Page - 22 - in Radical Solutions and Open Science - An Open Approach to Boost Higher Education

2 OpenScienceandRadicalSolutions forDiversity,Equity… 23 methods and data and the resulting findings can be called reliable and valid rep- resentations of the reality (Carnine, 1997;Odomet al., 2005). In this perspective, trustworthinesscontributesdirectlytothefirstobjectiveofOpenScience(i.e.,higher reliabilityandvalidity).Andinthelong-termview,itshouldsupportthesecondobjec- tive ofOpenScience (i.e., greater trust and confidence in research) by convincing both, the researchers, aswell as the citizens and thewhole society.Thisway,Open Sciencecouldplayan important role toovercome fakenewsand tobuild a societal consensusandknowledgecommunity. 2.3.3 OpenScience inScientificResearchandDimensions OpenScience is strongly growing currently and the term is used to describemany different concepts,means andpractices across thewhole science.Next to the com- mercialization of (higher) education, more problems were appearing in scientific practices andpublications during the last decades (Chambers, Feredoes,Muthuku- maraswamy&Etchells, 2014;Cook et al., 2018). There are also general concerns aboutwhether science is self-correctingand that theprogressof research is uneven (Shavelson&Towne,2002). Contrary interests of researchers against Open Science are secrecy, particular- ism, self-interestedness andorganizeddogmatism(Anderson,Ronning,DeVries& Martinson, 2007).Theywerefirst discoveredbyMitroff (1974) through interviews withelitescientistsfromtheApollolunarmissionswhoconductedresearchindirect contradiction to theMerton’snorms.Theconnectedproblemsofpressures forpub- lications and funding acquisition are demanding for researchers and under broad discussion (Casadevall&Fang, 2012;Giner-Sorolla, 2012;Gunsalus&Robinson, 2018;Nosek,Spies&Motyl, 2012). Inaddition,itisproventhatresearchershavegreatfreedomtomanipulateresearch analysis andfindings to achieve themost attractive and interesting results for easy publication andbest recognition (Simmons,Nelson&Simohnson, 2011;Wicherts et al., 2016). Normally, researchers do not falsify data as it would be accused as scientific misconduct but several manipulations can easily be conducted and are reportedasresearchpracticessuchasdatafishingandp-hacking(seehttps://projects. fivethirtyeight.com/phacking for an interactive demonstration), hypothesizing after results are known (calledHARKing), and selectively reporting analyses and pub- lishingstudieswithpositive results labelledas reportingandpublicationbias (John, Loewenstein&Prelec,2012;Simmonset al., 2011;Cooket al., 2018). Replicationstudiesarenotoftenpracticedandresulting infailuresfor thevalida- tionof theoriginalfindings (Camereretal.,2016;Ebersoleetal.,2016;Kleinetal., 2014; Open Science Collaboration, 2015). One first major replication study tried to repeat 100 studies in psychologywith 97 significant findings and could validate only 36 of them (OpenScienceCollaboration, 2015). That does notmean that the conclusions of the other studies were false-positive but that the reproducibility is