7 Matching Annotations
  1. Oct 2023
    1. since the most commonly  used models are underdispersed relative to the data, standard errors are almost always zero. The result is that ecologists see incredibly small p values, but replicating results is rare.

      Classical ecology models that work for 100-200 specimens don't work with microbial high throughput sequencing of thousands or millions of sequences, due to their less complexity and the underdispersed nature - Read more about the dispersion

  2. Oct 2021
    1. One obstacle to improvement is a perceived dichotomy between software and data, which has its roots in computing technology but has no counterpart in the structure of scientific knowledge.
  3. Apr 2020
  4. Jan 2020
  5. Jan 2019