Introductory Workshop: Analysing Complex
Sample Design in ESDS Government Datasets
Monday 20 July 2009
University of Manchester
Introduction
Standard statistical
software commands typically treat data as simple random samples.
A large proportion of Government surveys however employ complex
sample design features such as stratification and clustering.
Rather than treating such data as simple random samples, techniques,
or approaches to analysis are required that take into account
these design features. Focussing on design-based approaches, this
introductory workshop provides grounding in elementary concepts
in complex sample design and the analysis of survey data using
Stata. A new guide to analysing complex survey design in ESDS
Government datasets is further introduced.
Overview
The morning sessions
focus on theory and statistical concepts involved in the design
and the analysis of complex surveys. In the afternoon, hands on
workshops introduce design-based approaches to survey estimation
using the ‘svy’ suite of commands in Stata. Topics
covered include producing basic descriptive statistics, sub-population
analysis, estimating design effects, and basic multivariate analysis.
Practical exemplars are given using the Health
Survey for England (HSE).
Programme
| 10.00 |
Coffee and registration |
| |
|
| 10.15 |
Introduction
to complex survey design
Anthony Rafferty,
University of Manchester |
| |
|
| 11.00 |
Coffee |
| |
|
| 11.15 |
Estimation and
complex survey design (part 1)
Ian Plewis,
University of Manchester |
| |
|
| 12.15 |
Questions and
discussion |
| |
|
| 12.30 |
Lunch |
| |
|
| 13.00 |
Estimation and
complex survey design (part 2)
Ian Plewis,
University of Manchester
Estimation
and complex survey design: Slides |
| |
|
| 13.30 |
Workshop 1
Using svyset
commands in Stata: Weighting and Clustering
Anthony Rafferty,
University of Manchester |
| |
|
| 14.30 |
Tea |
| |
|
| 14.45 |
Workshop 2
Stratification
Workshop 3
Further topics
Anthony Rafferty,
University of Manchester
Introduction
to Complex Sample Design: Workbook
|
| |
|
| 15.45 |
Wrapping up |
| |
|
| 16.00 |
End |
| |
|
Course
Requirements:
• A basic
knowledge of the user interface and commands in Stata
is required. Those unfamiliar with Stata are referred
pre-course to the first few sections of the guide on Analysing
the Labour Force Survey using Stata.
• An understanding of elementary concepts in sampling
theory and estimation will also be helpful (e.g. what
a simple random is; knowing the distinction between a
population parameter and a sample estimate (such as sample
and population means); what are standard errors and confidence
intervals).
|
|
