A new, sensitive graphical method for detecting deviations from the normal distribution of drug responses: the NTV...

Area of Science:

• Biostatistics
• Statistical modeling
• Data analysis

Background:

• Accurate assessment of data distribution is crucial in biological and medical research.
• Existing methods like probit plots have limitations in detecting subtle deviations from normality.

Purpose of the Study:

• To introduce and evaluate a new graphical method, the normal test variable (NTV) plot, for detecting deviations from normal distribution.
• To compare the sensitivity of the NTV plot against traditional probit plots and probability density functions.

Main Methods:

• Developed a graphical method leveraging the similarity between graded dose-response relationships and cumulative normal distributions.
• Evaluated the normal test variable (NTV) plot using various simulated distributions, including skewed and composite normal distributions.
• Compared NTV plot performance with probit plots and probability density functions.

Main Results:

• The NTV plot demonstrated generally higher sensitivity in detecting deviations from normal distribution compared to the probit plot.
• Both NTV and probit plots showed potential for identifying bimodality through complementary analysis.
• The study illustrated the graphical characteristics using simulated data and a real-world example of sparteine metabolism.

Conclusions:

• The normal test variable (NTV) plot offers a more sensitive graphical approach for assessing normality.
• This method provides a valuable tool for researchers analyzing dose-response data and identifying complex distribution patterns.
• The NTV plot complements existing methods for robust statistical analysis in biological sciences.