The term “design-based” is used to describe newer methods in stereology whose probes and sampling schemes are ‘designed’, that is, defined a priori, so that the methods are independent of the size, shape, spatial orientation, and spatial distribution of the geometrical features to be investigated. Rules must be followed to achieve this independence, and if they are followed correctly than bias is eliminated. Older stereological methods were “model-based”, this means that they used models based on the geometric properties of the objects being studied. The design-based methods eliminate the need for using information about the geometry of the objects to be investigated, resulting in more robust data because potential sources of systematic errors in the calculations are eliminated.
Design-based stereology considers solutions in which models are avoided. The design includes making sure the combination of the tissue collection (e.g. sectioning method) and the nature of and orientation of the virtual geometric shapes used to probe the tissue obviate the need for knowledge about the geometrical features being studied to make a model. There are four basic assumptions that are avoided by design-based approaches:
- No assumptions about shape
- No assumptions about size
- No assumptions about orientation
- No assumptions about distribution