Over-optimism Among Experts in Assessment and Foresight

Abstract

It is still disputed whether foresight exercises should be based on top-expert assessments or on a broader base of less specialised experts, and whether the self-rating of experts is an acceptable method. Using the German 1993 and the Austrian 1998 Technology Delphis, this study addresses both questions: Self-rating is in fact an appropriate method for selecting experts. But the assessment of self-rated top-experts tend to suffer from an optimism bias, due to the experts' involvement and their underestimation of realisation and diffusion problems. The degree of optimism is positively correlated with the degree of self-rated knowledge, and it is more pronounced for the least pioneering and for organisational innovations. Experts with top self-ratings working in business have a stronger optimism bias than those working in academia or in the administration. Consistent with the insider hypothesis, they are most optimistic with regard to realisation, innovativeness, and potential leadership in economic exploitation. Given the optimism bias, foresight exercises should base their panels on a fair mixture of experts of different grades, with different types of knowledge and affiliation, and not only on top specialists of the respective field. Delphi-type exercises, therefore, offer an advantage relative to forum groups or small panels of specialists.

Genetic Engineering and Moral Responsibility

“…Scientists tend to have their focus on the good that could be accomplished from their discoveries (Small, 2011) Indeed, top experts tend to be overly optimistic about the technologies they are developing, often underestimating the risk associated with their field of work and underestimating realisation and diffusion problems (National Science Board, 1977;Rollin, 1996;Ticky, 2004). Less specialised experts were found to be less optimistic than top experts (Ticky, 2004). Small and Jollands (2006) also identified a third way in which technology may cause harm to nature and humans.…”

The potential for the clustering of the maritime transport sector in the Greater Dublin Region

“…1998). The subsequent Delphi node is the Decisio~l Delphi in which the consequence of the research question is not to predict, forecast or describe in terms of a reality (Tichy, 2004). The Decisioil Delphi aligns its objective to that of the approach of soft National College of Ireland…”

“…The literature supports the application of a heterogeneous Delphi panel (MTiersma a n d J u r s , 2005). Research h a s also been conducted i n the a r e a of o p t i~n i s m of rated expert panels (Tichy, 2004) which concluded t h a t there is a higher level of optiinisnl in opinion statements from selfr a t e d experts leading to over-optimistic results. Tichy (2004) therefore argues t h a t Delphi (especially foresight Delphi's) should include top experts b u t also experts with a more broad range of linowledge.…”

“…e t al, 2001) a n d helps to remove potential bias a s the National College of Ireland Maritime Clusters respo~ldeilts a r e not known to each other, although there a r e Delphi studies where experts have been ranked or self rated (Tichy, 2004). Here a higher weighting is given to experts whose opinion is deemed to be more valuable.…”

Using early concept narratives to collect valid customer input about breakthrough technologies: The effect of application visualization on transportation

“…Unfortunately for the problem at hand, there are three major problems with using the input of lead users to forecast the success of technological breakthroughs. First, expert users tend to be over-optimistic about the market chances of innovations [9]. Second, lead users might not be representative for the average customer; their needs might never become the needs of the majority of the customers.…”