floor effect

{{short description|Source of bias during data gathering in statistics}}

In statistics, a floor effect (also known as a basement effect) arises when a data-gathering instrument has a lower limit to the data values it can reliably specify.{{Cite journal|last1=Lim|first1=Christopher R.|last2=Harris|first2=Kristina|last3=Dawson|first3=Jill|last4=Beard|first4=David J.|last5=Fitzpatrick|first5=Ray|last6=Price|first6=Andrew J.|date=2015-07-01|title=Floor and ceiling effects in the OHS: an analysis of the NHS PROMs data set|url= |journal=BMJ Open|language=en|volume=5|issue=7|pages=e007765|doi=10.1136/bmjopen-2015-007765|issn=2044-6055|pmid=26216152|pmc=4521553}} This lower limit is known as the "floor".{{Cite web|url=https://www.statisticshowto.com/floor-effect/|title=Floor Effect / Basement Effect: Definition|last=Stephanie|date=2017-09-10|website=Statistics How To|language=en-US|access-date=2020-04-14}} The "floor effect" is one type of scale attenuation effect;{{Cite web|title=Scale Attenuation Effect - SAGE Research Methods|url=https://methods.sagepub.com/reference/dictionary-of-statistics-methodology/n1744.xml|access-date=2020-10-22|website=methods.sagepub.com|language=en}} the other scale attenuation effect is the "ceiling effect". Floor effects are occasionally encountered in psychological testing, when a test designed to estimate some psychological trait has a minimum standard score that may not distinguish some test-takers who differ in their responses on the test item content.{{Cite journal|last1=Zhu|first1=Leina|last2=Gonzalez|first2=Jorge|date=2017|title=Modeling Floor Effects in Standardized Vocabulary Test Scores in a Sample of Low SES Hispanic Preschool Children under the Multilevel Structural Equation Modeling Framework|journal=Frontiers in Psychology|language=en|volume=8|page=2146|doi=10.3389/fpsyg.2017.02146|pmid=29312033|pmc=5732956|issn=1664-1078|doi-access=free}} Giving preschool children an IQ test designed for adults would likely show many of the test-takers with scores near the lowest standard score for adult test-takers (IQ 40 on most tests that were currently normed as of 2010). To indicate differences in current intellectual functioning among young children, IQ tests{{Cite journal|last1=Sansone|first1=Stephanie M|last2=Schneider|first2=Andrea|last3=Bickel|first3=Erika|last4=Berry-Kravis|first4=Elizabeth|last5=Prescott|first5=Christina|last6=Hessl|first6=David|date=2014|title=Improving IQ measurement in intellectual disabilities using true deviation from population norms|journal=Journal of Neurodevelopmental Disorders|volume=6|issue=1|pages=16|doi=10.1186/1866-1955-6-16|issn=1866-1947|pmc=4613563|pmid=26491488 |doi-access=free }} specifically for young children are developed, on which many test-takers can score well above the floor score. An IQ test designed to help assess intellectually disabled persons might intentionally be designed with easier item content and a lower floor score to better distinguish among individuals taking the test as part of an assessment process.{{Cite web|url=https://health.ucdavis.edu/welcome/features/2014-2015/06/20150618_IQ-testing-Hessl.html|title=IQ testing in individuals with intellectual disability|website=health.ucdavis.edu|access-date=2020-04-14}}