BSRT Graduate School

Statistic

Title of Event: 
Statistics in Basic Research using R as a powerful tool for large data sets
Optional or Mandatory: 
This course is optional for doctoral researchers at the BSRT
booked/available (including waiting list): 
0/10
Description: 

Part I and Part II will blend into each other.

 

Part I Statistic in Basic Research

We have the duty of formulating, of summarizing, and of communicating our conclusions, in intelligible form, in recognition of the right of other free minds to utilize them in making their own decisions.(R.A.Fisher 1955)

 

Research is related to statistics at every single step: Papers you read to develop your hypotheses report various statistics to describe data and justify conclusions, in planning your experiments you apply statistical tools to get an idea about sample size, once you have some results you have to summarize your data and make decisions with respect to your research question. This introduction seminar will provide you with the necessary knowledge about underlying concepts, relevant technical terms, and frequently used tests. You will learn how to select the appropriate statistical procedures for your own data. The seminar will focus on understanding the ideas of statistics rather than the math behind it and will have the following topics:

  • Significance / Power / sample size
  • Outlier / data checks
  • Classification levels / scales
  • Cross-tables /li>
  • Normal distribution
  • Means and deviations
  • Tests for mean differences; one- vs. two-tailed significance
  • Univariate ANOVA / ANCOVA / repeated Measures ANOVA
  • Correlation / regression
  • Survival Analysis (Kaplan Meier / Cox regression)

 

Part II Introduction to R

R is a scripting language for statistical analysis and data visualization, scary for the beginner yet extremely powerful once you get the hang of it. This seminar will guide you through the first steps in using R.
 
There are two distinct philosophies when it comes to statistics software: Point&click as in SPSS vs. plain text scripting as in R. For the occasional quick comparison of two groups SPSS is fine, but selection among hundreds or thousands of variables easily becomes a real problem. And don’t expect publication ready tables and figures. If your research involves larger data sets (in terms of variables, not necessarily subjects) as typically found in Omics, if you want to automatize analyses, or need to visualize complex interactions, then you should give R some consideration. There is almost no statistical procedure that has not been implemented in a package for R, powerful import and export functions allow you to get publication ready tables without the usual copy/paste/reformat cycles. If on the other hand you are scared of working with a computer if it goes beyond swiping on your smartphone, R may not be the best choice for you! The seminar will focus on the practical use of R, BYO (bring your own data) is encouraged! Some basic knowledge in statistics is required, as we will not go into the theory behind the procedures used! 
  • Introduction to R
  • Data types
  • Control structures (loops, ifelse etc)
  • Working with extension packages
  • Import/export
  • Data manipulation (recoding, transformation, selection)
  • Descriptive statistics in R
  • Graphics (basic functions, short introduction for ggplot)
  • Statistical tests for mean differences
  • Regression

Please bring your own Laptop!

 

 

Credits: 
2.0