Search for notes by fellow students, in your own course and all over the country.
Browse our notes for titles which look like what you need, you can preview any of the notes via a sample of the contents. After you're happy these are the notes you're after simply pop them into your shopping cart.
Document Preview
Extracts from the notes are below, to see the PDF you'll receive please use the links above
Chapter 1
Data and Models
1
...
2
...
4
...
6
...
Collect appropriate data
...
Initial examination of the data (to include descriptive and graphical techniques)
...
Interpretation and communication of the results and conclusions
...
The distributional assumptions should be considered at this stage, e
...
, are the data
Normally distributed? The relationships between different variables may be examined
by graphical means
...
This is usually
based on a statistical model; the model being chosen because it ‘best’ describes the data
...
2
Experiments and Surveys
1
...
1 Surveys
Surveys are conducted to discover respondents’ attitudes, beliefs or opinions
...
1
...
2 Experiments
In a survey, the respondents and their opinions are sought and recorded but there is no
control over either the respondents or their opinions
...
1
...
3 Quasi-experiments
Quasi experiments are experiments where the experimenter does not have complete
control over the allocation of treatments to participants
...
2
...
g
...
In this example, a controlled experiment would
not be ethical- to perform a controlled experiment, expectant mothers would have to be
randomly assigned to a no alcohol group, low alcohol group, moderate alcohol group,
etc
...
1
...
Measurements are made on a scale and the type of scale
will, to some extent, influence the way in which the data are analysed
...
3
...
eg
...
The data categories should be mutually exclusive (i
...
, no overlap)
The data categories are labels and do not represent any order
...
3
...
Again the alternatives should be mutually
exclusive
...
g
...
10 scale
However it is unlikely that the difference between points 1 and 2 on the depression
scale represents exactly the same thing as a difference between 0 and 1
...
3
...
categories are mutually exclusive
(as for nominal and ordinal)
2
...
categories are scaled according to the characteristic (as for ordinal)
4
...
5
...
The standard example of this type of scale is temperature
...
3
...
The
properties of the ratio scale are as those of the interval scale with this additional
2
property
...
3
...
2
...
Categorical, discrete, continuous
...
4 Principles of Experimental Design
The features underpinning experimental design include:
·
·
·
·
Randomisation
Replication
Controls
Blocking
Randomisation implies the random allocation of treatments to subjects
...
Replication enables the experimental error or variability to be measured
...
Blocking is a type of design whereby ‘nuisance’ factors are allowed for in the
experiment thus allowing the effects of the treatments to be measured more precisely
...
The overall objective when designing an experiment is to measure the effect of a
treatment as precisely as possible using the minimum resources (resources include time
and number of experimental subjects)