Introduction to Nonparametric Statistics for the Biological Sciences Using R

ISBN-13: 9783319306339 / Angielski / Twarda / 2016 / 329 str.

Thomas W. Macfarland; Jan M. Yates

Opis

This book contains a rich set of tools for nonparametric analyses, and the purpose of this text is to provide guidance to students and professional researchers on how R is used for nonparametric data analysis in the biological sciences:

• To introduce when nonparametric approaches to data analysis are appropriate
  
• To introduce the leading nonparametric tests commonly used in biostatistics and how R is used to generate appropriate statistics for each test
  
• To introduce common figures typically associated with nonparametric data analysis and how R is used to generate appropriate figures in support of each data set
  

The book focuses on how R is used to distinguish between data that could be classified as nonparametric as opposed to data that could be classified as parametric, with both approaches to data classification covered extensively. Following an introductory lesson on nonparametric statistics for the biological sciences, the book is organized into eight self-contained lessons on various analyses and tests using R to broadly compare differences between data sets and statistical approach.

Spis treści

1 Nonparametric Statistics for the Biological Sciences

1.1 Purpose of This Lesson 

1.2 Data Types 

1.2.1 Nominal Data 

1.2.2 Ordinal Data

1.2.3 Interval Data and Ratio Data

1.3 Graphical Presentation of Populations

1.3.1 Samples that Exhibit Normal Distribution

1.3.2 Samples that Fail to Exhibit Normal Distribution

1.4 R and Nonparametric Analyses

1.4.1 Precision of Scales: Ordinal v Interval

1.4.2 Deviation from Normal Distribution

1.4.3 Sample Size: Number of Subjects for Each Breakout Group

1.5 Definition of Nonparametric Analysis

1.6 Statistical Tests and Graphics Associated with Distribution Patterns

1.7 Prepare to Exit, Save, and Later Retrieve This R Session

2 Sign Test

2.1 Background on This Lesson

2.1.1 Description of the Data

2.1.2 Null Hypothesis (Ho)2.2 Data Import of a .csv Spreadsheet-Type Data File into R

2.3 Organize the Data and Display the Code Book

2.4 Conduct a Visual Data Check

2.5 Descriptive Analysis of the Data

2.6 Conduct the Statistical Analysis

2.7 Summary 

2.8 Prepare to Exit, Save, and Later Retrieve This R Session

3 Chi-Square

3.1 Background on This Lesson

3.1.1 Description of the Data

3.1.2 Null Hypothesis (Ho)

3.2 Data Import of a .csv Spreadsheet-Type Data File into R

3.3 Organize the Data and Display the Code Book

3.4 Conduct a Visual Data Check

3.5 Descriptive Analysis of the Data

3.6 Conduct the Statistical Analysis

3.7 Summary

3.8 Addendum: Calculate the Chi-Square Statistic from Contingency Tables

3.9 Prepare to Exit, Save, and Later Retrieve This R Session

4 Mann-Whitney U Test

4.1 Background on This Lesson

4.1.1 Description of the Data

4.1.2 Null Hypothesis (Ho)

4.2 Data Import of a .csv Spreadsheet-Type Data File into R

4.3 Organize the Data and Display the Code Book

4.4 Conduct a Visual Data Check

4.5 Descriptive Analysis of the Data

4.6 Conduct the Statistical Analysis

4.7 Summary

4.8 Addendum: Stacked Data v Unstacked Data

4.9 Prepare to Exit, Save, and Later Retrieve This R Session

5 Wilcoxon Matched-Pairs Signed-Ranks Test

5.1 Background on This Lesson

5.1.1 Description of the Data

5.1.2 Null Hypothesis (Ho)

5.2 Data Import of a .csv Spreadsheet-Type Data File into R

5.3 Organize the Data and Display the Code Book

5.4 Conduct a Visual Data Check

5.5 Descriptive Analysis of the Data

5.6 Conduct the Statistical Analysis

5.7 Summary<5.8 Addendum 1: Stacked Data and the Wilcoxon Matched-Pairs Signed Ranks

Test

5.9 Addendum 2: Similar Functions from Different Packages

5.10 Addendum 3: Nonparamteric v Parametric Confirmation of Outcomes

5.11 Prepare to Exit, Save, and Later Retrieve This R Session

6 Kruskal-Wallis H-Test for Oneway Analysis of Variance (ANOVA) by Ranks

6.1 Background on This Lesson

6.1.1 Description of the Data

6.1.2 Null Hypothesis (Ho)

6.2 Data Import of a .csv Spreadsheet-Type Data File into R

6.3 Organize the Data and Display the Code Book

6.4 Conduct a Visual Data Check

6.5 Descriptive Analysis of the Data

6.6 Conduct the Statistical Analysis

6.7 Summary

6.8 Addendum: Comparison of Kruskal-Wallis Test Differences by Breakout Group

6.9 Prepare to Exit, Save, and Later Retrieve This R Session

7 Friedman Two Way Analysis of Variance (ANOVA) by Ranks

7.1 Background on This Lesson

7.1.1 Description of the Data

7.1.2 Null Hypothesis (Ho)

7.2 Data Import of a .csv Spreadsheet-Type Data File into R

7.3 Organize the Data and Display the Code Book

7.4 Conduct a Visual Data Check

7.5 Descriptive Analysis of the Data

7.6 Conduct the Statistical Analysis

7.7 Summary

7.8 Addendum: Similar Functions from Different Packages

7.9 Prepare to Exit, Save, and Later Retrieve This R Session

8 Spearman's Rank-Difference Coefficient of Correlation

8.1 Background on This Lesson

8.1.1 Description of the Data

8.1.2 Null Hypothesis (Ho)

8.2 Data Import of a .csv Spreadsheet-Type Data File into R

8.3 Organize the Data and Display the Code Book

8.4 Conduct a Visual Data Check

8.5 Descriptive Analysis of the Data

8.6 Conduct the Statistical Analysis

8.7 Summary

8.8 Addendum: Kendall Rank Correlation

8.9 Prepare to Exit, Save, and Later Retrieve This R Session 

9 Other Nonparametric Tests for the Biological Sciences

9.1 Binominal Test

9.2 Walsh Test

9.3 Kolmogoroy-Smirnov Two-Sample Test

9.4 Binominal Logistic Regression

9.5 Future Applications of Nonparametric Statistics

9.6 Contact the Authors

O autorze

Thomas W. MacFarland, Ed.D., is Associate Professor (Computer Technology) at Nova Southeastern University in Fort Lauderdale, Florida.  He joined the Graduate School of Computer and Information Sciences in 1988 and provides consulting services to the university community on research methods and statistical design as well as individual research on institutional concerns and assessment of student learning.  Dr. MacFarland's areas of research include institutional research, assessment of student learning outcomes, federal data resources, and K-12 computer science education.

Jan Yates, Ph.D., is Associate Professor of Educational Media and Computer Science Education at Nova Southeastern University's Abraham S. Fischler College of Education in Fort Lauderdale, Florida. Since 2001, she has worked in the areas of curriculum development, program assessment and review, and accreditation.

Blurb

This book contains a rich set of tools for nonparametric analyses, and the purpose of this supplemental text is to provide guidance to students and professional researchers on how R is used for nonparametric data analysis in the biological sciences:

• To introduce when nonparametric approaches to data analysis are appropriate
  
• To introduce the leading nonparametric tests commonly used in biostatistics and how R is used to generate appropriate statistics for each test
  
• To introduce common figures typically associated with nonparametric data analysis and how R is used to generate appropriate figures in support of each data set
  

The book focuses on how R is used to distinguish between data that could be classified as nonparametric as opposed to data that could be classified as parametric, with both approaches to data classification covered extensively.

Following an introductory lesson on nonparametric statistics for the biological sciences, the book is organized into eight self-contained lessons on various analyses and tests using R to broadly compare differences between data sets and statistical approach.

This supplemental text is intended for:

• Upper-level undergraduate and graduate students majoring in the biological sciences, specifically those in agriculture, biology, and health science - both students in lecture-type courses and also those engaged in research projects, such as a master's thesis or a doctoral dissertation
  
• And biological researchers at the professional level without a nonparametric statistics background but who regularly work with data more suitable to a nonparametric approach to data analysis