We argue for a new understanding of scientific objectivity, especially in the context of statistical inference. While previous research focused on objectivity as value freedom and correspondence to facts, we point out that important aspects of being objective involve questions such as transparency about assumptions and robustness to varying these assumptions. This perspective leads to a re-assessment of the objectivity of scientific inference procedures. For example, we conclude, seemingly paradoxically, that statistical inference based on subjective degrees of belief can be objective, too, and often it is even more objective than the standard method in the experimental sciences.
We then use our conceptual analysis of scientific objectivity to develop a new logic of hypothesis testing that balances philosophical considerations with practical requirements. Specifically, we show how our proposals for better statistical inference and a new credit reward system in science can help to overcome the replication crisis in science.
From a practical point of view, another important result is that we have given an operational meaning to objectivity in inference: that is, we have elaborated a checklist for the experimenter that he or she should follow when striving for objectivity in inference (e.g. in the design and analysis of experiments). Thus we have shown that quite abstract philosophical analysis is not only important for specialists, but it has a positive impact on very practical questions, too.
