Recording and communicating uncertainty in science: how geologists manage variability in spatial data

Area of Science:

• Geosciences
• Data Science
• Scientific Methodology

Background:

• Scientific inference relies on managing data variability.
• In observational geology, variability influences data recording and publication decisions.
• High variability may lead to observations being disregarded.

Purpose of the Study:

• Assess how data variability affects geologists' decisions to record and publish observations.
• Investigate geologists' ability to make inferences from variable data.
• Examine the impact of variability levels on inference accuracy.

Main Methods:

• Conducted three experiments with 97 geologists.
• Presented geologists with geological data on planar feature orientation with manipulated variability.
• Assessed willingness to record/publish data and accuracy of orientation estimates.

Main Results:

• Significant individual differences in variability tolerance were observed.
• High-criterion geologists were less likely to record/publish variable data.
• Low-criterion geologists were more willing to record/publish variable data.
• Geologists accurately estimated orientations from variable data, irrespective of variability level.

Conclusions:

• Geologists' criteria for data variability impact data recording and publication.
• Scientists may avoid publishing variable data even when accurate inferences are possible.
• Understanding variability tolerance is crucial for data sharing and scientific progress.