On using expert opinion in ecological analyses: a frequentist approach
Bibliographic record
Abstract
Abstract Many ecological studies are characterized by paucity of hard data. Statistical analysis in such situations leads to flat‐likelihood functions and wide confidence intervals. Although, there is paucity of hard data, expert knowledge about the phenomenon under study is many times available. Such expert opinion may be used to strengthen statistical inference in these situations. Subjective Bayesian is one approach to incorporate expert opinion in statistical studies. This approach, aside from the subjectivity, also faces operational problems. Elicitation of the prior is the most difficult step. Another is the lack of a precise quantitative definition of what characterizes an expert. In this paper, we discuss a different approach to incorporating subjective expert opinion in statistical analyses. We argue that it is easier to elicit data than to elicita prior. Such elicited data can then be used to supplement the hard, observed data to possibly improve precision of statistical analyses. The approach suggested here also leads to a natural definition of what constitutes a useful expert. We define a useful expert as one whose opinion adds information over and above what is provided by the observed data. This can be quantified in terms of the change in the Fisher information before and after using the expert opinion. One can, thus, avoid the real possibility of using an expert opinion that adds noise, instead of information, to the hard data. We illustrate this approach using an ecological problem of modeling and predicting occurrence of species. An interesting outcome of this analysis is that statistical thinking helps discriminate between a useful expert and a not so useful expert; expertness need not be decided purely on the basis of experience, fame, or such qualitative characteristics. Copyright © 2006 John Wiley & Sons, Ltd.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.079 | 0.215 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.002 |
| Bibliometrics | 0.009 | 0.004 |
| Science and technology studies | 0.003 | 0.010 |
| Scholarly communication | 0.006 | 0.009 |
| Open science | 0.005 | 0.004 |
| Research integrity | 0.005 | 0.004 |
| Insufficient payload (model declined to judge) | 0.004 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".