31 March - 1 April 2016

Saïd Business School, Oxford, UK

The Survey Octopus: Getting valid data from surveys

Caroline Jarrett Effortmark Ltd

Session type: Tutorial
Session duration: 60 minutes

Video of session

About this Tutorial

To help us get the best out of this tricky research method, Caroline will describe the Survey Octopus, a friendly creature that helps her to tackle all the issues that may lie between ""what we want to ask"" and ""who we want to ask"", and a solid, reliable number that can be used to make decisions.

Along the way, we'll encounter the key concept in survey methodology: Total Survey Error, and the various types of error that can affect your survey.

"Surveys are probably the research method that is most used in general, but UX practitioners often avoid them due to the complications and because we see so many bad surveys.

This talk introduces the challenging concept of Total Survey Error (TSE), a combination of validity, measurement error, processing error, coverage error, sampling error, non-response error and adjustment error.

By using the Survey Octopus, attendees get to know about all the choices that affect TSE but without having to get their heads around all the jargon all at once.

Caroline will also include the idea of the ‘Light Touch’ survey, a shorter, simpler type of survey that fits better into the iterative / Agile processes that most of us are now using in our design and development work.

About the Speaker

Caroline Jarrett is a user researcher and forms specialist. She helps organisations to make their forms easier to fill in, and to make business processes more effective.

She helps teams to:

  • Find out about their users through user research
  • Learn about best practices
  • Evaluate whether the ideas they’ve found on the Internet will work for them
  • Create, communicate, and evaluate their research.

Her book "Surveys that work" will be published by Rosenfeld Media in summer 2016.


Session Types

Need help planning which sessions to attend? We've provided a breakdown of our various session types below.

Case Study/Experience Report

A presentation and discussion of real-life (not theoretical) experiences of the application (or mis-application) of service design techniques. Case studies and experience reports include some discussion of lessons learned and an indication of how novel the work is.


Participants learn a new approach, tool or technology through using it to solve one or more practical exercises. Any software/hardware requirements are disclosed in the session description.


A session focused around some specific tool, technique or issue. Primarily led by the speaker, tutorials usually include some elements of interactivity or individual / group exercise.


An in-depth working session on a specific topic. May include paper presentations.

Lightning Talks

Short talks (max 10 minutes) on almost any topic.


The information for this talk hasn't yet been confirmed.