Australasian Science: Australia's authority on science since 1938

What’s in a Name?

By Hannah Fraser

Inconsistent classification of species introduces systematic bias to ecological studies.

Woodland birds are bird species that depend on native woodlands. Unfortunately, woodlands have been widely cleared for agriculture and urban development, leading to a widespread belief that woodland birds must be declining.

Many have studied the decline of woodland birds, most commonly the effect of changing tree cover and fragmentation. The results of these studies vary. Some find evidence of decline; others dispute that a decline is taking place.

Similarly, the nature of the relationship between woodland birds and tree cover and fragmentation varies substantially too. These differences might be due to regional or scale differences between studies, but could there also be underlying disagreement about what actually constitutes a “woodland bird”?

In ecology, there have been sporadic efforts to promote consistency in terminology but little progress. Inconsistent terminology can lead to a range of problems, including difficulties in finding relevant studies, redundant investigations and an inability to synthesise across studies. It can also create problems when communicating findings to other scientists, policy-makers and the public.

How important is consistent terminology when it comes to determining the conservation status and trends of a group of birds loosely referred to as “woodland birds”? To answer this question, I led an investigation that systematically reviewed the literature and compiled a set of 38 lists of woodland birds. This allowed us to work out how consistently each species was classified as a woodland bird.

We found that eight species were always classified as woodland birds and 13 species were always classified as non-woodland birds. The remaining 144 species were sometimes classified as woodland birds and sometimes as non-woodland birds. This surprised us, as we had expected that only a few, less-understood species would be classified inconsistently.

We surveyed the authors of the papers we had reviewed and found that the main reasons researchers classify different species as woodland birds were:

  • different aims of research: researchers tailor their list of “woodland birds” to include species they expect to respond most strongly to the phenomenon they are interested in;
  • disagreement about what a woodland is; and
  • disagreement about how to determine which birds depend on woodlands.

What impact does this have? Colleagues have previously used a subset of species that they considered to be woodland birds to model the effect of habitat aggregation (the inverse of habitat fragmentation) on the occurrence of woodland birds. We re-ran their model, first using the entire complement of species and then using different subsets to emulate the effect of being increasingly selective about which species are considered woodland birds.

We found that as we become more selective about which species are included, the estimated effect of tree cover aggregation increases. Our analysis revealed a systematic bias in results whereby studies that are less selective about which species are woodland birds are likely to obtain different results (probably with lower effect sizes) than those that are very selective about their classification. In other words, how you define woodland bird species has an important bearing on the results you obtain.

What does this mean? When comparing results from studies using different classifications it’s impossible to know whether differences are attributable to data collection, the survey area or analyses, or whether they are due to differences in classification. This essentially renders all studies with non-identical lists incomparable.

This is particularly problematic when trying to understand woodland bird ecology or predict how they will react to management. Only a small subset of research uses identical lists of woodland birds, so researchers must choose between including all available information (which risks confounding results due to differences in classification) or only including studies that use the same list of woodland birds (which risks the exclusion of valuable insights from other studies).

So what should we do? Our analysis leads us to suggest that woodland bird researchers should unite behind a single definition and list of woodland birds. We also believe that this is an approach that would be beneficial to ecology as a whole.

It’s unlikely that woodland bird research is the only realm of ecology where terms are being used inconsistently and clouding results. If our findings hold true for other terms, we believe that it’s extremely important to develop a consistent definitions of these terms.

Hannah Fraser is a member of the ARC Centre of Excellence for Environmental Decisions (CEED). She was based at the University of Melbourne for this research.