2022 Early Hearing Detection & Intervention Virtual Conference

March 13 - 15, 2022

<< BACK TO AGENDA

6/06/2017  |   2:45 PM - 3:00 PM   |  The perils of statistical hypothesis testing in freshwater ecological studies: the Raven’s paradox and hidden multiple comparisons traps   |  301A

The perils of statistical hypothesis testing in freshwater ecological studies: the Raven’s paradox and hidden multiple comparisons traps

Statistical hypothesis testing is a long-debated topic among freshwater scientists who often cannot easily conduct randomized experiments. Issues related to the use of an arbitrary significance level (0.05) and problems associated with low statistical power are often discussed. However, two common errors, the Raven’s paradox and the hidden multiple comparisons trap, have received little discussion. The Raven’s paradox involves supporting a hypothesis with evidence that is often irrelevant to the hypothesis being tested. The hidden multiple comparisons trap is a common issue in tests using computer-intensive methods that hide the underlying multiple comparisons and result in a high false positive (type I) error. These two errors often occur at the same time and lead to misleading results. We illustrate the two errors and how to expose and avoid them using two studies that estimate nutrient thresholds in aquatic systems using macroinvertebrate abundance and composition data.

  • S24 Towards a predictive freshwater ecology: using time-series data to understand and forecast responses to a changing environment
  • C19 Causal Assessment
  • S27 Advancing Consistency in Ecological Assessments

Presentation:
This presentation has not yet been uploaded.

Handouts:
Handout is not Available

Transcripts:
CART transcripts are NOT YET available, but will be posted shortly after the conference


Presenters/Authors

Song Qian (), The University of Toledo, song.qian@utoledo.edu;


ASHA DISCLOSURE:

Financial -

Nonfinancial -

Thomas Cuffney (), U.S. Geological Survey, South Atlantic Water Science Center, 3916 Sunset Ridge Rd., Raleigh, NC 27607, tcuffney@usgs.gov;


ASHA DISCLOSURE:

Financial -

Nonfinancial -