Manipulating the Alpha Level Cannot Cure Significance Testing
Explore this paper's citation graph
Summary
It is argued that making accept/reject decisions on scientific hypotheses, including a recent call for changing the canonical alpha level from p =0.05 to p = 0.005, is deleterious for the finding of new discoveries and the progress of science.
- Type
- preprint
- Published
- 2017-11-14
- Cited by
- 105
- References
- 68
- Access
- Open access
- OpenAlex
- https://openalex.org/W2770766519
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:44154310
Keywords
Null hypothesis, Statistical significance, Significance testing, Statistical hypothesis testing, p-value
References
- Beyond statistical significance: clinical interpretation of rehabilitation research literature.
- Lash Responds to "Is Reproducibility Thwarted by Hypothesis Testing?" and "The Need for Cognitive Science in Methodology".
- The Precautionary Principle: Scientific Uncertainty and Type I and Type II Errors
- Deconstructing arguments from the case against hypothesis testing
- Estimating the reproducibility of psychological science
- Robust Statistical Estimation
- Understanding the New Statistics: Effect Sizes, Confidence Intervals, and Meta-Analysis
- Theory-Testing in Psychology and Physics: A Methodological Paradox
- How many people have to die over a type II error?
- Clinical Versus Statistical Significance: Interpreting P Values and Confidence Intervals Related to Measures of Association to Guide Decision Making
- Setting an Optimal α That Minimizes Errors in Null Hypothesis Significance Tests
- The Controversy over Null Hypothesis Significance Testing Revisited
- On the Persistence of Low Power in Psychological Science
- Statistical and clinical significance, and how to use confidence intervals to help interpret both.
- The earth is round (p < .05)
- “On the Relative Motion of the Earth and the Luminiferous Ether”
- Tutorial to robust statistics
- A Robust Journey in the New Millennium
- The fickle P value generates irreproducible results
- A Precautionary Approach to Genetically Modified Organisms: Challenges and Implications for Policy and Science
Cited by
- Why "Redefining Statistical Significance" Will Not Improve Reproducibility and Could Make the Replication Crisis Worse
- Retract p < 0.005 and propose using JASP, instead
- Twenty Steps Towards an Adequate Inferential Interpretation of p-Values in Econometrics
- The replication crisis in epidemiology: snowball, snow job, or winter solstice?
- MicroRNA‐184 and its long noncoding RNA sponge urothelial carcinoma associated 1 are induced in wounded keratinocytes in a store‐operated calcium entry‐dependent manner
- The Impact of P-hacking on “Redefine Statistical Significance”
- Reconnecting p-Value and Posterior Probability Under One- and Two-Sided Tests
- Pitfalls of significance testing and p-value variability: An econometrics perspective
- Paving the Way to Eureka—Introducing “Dira” as an Experimental Paradigm to Observe the Process of Creative Problem Solving
- Investigating the moment when solutions emerge in problem solving
- Knowing how effective an intervention, treatment, or manipulation is and increasing replication rates: accuracy in parameter estimation as a partial solution to the replication crisis
- Effect of fine and gross motor training or motor imagery, delivered via novel or routine modes, on cognitive function
- Inferential Statistics as Descriptive Statistics: There Is No Replication Crisis if We Don’t Expect Replication
- Beyond p values: utilizing multiple methods to evaluate evidence
- A Proposed Hybrid Effect Size Plus p-Value Criterion: Empirical Evidence Supporting its Use
- 心理学の社会貢献に関する私見:教育・発達領域の論文を読んで
- Brain event-related potentials predict individual differences in inhibitory control.
- Interpretable hypothesis tests
- Sparse Sampling of Silence Type I Errors With an Emphasis on Primary Auditory Cortex
- False confidence, non-additive beliefs, and valid statistical inference
Related papers
- Issues in biomedical statistics: statistical inference.
- Are Only p-Values Less Than 0.05 Significant? A p-Value Greater Than 0.05 Is Also Significant!
- Review on Problems with Null Hypothesis Significance Testing in Dental Research and Its Alternatives
- Significance Testing in Accounting Research: A Critical Evaluation Based on Evidence
- Problems and alternatives of testing significance using null hypothesis and P-value in food research