A historical review was given of the development of rice cultivation in Java, of the influence of the Netherlands colonial government on the agricultural methods by advice before and after local research, and of the development of field trial techniques and the organization of systematic research. The principle of randomization was not applied at that time. Tests of significance were made by comparing a given difference with three times its estimated standard deviation. To reduce standard deviation the elimination of a linear trend in fertility for different places was discussed in detail. From the results of field experiments between 1911 and 1926 the relation between expected yield and the standard deviation was studied for different soil types and only the data were used of trials, where fertility of the area was even enough to be chosen as the place for a trial. This relation seemed to be reasonably linear. Attention was paid to the power of a test in relation to standard deviation. In the period considered investigators were more concerned with lack of bias in estimators than their efficiency. In this thesis the importance of efficiency is stressed and more insight is given into the possibilities offered by field experiments.
There are no comments yet. You can post the first one!
Post a comment
Please log in to use this service. Login as Wageningen University & Research user or guest user in upper right hand corner of this page.