|
Why you need to care about design |
|
|
|
Why experiments need to be designed |
|
|
1 | (2) |
|
|
|
3 | (2) |
|
|
|
3 | (1) |
|
|
|
4 | (1) |
|
The relationship between experimental design and statistics |
|
|
5 | (1) |
|
Why good experimental design is particularly important to life scientists |
|
|
5 | (4) |
|
|
|
6 | (1) |
|
|
|
6 | (1) |
|
|
|
7 | (2) |
|
Starting with a well-defined hypothesis |
|
|
|
Why your experiment should be focused: questions, hypotheses and predictions |
|
|
9 | (5) |
|
An example of moving from a question to hypotheses, and then to an experimental design |
|
|
11 | (1) |
|
An example of multiple hypotheses |
|
|
11 | (3) |
|
Producing the strongest evidence with which to challenge a hypothesis |
|
|
14 | (3) |
|
|
|
15 | (1) |
|
Considering all possible outcomes of an experiment |
|
|
16 | (1) |
|
Satisfying sceptics: the Devil's advocate |
|
|
17 | (1) |
|
The importance of a pilot study and preliminary data |
|
|
18 | (4) |
|
Making sure that you are asking a sensible question |
|
|
18 | (2) |
|
Making sure that your techniques work |
|
|
20 | (2) |
|
Experimental manipulation versus natural variation |
|
|
22 | (8) |
|
An example of a hypothesis that could be tackled by either manipulation or correlation |
|
|
22 | (1) |
|
Arguments for doing a correlational study |
|
|
23 | (2) |
|
Arguments for doing a manipulative study |
|
|
25 | (3) |
|
Situations where manipulation is impossible |
|
|
28 | (2) |
|
Deciding whether to work in the field or the laboratory |
|
|
30 | (2) |
|
In vivo versus in vitro studies |
|
|
32 | (1) |
|
There is no perfect study |
|
|
33 | (3) |
|
|
|
34 | (2) |
|
Between-individual variation, replication and sampling |
|
|
|
Between-individual variation |
|
|
36 | (1) |
|
|
|
37 | (6) |
|
|
|
43 | (11) |
|
Explaining what pseudoreplication is |
|
|
43 | (2) |
|
Common sources of pseudoreplication |
|
|
45 | (4) |
|
Dealing with pseudoreplication |
|
|
49 | (2) |
|
Accepting that sometimes pseudoreplication is unavoidable |
|
|
51 | (1) |
|
Pseudoreplication, third variables and confounding variables |
|
|
52 | (1) |
|
Cohort effects, confounding variables and cross-sectional studies |
|
|
53 | (1) |
|
|
|
54 | (6) |
|
Why you often want a random sample |
|
|
55 | (1) |
|
|
|
56 | (1) |
|
|
|
57 | (1) |
|
Some pitfalls associated with randomization procedures |
|
|
58 | (1) |
|
Randomizing the order in which you treat replicates |
|
|
58 | (1) |
|
Random samples and representative samples |
|
|
59 | (1) |
|
Selecting the appropriate number of replicates |
|
|
60 | (9) |
|
|
|
61 | (1) |
|
|
|
61 | (1) |
|
Factors affecting the power of an experiment |
|
|
62 | (1) |
|
Relationship between power and type I and type II errors |
|
|
63 | (5) |
|
|
|
68 | (1) |
|
Different experimental designs |
|
|
|
|
|
69 | (7) |
|
Different types of control |
|
|
70 | (2) |
|
|
|
72 | (1) |
|
Making sure that the control is as reliable as possible |
|
|
73 | (1) |
|
The ethics of controlling |
|
|
74 | (1) |
|
Situations where a control is not required |
|
|
75 | (1) |
|
Completely randomized and factorial experiments |
|
|
76 | (7) |
|
Experiments with several factors |
|
|
77 | (1) |
|
Confusing levels and factors |
|
|
78 | (3) |
|
Pros and cons of complete randomization |
|
|
81 | (2) |
|
|
|
83 | (6) |
|
Blocking on individual characters, space and time |
|
|
85 | (1) |
|
The pros and cons of blocking |
|
|
86 | (1) |
|
|
|
87 | (1) |
|
|
|
87 | (1) |
|
|
|
88 | (1) |
|
|
|
89 | (6) |
|
The advantages of a within-subject design |
|
|
90 | (1) |
|
The disadvantages of a within-subject design |
|
|
91 | (1) |
|
Isn't repeatedly measuring the same individual pseudoreplication? |
|
|
92 | (1) |
|
With multiple treatments, within-subject experiments can take a long time |
|
|
93 | (1) |
|
Which sequences should you use? |
|
|
94 | (1) |
|
Split-plot designs (sometimes called split-unit designs) |
|
|
95 | (2) |
|
Thinking about the statistics |
|
|
97 | (5) |
|
|
|
100 | (2) |
|
|
|
|
|
|
102 | (1) |
|
Inaccuracy and imprecision |
|
|
103 | (2) |
|
Intra-observer variability |
|
|
105 | (4) |
|
|
|
105 | (1) |
|
|
|
106 | (1) |
|
|
|
106 | (2) |
|
Remember, you can be consistent but still consistently wrong |
|
|
108 | (1) |
|
Inter-observer variability |
|
|
109 | (1) |
|
|
|
109 | (1) |
|
|
|
109 | (1) |
|
|
|
110 | (1) |
|
|
|
110 | (2) |
|
|
|
112 | (2) |
|
Don't try to record too much information at once |
|
|
112 | (1) |
|
Beware of shorthand codes |
|
|
112 | (1) |
|
Keep more than one copy of your data |
|
|
113 | (1) |
|
Write out your experimental protocol formally and in detail, and keep a detailed field journal or lab book |
|
|
113 | (1) |
|
|
|
113 | (1) |
|
Computers and automated data collection |
|
|
114 | (1) |
|
Floor and ceiling effects |
|
|
114 | (2) |
|
|
|
116 | (1) |
|
Taking measurements of humans and animals in the laboratory |
|
|
116 | (3) |
|
|
|
117 | (2) |
|
|
|
|
How to select the levels for a treatment |
|
|
119 | (2) |
|
Subsampling: more woods or more trees? |
|
|
121 | (2) |
|
Using unbalanced groups for ethical reasons |
|
|
123 | (2) |
|
|
|
125 | (4) |
|
|
|
126 | (1) |
|
|
|
126 | (1) |
|
|
|
127 | (2) |
|
|
|
129 | (1) |
|
|
|
130 | (4) |
|
Covariates can interact too |
|
|
130 | (2) |
|
The importance of interactions (and the interaction fallacy) |
|
|
132 | (2) |
|
Dealing with human subjects |
|
|
134 | (8) |
|
|
|
136 | (1) |
|
Collecting data without permission |
|
|
137 | (1) |
|
|
|
137 | (1) |
|
|
|
138 | (1) |
|
|
|
138 | (1) |
|
|
|
139 | (1) |
|
|
|
139 | (1) |
|
There is no perfect study: a reprise |
|
|
140 | (1) |
|
|
|
140 | (2) |
| Sample answers to self-test questions |
|
142 | (11) |
| Flow chart on experimental design |
|
153 | (5) |
| Bibliography |
|
158 | (3) |
| Index |
|
161 | |
Other versions by this Author
