Muutke küpsiste eelistusi

Introduction to Data Analysis with R for Forensic Scientists [Kõva köide]

4.50/5 (4 hinnangut Goodreads-ist)
(University of Auckland, New Zealand)
  • Formaat: Hardback, 331 pages, kõrgus x laius: 234x156 mm, kaal: 770 g, 52 Tables, black and white; 95 Illustrations, black and white
  • Sari: International Forensic Science and Investigation
  • Ilmumisaeg: 30-Jul-2010
  • Kirjastus: CRC Press Inc
  • ISBN-10: 1420088262
  • ISBN-13: 9781420088267
Teised raamatud teemal:
Introduction to Data Analysis with R for Forensic ScientistsSuurem pilt
  • Formaat: Hardback, 331 pages, kõrgus x laius: 234x156 mm, kaal: 770 g, 52 Tables, black and white; 95 Illustrations, black and white
  • Sari: International Forensic Science and Investigation
  • Ilmumisaeg: 30-Jul-2010
  • Kirjastus: CRC Press Inc
  • ISBN-10: 1420088262
  • ISBN-13: 9781420088267
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9781420088267.html
Märksõnad:
Statistical methods provide a logical, coherent framework in which data from experimental science can be analyzed. However, many researchers lack the statistical skills or resources that would allow them to explore their data to its full potential. Introduction to Data Analysis with R for Forensic Sciences minimizes theory and mathematics and focuses on the application and practice of statistics to provide researchers with the dexterity necessary to systematically analyze data discovered from the fruits of their research.

Using traditional techniques and employing examples and tutorials with real data collected from experiments, this book presents the following critical information necessary for researchers:











A refresher on basic statistics and an introduction to R Considerations and techniques for the visual display of data through graphics An overview of statistical hypothesis tests and the reasoning behind them A comprehensive guide to the use of the linear model, the foundation of most statistics encountered An introduction to extensions to the linear model for commonly encountered scenarios, including logistic and Poisson regression Instruction on how to plan and design experiments in a way that minimizes cost and maximizes the chances of finding differences that may exist

Focusing on forensic examples but useful for anyone working in a laboratory, this volume enables researchers to get the most out of their experiments by allowing them to cogently analyze the data they have collected, saving valuable time and effort.
1 Introduction
1(6)
1.1 Who is this book for?
1(1)
1.2 What this book is not about
1(1)
1.3 How to read this book
2(2)
1.3.1 Examples and tutorials
3(1)
1.4 How this book was written
4(1)
1.5 Why R?
4(3)
1.5.1 R is free
4(1)
1.5.2 R does not have to be installed into system directories
5(1)
1.5.3 R is extensible
5(1)
1.5.4 R has a high-quality graphics system
5(1)
1.5.5 R allows you to share your analyses with others
6(1)
2 Basic statistics
7(38)
2.1 Who should read this chapter?
7(1)
2.2 Introduction
7(1)
2.3 Definitions
8(1)
2.3.1 Data sets, observations, and variables
8(1)
2.3.2 Types of variables
8(1)
2.3.2.1 Quantitative or qualitative
8(1)
2.3.2.2 Continuous, discrete, nominal, and ordinal
9(1)
2.4 Simple descriptive statistics
9(4)
2.4.1 Labeling the observations
10(1)
2.4.2 The sample mean, standard deviation, and variance
10(2)
2.4.3 Order statistics, medians, quartiles, and quantiles
12(1)
2.5 Summarizing data
13(14)
2.5.1 An important question
13(1)
2.5.2 Univariate data analysis
14(1)
2.5.3 Three situations
14(1)
2.5.4 Two categorical variables
14(2)
2.5.4.1 Comparing two proportions
16(1)
2.5.5 Comparing groups
17(1)
2.5.5.1 Measures of location or center
17(1)
2.5.5.2 Measures of scale or spread
18(2)
2.5.5.3 Distributional shape and other features
20(1)
2.5.5.4 Example 2.1---Comparing grouped data
21(2)
2.5.6 Two quantitative variables
23(2)
2.5.6.1 Two quantitative variables---a case study
25(2)
2.5.7 Closing remarks for the chapter
27(1)
2.6 Installing R on your computer
27(1)
2.7 Reading data into R
28(4)
2.7.1 read. csv
29(2)
2.7.1.1 Checking your data has loaded correctly
31(1)
2.7.2 Scan and Others
31(1)
2.8 The dafs package
32(1)
2.9 R tutorial
33(12)
2.9.1 Three simple things
33(1)
2.9.1.1 Tutorial
34(4)
2.9.2 R data types and manipulating R objects
38(2)
2.9.2.1 Tutorial
40(5)
3 Graphics
45(34)
3.1 Who should read this chapter?
45(1)
3.2 Introduction
45(1)
3.2.1 A little bit of language
45(1)
3.3 Why are we doing this?
46(1)
3.4 Flexible versus "canned"
46(1)
3.5 Drawing simple graphs
46(14)
3.5.1 Basic plotting tools
47(1)
3.5.2 The hitogram
47(1)
3.5.3 Kernel density estimates
47(3)
3.5.4 Box plots
50(1)
3.5.5 Scatter plots
51(1)
3.5.6 Plotting categorical data
51(2)
3.5.6.1 Plotting groups
53(1)
3.5.6.2 Pie graphs, perspective, and other distractions
53(3)
3.5.7 One categorical and one continuous variable
56(1)
3.5.7.1 Comparing distributional shape
57(2)
3.5.8 Two quantitative variables
59(1)
3.6 Annotating and embellishing plots
60(4)
3.6.1 Legends
60(1)
3.6.2 Lines and smoothers
61(1)
3.6.2.1 Smoothers
61(1)
3.6.3 Text and point highlighting
62(1)
3.6.4 Color
63(1)
3.6.5 Arrows, circles, and everything else
63(1)
3.7 R graphics tutorial
64(12)
3.7.1 Drawing bar plots
64(4)
3.7.2 Drawing histograms and kernel density estimates
68(2)
3.7.3 Drawing box plots
70(1)
3.7.4 Drawing scatter plots
71(2)
3.7.5 Getting your graph out of R and into another program
73(1)
3.7.5.1 Bitmap and vector graphic file formats
74(1)
3.7.5.2 Using R commands to save graphs
75(1)
3.8 Further reading
76(3)
4 Hypothesis tests and sampling theory
79(38)
4.1 Who should read this chapter?
79(1)
4.2 Topics covered in this chapter
79(1)
4.3 Additional reading
80(1)
4.4 Statistical distributions
80(8)
4.4.1 Some concepts and notation
80(2)
4.4.2 The normal distribution
82(3)
4.4.3 Student's t-distribution
85(1)
4.4.4 The binomial distribution
86(1)
4.4.5 The Poisson distribution
87(1)
4.4.6 The x2-distribution
87(1)
4.4.7 The F-distribution
87(1)
4.4.8 Distribution terminology
87(1)
4.5 Introduction to statistical hypothesis testing
88(23)
4.5.1 Statistical inference
88(1)
4.5.1.1 Notation
88(1)
4.5.2 A general framework for hypothesis tests
89(2)
4.5.3 Confidence intervals
91(1)
4.5.3.1 The relationship between hypothesis tests and confidence intervals
92(1)
4.5.4 Statistically significant, significance level, significantly different, confidence, and other confusing phrases
93(1)
4.5.5 The two sample t-test
94(1)
4.5.5.1 Example 4.1---Differences in RI of different glass strata
94(2)
4.5.5.2 Example 4.2---Difference in RI between bulk and near-float surface glass
96(2)
4.5.6 The sampling distribution of the sample mean and other statistics
98(4)
4.5.7 The X2-test of independence
102(2)
4.5.7.1 Example 4.3---Occipital squamous bone widths
104(1)
4.5.7.2 Comparing two proportions
105(1)
4.5.7.3 Example 4.4---Comparing two proportions relating to occipital squamous bones
106(1)
4.5.7.4 Example 4.5---SIDS and extramedullary haematopoiesis
107(1)
4.5.7.5 Fisher's exact test
107(1)
4.5.7.6 Example 4.6---Using Fisher's exact test
108(2)
4.5.7.7 Example 4.7---Age and gender of victims of crime
110(1)
4.6 Tutorial
111(6)
5 The linear model
117(94)
5.1 Who should read this?
117(1)
5.2 How to read this chapter
117(1)
5.3 Simple linear regression
118(15)
5.3.1 Example 5.1---Manganese and barium
119(2)
5.3.2 Example 5.2---DPD and age estimation
121(6)
5.3.2.1 The normal Q-Q plot
127(1)
5.3.3 Zero intercept models or regression through the origin
128(1)
5.3.4 Tutorial
129(4)
5.4 Multiple linear regression
133(18)
5.4.1 Example 5.3---Range of fire estimation
133(7)
5.4.2 Example 5.4---Elemental concentration in beer bottles
140(2)
5.4.3 Example 5.5---Age estimation from teeth
142(4)
5.4.4 Example 5.6---Regression with derived variables
146(1)
5.4.5 Tutorial
146(5)
5.5 Calibration in the simple linear regression case
151(9)
5.5.1 Example 5.7---Calibration of RI measurements
153(2)
5.5.2 Example 5.8---Calibration in range of fire experiments
155(1)
5.5.3 Tutorial
156(4)
5.6 Regression with factors
160(8)
5.6.1 Example 5.9---Dummy variables in regression
163(1)
5.6.2 Example 5.10---Dummy variables in regression II
164(2)
5.6.3 A pitfall for the unwary
166(1)
5.6.4 Tutorial
167(1)
5.7 Linear models for grouped data---One-way ANOVA
168(25)
5.7.1 Example 5.11---RI differences
170(4)
5.7.2 Three procedures for multiple comparisons
174(1)
5.7.2.1 Bonferroni's correction
174(1)
5.7.2.2 Fisher's protected least significant difference (LSD)
175(1)
5.7.2.3 Tukey's Honestly Significant Difference (HSD) or the Tukey-Kramer method
176(1)
5.7.2.4 Which method?
177(1)
5.7.2.5 Linear contrasts
178(2)
5.7.3 Dropping the assumption of equal variances
180(1)
5.7.3.1 Example 5.12---GHB concentration in urine
181(1)
5.7.3.2 An alternative procedure for estimating the weights
182(1)
5.7.3.3 Example 5.13---Weighted least squares
183(1)
5.7.4 Tutorial
183(10)
5.8 Two-way ANOVA
193(15)
5.8.1 The hypotheses for two-way ANOVA models
195(1)
5.8.2 Example 5.14---DNA left on drinking containers
196(5)
5.8.3 Tutorial
201(7)
5.9 Unifying the linear model
208(3)
5.9.1 The ANOVA identity
208(3)
6 Modeling count and proportion data
211(46)
6.1 Who should read this?
211(1)
6.2 How to read this chapter
211(1)
6.3 Introduction to GLMs
212(1)
6.4 Poisson regression or Poisson GLMs
213(6)
6.4.1 Example 6.1---Glass fragments on the ground
213(6)
6.5 The negative binomial GLM
219(15)
6.5.1 Example 6.2---Over-dispersed data
220(3)
6.5.2 Example 6.3---Thoracic injuries in car crashes
223(1)
6.5.3 Example 6.4---Over-dispersion in car crash data
223(2)
6.5.4 Tutorial
225(9)
6.6 Logistic regression or the binomial GLM
234(19)
6.6.1 Example 6.5---Logistic regression for SIDS risks
236(1)
6.6.2 Logistic regression with quantitative explanatory variables
237(1)
6.6.3 Example 6.6---Carbohydrate deficient transferrin as a predictor of alcohol abuse
237(3)
6.6.4 Example 6.7---Morphine concentration ratios as a predictor of acute morphine deaths
240(2)
6.6.5 Example 6.8---Risk factors for thoracic injuries
242(1)
6.6.6 Pitfalls for the unwary
243(2)
6.6.7 Example 6.9---Complete separation of the response in logistic regression
245(1)
6.6.8 Tutorial
245(8)
6.7 Deviance
253(4)
7 The design of experiments
257(38)
7.1 Introduction
257(1)
7.2 Who should read this chapter?
258(1)
7.3 What is an experiment?
258(1)
7.4 The components of an experiment
259(2)
7.4.1 Questions of interest?
259(1)
7.4.2 Response variables
259(1)
7.4.3 Treatment factors
260(1)
7.4.4 Experimental units
260(1)
7.4.5 Structure in experimental units
261(1)
7.4.6 Assignment of treatments to experimental units
261(1)
7.5 The principles of experimental design
261(2)
7.5.1 Replication
262(1)
7.5.2 Blocking
262(1)
7.5.3 Randomization
263(1)
7.6 The description and analysis of experiments
263(1)
7.7 Fixed and random effects
263(1)
7.8 Completely randomized designs
264(8)
7.8.1 Examples
264(1)
7.8.1.1 Block structure
265(1)
7.8.1.2 Treatment structure
265(1)
7.8.1.3 Randomization
265(1)
7.8.1.4 Analysis in R
266(1)
7.8.1.5 Factorial treatment structure
267(2)
7.8.1.6 Interaction plots
269(1)
7.8.1.7 Quantitative factors
269(3)
7.9 Randomized complete block designs
272(18)
7.9.1 Block structure
272(1)
7.9.2 Data model for RCBDs
272(1)
7.9.2.1 Example 7.1---Annealing of glass
272(1)
7.9.2.2 Treatment structure
273(1)
7.9.2.3 Block structure
273(1)
7.9.2.4 Tutorial: Analysis in R
273(3)
7.9.2.5 Example 7.2---DNA left on drinking containers
276(2)
7.9.2.6 Example 7.3---Blood alcohol determination
278(1)
7.9.2.7 Treatment structure
278(1)
7.9.2.8 Block structure
278(1)
7.9.2.9 Tutorial - analysis in R
278(3)
7.9.3 Randomized block designs and repeated measures experiments
281(1)
7.9.3.1 Example 7.4---Musket shot
282(2)
7.9.3.2 Treatment structure
284(1)
7.9.3.3 Blocking structure
284(1)
7.9.3.4 Tutorial - Analysis in R
284(6)
7.10 Designs with fewer experimental units
290(2)
7.10.1 Balanced incomplete block designs
290(1)
7.10.1.1 Example 7.5---DNA left on drinking containers II
291(1)
7.10.2 2p factorial experiments
291(1)
7.11 Further reading
292(3)
Bibliography 295(6)
Index 301(8)
Example Index 309
James M. Curran is currently an Associate Professor of Statistics in the Department of Statistics at the University of Auckland (Auckland, New Zealand). Dr. Curran is also the co-director of the New Zealand Bioinformatics Institute at the University of Auckland (www.bioinformatics.org.nz).