“One way to ruin an otherwise well-conceived survey is to use a convenience sample rather than one which is based on a probability design” (Ferber et al, 1980). Do you have the right sampling design for your study?
This course introduces participants to what survey sampling is, why it is important, and how it is implemented. It focuses on the practical aspects as well as some of the mathematics.
● Types of samples (probability versus non-probability)
● How to construct a ‘sampling frame’
● Types of probability samples (e.g., simple random, systematic, stratified, multi-stage clustered, unequal probabilities of selection).
● What ‘sampling error’ is
● The role of sampling error in confidence intervals
● How to determine sample size
● A very brief introduction to the effects of different types of sample designs on confidence intervals
● Introduction to weighting
PARTICIPANTS SHOULD BRING A CALCULATOR WITH A SQUARE ROOT FUNCTION.
By the end of the course, participants will:
● Recognise the strengths and weaknesses of different sampling strategies
● Understand confidence intervals for means and proportions and how to select a sample size which will guarantee the desired width of the confidence interval after data collection
● Be able to critique aspects of existing survey samples
● Have introductory knowledge about how to draw their own survey sample
Who will benefit?
● The course will benefit anyone who wishes to conduct their own survey or commission a survey.
● Ideally participants should have had at least some prior experience of statistics.
● Participants will have a good awareness of the key aspects of designing a survey sample.
The day will be led by Dr Pamela Campanelli. Pamela is a Survey Methods Consultant and U.K. Chartered Statistician and Chartered Scientist. She received her Ph.D. in statistics from the London School of Economics, and an M.A. in survey research methods and B.A. in psychology from the University of Michigan. Prior to becoming an independent consultant, she was a Research Associate at the at the University of Michigan, a Survey Statistician at the U.S. Bureau of the Census, Chief Research Officer at the UK Institute for Social and Economic Research at the University of Essex, and a Research Director at the Survey Methods Centre at the National Centre for Social Research, London.
Her main interests and publications are in the study of survey error and data quality issues, with special emphasis on questionnaire design, question testing strategies, survey sampling and survey analysis. In addition to her consultancy work, she has led a UK ESRC grant on survey nonresponse and been a team member of a UK ESRC grant to explore measurement error in mixed mode surveys. She regularly teaches short courses in the UK for the SRA, the Cathie Marsh Institute for Social Research and the Royal Statistical Society. She also provides courses both in the UK and internationally for government departments (e.g., Australian Bureau of Statistics), UK Survey Research companies (e.g., TNS-BMRB), universities (e.g., University of Wollongong Australia, University of Hong Kong, University of Michigan Summer Institute), as well as for various other institutions and businesses (e.g., Brazilian Network Information Center; UNESCO Asia Pacific, Civil Service College Singapore) (see www.thesurveycoach.com).
This course contributes 6 hours to the MRS CPD programme
Price for SRA members: £195. Use the discount code emailed to you.
If you are unsure of your code, or haven't received one, please contact us.
Registration on the day is from 9.00 - 9.30 am
Cancellation policy: Full refund if cancelled 21 days or more from the course date. No refund if cancelled 20 days or fewer from the course date.
Registrations for this course will not be accepted after 30 November 2017 at noon.