TRAINING KEY 28: Analysing an SRS sample

Practical Methods for Design and Analysis of Complex Surveys.
Risto Lehtonen and Erkki Pahkinen

TRAINING KEY 28: Analysing an SRS sample



INTRODUCTION

In this Training Key we use the Province'91 data set as a frame population in analysing an SRSWOR sample. In part A we calculate the SRSWOR estimates for the total of UE91 (number of unemployed in the province) and for the total of LAB91 (number of work force in the province) and their corresponding standard errors from the selected sample where sample size is eight (n=8). In part B, we examine how sample size affects to the SRSWOR estimates and their standard errors by selecting different sample sizes. In part C we simulate SRSWOR samples by using Monte Carlo simulation techniques to examine the behaviour of the SRSWOR estimator. Part D is the option for interactive analysis.



A) ANALYSING AN SRS SAMPLE FROM THE PROVINCE'91 POPULATION

We calculate the SRSWOR total estimates of UE91 and LAB91, the ratio estimate (UE91/LAB91) and their corresponding standard error estimates. Further instructions will be given once you start.
 

SAS version 9
SAS/SUDAAN


B) SRS ESTIMATION WITH DIFFERENT SAMPLE SIZES

In this part, We examine how sample size affects to the SRSWOR total estimates and for the ratio estimate and their corresponding standard errors. Further instructions will be given once you start.
 

SAS version 9
SAS/SUDAAN


C) SIMULATING SRSWOR SAMPLES BY USING MONTE CARLO SIMULATION TECHNIQUES

Monte Carlo simulations are performed to examine the behavior of the SRSWOR estimator. Further instructions will be given once you start.
 

 


D) INTERACTIVE SAS USE

Please download the SAS code for your own further training. Analysing an SRS sample from the PROVINCE'91 population, SRS estimation with different sample sizes and examination of the behavior of SRSWOR estimator will be carried out. Instructions are given in the SAS code once you download.