories, the character of the research described in this paper is as follows. It was a relatively small research project, where the purpose of the research was primarily to gather general information about the state of working conditions and industrial relations at a Japanese transplant, Toyota Motor Manufacturing, U.S.A. (TMM), secondly to collect information adequate for the analysis of at least one work team within the factory.^[4] We knew from beforehand that the company pursued a policy that could be interpreted as control of relations between employees and outsiders. This policy is evident in the following TMM Employee Handbook formulation: 'if you receive a request for information from outside the Company, please check with your Group Leader/Supervisor before releasing any information' (TMM, 1988: p. 96). Employment relations in general may, in addition to issues of a technological or business nature, thus be regarded as sensitive topics in this case. This is evidenced by also other researchers having experienced access problems at the company in question (see for example Besser, 1996: p. 31). Although we anticipated problems we nevertheless wanted to try to obtain interviews with workers and supervisors independently from company influence, while we at the same time felt the need for management co-operation in terms of interviewing managers, receiving documents (such as the employee handbook, organisation charts etc.), as well as receiving permission to tour the plant. In other words we were, like the reviewed Adler (1993) and Kenney and Florida (1993), interested in utilising the first two of the strategies reviewed above in parallel. Each strategy has, as described, got its points of merit and demerit. In the case study which follows some of these points are illustrated as such. Furthermore, the special dilemmas emerging from wanting to use the two approaches in parallel are illustrated.

Finding and Contacting Interviewees

3.1

The research project in question was not a very large one by any standards. Financially speaking it made room for only very short stays the first two times we visited, and only about three weeks of stay in the area the third time for the purpose of contacting and interviewing company employees.^[5] Even with this inherent constraint in terms of time and funding, we regarded the method consisting in independently finding and contacting workers and supervisors for semi-structured interviews about their workplace life and employment relations (see interview guide) as an indispensable part of our research strategy. This was because we felt, based on previous research experiences within the industry in Japan, that interviewing away from the work context usually resulted in a more relaxed interview environment. Moreover, and as mentioned as a meritable point in the previous section, the freedom inherent in being able to select participants for interviewing instead of having potential interviewees appointed by the company was of great importance to us, since we then could avoid the risk of being presented with potentially 'biased', hand-picked persons (cf. Hammersley and Atkinson, 1995: p. 134). Related to this is the freedom that the interviewee is guaranteed by being able to be interviewed without the company knowing about it whatsoever.

3.2

This method was not without its difficulties, however. Without relying on the company for a ready line-up of interviewees, or without relying on lists with the names and addresses of persons employed at the company, the first problem was to find and contact the potential interviewees. Of three trips to TMM, the first two were very short trips with aims primarily other than employee interviews. The second trip did not involve any employee interviews at all and is therefore not treated in this paper. The third trip was organised with as extensive interviewing as possible in mind and is subsequently in focus in this paper. The principal method used in order to find potential interviewees the third time was to obtain copies of the company newspaper before and after arrival, and then make on basis of this a list of some 100 employee names. We then set out to investigate their phone numbers and addresses, starting with the phone directory of Georgetown, Kentucky, where the plant is located, and expanding to the directory of Lexington and other nearby towns and cities. Another method used was the snowball method, i.e. asking persons already interviewed or persons we met whether they knew anyone or had relatives at the plant.

3.3

Using these methods we got into contact with a little less than one fifth (19 persons) of all the persons on the list. Concerning the remaining persons we either felt that we did not have resources to pursue more contacts during this specific research trip, or we were unable to find reliable addresses and telephone numbers. Six of 13 unsuccessful interviews were persons who declined in an outright way, while the seven remaining unsuccessful interviews ended up that way for other reasons. The most usual reason for the latter group was that the agreement to be interviewed dissolved when we at a later stage were to agree on a meeting time and place. Either the person was not at home when we tried to reach him or her, or the person did not call us back, as was the appointment in some of the cases. Some of the cases within this category might thus actually have been declinations due to gatekeeper interference or due to second thoughts. But we have, except for cases which I will return to below, no evidence concerning the specific reason behind the broken appointments.

3.4

Three of the six persons who declined in an outright way did not state any specific reason for declining. For example, one of them just stated: 'I don't give interviews, thank you' in a very quick and determined way. One of the persons stating a reason first consented to an interview, and then changed his mind during the interview because he recalled that he had to see to some private matters. Another person confused us with union organisers (TMM is not unionised):

• There's too much happening with the union people. I would rather keep my job and my mouth shut. Sorry, sir.
• We don't have anything to do with the union.
• You're not? Well, anyway, I don't think I'll be able to help you.

3.5

The remaining case of decline as well as two of the broken appointments provide evidence of gatekeeper interference, and I shall in the next section describe these cases.

Conduct in the Field and Relationship with the Gatekeepers

4.1

As a general policy, TMM stresses that loyalty is required from its employees. This policy is stated one place in the Team Member Handbook in connection with rules and regulations regarding 'conflicts of interest': 'Examples of conflicts of interest include ... conduct that is disloyal, disruptive, competitive or damaging to the company' (TMM, 1988: pp. 24 - 25).^[6] As mentioned above, the company has in addition got a policy concerning employees giving statements to outsiders. This policy reads as follows:

From time to time during the course of your employment, you may be entrusted with some confidential Company information. This type of information may involve our production schedule, model change information, or even research and development activity. As you are aware, the automotive industry is extremely competitive and the release of confidential information may affect our ability to compete and your very job security. Therefore, we ask that you maintain confidential information within the gates of our plant. If you are not sure if something is confidential, or if you receive a request for information from outside the Company, please check with your Group Leader/Supervisor before releasing any information. (TMM, 1988: p. 96)^[7]

4.2

In the previous section, I mentioned some cases where the potential interviewee declined in an outright way and some cases where the interview never materialised. For the latter category we suspect that such company policy and/or gatekeeper interference is relevant. This was indeed confirmed by a manager of the company's public affairs section during a telephone conversation. We were told that TMM was not endorsing our interview activities, but at the same time not forbidding us doing it. The same point was made during a subsequent interview at the public affairs section:

As far as interviewing our team members, that has created some problems. We've had some comments from some of our employees who've called up. ... A lot of them have called their supervisors, and their supervisors called me saying: 'Is this OK?' I said we have formally not approved this, but I said, 'It's a free country,' and it's up to them whether they want to help. I would respectfully request that you do not call them. At least don't tell them that we've approved it from public affairs. But I would respectfully request that you don't make interviews of our employees, because they always call me and say: 'What is this all about? Who's doing these surveys?' It causes me a great deal of trouble. It makes me reluctant, frankly, to give you information that has our employees' names in it. ... I encourage you, if you do call employees, to be very clear about the way you are doing the survey. Some of them have come to me and said they think the UAW is calling. For your information, they misunderstand.

4.3

As for cases where problems appeared during the preliminary talks between potential interviewees and us as researchers, the following case involves a group leader: 'as [group leaders are] part of management, we are supposed to clear stuff with public relations before giving detailed explanations'. We agreed to call this person at an appointed date, but there was only an answering machine all day. Another example is an incident which occurred immediately before we received the telephone call from the company as mentioned at the beginning of this section. We were at the stage of discussing with a team leader where to meet, when he said that he first had to place a couple of other telephone calls, and that he would then call back. After having discussed with the public affairs section by phone (as described above), I tried to call the team leader again. A person who I presume to be his wife answered and said she would go and get him. However, a (circa) 10 year old child came on the phone and told us: 'He is not here'.

4.4

Still another example is from the first research trip in 1988. The person in question was absent, but we left a message with the spouse, who at this time stated that 'it is surely okay with an interview'. Later the same day the potential interviewee stated that it was needed 'to talk with public affairs first'. We then agreed that I would call public relation, and subsequently call the potential interviewee the following workday. In the conversation, the public relations representative stated that:

• We get many requests like this, and have said no to others. We just can't make it. If we say yes to one, we have to say yes to everybody. We are not able to respond to everybody.
• But n.n. seems to be willing to be interviewed.
• N.n. is not available today. We are all busy.

4.5

During the interview at the public affairs section we were also offered a kind of trade-off, in which we were going to receive permission to interview a panel of employees on the condition that direct contacting activities would cease:

• So is it something you intend to continue, or ...? I would request that you do not, because I think it raises so many questions in their minds: 'What's this all about? Who are these people?' Because they don't understand the academic research aspect of it. I could put together a panel of three or four team members and let you talk to them specifically at some later date. Would that suit your purpose?
• That would be very helpful and useful to us.
• Would it mean you would not have to do the interviews, or would you probably still do them? ... Well, the point I am making is it might be easier for me - save me time - if I could arrange such a team member panel and then not have to deal with all the questions I get from people who are called by you. So think about that.

4.6

This incident could, in other words, potentially have evolved into an example of the third kind of conditional access, where research design was altered in return for a particular form of access.

Research Ethics, Access Issues and Gatekeepers Relationships in the Case of Research on Sensitive Topics within Large Corporations

5.1

To sum up the case study we tried in this small research project to pursue a dual strategy of front door approach combined with direct relationships between researcher and interview respondent. This lead to certain dilemmas, most important of which were how to negotiate with the corporate gatekeepers at the same time as we did not back up on our wish of having direct relationship with the employees. The following reasons are those which the gatekeepers officially stated for not wanting researchers to conduct employee interviews on a direct basis:

• employees are busy
• it makes public relations and human resources departments too busy
• employees should be spared of being confused by uncertainty of whether it was researchers or union persons contacting them

5.2

The official reasons are essentially practically oriented. Although the reasons which were explicitly stated are of such a practical nature, is there still a basis for stating that we experienced a case of corporate gatekeepers attempting 'to exercise some degree of surveillance and control' (Hammersley and Atkinson, 1995: p. 66)? There is no reason to believe that the practically oriented reasons are invalid. But I think that the surveillance and control issue was nevertheless present. The reason for stating this is, firstly, based on reflections of an economical nature. Indeed, it is understandable that a company is not overjoyed of getting its daily routines interrupted by visiting researchers. Moreover, the company is evidently interested in keeping its expenditures for the public affairs section as low as possible. At the same time the company is obviously aware of the fact that the Japanese-invested transplants in U.S.A. have been, and still are, attracting a great deal of interest. TMM, for instance, has invested in a visitor's centre and implemented standardised tours for the purpose of accommodating the great number of visitors. In view of this it seems to be a strange and illogical strategy to keep the staffing of the public affairs section so low that most serious researchers have to be turned away, as long as the demand is there. Secondly, the surveillance and control issue becomes evident through looking at this case from an organisational perspective. In the employee handbook as well as within employee training, the confidentiality issue is being stressed. This is logical and natural in many respects. What is not so logical in this context is that employees are required to check with the 'Group Leader/Supervisor before releasing any information' each time there is 'a request for information from outside the Company' (TMM, 1988: p. 96). This is not organisationally wise because it signifies a low degree of trust in the employee's sense of judgement. It also triggers a tremendous amount of traffic directed towards superiors and the public affairs section in the case of such requests from outside the company. Having made these remarks it should be noted that it is amazing how well the reporting structure actually worked in this particular instance. We were involved directly with 19 persons, and this resulted in a major flow of communication from the employees to the supervisors and to the public affairs section.

5.3

As for learning experiences and ethical issues for us as researchers it goes without saying that we did not have any implicit right to get any interviews at all. We did after all intrude on a busy life and work situation, and the employees did right in contemplating the current company rules and regulations. But as long as we were not planning to ask questions of technical and confidential nature, it should not be regarded as unethical to at least ask for interviews, even under such constrained conditions. We were interested in keeping the delicate balance between pursuing an independent strategy at the same time as we kept up good front door relations with the gatekeepers. If we had accepted the offer of conditional access in the form of having a team member interview panel arranged we could have improved the relations with the corporation, and we would in the short run have benefited by getting to meet some team members on company grounds. But we would at the same time have had to agree to give up the strategy of independent interviewing. I think that it was ethically correct to reject this offer, in spite of the possibility that the rejection may to a certain degree have jeopardised our relationship with the gatekeepers. Our strategy was ethically correct because we valued the strategy of independent interviewing both as a methodology under any circumstances as well as the methodology of most value in relation to the specific questions within the research project.

5.4

What we could do differently in the future is to send a message in beforehand to the gatekeepers announcing our arrival and explaining our intention of independently contacting employees.We thereby risk that employees will be specifically warned against researchers, but we would also have the benefit of reducing the risk of incidents such as the ones described in this paper during the field period. In order to increase the success rate under constrained conditions such as these, it would be advisable to send letters of introduction by mail to potential interviewees in beforehand instead of just telephoning them.^[8] Finally, the account should have made it clear that the method of independent interviewing is physically and psychologically demanding. Research projects should be budgeted with considerably more field time and resources than what was the case here.

Appendix: see Interview Guide

Notes

¹This is not clearly stated, though (cf. Shimada, 1988: pp. 33 - 41). Within the Abo-group (Abo, 1994a and b), one research member utilises 'group discussions' with workers and their families as her method when researching attitudes towards Japan and the community impacts of Japanese direct investment (cf. Kumagaya, 1994: pp. 310-311).

² On Adler's methodology the source is P.S. Adler, personal communication, autumn 1996.

³ The proliferation of, and possible reasons for, this practice of concealing the firm's identity in the case of research by Japanese researchers in Japan is being discussed in Tsuji (1994).

⁴ The research was planned and conducted in conjunction with Katsuji Tsuji, Ritsumeikan University and then guest researcher at the University of Kentucky, with the assistance of Robynn Pease, then graduate student at the University of Kentucky. Acknowledgements to both in connection with the project in general, and acknowledgements to the former for commenting upon an earlier version of this paper. I also want to express my sincere gratitude to managers and workers interviewed at TMM for taking the time to do so. Managers have also been helpful in supplying company documentation. The portions of the paper which may be interpreted as somewhat critical towards certain aspects of TMM public relations practices are meant as constructive criticism. Preliminary results of our research have been published as Grønning (1994), which is a paper analysing the training system, and as Grønning and Tsuji (1997), which is a comparison of team organisation at TMM and Ford Motor's factory for light trucks in Kentucky. Further information on the Toyota Motor Corporation and TMM is available through the Internet, see Toyota Motor Corporation and the separate server containing information on activities in the U.S.A., Toyota U.S.A..

⁵ The project was in principle privately funded since it had no funding in the form of grants, except for a paid sabbatical for one of the researchers. In addition to the expenses for staying in the area the project also involved inter-continental travel to U.S.A. for two persons.

⁶ It should be noted that this is the only example with such a general wording. The three other examples concerning conflict of interests are all of an explicitly economical nature, i.e. 'acceptance of outside employment', 'financial interests in a firm that does business with TMM', and 'acceptance of gifts from any firm doing or seeking to do business with TMM' (TMM, 1988, p. 25).

⁷ Elsewhere in the handbook there is an item called 'Media Relations Policy': 'Establishing a trusting, beneficial relationship between TMM and the news media requires that all information be handled and distributed in the same manner. Individual team members may be contacted by the news media and asked for information or to respond about the Company's position on public issues. Team members should refer all requests from the news media to the TMM Public Affairs Section. In addition, team members may not release information to the news media about Company activities or the activities of other TMM or TMC team members. The TMM Public Affairs Section has established systems and procedures for responding to news media requests and for obtaining management approval for public statements. If an activity merits or requires public disclosure, its release will be handled by Public Affairs Section' (TMM, 1988, p. 58).

⁸ We were obviously aware of this method also in connection with this particular research, but were hindered from applying it because we lacked sufficient name and address information before arrival. A good idea in the case of such a letter would be to include a photocopy of the letter which is being sent to the public affairs section.

References

ABO, T. (editor) (1994a) Hybrid Factory: The Japanese Production System in the United States. New York: Oxford University Press.

ABO, T. (editor) (1994b) Nihon-teki keiei/seisan shisutemu to Amerika (The Japanese Management/Production System and U.S.A.). Tokyo: Minerva Shobô.

ADLER, P.S. (1993) 'The Learning Bureacracy: New United Motors Manufacturing, Inc.' in B. Staw and L. Cummings (editors) Research in Organizational Behaviour, vol. 15. Greenwich, CT: JAI Press.

ADLER, P.S. (forthcoming) 'Hybridization of Human Resource Management at Two Toyota Transplants' in J. Liker, M. Fruin and P.S. Adler (editors) Remade in America: Transplanting and Transforming Japanese Management Systems. New York: Oxford University Press.

ARGYRIS, C. (1969) 'Diagnosing Defenses Against the Outsider' in G.J. McCall and J.L. Simmons (editors) Issues in Participant Observation. Reading, MA: Addison-Wesley.

BABSON, S. (1995) 'Whose Team? Lean Production at Mazda U.S.A.' in S. Babson (editor) Lean Work: Empowerment and Exploitation in the Global Auto Industry. Detroit: Wayne State University Press, pp. 235 - 246.

BESSER, T. (1996) Team Toyota: Transplanting the Toyota Culture to the Camry Plant in Kentucky. Albany: State University of New York Press.

COLE, R.E. and DESKINS, D.R. Jr. (1988) 'Racial Factors in Site Location and Employment Patterns of Japanese Auto Firms in America', California Management Review, vol. 31, no. 1, pp. 9 - 22.

GRAHAM, L. (1994) 'How Does the Japanese Model Transfer to the United States?' in T. Elger and C. Smith (editors) Global japanization? The Transnational Transformation of the Labour Process. London and New York: Routledge.

GRAHAM, L. (1995) On the Line at Subaru-Isuzu: The Japanese Model and the American Worker. Ithaca: ILR Press/Cornell University Press.

GRONNING, T. (1994) 'Toyota no kaigai genchi seisan kôjô' (One of Toyota's Plants Abroad), Chapter 8 in Shokugyô Seikatsu Kenkyukai (editor) Kigyô shakai to ningen: Toyota no rôdô, seikatsu, chiiki (Corporate Society and Human Beings: Work, Living Conditions and Community at Toyota). Kyoto: Hôritsu Bunka-sha.

GRONNING, T. and TSUJI, K. (1997) 'Technical, Social and Political Dimensions of Labor Groupification: A Comparison of the Ford and Toyota Plants in Kentucky, U.S.A.,' paper presented at The 15th International Labour Process Conference, Edinburgh, 25 - 27 March, 1997.

HAMMERSLEY, M. and ATKINSON, P. (1995) Ethnography: Principles in Practice, Second Edition. London: Routledge.

KENNEY, M. and FLORIDA, R. (1993) Beyond Mass Production: The Japanese System and its Transfer to the U.S. New York & Oxford: Oxford University Press.

KLEIN, L. (1976) A Social Scientist in Industry. London: Gower.

KUMAGAYA, F. (1994) 'Amerika-jin jugyôin no genchi Nihon kôjô ninshiki' (U.S. Employees' Conception of the Loaclised Factory) in T. Abo (editor) Nihon-teki keiei/seisan shisutemu to Amerika (The Japanese Management/Production System and U.S.A.). Tokyo: Minerva Shobô.

LEE, R.M. (1993) Doing Research on Sensitive Topics. London: Sage.

PARKER, M. and SLAUGHTER, J. (1988) 'Mazda: Choosing Workers Who Fit' in Parker, M. and J. Slaughter (editors) Choosing Sides: Unions and the Team Concept. Detroit: Labor Notes.

PUNCH, M. (1994) 'Politics and Ethics in Qualitative Research' in N.K. Denzin and Y.S. Lincoln (editors) Handbook of Qualitative Research. Thousand Oaks: Sage.

RENZETTI, C.M. and LEE, R.M. (editors) (1993) Researching Sensitive Topics. Newbury Park: Sage.

SHAFFIR, W.B. and STEBBINS, R.A. (editors) (1991) Experiencing Fieldwork: An Inside View of Qualitative Research. Newbury Park, CA: Sage.

SHIMADA, H. (1988) Humanware no keizaigaku (The Economics of Humanware). Tokyo: Iwanami Shoten.

SUZUKI, N. (1991) Amerika shakai no naka no nikkei kigyô (Japanese-related Firms within American Society). Tokyo: Tôyô Keizai Shinpô-sha.

TMM (Toyota Motor Manufacturing, Inc., U.S.A.). 1988. Toyota Motor Manufacturing, Inc. Team Member Handbook. Georgetown, Kentucky: Toyota Motor Manufacturing, Inc., U.S.A.

TOTSUKA, H. (1995) 'The Transformation of Japanese Industrial Relations: A Case Study of the Automobile Industry' in S. Babson (editor) Lean Work: Empowerment and Exploitation in the Global Auto Industry. Detroit: Wayne State University Press.

TSUJI, K. (1994) 'Rôdô kenkyu no "kigyô shakai-ka" to sono dakai ni mukete' (On the 'corporate socialization' of labour research and for a reparation of the situation), Ritsumeikan Sangyô Shakai Ronshu, vol. 30, no. 3 [December, 1994], pp. 83-96)

WADDINGTON, D. (1994) 'Participant Observation' in Cassell, Catherine and G. Symon (editors) Qualitative Methods in Organizational Research: A Practical Guide. London: Sage.

WILLIAMS, K., MITSUI, I. and HASLAM, C. (1994) 'How Far from Japan? A Case Study of Japanese Press Shop Practice and Management Calculation' in T. Elger and C. Smith (editors) Global Japanization? The Transnational Transformation of the Labour Process. London and New York: Routledge.

Copyright Sociological Research Online, 1997