No collection method is universally the most reliable. Match the method to the question, population, measurement and practical constraints.
A useful data process connects a decision to reliable evidence and a clear action; a dashboard alone does not improve outcomes.
Organize data import, validation, analysis and reporting in a script so another person can understand what you did and repeat it.
Use a scatterplot and a suitable correlation method to describe an association, while recognizing that correlation does not prove causation.
A regression model describes an association under its assumptions; a fitted line alone does not establish causation or guarantee useful predictions.
R is the statistical computing environment; RStudio is an interface for working with it. Install R first, then a compatible RStudio Desktop build.
Use one-way ANOVA to compare group means when the design and model assumptions fit your data, then interpret the result with diagnostics and context.