Experimental design for the life sciences / Graeme D. Ruxton, Nick Colegrave. -- 4th ed. -- Oxford ; New York : Oxford University Press, 2016. -- (58.105 /R984 /4th ed.) |
Contents
I Why you should care about design
1.1
Why experiments need to be designed
1.2
The costs of poor design
1.3
The relationship between experimental design and statistics
1.4
Why good experimental design is particularly important to life
scientists
1.5
Subjects, experimental units, samples, and terminology
Summary
2 Starting with a well-defined hypothesis
2.1
Why your study should be focused: questions, hypotheses, and predictions
2.2
Producing the strongest evidence with which to challenge a hypothesis
2.3
Controls
2.4
The importance of a pilot study and preliminary data
Summary
3 Selecting the broad design of your study
3.1
Experimental manipulation versus natural variation
3.2
Deciding whether to work in the field or the laboratory
3.3
In vivo versus in vitro studies
3.4
There is no perfect study
Summary
4 Between-individual variation, replication,
and sampling
4.1
Between-individual variation
4.2
Replication
4.3
Selecting a sample
Summary
5 Pseudoreplication
5.1
Explaining what independence and pseudoreplication are
5.2
Common sources of pseudoreplication
5.3
Dealing with non-independence
5.4
Accepting that sometimes replication is impractical
5.5
Pseudoreplication, third variables, and confounding variables
5.6
Cohort effects, confounding variables, and cross-sectional studies
Summary
6 Sample size, power, and efficient design
6.1
Selecting the appropriate number of replicates
6.2
Factors affecting the power of an experiment
6.3
Working out the power of a planned study
6.4
Improving the power of a study
6.5
Comparing the power of different planned studies
Summary
7 The simplest type of experimental design:
completely randomized, single-factor
7.1
Completely randomized single-factor designs
7.2
Randomization
7.3
Factors with more than one level
7.4
Advantages and disadvantages of complete randomization
Summary
8 Experiments with several factors
(factorial designs)
8.1
Randomized designs with more than one factor
8.2
Interactions
8.3
Confusing levels and factors
8.4
Split-plot designs (sometimes ca[ted split-unit designs)
8.5
Latin square designs
8.6
Thinking about the statistics
Summary
9 Beyond complete randomization: blocking
and covariates
9.1
The concept of blocking on a particular variable
9.2
Blocking on individual characters, space, and time
9.3
The advantages and disadvantages of blocking
9.4
Paired designs
9.5
How to select blocks
9.6
Covariates
9.7
Interactions between covariates and factors
Summary
10 Within-subject designs
10.1 What do we mean by a within-subject
design?
10.2 The advantages of a within-subject
design
10.3 The disadvantages of a within-subject
design
10.4 Isn't repeatedly measuring the same
individual pseudoreplication?
10.5 With multiple treatments,
within-subject experiments can take a long time
10.6 Which sequences should you use?
10.7 Within-subject designs and randomized
block designs
10.8 It is possible to design experiments
that mix within-subject and between-subject effects
Summary
11 Taking measurements
11.1 Calibration
11.2 Inaccuracy and imprecision
11.3 Sensitivity and specificity
11.4 Intra-observer variability
11.5 Inter-observer variability
11.6 Deciding how to measure
11.7 Pitfalls to avoid when recording data
Summary
Sample answers to self-test questions
Flow chart on experimental design
Bibliography
Index