Skip to main content
Log in

Minimum Animal Units and the Standardized Count Problem

  • Published:
Journal of Archaeological Method and Theory Aims and scope Submit manuscript

Abstract

Minimum animal units (MAU) are central to the study of skeletal part profiles in zooarchaeology. This measure standardizes skeletal part counts by their anatomical frequencies in a complete animal, transforming those counts into a series of values—one for each type of skeletal part. Zooarchaeologists often treat MAU as ordinal scale and use rank order statistics to compare MAU values against measures of dietary utility and bone density. Using simulation, I show that these standardized values erase critical sample size information and lead to biased ordinal correlations, preventing reliable inferences about the fossil populations from which the samples were drawn. Given the sample sizes typical of zooarchaeological work, the standardized count problem probably misguides many interpretations of taphonomy and human subsistence. The problem can be circumvented by using Poisson regression, a simple statistical method that provides conservative inferences for relationships between skeletal part profiles and measures of bone density and dietary utility, especially when implemented in a Bayesian framework. The regression approach treats skeletal part data as counts rather than ranks, while also retaining sample size information. I demonstrate the method with two archaeofaunal examples. Poisson regression allows for reliable inferences about fossil assemblages, although extending those inferences back to past animal communities or death assemblages presents additional challenges. Insights into these communities and assemblages require that zooarchaeologists carefully consider the relationship between statistical model specification and causation.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15

Similar content being viewed by others

Notes

  1. Binford (1978; Binford and Bertram, 1977) originally used the term minimum number of individuals (MNI) rather than MAU, conflicting with existing use of the term MNI in zooarchaeology. This conflict was resolved in his book Faunal Remains from Klasies River Mouth, where Binford (1984: 50) began to use the term MAU. Calculation of the values represented by MAU/MNI does not differ between his two terms.

  2. Poisson regression is a type of generalized linear model. Generalized linear models are regression methods that employ distributions other than the normal distribution for the outcome variable. Although these models are widespread in the natural sciences, their use in zooarchaeology is more limited (Carlson, 2017: 232–243 provides an introduction for archaeologists).

  3. Bayesian Poisson regression is supported in numerous R packages, including rethinking (McElreath, 2020; used here), brms (Bürkner, 2017), and rstanarm (Goodrich et al., 2020). Python users can implement the approach with the PyMC3 package (Salvatier et al., 2016). Performing these analyses in R or Python requires only a few lines of code, and novice users of either programming language should find these packages accessible. For those who prefer a graphical interface, recent versions of SPSS (IBM Corp., 2020) also support the method.

References

Download references

Acknowledgements

I thank Karen Lupo and Ian Jorgeson for feedback and suggestions. R. Lee Lyman and an anonymous reviewer provided comments that greatly improved this paper. All simulations were performed on the ManeFrame II computing cluster at Southern Methodist University.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Ryan P. Breslawski.

Ethics declarations

Competing Interests

The author declares no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Breslawski, R.P. Minimum Animal Units and the Standardized Count Problem. J Archaeol Method Theory 30, 268–309 (2023). https://doi.org/10.1007/s10816-022-09563-9

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10816-022-09563-9

Keywords

Navigation