Record Details

Brazilian Reproducibility Initiative Survey of Beliefs Data

Harvard Dataverse (Africa Rice Center, Bioversity International, CCAFS, CIAT, IFPRI, IRRI and WorldFish)

View Archive Info
 
 
Field Value
 
Title Brazilian Reproducibility Initiative Survey of Beliefs Data
 
Identifier https://doi.org/10.7910/DVN/2RLSMG
 
Creator Amaral, Olavo B
Valério-Gomes, Bruna
Wasilewska-Sampaio, Ana Paula
Abreu, Mariana
Neves, Kleber
Carneiro, Clarissa F D
Tan, Pedro B
 
Publisher Harvard Dataverse
 
Description Data from the BRI Survey of Beliefs. Data is currently under embargo, and will become public at the end of 2022. For more information about the project, check the OSF page at https://osf.io/6av7k/. A brief description of the methods is provided below:

 


Participants were recruited via open invites in social media and institutional emails. Recruitment criteria were age above 18 years old and having previous or current research experience. Only participants that completed the survey were included (69 total, where 20 are members from the BRI), and these were subsequently invited to prediction markets.

 


Participants could choose between the 3 techniques that were selected for replications within the BRI: (1) MTT, an assay for mitochondrial activity/cell viability; (2) reverse transcriptase PCR (RT-PCR), an assay to measure relative RNA levels); and (3) Elevated Plus Maze (EPM), a behavioral task to assess anxiety in rodents. Each of these techniques contained 20 studies to be replicated. Participants made predictions for all 20 studies, except for members of the BRI, which did not forecast experiments their lab was involved in replicating.

 


At the beginning of the survey, participants filled in personal information about their research experience and education. They also self-rated their theoretical knowledge of the technique, their practical experience and their knowledge on statistics and research methodology in a scale of 1-5.

 


For each study, they were shown summary information containing a brief description of the experiment, the original figure or table, and information on effect size, sample size and statistics. Different survey versions were created where 10 of the studies had the link to the original paper and author/journal information, while the other 10 had none of this information. This means participants were blinded to additional information from the paper for half of their predictions, having access only to the abstract, figure and the summary.

 


For each study, participants were asked the following questions:

*For the purpose of the answers below, please consider the effect size and significance estimates of a fixed-effects meta-analysis of the 3 replications

 


(1) "In your opinion, what is the probability that the replication will obtain an effect significantly different from 0 (p
 
Subject Medicine, Health and Life Sciences
Reproducibility, replication, Brazil, MTT, RT-PCR, elevated plus maze
 
Language English
 
Contributor Amaral, Olavo
 
Type survey data