Training Course on Analysis of Complex Sample Survey Data

INTRODUCTION

Standard courses on statistical analysis assume that survey data arise from a simple random sample of the target population. Little attention is given to characteristics often associated with survey data, including missing data, unequal probabilities of selection, stratified multistage sample designs, and measurement errors. Most standard statistical procedures in software packages commonly used for data analysis (e.g. SAS, SPSS, and Stata) do not allow the analyst to take these properties of survey data into account unless specialized survey procedures are used. Failure to do so can have an important impact on the results of all types of analyses, ranging from simple descriptive statistics to estimates of parameters of multivariate models.

This course provides an introduction to specialized software procedures that have been developed for the analysis of complex sample survey data including testing for between-group differences in means and proportions, regression analysis, logistic regression and multilevel modeling. We will also consider the consequences of non response and missing data on survey analysis and methods for dealing with missing data. Specialized procedures for survey data analysis from the Stata systems for data management and analysis will be used to develop course examples and exercises

DURATION

10 Days

WHO SHOULD ATTEND?

The course does not require rigorous training in mathematics; however, proficiency in basic mathematics, including algebra and functions, is essential. Survey sampling methods and a basic understanding of sampling concepts such as stratification, cluster sampling and weighting is required. Participants should also have familiarity with basic statistical concepts, including point estimates, sampling variance, confidence intervals, p-values, the maximum likelihood method of estimation and simple linear and logistic regression models.

TOPICS COVERED

• Survey estimation and inference for complex designs
  • Complex sample designs, survey estimation and inference
  • Multi-stage designs, stratification, cluster sampling, weighting, item missing data, finite population corrections
  • Models and assumptions for inference from complex sample survey data
  • Sampling distributions, confidence intervals
  • Design effects.
• Sampling error calculation models; ultimate clusters
• Sampling error estimation for descriptive statistics
• Replication Methods for Variance Estimation
• Estimation and inference for special statistics (percentiles, indices)
• Methods for Categorical Data
• Linear Regression Analysis
• Logistic Regression Analysis
• Multinomial, ordinal logistic regression
• Poisson and negative binomial regression
• Survival analysis and event history analysis
• Multiple imputation inference for survey data
• Multi-level models for complex sample survey data.

Others Details

The course does not require rigorous training in mathematics; however, proficiency in basic mathematics, including algebra and functions, is essential. Survey sampling methods and a basic understanding of sampling concepts such as stratification, cluster sampling and weighting is required. For more details and updates please visit website