**A Software to chosee between **

**Sampling Probabilities Plans **

Espace oriented to statistical sampling, sampling plan, survey sampling, sampling techniques, sampling methods: systematic sampling, stratified sampling, random sampling, probabilities proportional to size.

POSDEM, allows for several sampling methods, answering among other questions: what is the best sampling plan for a specific population frame?; which is the sampling size related for a significance level?; what is the mean square error expected?; what are the units that must be investigated?; what happens if ...?

The software is available; to be used in research projects, for educational purposes and in the preparatory legwork of the surveys based on samples. Allows one to carry out sampling plans adapted to each investigation and incorporating the expert sampling knowledge, without the person carrying out the sampling design, necessarily needing that knowledge At the same time the help files have a pedagogic purpose and could be useful like a book about principles of sampling techniques.

Software, POSDEM,
permits the evaluation of sampling methods. To inform about citation, authors or problems visit. POSDEM authors and citation page |

This program is distributed free, writing to: colindres10@hotmail.com POSDEM allows to work with populations of up to 30,000 units. The authors of POSDEM decline all
responsibility for the use and the consequences that
could be made of this program on the part of the users. The POSDEM application runs 386 or superior, low Windows 3.1, 3.11 or Windows 95 and superiors in personal computers with processor. They are needed a minimum of 8 Megabytes of RAM memory and 5 free Megabytes in the hard disk. Do you want to see some screens of POSDEM?
Starting with a population frame it allows to generate defined variables to measure and generate random populations under a wide range of possibilities: randomness, dispersion, tendency and cycle. So that it is possible to simulate a great variety of populations that one could find in the practice of the sampling surveys. This possibility is used mainly inside the focus of superpopulation models. The program has seventeen methods of selection and twenty methods of different estimate. From the procedures of selection ,there exists two well differentiated groups: first group with twelve procedures that have in common that the selection is carried out with equal probabilities, and one second group of three methods with unequals probabilities, sometimes proportional to the value of the variable and another proportional to an auxiliary variable, generally the size. The program calculates the variance, typical deviation, variation coefficient and confidence limits for all the estimators calculated in each sample. So we have the possibility to studying the variance of the estimator and also the variance of the variance also under a superpopulation aproach.. The application uses by default quantitative variables, although it also allows to process qualitative data, and different kinds of estimators. Inside the philosophy of the program, a basic point is that it permits to carry out comparisons between several methods of selection, in order to choose the most convenient to certain structure of the population. The program allows to study the results of each experiment in analytical or graphical way. |