Statistical Planning
The Chapter describes main steps in the sample survey design, which include: statement of objectives, definition of target population, defining the sampling frame, formulating the sampling design and determining the sampling Intensity. This is followed by
- PDF / 1,683,626 Bytes
- 12 Pages / 439.37 x 666.142 pts Page_size
- 104 Downloads / 203 Views
Statistical Planning
6.1 The Purpose of Statistical Planning The main purpose of statistical planning is to ensure that survey estimates are reliable (i.e. they fall within the stated confidence interval); and obtained at the lowest cost for a given precision. In addition, some other benefits are the following: • Less Time: because only a fraction of the parent population is measured, the time involved in providing results is much less compared to a Census; • More Adaptability: Survey design and measurement procedures can be adapted to accessibility conditions and the available manpower; • Wider Scope: Innovative studies may be carried out by including additional measurements over only a subset of the sampled population. Fisher (1950) summarized the advantages of sampling in the following words: ‘‘I have made four claims for the sampling procedure. About the first three, adaptability, speed and economy, I need say nothing further. Too many examples are already available to show how much the method has to give in these ways. But, why do I say that it is more scientific than the only procedure with which it may sometimes be in competition, complete enumeration? The answer, in my view, lies in the primary process of designing and planning an enquiry by sampling. Rooted as it is in the mathematical theory of the errors of random sampling, the idea of precision is from the first in the forefront. The director of the survey plans from the first for a predetermined and known level of precision; it is a consideration of which he never loses sight, and precision actually attained, subject to wellunderstood precautions, is manifest from the results of the enquiry.’’
K. D. Singh, Capacity Building for the Planning, Assessment, and Systematic Observations of Forests, Environmental Science and Engineering, DOI: 10.1007/978-3-642-32292-1_6, Springer-Verlag Berlin Heidelberg 2013
51
52
6 Statistical Planning
6.2 Role of the Forest Statistician As the complexity of the survey increases, consultations with a qualified statistician will be found more and more useful: not only in the design of survey and analysis of data, but also consultations/discussions in various phases of the inventory. A quote from Cox (1951) is appropriate: ‘‘the statistician who expects that his contribution to the planning will involve statistical theory finds repeatedly that he makes his most valuable contribution simply by getting the investigator to explain why he is doing the experiment, to justify the experimental treatments and to defend his claim that the experiment, when completed will enable its objectives to be realized.’’ Besides specialized knowledge of statistics/sampling, a good knowledge of spatial variation patterns in the survey area and costs associated with alternative survey designs is required for a good planning of forest inventory. As an example, Chap. 11 presents spatial patterns of variation for tropical rainforests based on data collected from past inventories. The needed budget, manpower and available time for fieldwork ar
Data Loading...
