Breast cancer: reporting results with inflationary arithmetic

Area of Science:

• Biostatistics
• Medical Research Methodology
• Oncology

Background:

• Numeric data from scientific experiments are rarely published in their raw form.
• Data reduction techniques are common but can introduce author bias.
• Publication bias in scientific data presentation is an ongoing concern.

Purpose of the Study:

• To review methods of data presentation in breast cancer research.
• To highlight potential author biases in scientific publications.
• To discuss the implications of data reduction techniques on study interpretation.

Main Methods:

• Review of common data reduction techniques.
• Analysis of how data presentation choices can influence perception.
• Discussion of specific examples in breast cancer research.

Main Results:

• Data reduction is unavoidable, and various methods exist.
• The choice of data presentation can reflect and convey author bias.
• Techniques like ratio selection and graphic display can subtly alter interpretations.

Conclusions:

• Authors must be aware of the inherent biases in data presentation.
• Critical evaluation of presented data, including denominators and graphical representations, is essential.
• Understanding data reduction methods aids in accurate interpretation of breast cancer research outcomes.