Content uploaded by Mark Akos Sarvary
Author content
All content in this area was uploaded by Mark Akos Sarvary on Mar 13, 2015
Content may be subject to copyright.
Tested Studies for Laboratory Teaching
Proceedings of the Association for Biology Laboratory Education
Vol. 35, 405-407, 2014
405
Biostatistics in the Classroom: Teaching Introductory
Biology Students How to Use the Statistical Software ‘R’
Effectively
Mark A. Sarvary
Cornell University, Investigative Biology Laboratory, Comstock Hall, Ithaca NY 14853 USA
(mas245@cornell.edu)
Firstpage
Extended Abstract
Background
The Investigative Biology Course at Cornell University is designed for biology majors to provide lab experience
with emphasis on processes of scientic investigation and students gain expertise in scientic methods and instru-
mentation. The course modules follow the “crawl, walk, run” approach to develop the capacity of students to solve
increasingly challenging problems with greater independence. During the “crawl” phase students ll their scientic
toolbox, learn instrumentation and skills to be able to design and conduct their own experiments and analyze and
communicate their results. For multiple years students analyzed the results of their experiments by hand, or by the
assistance of a Laboratory Instructor who may or may not had been procient in using statistical software. To “even
out the playing eld” and to provide an additional tool to the students, we started to teach freshmen and sophomores
how to use the statistical program ‘R’.
‘R’ is a comprehensive, programming, graphical and statistical software that will provide signicant help with
the statistical analysis needed for the research projects conducted in biology labs. It became one of the most popu-
lar statistical software in the biological sciences in the past few years. ‘R’ is an implementation of the statistical
programming language “S”, which was developed at the Bell Laboratories. Every semester, over 400 students learn
how to use ‘R’ in our investigative biology laboratories. After a one-semester biology lab our students do not only
walk away with laboratory skills, but by using ‘R’, they improve their statistical, mathematical and analytical skills
as well. This new addition to our labs has been very successful, and the feedback from the students, TAs and other
faculty has been very positive. ‘R’ is a freeware, and compatible with Unix, Apple and Windows operating systems.
The freeware can be downloaded from www.r-project.org. In addition, we also teach students how to use R-studio
(www.rstudio.com) that organizes graphs, scripts and variables in a user friendly way. R-studio and R-project give
the same results; both software use the same commands; and students can use either of them for their data analysis.
Organization
Students or groups should bring their own laptops to class, and download the software. After collecting simple
data, students either enter their data in an excel le, and save it as “tab delimited text” in their working directory,
or read their data directly into ‘R’. It is very important that students become familiar with data entry right at the
beginning. Students will explore and understand the program on their own, but struggle with data entry. Working
in groups helps students solve problems and troubleshoot faster, with less help from the instructor. In the rst lab
students start analyzing their data by calculating mean, standard deviation, median and IQR and learn how to draw
a boxplot.
Command lines can be entered and executed in the console window. Students can also type a command in the
script window, highlight it, and execute it from there. The script window functions as a word processor with edit-
able command lines, while in the console window each command needs to be retyped or recalled by using the ar-
rows on the keyboard. During the rst lab only the console window and r-project is being used. This helps students
understand the program better. Students are introduced to scripts and R-studio only during their third week.
© 2014 by Mark A. Sarvary
406 Tested Studies for Laboratory Teaching
Sarvary
Pedagogy
‘R’ is a “new language” for the students, especially for those who never used programming and statistics before,
and therefore it should be taught in the crawl phase, simultaneously with basic statistical methods. This helps stu-
dents make the immediate connection between the statistical methods and the software. If both statistics and ‘R’ are
taught during the rst lab section, it becomes a tool that students are familiar with and will more likely use. There
is a simple worksheet due at the beginning of their second lab, which ensures that they understand the rst basic
steps (data entry, descriptive statistics and graphing). Students may draw a boxplot or calculate means and standard
deviations by hand rst, and do the same immediately after that in ‘R’. This exercise gives students the “aha” mo-
ment, and convinces even the most skeptical students that is it worth it to learn this new statistical programming
language.
In order to have students become familiar with many aspects of the software, we have them hand type all the
commands and datasets. Later, students can receive scripts from their instructors or share scripts among them-
selves. Instructors do not need to be statisticians or experts in ‘R’ in order to be able to teach the basics. This gen-
eration of students only need help with the rst few steps, and after that they go and explore this software on their
own.
Assessment
A worksheet or assignment is very important during the rst labs; otherwise the students will not spend time
with the software outside of the classroom. Simple questions on prelims and exams are also asked; however,
students are not required to memorize any of the codes. The real assessment is using the software for their data
analysis, providing the code in the appendix of their papers and using ‘R’ to create the gures for their lab reports
and posters.
Examples
There are several great statistical books and online resources describing how to use ‘R’. Since this is a freeware,
many developers freely share their scripts online. Unfortunately, many of these books and online resources are too
complex for rst and second year students; therefore we write our own lab manual chapter with ‘R’ codes targeting
the level of understanding of the students in our course.
In addition, students can use hash tags (#) to make notes for every command they type in the script window.
These notes remind them what each code stands for, and therefore they create their own teaching material.
The following is a simple example we provide to our students. They can type these lines word by word (includ-
ing #):
#Enter data manually into R. You have 5 data points for kanamycin and another 5 data points for ampicillin:
count<-c(5,7,8,19,22,11,22,33,44,55)
antibiotic<- c(rep (“kanamycin”, 5), rep (“ampicillin”, 5))
#Combine the two variables into one data table (data frame), and call it Example1:
Example1<-data.frame(count,antibiotic)
Example1
#Identify variables, calling the rst 5 data points kan and the other 5 amp:
kan=(count[1:5])
amp=(count[6:10])
#Explore the data with descriptive statistics, rst calculating the mean and standard deviation:
tapply(count,antibiotic, mean)
tapply(count,antibiotic, sd)
#Graphing. Draw two boxplots and color them orange and red:
boxplot(count~antibiotic, col=c(“Orange”, “Red”), ylab=”number of cells”, main=”Serratia cell count”)
Keywords: biostatistics, data analysis, graphing
Proceedings of the Association for Biology Laboratory Education, Volume 35, 2014 407
Mini Workshop: Biostatistics with ‘R’
Mission, Review Process & Disclaimer
The Association for Biology Laboratory Education (ABLE) was founded in 1979 to promote information exchange among
university and college educators actively concerned with teaching biology in a laboratory setting. The focus of ABLE is to
improve the undergraduate biology laboratory experience by promoting the development and dissemination of interesting, in-
novative, and reliable laboratory exercises. For more information about ABLE, please visit http://www.ableweb.org/.
Papers published in Tested Studies for Laboratory Teaching: Peer-Reviewed Proceedings of the Conference of the Associa-
tion for Biology Laboratory Education are evaluated and selected by a committee prior to presentation at the conference, peer-
reviewed by participants at the conference, and edited by members of the ABLE Editorial Board.
Citing This Article
Sarvary, M.A. 2014. Biostatistics in the Classroom: Teaching Introductory Biology Student How to Use the Statistical Software
‘R’ Effectively. Pages 405-407 in Tested Studies for Laboratory Teaching, Volume 35 (K. McMahon, Editor). Proceedings
of the 35th Conference of the Association for Biology Laboratory Education (ABLE), 477 pages. http://www.ableweb.org/
volumes/vol-35/v35reprint.php?ch=42
Compilation © 2014 by the Association for Biology Laboratory Education, ISBN 1-890444-17-0. All rights reserved. No
part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic,
mechanical, photocopying, recording, or otherwise, without the prior written permission of the copyright owner.
ABLE strongly encourages individuals to use the exercises in this proceedings volume in their teaching program. If this
exercise is used solely at one’s own institution with no intent for prot, it is excluded from the preceding copyright restriction,
unless otherwise noted on the copyright notice of the individual chapter in this volume. Proper credit to this publication must
be included in your laboratory outline for each use; a sample citation is given above.
Endpage
