Model Uncertainty and the Crisis in Science

Model Uncertainty and the Crisis in Science. Young, C. Socius: Sociological Research for a Dynamic World, 4:237802311773720, January, 2018. 00023

Paper doi abstract bibtex

The “crisis in science” today is rooted in genuine problems of model uncertainty and lack of transparency. Researchers estimate a large number of models in the course of their research but only publish a small number of preferred results. Authors have much influence on the results of an empirical study through their choices about model specification. I advance methods to quantify the influence of the author—or at least demonstrate the scope an author has to choose a preferred result. Multimodel analysis, combined with modern computational power, allows authors to present their preferred estimate alongside a distribution of estimates from many other plausible models. I demonstrate the method using new software and applied empirical examples. When evaluating research results, accounting for model uncertainty and model robustness is at least as important as statistical significance.

@article{young_model_2018,
	title = {Model {Uncertainty} and the {Crisis} in {Science}},
	volume = {4},
	issn = {2378-0231, 2378-0231},
	url = {http://journals.sagepub.com/doi/10.1177/2378023117737206},
	doi = {10.1177/2378023117737206},
	abstract = {The “crisis in science” today is rooted in genuine problems of model uncertainty and lack of transparency. Researchers estimate a large number of models in the course of their research but only publish a small number of preferred results. Authors have much influence on the results of an empirical study through their choices about model specification. I advance methods to quantify the influence of the author—or at least demonstrate the scope an author has to choose a preferred result. Multimodel analysis, combined with modern computational power, allows authors to present their preferred estimate alongside a distribution of estimates from many other plausible models. I demonstrate the method using new software and applied empirical examples. When evaluating research results, accounting for model uncertainty and model robustness is at least as important as statistical significance.},
	language = {en},
	urldate = {2020-10-02},
	journal = {Socius: Sociological Research for a Dynamic World},
	author = {Young, Cristobal},
	month = jan,
	year = {2018},
	note = {00023},
	pages = {237802311773720},
}

Downloads: 0

{"_id":"EjAWiXniGzPdHtBXF","bibbaseid":"young-modeluncertaintyandthecrisisinscience-2018","authorIDs":[],"author_short":["Young, C."],"bibdata":{"bibtype":"article","type":"article","title":"Model Uncertainty and the Crisis in Science","volume":"4","issn":"2378-0231, 2378-0231","url":"http://journals.sagepub.com/doi/10.1177/2378023117737206","doi":"10.1177/2378023117737206","abstract":"The “crisis in science” today is rooted in genuine problems of model uncertainty and lack of transparency. Researchers estimate a large number of models in the course of their research but only publish a small number of preferred results. Authors have much influence on the results of an empirical study through their choices about model specification. I advance methods to quantify the influence of the author—or at least demonstrate the scope an author has to choose a preferred result. Multimodel analysis, combined with modern computational power, allows authors to present their preferred estimate alongside a distribution of estimates from many other plausible models. I demonstrate the method using new software and applied empirical examples. When evaluating research results, accounting for model uncertainty and model robustness is at least as important as statistical significance.","language":"en","urldate":"2020-10-02","journal":"Socius: Sociological Research for a Dynamic World","author":[{"propositions":[],"lastnames":["Young"],"firstnames":["Cristobal"],"suffixes":[]}],"month":"January","year":"2018","note":"00023","pages":"237802311773720","bibtex":"@article{young_model_2018,\n\ttitle = {Model {Uncertainty} and the {Crisis} in {Science}},\n\tvolume = {4},\n\tissn = {2378-0231, 2378-0231},\n\turl = {http://journals.sagepub.com/doi/10.1177/2378023117737206},\n\tdoi = {10.1177/2378023117737206},\n\tabstract = {The “crisis in science” today is rooted in genuine problems of model uncertainty and lack of transparency. Researchers estimate a large number of models in the course of their research but only publish a small number of preferred results. Authors have much influence on the results of an empirical study through their choices about model specification. I advance methods to quantify the influence of the author—or at least demonstrate the scope an author has to choose a preferred result. Multimodel analysis, combined with modern computational power, allows authors to present their preferred estimate alongside a distribution of estimates from many other plausible models. I demonstrate the method using new software and applied empirical examples. When evaluating research results, accounting for model uncertainty and model robustness is at least as important as statistical significance.},\n\tlanguage = {en},\n\turldate = {2020-10-02},\n\tjournal = {Socius: Sociological Research for a Dynamic World},\n\tauthor = {Young, Cristobal},\n\tmonth = jan,\n\tyear = {2018},\n\tnote = {00023},\n\tpages = {237802311773720},\n}\n\n","author_short":["Young, C."],"key":"young_model_2018","id":"young_model_2018","bibbaseid":"young-modeluncertaintyandthecrisisinscience-2018","role":"author","urls":{"Paper":"http://journals.sagepub.com/doi/10.1177/2378023117737206"},"downloads":0},"bibtype":"article","biburl":"https://api.zotero.org/users/5133066/collections/6HGF9IGG/items?key=WPTCej56hwYq0I87MNGVKpeu&format=bibtex&limit=100","creationDate":"2020-11-16T15:04:17.866Z","downloads":0,"keywords":[],"search_terms":["model","uncertainty","crisis","science","young"],"title":"Model Uncertainty and the Crisis in Science","year":2018,"dataSources":["iJmFh6MHv8DSpWmtJ"]}