Back

Sampling bias and the robustness of ecological metrics for plant-damage-type association networks: Comment

Schachat, S. R.

2023-02-15 ecology
10.1101/2023.02.14.528448 bioRxiv
Show abstract

Bipartite network metrics, which link taxa at two trophic levels, are notoriously biased when sampling is incomplete or uneven (Bluthgen et al., 2008; Dormann and Bluthgen, 2017; Frund et al., 2016). Yet a new contribution (Swain et al., 2023, henceforth SEA) claims the opposite: that bipartite network metrics are minimally sensitive to incomplete sampling and, in fact, perform better at low sample sizes than traditional richness metrics. Here I show that SEA achieved this extraordinary finding by abandoning accepted practices, including practices from the authors previous papers.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.