اعتبار کنونی روش دلفی در علوم اجتماعی

The Delphi method is a popular technique for forecasting and an aid in decision-making based on the opinions of experts, which has been in existence for over half a century. This work evaluates its methodology and reviews its validity in the present day, especially in the area of Social Sciences. Three recent applications in this field are also explained, professional in nature, which have some characteristics that are not frequent with respect to other Delphi studies published. The main aim of two of these studies was to provide input for economic or statistical quantitative models, using the judgement of expert groups, while the third study aimed to analyse a complex social reality by means of a Policy Delphi in order to obtain reliable information before taking a policy decision. These applications highlight how this technique may be adapted to different social realities and requirements, making a positive contribution to social progress, provided it is applied with the necessary methodological rigour and with a good knowledge of the social medium in which it is being applied. Finally, there is an explanation of a number of lessons learned from the theory and aforementioned experiences, which may contribute to the successful outcome of a Delphi exercise.

More than 50 years have passed since the first Delphi experiment took place and more than 40 since the first article came to light describing its use and procedure. Since then, the Delphi Method has become a widely used and recognised instrument to make predictions and help decision-making. In this process of diffusion it has gone through several different stages since its birth. From the first stage of secrecy that accompanied its genesis with military aims, it has advanced through successive phases of novelty, popularity, criticism and re-examination, until reaching the current phase of continual but relatively infrequent use, in which it has remained since the 80s [1]. Throughout these years, the Delphi Method has broadened its fields of application and has been the object of fierce criticism and numerous evaluations, in which its weaknesses and strengths have been highlighted, as well as its potential. It is possibly the best known forecasting technique with its own name, based on the opinion of experts, and about which most has been written, but is it still valid? Is it still being used for academic and professional purposes or does it now form part of history? Has it overcome its weaknesses? What has been learned throughout these years? In this article we will note some answers to these questions, paying particular attention to the use of this technique in social sciences and also explaining some recent applications run by the author, which will serve to compare the contributions of other authors and to present possible fields of development and application for the technique.

By searching through and reviewing the literature, we have been able to confirm that the Delphi method continues to be used and is a valid instrument for forecasting and supporting decision-making. The comparisons made to date with other techniques that share its scope and conditions of action do not provide decisive findings either in favour of or against the Delphi method, although comparisons with statistical groups and with classic groups using direct interaction do show results that are, in their majority, positive for this technique. It extent of use in research has not fallen over the last 30 years, and over the last 5 years there has even been a greater proliferation of articles using this technique as an instrument, particularly in the fields of Social Sciences (Business, Economics, Psychology) and Health Sciences (Medicine and Nursing). It is a usual instrument in the areas of technological and social forecasting, social diagnosis, consensus interpretations of social or health realities, communication and participation. We should also note that, in spite of the technique's maturity, articles are still being published about the Delphi methodology. However, there are characteristics or weaknesses inherent in the methodology or its application that have not been completely corrected. These problems are greater in the area of Social Sciences, especially in professional applications carried out on experts who do not have an emotional or professional link or commitment to those running the study. The experiences we have presented here have provided us with evidence of the present and potential of the Delphi method in the areas of input for quantitative models by means of expert opinion, diagnosis of complex social situations, social forecasting, obtaining a consensus with respect to social needs and improvement in institutional participation and communication. On the other hand, the aforementioned experiences, and other earlier ones, have allowed us to learn certain lessons about effectively running professional Delphi exercises in Social Sciences, some of them already discussed in classic publications [50]. In summary, the lessons or aspects we feel particular attention should be paid to are as follows: 1. Having institutional support to support and guarantee the beneficial purpose for society (and not an exclusively profit-seeking aim for the team running the study) facilitates expert collaboration. This support must be visible in the acts and instruments of communication with the experts. In most cases, experts take part due to their pride and their desire to collaborate in research that benefits society. 2. Setting up a plural team to run the study with a perfect understanding of the technique but also with a very good knowledge of the area being studied, the experts in the area and their motivations and psychology. 3. Selecting a panel of experts who can make an effective contribution to the theme being studied, with their degree of motivation to take part also being a key criterion. 4. Putting oneself in the place of the expert. Thinking about what could motivate him or her to take part and to collaborate actively until the end of the process, what we could offer him or her and the effort involved for the expert. Generally it is necessary to sacrifice questions and rounds in order to guarantee panel participation and continuity. 5. When possible, it is advisable to carry out a presentation of the study and emphasise its importance with the attendance of the experts. The expert becomes aware of the seriousness and relevance of the study, sees the people who are behind it, understands its methodology, any doubts can be clarified and fears removed, the experts can meet each other and commit themselves to taking part. Participants appreciate personal contact [51]. This meeting also serves to differentiate, in principle, those who have a real interest in taking part (they come) from those who don't. 6. Carrying out a pilot application to improve the precision and comprehension of the questionnaire and to calibrate any difficulties in managing and motivating the panel of experts and in administrating the study [52]. 7. Encouraging the contribution and sending of qualitative feedback: comments, explanations and suggestions from the experts. Feeding this back so that the change in the estimate, if it takes place, is due to a real change or modification in opinion, caused by the new information. In other words, the interaction of qualitative information. This must help, on the one hand, in improving the quality of the group opinion [40] and, on the other hand, in compensating emotionally those who contribute their knowledge, as in the following round they see their suggestions included in the questionnaire sent to all the experts. The expert must have a real sensation of taking part, both in providing information and constructing the study in collaboration with the rest of the panel and the people running the research. 8. The study must not finish for the expert when the last completed questionnaire has been sent in but the expert must be aware when the study has finally finished and that his or her contribution has been of some use. It is vital to send the experts the results of the study as soon as possible, with a personalised letter of thanks and, if possible, holding an event to present the results by way of gratitude and as a farewell. Our experiences have shown us that, together with the methodological guidelines usually contained in the literature on the selection of experts, on the formulation of questions and processing of data, following the logical guidelines of conduct indicated in this section means that both the researchers and the experts feel satisfied with having taken part in a Delphi study that has been effectively carried out, also achieving relatively high levels of reliability and validity for a technique of these characteristics. As a final conclusion we may state that the opinion of experts and the techniques that help to improve the selection of experts, the externalising of their know-how and the integration and improvement thereof, as is the case of the Delphi method, are still fully valid in a context where the speed of change means that the future depends increasingly less on the past and more on the will of the agents of the present. This is particularly true in the field of Social Sciences, where the intervention of human beings, with all their complexity and variability, means that, on many occasions, objective data and relations and models based on these are insufficient to explain and forecast social actions.