be due to error. Therefore, differences in production between flasks were considered
to be caused by accident (i.e., error) and not by differences between the characteristics of the flasks. The outline of the branching-type ANOVA technique is shown in
Fig. 6.4.
*A two-way parameter model is a method that assumes that the amount of change
with respect to individual differences and the changes in response over time may be
expressed as constants and that the population mean of each measured value can be
expressed as the sum of these constants.
The microcosm is cultured under the same conditions allotted to either a control
system or an addition system. While identical microcosms should have originally
been generated, some differences are observed when they are actually measured
because there are many coexisting species within the original system. These differences are not based upon any fundamental characteristics, and it is thought that they
occur because of various accidental factors. Namely, such differences are assumed to
be related to various accidental products and probabilistic errors, and the experimental microcosms are randomly assigned to either a control system or an addition
system. From these randomly assigned microcosms, daily fluctuations are measured
continuously.
In contrast with the individual differences among the microcosms described above,
it is thought that some of the daily fluctuations in the microcosm data are in accordance
with principles underlying the characteristics of the microcosms. In other words, when
experimental microcosms from the original population are randomly assigned to the
control system and the addition system, it is assumed that the differences between the
flasks will fluctuate non-stochastically (i.e., nonrandomly). The branching-type
ANOVA is the analytical method for measuring 14-day data that assumes the experimental conditions shown in Fig. 6.4. Additionally, this analysis is called a
“branching-type” ANOVA (of the population from the above group of microcosms)
because it branches, as shown diagrammatically in Fig. 6.4.
Fig. 6.4 Outline of branching-type ANOVA in microcosm test
6 Estimation Using the Microcosm N-System
53
to be caused by accident (i.e., error) and not by differences between the characteristics of the flasks. The outline of the branching-type ANOVA technique is shown in
Fig. 6.4.
*A two-way parameter model is a method that assumes that the amount of change
with respect to individual differences and the changes in response over time may be
expressed as constants and that the population mean of each measured value can be
expressed as the sum of these constants.
The microcosm is cultured under the same conditions allotted to either a control
system or an addition system. While identical microcosms should have originally
been generated, some differences are observed when they are actually measured
because there are many coexisting species within the original system. These differences are not based upon any fundamental characteristics, and it is thought that they
occur because of various accidental factors. Namely, such differences are assumed to
be related to various accidental products and probabilistic errors, and the experimental microcosms are randomly assigned to either a control system or an addition
system. From these randomly assigned microcosms, daily fluctuations are measured
continuously.
In contrast with the individual differences among the microcosms described above,
it is thought that some of the daily fluctuations in the microcosm data are in accordance
with principles underlying the characteristics of the microcosms. In other words, when
experimental microcosms from the original population are randomly assigned to the
control system and the addition system, it is assumed that the differences between the
flasks will fluctuate non-stochastically (i.e., nonrandomly). The branching-type
ANOVA is the analytical method for measuring 14-day data that assumes the experimental conditions shown in Fig. 6.4. Additionally, this analysis is called a
“branching-type” ANOVA (of the population from the above group of microcosms)
because it branches, as shown diagrammatically in Fig. 6.4.
Fig. 6.4 Outline of branching-type ANOVA in microcosm test
6 Estimation Using the Microcosm N-System
53