This site deals only with the corporate corruption of science, and makes no inference about the motives or activities of individuals involved.
There are many reasons why individuals become embroiled in corporate corruption activities - from political zealotry to over-enthusiastic activism; from gullibility to greed.
Please read the OVERVIEW carefully, and make up your own mind.
Social & Community Planning Research
— A genuine research group in the UK which demonstrates how commercialism and arrogance by scientists can be used by experienced scientific lobbyists/manipulators to generate misleading 'facts'. —
Philip Morris commissioned a major study on bias in epidemiology in order to counter the regular findings of epidemiologists that smokers had more lung cancer and coronary heart disease (CHD) than non smokers — and that they died earlier, etc.
The aim of the tobacco industry was to develop some 'scientific proof' that the epidemiology was wrong, and that these fallacious results were a natural consequence of the anti-smoking prejudices of the researchers.
There is no doubting that Social & Community Planning Research (SCPR) was a genuine non-profit (for the group ... not for those employed) research centre, which probably deserved its good reputation.
However, as with any organisation of this size and complexity, it needed regular work to sustain its staff. This made it vulnerable to the con-artists and scientific manipulators who ran organisations like The Weinberg Group (TWG - later WashTech), and various subdivisions of the tobacco industry, such as Philip Morris's Worldwide Scientific Affairs (WSA)
In this case SCRPR was being paid very generously to conduct an epidemiological study made by interviews and questionnaires among 2000 randomly-selected members of the public in order to prove that the anti-smoking bias of the interviewer could effect the outcome of this branch of the health sciences.
It is important to realise that the 2000 members of the public were not the primary subjects of this study — they were just being used. The study subjects were the 100 semi-professional interviewers who were paid to conduct what was essentially spurious research. How much could SCPR influence the outcome of a study, by pre-conditioning its own research staff ? It did this by feeding them some emotionally-charged (but fictional) health information that might lead them to bias their records of interview.
So the essential component was the pre-conditioning story which was chosen to be implanted in half the interviewers. [The other 50 were controls.] Other than this distortion, the briefing of all interviewers needed to be standarised.
SCPR had to ensure that with the concocted and implanted bias,
the "theory" was a plausible one and may well be correct to some extent. However, what we are interested in is, whether there is a differential effect on the response.... [between those with the implanted bias, and those without] The simple fact that Philip Morris had control over this pre-conditioning brings the whole study into question — and with it, the reputation of SCPR.
This case-study in tobacco science distortion immediately raises doubts about the trustworthiness of the other research the SCPR group were doing with European Union orgainisations and for the UK Department of Health and Social Security, etc.
Founded in 1969, SCPR is an independent non-profit institute specialising in social research and surveys, with its own research, interviewing, coding and computing resources. It is registered as a charitable trust.|
Some of SCPR's studies are initiated by the institute itself and grant-funded by research councils or foundations. Other work is initiated by govenment departments, local authorities or quasi-govenment organisations to collect and interpret information on aspects of social or economic policy. SCPR also works frequently with other research institutes and academics providing them with research and survey facilities for their own grant-aided work.
The institute has set up a number of specialist units whose work parallels, and is on occasions integrated with, the main body of quantitative survey work:
SCPR is also one of the founder members of COMPASS (the European Consortium for Comparative Social Surveys), a five-nation grouping of research institutes within the EU which collaborate on cross-national studies.
- The Survey Methods Centre has a programme of experimental methodological research, advice and teaching. Together with the University of Southampton it is also responsible for the Centre for Applied Social Surveys (CASS), an ESRC Resource Centre devoted to a training and publications programme to improve standards of quantitative social measurement.
- The Qualitative Research Unit specialises in the design, conduct and interpretation of group discussions, depth interviews and other non-quantitative methods of social research.
- The Centre for Research into Elections and Social Trends (CREST), is an ESRC Research Centre linking SCPR and Nuffield College, Oxford in a research programme analysing and interpreting social and political trends.
- The Joint Health Surveys Unit is a research centre linking SCPR and University College London in a programme of health policy and. epidemiological research and surveys.
Some key documents
1997 Feb 12: SCPR head, Roger Thomas of the Survey Methods Centre, replies to Myron Weinberg of The Weinberg Group (at its Brussels Office). They are sending him the protocol for the [later called] "Inteview-bias in Epidemiology" study involving 30 minutes of interview with 2000 subjects in their homes.
These experienced interviewers would create questionnaires for later evaluation. They propose to use 100 interviewers, 50 of whom are deliberately given a biased view about some 'health-related matter', and therefore likely to influence the outcome. The selection of subjects and inteviewer is said to be at random.
It is considered that interviewer/respondent artefacts, if lhey occur, may arise through any of a number of different interviewer/respondent interaction or response-recording processes which may occur in face-to-face interview surveys. The source of artefacts might, for example, be: SCPR saw this as a pilot study that may lead to a more significant study with more robust results later.
- some form of unconscious verbal or non-verbal communication bias, such that some respondents are encouraged to give particular responses by interviewers who have a particular belief. It might also
- be an interviewer bias in the probing or interpretation and recording of responses.
- Effects are probably most likely to occur where there is for some respondents a degree of ambiguity or vagueness about questions or responses.
The original protocol is here [Modified later by Sanders]
[Note that they propose to deal with large number of subjects because "the size of any effect [of bias] is expected to be small" which probably "cannot be shown to be significant at the 95% confidence level" — which is not how the tobacco industry would later portray the results.]
1997 Apr 9: Roger Thomas replies to Myron Weinberg (and through him to Ted Sanders at Philip Morris). The SCPR's protocol has been accepted [subject to Sander's modifications] and the project is on the move awaiting a formal contract. There is now to be a pre-test in June-July of Sanders changes to the interview-bias questions.
This letter follows up on our letters to you dated 13 and 14 February and 6 March, and also on our meeting with yourself and Dr Sanders on 7 April.
- Ted Sanders of Philip Morris is noted repeatedly in the memo;
- The total costs will total £169,832.
- The final report will be produced in April 1998
What I am now hoping to receive from you and Dr Sanders is, first, an indication that these plans and costs are acceptable and, second, a letter setting out the proposed terms of contract between SCPR and the Weinbeig Group which takes account of all the other details set out in my letter of 14 February and other correspondence. If you or Dr Sanders require any further information, please contact me.
1997 Aug: /E The Weinberg Group is being paid $140,000 per year [now second year] for its European operations on behalf of Philip Morris International. Part of this is to manage the SCPR "Interview-bias in epidemiology" project in London
The reason that the Weinberg Group was asked by WSA [Worldwide Scientific Affairs of Philip Morris] to manage this project was twofold.
It should be noted that SCPR is fully aware that PM is funding the project and has also agreed to provide attribution on all publications, presentations and report.
- First It was the Weinberg Group who located SCPR.
- Secondly, it was felt that it was important to insulate PM as much as possible from the administration of the project to ensure maximum credibility.
Ted Sanders can and does provide input to the manner in which the study is carried out, although SCPR is under no obligation to accept such input.
Page 1 of the questionnaire [others destroyed].
[It is important to note a couple of things about PM's influence over the SCPR results.
This is the sort of mindless crap that the top industry executives put out to salve any pangs of conscience their minions may have further down the employee guilt trail.]
- If SCPR wants further work from The Weinberg Group, then, rather than isolating it from tobacco industry pressure, TWG is another force applying that pressure. TWG also worked for chemical and pharmacuetical companies, so it was more important to SCPR than just the tobacco industry.
- If SCPR knows that PM funding is behind the project, then it knows instinctively what is the most desirable outcome of their study. Epidemiology was a major threat to the tobacco industry, and the idea that interviewers could be biased against smoking was one of the industry's few part-rational counter-claims.
- If PM scientific lobbyist Ted Sanders was actively involved in setting parameters for the study and making changes to the protocols, then the claim that SCPR has "no obligation to accept" his requirements is meaningless.
- The funding company always has control over attribution after the study is complete, and before publication. So PM can still choose to have its funding recognised or not.
- Nothing in such agreements guarantees that the study will be published. If it finds NO investigator bias, it will simply disappear, and so never be available to contradict their other studies made specifically to prove that this bias distorted the results of anti-smoking studies.