Value Conflicts in Computational Modeling

What value conflicts influence the success of the modeling process and of models as products? Clients commission models, modelers develop models, users use models, and many other members of society are affected by models. These different groups all have different values, leading to value conflicts that influence the success of both the modeling process and of models as products. Data collection included in-depth, semi-structured interviews at three field sites: a corporate research laboratory, an academic research laboratory, and a government research laboratory. Three key value conflicts influence the success of the modeling process: timeliness versus completeness, innovation versus reliability, and honesty versus obedience. Value conflicts among different groups influence the success of models as products, including: the modelers versus their organization, the modelers versus the clients, and the clients versus the users. Thus, understanding value conflicts can help us to understand the ethical implications of models.