1st International Congress on Environmental Modelling and Software - Lugano, Switzerland - June 2002
Keywords
modelling, choice processes, bias, subjectivity, perspectives
Start Date
1-7-2002 12:00 AM
Abstract
When building a (interdisciplinary) model for policy support, modellers are faced with many choices that influence the model, and may influence the model outcomes. When a personal judgment is involved in a choice process, this can make the model biased. A model and its outcomes may be unacceptable to the users and stakeholders and may lead to conflict if the model does not adequately take their knowledge and perspectives into account. This study explored how choice processes in the modelling practice take place and what biases may occur that may influence the knowledge and perspectives incorporated in a model. Based on qualitative interviews with modellers at IIASA, it was analysed how modelers deal with choices on problem framing, variables and indicators, uncertainties, computational limitations and interdisciplinary modelling. This paper shows that in the course of modelling for policy support many choice moments are encountered. Moments at which bias may occur are: when determining that an issue requires choices to be taken; while making an inventory of options to choose from; while making the actual choice; and while evaluating the choice made. Goals, restrictions, common practice, the values of the person making the choice, and the opinions of users, stakeholders and peers seem to influence the eventual choices made. Insight into how choices are made, and into what biases may be introduced in a model, may help modellers in treating the incorporation of knowledge and perspectives in their models more consciously and more transparent. This analysis of choice processes in modelling is a first onset of a checklist on choices in modelling assisting this task.
Choice processes in modelling for policy support
When building a (interdisciplinary) model for policy support, modellers are faced with many choices that influence the model, and may influence the model outcomes. When a personal judgment is involved in a choice process, this can make the model biased. A model and its outcomes may be unacceptable to the users and stakeholders and may lead to conflict if the model does not adequately take their knowledge and perspectives into account. This study explored how choice processes in the modelling practice take place and what biases may occur that may influence the knowledge and perspectives incorporated in a model. Based on qualitative interviews with modellers at IIASA, it was analysed how modelers deal with choices on problem framing, variables and indicators, uncertainties, computational limitations and interdisciplinary modelling. This paper shows that in the course of modelling for policy support many choice moments are encountered. Moments at which bias may occur are: when determining that an issue requires choices to be taken; while making an inventory of options to choose from; while making the actual choice; and while evaluating the choice made. Goals, restrictions, common practice, the values of the person making the choice, and the opinions of users, stakeholders and peers seem to influence the eventual choices made. Insight into how choices are made, and into what biases may be introduced in a model, may help modellers in treating the incorporation of knowledge and perspectives in their models more consciously and more transparent. This analysis of choice processes in modelling is a first onset of a checklist on choices in modelling assisting this task.
