MCET04: Engineering Data Analysis
Appendix 4
UNIT IV. DESIGN OF EXPERIMENTS AND ANALYSIS OF VARIANCE
Lesson 1. Principles of Experimental Design
Target Outcomes
At the end of the lesson, the students should be able to:
a. Know the different types of experimental design
b. Identify the best experimental design for a certain problem or study
Abstraction
In agricultural and biosystems engineering, most of the research areas deals with
applied research. And these comes in the form of an experiments. The previous lessons,
introduces us to hypothesis testing and how to interpret results made after the computations.
This unit, will introduce us to the concepts and application of experimental designs and
analysis of variance. In this lesson, we will focus more on the different types of experimental
design. Moreover, we will give emphasis more on the single-factor experiments. Furthermore,
we will be able to decipher, which type of experimental design is best suited for a specific
problem. The following are some of the typical experimental designs and how they are
performed. Most of the contents here are sourced out from Gomez and Gomez (1984).
Basic Terminologies
*The following definitions are based on the assumption that you will conduct a field
experiment
Experimental Error - the difference among the experimental plots treated alike. Example.
Even if the same fertilizer inputs or treatment is done in say, 3 plots, there will still be
difference in the yield gathered.
Replication - to obtain a measure of the experimental error, replication is done. Example. A
treatment of 120-60-30 kg/ha of fertilizer material is applied as one treatment. You should
have at least three or four experimental plots having the same treatment. This is to validate
that the results of the varying treatments are real
Randomization - done in designing the experiment to avoid bias. This ensures that each
treatment will have an equal chance of being assigned to any experimental plot
, MCET04: Engineering Data Analysis
In order to control error, blocking, proper plot techniques and data analysis are common
techniques employed in agricultural researches.
Blocking
Putting experimental units that are similar as possible together in the same group (referred to
as a block) and by assigning all treatments into each block separately and independently,
variation among blocks can be measured and removed from experimental error.
Proper Plot Techniques
All other factors aside from those considered as treatments be maintained uniformly for all
experimental units. E.g. variety trials where treatments consist solely of the test varieties, it is
required that nutrients, solar energy, plant population, pest incidence,and almost infinite
number of other environmental factors are maintained uniformly for all plots in the
experiment. But, this is almost impossible to satisfy. However, the most important ones be
watched closely to ensure that variability among experimental plots is minimized
.
Data Analysis
Proper choice of data analyis. Covariance analysis as the most commonly used. By measuring
one or more covariates - the characters whose functional relationships to the character of
primary interest are known - the analysis of covariance can reduce the variability among
experimental units bu adjusting the the values to a common value of the covariates.
Proper interpretation of the results of experiments is also a very vital element in experiment.
As this considers the whole story of your experiment. The result of the experiment in a certain
field may not be applicable or useful to other field especially if they have different ecological
or environmental aspects.
Single-Factor Experiments
Experiments in which only a single varies while all others are kept constant
Completely Randomized Design (CRD)
The treatments are assigned completely at random so that each experimental unit has the
same chance of receiving any one treatment
Only appropriate for experiments with homogeneous experimental units, such as laboratory
experiments, where environmental effects are relatively easy to control
Appendix 4
UNIT IV. DESIGN OF EXPERIMENTS AND ANALYSIS OF VARIANCE
Lesson 1. Principles of Experimental Design
Target Outcomes
At the end of the lesson, the students should be able to:
a. Know the different types of experimental design
b. Identify the best experimental design for a certain problem or study
Abstraction
In agricultural and biosystems engineering, most of the research areas deals with
applied research. And these comes in the form of an experiments. The previous lessons,
introduces us to hypothesis testing and how to interpret results made after the computations.
This unit, will introduce us to the concepts and application of experimental designs and
analysis of variance. In this lesson, we will focus more on the different types of experimental
design. Moreover, we will give emphasis more on the single-factor experiments. Furthermore,
we will be able to decipher, which type of experimental design is best suited for a specific
problem. The following are some of the typical experimental designs and how they are
performed. Most of the contents here are sourced out from Gomez and Gomez (1984).
Basic Terminologies
*The following definitions are based on the assumption that you will conduct a field
experiment
Experimental Error - the difference among the experimental plots treated alike. Example.
Even if the same fertilizer inputs or treatment is done in say, 3 plots, there will still be
difference in the yield gathered.
Replication - to obtain a measure of the experimental error, replication is done. Example. A
treatment of 120-60-30 kg/ha of fertilizer material is applied as one treatment. You should
have at least three or four experimental plots having the same treatment. This is to validate
that the results of the varying treatments are real
Randomization - done in designing the experiment to avoid bias. This ensures that each
treatment will have an equal chance of being assigned to any experimental plot
, MCET04: Engineering Data Analysis
In order to control error, blocking, proper plot techniques and data analysis are common
techniques employed in agricultural researches.
Blocking
Putting experimental units that are similar as possible together in the same group (referred to
as a block) and by assigning all treatments into each block separately and independently,
variation among blocks can be measured and removed from experimental error.
Proper Plot Techniques
All other factors aside from those considered as treatments be maintained uniformly for all
experimental units. E.g. variety trials where treatments consist solely of the test varieties, it is
required that nutrients, solar energy, plant population, pest incidence,and almost infinite
number of other environmental factors are maintained uniformly for all plots in the
experiment. But, this is almost impossible to satisfy. However, the most important ones be
watched closely to ensure that variability among experimental plots is minimized
.
Data Analysis
Proper choice of data analyis. Covariance analysis as the most commonly used. By measuring
one or more covariates - the characters whose functional relationships to the character of
primary interest are known - the analysis of covariance can reduce the variability among
experimental units bu adjusting the the values to a common value of the covariates.
Proper interpretation of the results of experiments is also a very vital element in experiment.
As this considers the whole story of your experiment. The result of the experiment in a certain
field may not be applicable or useful to other field especially if they have different ecological
or environmental aspects.
Single-Factor Experiments
Experiments in which only a single varies while all others are kept constant
Completely Randomized Design (CRD)
The treatments are assigned completely at random so that each experimental unit has the
same chance of receiving any one treatment
Only appropriate for experiments with homogeneous experimental units, such as laboratory
experiments, where environmental effects are relatively easy to control
