home.social

#hiddenuncertainty — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #hiddenuncertainty, aggregated by home.social.

fetched live
  1. Daniele de Rigo @[email protected] ·

    6/

    The #HiddenUncertainty exemplified in [2] may suggest:

    - each step (including "hidden steps") of the research process should be seen as non-trivial [youtube.com/watch?v=Hr4K5WdV8t ]

    - epistemic humility: different researchers performing "identical tasks" may not converge

    (besides, this implies that each substantial data/model preparation/manipulation/validation deserves #authorship, as these steps are critical for the research findings and #accountability is needed: see credit.niso.org/ )

    #hiddenuncertainty #authorship #accountability
  2. Daniele de Rigo @[email protected] ·

    6/

    The #HiddenUncertainty exemplified in [2] may suggest:

    - each step (including "hidden steps") of the research process should be seen as non-trivial [youtube.com/watch?v=Hr4K5WdV8t ]

    - epistemic humility: different researchers performing "identical tasks" may not converge

    (besides, this implies that each substantial data/model preparation/manipulation/validation deserves #authorship, as these steps are critical for the research findings and #accountability is needed: see credit.niso.org/ )

    #hiddenuncertainty #authorship #accountability
  3. Daniele de Rigo @[email protected] ·

    5/

    A presentation on [2] by Breznau is available at youtube.com/watch?v=Hr4K5WdV8t (archived: web.archive.org/web/2024010718 )

    where a clear #counterexample (a key finding which does not suffer from uncertainty) is also provided -
    to remark how, although #HiddenUncertainty into non-trivial data and models is a frequent issue scientists need to be aware of, important cases exist where core findings are evident and unambiguous:
    human-driven #ClimateChange is a key one

    (screenshot: youtube.com/watch?v=Hr4K5WdV8t )

    #counterexample #hiddenuncertainty #climatechange
  4. Daniele de Rigo @[email protected] ·

    5/

    A presentation on [2] by Breznau is available at youtube.com/watch?v=Hr4K5WdV8t (archived: web.archive.org/web/2024010718 )

    where a clear #counterexample (a key finding which does not suffer from uncertainty) is also provided -
    to remark how, although #HiddenUncertainty into non-trivial data and models is a frequent issue scientists need to be aware of, important cases exist where core findings are evident and unambiguous:
    human-driven #ClimateChange is a key one

    (screenshot: youtube.com/watch?v=Hr4K5WdV8t )

    #counterexample #hiddenuncertainty #climatechange
  5. Daniele de Rigo @[email protected] ·

    3/

    A partial illustration on #HiddenUncertainty influencing attempts to compress #complexity into #OneDimensionalMetrics may be "a large-scale crowdsourced research effort involving 73 teams" which found "that analyzing the same hypothesis with the same data can lead to substantial differences in statistical estimates and substantive conclusions" [2]

    "Instead of convergence, teams’ results varied greatly, ranging from large negative to large positive effects"

    See Fig 1
    pnas.org/doi/10.1073/pnas.2203

    #hiddenuncertainty #complexity #onedimensionalmetrics
  6. Daniele de Rigo @[email protected] ·

    3/

    A partial illustration on #HiddenUncertainty influencing attempts to compress #complexity into #OneDimensionalMetrics may be "a large-scale crowdsourced research effort involving 73 teams" which found "that analyzing the same hypothesis with the same data can lead to substantial differences in statistical estimates and substantive conclusions" [2]

    "Instead of convergence, teams’ results varied greatly, ranging from large negative to large positive effects"

    See Fig 1
    pnas.org/doi/10.1073/pnas.2203

    #hiddenuncertainty #complexity #onedimensionalmetrics