Would anyone be able to recommend me any reference dedicated to determining False Discovery Rate? In the literature (social sciences), it is set at 0.05-0.1; I am mainly interested in learning in which situations it can be less conservative.
Setting the False Discovery Rate (FDR) at 0.1 can be appropriate in several situations, especially when the research context prioritizes a higher sensitivity to detect potentially relevant findings. Here are some scenarios where a less conservative FDR threshold of 0.1 might be applicable:
Exploratory Analysis: In exploratory research or when exploring a new area of study, a higher FDR threshold allows for a broader exploration of potential associations or relationships. It helps generate hypotheses and identify areas for further investigation.
Large-Scale Studies: In studies with a large number of statistical tests, such as genome-wide association studies (GWAS) or high-dimensional data analysis, a more relaxed FDR threshold may be used to control the number of false discoveries while still capturing a substantial number of true positives.
Early-Stage Research: In the early stages of research or when preliminary findings are being examined, a higher FDR threshold can be used to identify potential leads or patterns that can guide further research or hypothesis generation.
Data Screening: When screening data for potential associations or patterns, a less conservative FDR threshold can be employed to identify initial candidates for further investigation.
It is important to note that the choice of FDR threshold should be made based on the specific research context, objectives, and the acceptable balance between false positives and false negatives. The appropriateness of a particular threshold can vary across disciplines and research fields.
Regarding references on determining the False Discovery Rate, here are a few resources that can provide more insights into the topic:
Benjamini, Y., & Hochberg, Y. (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B (Methodological), 57(1), 289-300.
Storey, J. D. (2002). A direct approach to false discovery rates. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 64(3), 479-498.
Benjamini, Y., & Yekutieli, D. (2005). False discovery rate-adjusted multiple confidence intervals for selected parameters. Journal of the American Statistical Association, 100(469), 71-81.
These references discuss the theoretical foundations and methodologies for controlling the FDR in statistical analysis. They can provide a solid starting point for understanding and determining the False Discovery Rate in your research.