Reliability of MTurk Data From Masters and Workers

Autor:	Steven V. Rouse
Rok vydání:	2020
Předmět:	Data collection Psychometrics 05 social sciences Applied psychology 050109 social psychology 0501 psychology and cognitive sciences Psychology 050105 experimental psychology Biological Psychiatry General Psychology Reliability (statistics)
Zdroj:	Journal of Individual Differences. 41:30-36
ISSN:	2151-2299 1614-0001
DOI:	10.1027/1614-0001/a000300
Popis:	Abstract. Previous research has supported the use of Amazon’s Mechanical Turk (MTurk) for online data collection in individual differences research. Although MTurk Masters have reached an elite status because of strong approval ratings on previous tasks (and therefore gain higher payment for their work) no research has empirically examined whether researchers actually obtain higher quality data when they require that their MTurk Workers have Master status. In two different online survey studies (one using a personality test and one using a cognitive abilities test), the psychometric reliability of MTurk data was compared between a sample that required a Master qualification type and a sample that placed no status-level qualification requirement. In both studies, the Master samples failed to outperform the standard samples.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::83739e6f6e014a6f4cb1470cfe598bff https://doi.org/10.1027/1614-0001/a000300 Zobrazit plný text záznamu