Statistical guidelines for clinical studies of human vision

Area of Science:

• Ophthalmology
• Optometry
• Biostatistics

Background:

• Statistical analysis in vision science can be complex, leading to disagreements among researchers, editors, and reviewers.
• Clear statistical reporting is crucial for the validity and reproducibility of optometric and ophthalmic research.

Purpose of the Study:

• To provide statistical guidance for authors submitting to optometric and ophthalmic journals.
• To offer advice tailored to clinical studies involving human vision.
• To recommend appropriate statistical analyses for various research scenarios.

Main Methods:

• General statistical concepts discussed include sample size, statistical power, hypothesis testing, parametric/non-parametric variables, bootstrap methods, one/two-tail testing, and Bonferroni correction.
• Specific statistical procedures are recommended based on data types encountered in vision research.
• Examples of correct statistical practice are drawn from recent optometric and ophthalmic literature.

Main Results:

• Authors should clearly describe and justify all statistical procedures used, especially non-standard analyses.
• Understanding and applying appropriate statistical methods enhance the quality of submitted manuscripts.
• Consistent and transparent statistical reporting facilitates peer review and scientific discourse.

Conclusions:

• Adherence to recommended statistical practices improves the rigor of optometric and ophthalmic research.
• Clear statistical communication is essential for resolving conflicts and advancing vision science.
• This guidance aims to empower researchers to present their quantitative data effectively and confidently.