Online Bespoke Course on Survey Methodology

Course Description

Surveys are a key component of quantitative research whether this is for academia, research institutes, government or other organisations. Learn the art and science of creating an effective survey from initial planning through questionnaire design and testing, data collection and processing to analysis and reporting.

This course looks at each of the main steps in having a successful quantitative survey. Drawing on both practical experience and research findings from survey experiments, this course is full of insightful and pragmatic advice.

The course will take place online over 3 afternoons which start on Monday 6th October, completing on Monday 20th October, each session starts at 12.30 and finishes at 16.30. There will be x3 breaks in each session.

About the Instructor

Dr Pamela Campanelli is an international trainer, survey methods specialist and consultant. She is also a UK Chartered Statistician and Fellow of the Academy of Social Sciences. Prior to creating her own business, The Survey Coach, she has worked at the University of Michigan, the U.S. Bureau of the Census, ISER at the University of Essex, and NatCen Social Research. She regularly teaches short courses on various survey related topics in the UK and abroad for government departments, survey research companies, universities, as well as for various other institutions and businesses.

She believes in delivering courses that are lively and engaging, foster an informal and interactive atmosphere, communicate concepts in a straightforward and accessible way, illustrate the material using ‘real life’ examples, and provide workshops to put theory into practice.

Topics Covered on the Course

An overview of quantitative survey design (with DISCUSSION) (including what is involved in a survey, types of survey designs, project management, timetabling, budgeting, quality and best practice approach)

Mini appendix on duties and responsibilities of survey researchers / survey ethics

Key principles of questionnaire design - getting started with new questionnaire, trade-offs – short and simple versus clear, four cognitive stages a respondent goes through in answering a survey question, solutions to ambiguous terms, and question wording guidelines

WORKSHOP 1 on critiquing a survey question

End of day 1 appendix - Comprehension and Answer Categories - Update on Fowler (1995, p. 39), aids to improve recall (basic ones - advanced methods available upon request), and solutions to sensitive questions (basic ones - advanced methods available upon request)

A bit more about factual questions, mini appendix on actual versus usual behaviour and a bit more about subjective questions

Mini Appendix - Demographic questions are always the most difficult to write

Knowing the pros and cons of open versus closed questions (this about important less well-known issues)

Looking at the 8 key problems with agree/disagree questions

Mini appendix on other problematic question formats to avoid or be beware of (avoid satisfaction, net promoter scale, and tick all that apply; beware of ranking and hypothetical questions)

WORKSHOP 2 writing a survey question

• Mini appendix on mode differences

• The importance of knowing mode effects

• Mini appendix on mode effect examples

• Mini appendix on the questionnaire with a focus on web surveys

• Mini appendix on the visual side of web surveys

• Mini appendix on key points for web surveys on mobile phones

• Mini appendix on testing your survey questions

• What to look for in a good sample design includes: what is sampling, well-defined population, census versus sample, types of samples - probability versus nonprobability, where are your respondents coming from and the advantages of a probability sample

• Mini appendix of types of probability samples

• Mini appendix on sample standard deviation and variance, sampling standard error and variance, and the proportion as a special case of the mean

• Confidence intervals - interpretation and calculation

WORKSHOP 3 on calculating a confidence interval

• Mini appendix with more details on confidence intervals - 1.96 comes from normal distribution, why we can use numbers from the normal distribution, central limit theorem and the standard error is the standard deviation of the sampling variance

• Mini appendix on determining sample size for desired confidence interval width for probability samples

• Field work tips - Initial contact and reminders

• Understanding nonresponse / nonresponse bias - types of nonresponse, response rate calculation and interpretation, response rates versus bias, nonresponse bias formula

• Mini appendix on study design for determining who the nonrespondents are and correlates of nonresponse bias

• Minimising nonresponse before it happens - strategies aimed at the respondent - well-designed questionnaire, media / publicity, utilise sponsorship by a legitimate organisation, confidentiality assurance, keeping respondent burden down, fact sheet / hyperlink with answers to frequently asked questions (covered in session on the questionnaire as a whole), psychology of compliance / persuasion, incentives, and multiple reminders (covered in session of fieldwork tips)

• Mini appendix on cover letters

Coding open questions with WORKSHOP 4

Mini appendix on other data processing

Mini appendix on imputation

Mini appendix on introduction to weighting

Survey analysis / descriptive versus inferential (descriptive including mean / median, measures of association - tables and graphs covered in reporting session), (inferential including significance testing, chi-square test with WORKSHOP 5 - how to report confidence intervals covered in reporting session), and table and logistic regression examples from Smoking, Drinking and Drugs Study

Highlights from mini appendix on writing survey reports

Mini appendix - how the data were selected, survey methodology annex, and quality check list for reports

Learning Outcomes

Participants will:

Have a better awareness of the key aspects involved in the creation and implementation of a quantitative survey

Be able to critique aspects of existing surveys

Have knowledge to conduct or improve the quality of their own surveys

Knowledge Assumed

It is helpful, but not required, that participants have a basic understanding of producing surveys/doing research. For the sampling part of the course, it is useful for participants to have some familiarity with basic statistical concepts such as normal distribution, mean, standard deviation and standard error.