This preview shows pages 1–5. Sign up to view the full content.
Chapter 1  Research Design Principles
1
The Legacy of Ronald A. Fisher
Section 1.1 1.3 pp 15
“
Fairly large print is a real antidote to stiff reading.
”
31 May 1929, in a letter to K.Sisam, Oxford University Press.
Printed in Natural Selection, Heredity, and Eugenics, p.20,
J.H.Bennett, Oxford: Clarendon Press, 1983.
“
To consult the statistician after an experiment is finished is
often merely to ask him to conduct a post mortem examination.
He can perhaps say what the experiment died of.”
Presidential Address to the First Indian Stat Congress, 1938.
“The analysis of variance is not a mathematical theorem, but
rather a convenient method of arranging the arithmetic.”
Discussion to ‘Statistics in agricultural research’ by J.Wishart,
Journal of the Royal Statistical Society, Supplement, 1, 2661,
1934.
Three fundamental components of field trials:
1) local control for the reduction of experimental error.
This was familiar.
2) replication for the estimation of experimental error.
This was familiar.
3) randomization for valid estimation of experimental error.
They hated this and they still do.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document Chapter 1  Research Design Principles
2
Research Hypotheses Generate Treatment Designs
Section 1.4 pp 58
Treatments  the set of circumstances created for the experiment
in response to research hypotheses.
Examples are nitrogen rate
to corn, diets to beef cattle, etc.
Experimental Unit  the physical entity or subject exposed to the
treatment independently of other units.
Experimental Error  the variation among identically and
independently treated experimental units. It is best thought of as
the background noise, but includes.
..
1. Natural variation among experimental units.
2. Variability in measurement of the response.
3. Inability to reproduced the treatment conditions exactly
from one experimental unit to another.
4. Interaction of treatments and experimental units.
5. Other extraneous factors that influence the measured
characteristic.
Chapter 1  Research Design Principles
3
Replication  the experimental unit, upon exposure to the
treatment, constitutes a single replication of the experiment.
Pseudoreplication or subsampling (The rats in cages problem.)
Factor  a particular group of treatments.
For example,
temperature, moisture and soil type are all factors.
Level  the categories of a factor.
For example, 20, 30 and 40 C
o
are all levels of the temperature factor while normal, saline and
sodic might be levels of the soil type factor.
Quantitative Factor  the levels are ordered points on some
metrical scale.
For example, temperature is a quantitative factor.
Qualitative Factor  the levels are distinct categories or
classifications that cannot be arranged in order of magnitude.
For example, soil type is a qualitative factor.
Local Control of Experimental Errors
Section 1.5 pp 816
Local control is the effort by the investigator to reduce or control
experimental error, increase accuracy of observations and
establish the inference base of a study. This is done by.
..
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document Chapter 1  Research Design Principles
4
1) Technique  be neat, be uniform, be a bit anal retentive.
This is the end of the preview. Sign up
to
access the rest of the document.
This note was uploaded on 12/26/2010 for the course CPSC 540 taught by Professor Bullock,d during the Spring '08 term at University of Illinois, Urbana Champaign.
 Spring '08
 Bullock,D

Click to edit the document details