Statistical principles for writing experimental papers
 |
Description
Book Introduction
One of the things that surprised me while attending graduate school was that many researchers analyzed data and wrote papers without any basic knowledge of statistics.
They select two of the three experimental groups, perform a t-test without a normality test, and perform linear regression without analyzing the residuals (if you think this is okay, you need to read this book).
Such primitive and elementary errors are rampant in our graduate schools, and if you write a paper by twisting statistics like that, a good journal will reject it right away.
We can't just blame the author for this, because these practical statistical techniques aren't taught well in graduate schools.
Although the Department of Statistics offers several classes, these classes are generally focused on theoretical aspects.
To begin with, statistics instructors often have little experimental experience, often explaining formulas and telling stories that are far removed from reality.
To properly teach practical statistics, you need to not only have statistical knowledge, but also have experience designing and conducting experiments, collecting data overnight, and writing and presenting experimental papers using that data.
Because lecturers with these qualifications are rare, there are not many opportunities to learn the statistical techniques practically necessary for writing papers.
Statistical analysis is not a sideline to research, but rather a core element, and it is regrettable to overlook it in writing a paper.
With this awareness in mind, I have been conducting special lectures on practical statistics for graduate students in the Department of Life Sciences since 2020.
As word of mouth spread about the easy and friendly classes, requests to lecture came from the Department of Chemistry at Seoul National University and Myongji Hospital in Goyang.
Every time I held a special lecture on statistics, there were so many applicants that I couldn't give everyone a chance to take the class.
Some students from other departments even sent desperate emails expressing their desire to take a special statistics course.
I also felt sorry that I couldn't tell all the statistics due to limited time.
Because I wanted to make these practical statistical techniques more widely available, I compiled the statistical techniques needed to write experimental papers into a single book.
The goal of this book is to convey the philosophy and techniques of statistics as easily as possible, so that readers familiar with experimental research can analyze their own data.
It is also important to know how to write the results of the analysis in a paper.
This book is not one that just presents complex formulas, but rather one that is faithful to the theory while conveying the content in an easy-to-understand manner.
I hope that researchers who read this book will confidently and accurately use the statistical testing method most appropriate for their data and publish their papers in reputable journals.
I would like to express my gratitude to Jaeseung Kim of the Department of Life Sciences at Seoul National University for reviewing this textbook and correcting several errors.
I would like to thank Park Young-sa and Jeong Yeon-hwan for selecting this book, and Kim Min-jo for editing it.
Above all, I dedicate this book to my family, who have always been my strength in times of difficulty and joy.
Author Choi Ji-beom
They select two of the three experimental groups, perform a t-test without a normality test, and perform linear regression without analyzing the residuals (if you think this is okay, you need to read this book).
Such primitive and elementary errors are rampant in our graduate schools, and if you write a paper by twisting statistics like that, a good journal will reject it right away.
We can't just blame the author for this, because these practical statistical techniques aren't taught well in graduate schools.
Although the Department of Statistics offers several classes, these classes are generally focused on theoretical aspects.
To begin with, statistics instructors often have little experimental experience, often explaining formulas and telling stories that are far removed from reality.
To properly teach practical statistics, you need to not only have statistical knowledge, but also have experience designing and conducting experiments, collecting data overnight, and writing and presenting experimental papers using that data.
Because lecturers with these qualifications are rare, there are not many opportunities to learn the statistical techniques practically necessary for writing papers.
Statistical analysis is not a sideline to research, but rather a core element, and it is regrettable to overlook it in writing a paper.
With this awareness in mind, I have been conducting special lectures on practical statistics for graduate students in the Department of Life Sciences since 2020.
As word of mouth spread about the easy and friendly classes, requests to lecture came from the Department of Chemistry at Seoul National University and Myongji Hospital in Goyang.
Every time I held a special lecture on statistics, there were so many applicants that I couldn't give everyone a chance to take the class.
Some students from other departments even sent desperate emails expressing their desire to take a special statistics course.
I also felt sorry that I couldn't tell all the statistics due to limited time.
Because I wanted to make these practical statistical techniques more widely available, I compiled the statistical techniques needed to write experimental papers into a single book.
The goal of this book is to convey the philosophy and techniques of statistics as easily as possible, so that readers familiar with experimental research can analyze their own data.
It is also important to know how to write the results of the analysis in a paper.
This book is not one that just presents complex formulas, but rather one that is faithful to the theory while conveying the content in an easy-to-understand manner.
I hope that researchers who read this book will confidently and accurately use the statistical testing method most appropriate for their data and publish their papers in reputable journals.
I would like to express my gratitude to Jaeseung Kim of the Department of Life Sciences at Seoul National University for reviewing this textbook and correcting several errors.
I would like to thank Park Young-sa and Jeong Yeon-hwan for selecting this book, and Kim Min-jo for editing it.
Above all, I dedicate this book to my family, who have always been my strength in times of difficulty and joy.
Author Choi Ji-beom
index
Chapter 01 The Zen of Statistics
1.1 What is a random variable? 11
1.2 Probability Density Function 12
1.3 Mean and Variance 15
1.4 Median and IQR 17
1.5 Mode and Skewness 19
1.6 Properties of Mean and Variance 1 21
1.7 Properties of Mean and Variance 2 24
1.8 Independence of Random Variables 28
1.9 Covariance and Correlation 31
1.10 (Reference) Pearson Correlation Coefficient and Cauchy-Schwarz Inequality 33
1.11 Population and Sample 36
1.12 Calculating Sample Variance 39
1.13 Binomial and Normal Distributions 42
1.14 Misconceptions about the Central Limit Theorem and the Normal Distribution 45
1.15 Significant Figures and Precision 47
Chapter 02 Hypothesis Testing
2.1 Types of Errors 53
2.2 Meaning of p-value 55
2.3 Interpreting p-values 58
2.4 If the p-value is large, 60
2.5 If the p-value is small, 62
2.6 p-value simulation 63
2.7 Finding the p-value using the proportion test 64
2.8 One-tailed and two-tailed tests 68
2.9 Sensitivity and Specificity 70
2.10 Publication Bias and the Funnel Diagram 72
Chapter 03 t-test, F-test
3.1 Assumptions of Statistical Testing Methods 77
3.2 Chi-square distribution 81
3.3 Relationship between Population Variance and Sample Variance 83
3.4 Reasons for using the t-distribution and its characteristics 86
3.5 One-sample t-test and Paired t-test 88
3.6 Definition of the F-Distribution 91
3.7 Comparing Data for Multiple Groups 93
3.8 Two-sample t-test 96
3.9 Prerequisites for the t-test 99
3.10 Checking Variance Using the F-Test 104
3.11 Determining the Appropriate Statistical Method 106
3.12 Confidence interval 108
Chapter 04 One-way ANOVA
4.1 Propositions and Logic 115
4.2 Problems with Multiple Comparisons 117
4.3 Can't we just assume it doesn't exist? 120
4.4 Factors and Levels 121
4.5 Structure of ANOVA 123
4.6 One-way ANOVA Assumptions and Terminology 125
4.7 Statistical Tests of One-Way ANOVA 128
4.8 Calculating the p-value for one-way ANOVA 131
4.9 What exactly are degrees of freedom? 133
4.10 Example of One-Way ANOVA 135
4.11 Post hoc analysis 138
4.12 Intra-Ocular Trauma Test 140
4.13 Simple Post-Hoc Test: Bonferroni Correction 141
4.14 The Need for Repeated Measures ANOVA 142
4.15 Calculating Repeated Measures ANOVA 143
4.16 Example of Repeated Measures ANOVA 146
4.17 Sphericity Assumption 149
4.18 Handling Missing Data 150
Chapter 05 Two-way ANOVA
5.1 Structure of Two-Way ANOVA 158
5.2 Meaning of Two-way ANOVA Calculations 159
5.3 Degrees of Freedom in Two-Way ANOVA 162
5.4 Two-way ANOVA Example 165
5.5 When the interaction is significant 167
5.6 Two-way Repeated measures ANOVA 168
5.7 Factorial ANOVA 170
Chapter 06 Regression Analysis
6.1 Types of Data and Statistical Tests Based on Them 175
6.2 Correlation and Causality 176
6.3 Anscombe's quartet 178
6.4 Linear relationship 182
6.5 Estimating the Slope and Intercept 184
6.6 (Reference) Why minimize the square of the error rather than its absolute value? 186
6.7 Statistical Significance of the Slope 187
6.8 Degrees of Freedom in Regression and the F-Test 188
6.9 What to check after performing regression 191
6.10 Concave and Convex Functions 193
6.11 Transformation using degree 194
6.12 Correlation Coefficients and Regression Coefficients 197
Chapter 07 Advanced Regression
7.1 Multiple regression analysis 201
7.2 Representing Regression as a Matrix 204
7.3 t-test for regression coefficients 205
7.4 Variance Inflation Factor (VIF) 207
7.5 Interaction and Quadratic Models 208
7.6 Generalized Linear and Nonlinear Models 209
7.7 Logistic regression 211
7.8 Stepwise model selection 214
7.9 Principles of Building Models 217
7.10 ANCOVA 219
Chapter 08 Non-Parametric Tests
8.1 Creating a p-value close to 0 with 4 points 225
8.2 A Note on Normality Testing 226
8.3 Kurtosis and Skewness 227
8.4 Chi-square test 230
8.5 Rank sum test (Mann-Whitney U test) 233
8.6 Wilcoxon signed-rank test 235
8.7 Nonparametric Tests Alternative to ANOVA 236
Chapter 09 Problems and Cases
1.1 What is a random variable? 11
1.2 Probability Density Function 12
1.3 Mean and Variance 15
1.4 Median and IQR 17
1.5 Mode and Skewness 19
1.6 Properties of Mean and Variance 1 21
1.7 Properties of Mean and Variance 2 24
1.8 Independence of Random Variables 28
1.9 Covariance and Correlation 31
1.10 (Reference) Pearson Correlation Coefficient and Cauchy-Schwarz Inequality 33
1.11 Population and Sample 36
1.12 Calculating Sample Variance 39
1.13 Binomial and Normal Distributions 42
1.14 Misconceptions about the Central Limit Theorem and the Normal Distribution 45
1.15 Significant Figures and Precision 47
Chapter 02 Hypothesis Testing
2.1 Types of Errors 53
2.2 Meaning of p-value 55
2.3 Interpreting p-values 58
2.4 If the p-value is large, 60
2.5 If the p-value is small, 62
2.6 p-value simulation 63
2.7 Finding the p-value using the proportion test 64
2.8 One-tailed and two-tailed tests 68
2.9 Sensitivity and Specificity 70
2.10 Publication Bias and the Funnel Diagram 72
Chapter 03 t-test, F-test
3.1 Assumptions of Statistical Testing Methods 77
3.2 Chi-square distribution 81
3.3 Relationship between Population Variance and Sample Variance 83
3.4 Reasons for using the t-distribution and its characteristics 86
3.5 One-sample t-test and Paired t-test 88
3.6 Definition of the F-Distribution 91
3.7 Comparing Data for Multiple Groups 93
3.8 Two-sample t-test 96
3.9 Prerequisites for the t-test 99
3.10 Checking Variance Using the F-Test 104
3.11 Determining the Appropriate Statistical Method 106
3.12 Confidence interval 108
Chapter 04 One-way ANOVA
4.1 Propositions and Logic 115
4.2 Problems with Multiple Comparisons 117
4.3 Can't we just assume it doesn't exist? 120
4.4 Factors and Levels 121
4.5 Structure of ANOVA 123
4.6 One-way ANOVA Assumptions and Terminology 125
4.7 Statistical Tests of One-Way ANOVA 128
4.8 Calculating the p-value for one-way ANOVA 131
4.9 What exactly are degrees of freedom? 133
4.10 Example of One-Way ANOVA 135
4.11 Post hoc analysis 138
4.12 Intra-Ocular Trauma Test 140
4.13 Simple Post-Hoc Test: Bonferroni Correction 141
4.14 The Need for Repeated Measures ANOVA 142
4.15 Calculating Repeated Measures ANOVA 143
4.16 Example of Repeated Measures ANOVA 146
4.17 Sphericity Assumption 149
4.18 Handling Missing Data 150
Chapter 05 Two-way ANOVA
5.1 Structure of Two-Way ANOVA 158
5.2 Meaning of Two-way ANOVA Calculations 159
5.3 Degrees of Freedom in Two-Way ANOVA 162
5.4 Two-way ANOVA Example 165
5.5 When the interaction is significant 167
5.6 Two-way Repeated measures ANOVA 168
5.7 Factorial ANOVA 170
Chapter 06 Regression Analysis
6.1 Types of Data and Statistical Tests Based on Them 175
6.2 Correlation and Causality 176
6.3 Anscombe's quartet 178
6.4 Linear relationship 182
6.5 Estimating the Slope and Intercept 184
6.6 (Reference) Why minimize the square of the error rather than its absolute value? 186
6.7 Statistical Significance of the Slope 187
6.8 Degrees of Freedom in Regression and the F-Test 188
6.9 What to check after performing regression 191
6.10 Concave and Convex Functions 193
6.11 Transformation using degree 194
6.12 Correlation Coefficients and Regression Coefficients 197
Chapter 07 Advanced Regression
7.1 Multiple regression analysis 201
7.2 Representing Regression as a Matrix 204
7.3 t-test for regression coefficients 205
7.4 Variance Inflation Factor (VIF) 207
7.5 Interaction and Quadratic Models 208
7.6 Generalized Linear and Nonlinear Models 209
7.7 Logistic regression 211
7.8 Stepwise model selection 214
7.9 Principles of Building Models 217
7.10 ANCOVA 219
Chapter 08 Non-Parametric Tests
8.1 Creating a p-value close to 0 with 4 points 225
8.2 A Note on Normality Testing 226
8.3 Kurtosis and Skewness 227
8.4 Chi-square test 230
8.5 Rank sum test (Mann-Whitney U test) 233
8.6 Wilcoxon signed-rank test 235
8.7 Nonparametric Tests Alternative to ANOVA 236
Chapter 09 Problems and Cases
GOODS SPECIFICS
- Date of issue: August 22, 2023
- Page count, weight, size: 275 pages | 171*244*2mm
- ISBN13: 9791130318127
You may also like
카테고리
korean
korean
![ELLE 엘르 스페셜 에디션 A형 : 12월 [2025]](http://librairie.coreenne.fr/cdn/shop/files/b8e27a3de6c9538896439686c6b0e8fb.jpg?v=1766436872&width=3840)
