ADDResponse appended around 400 auxiliary variables from 20 different sources to the European Social Survey Round 6 UK dataset.
The project technical report giving details of all of the data sources used is now available.
A subset of the ADDResponse data, including ESS survey and paradata and appended small-area administrative data is available, for ESS respondents only, from the UK Data Archive via secure access. Find out more
Researchers interested in accessing the small-area administrative data used in ADDResponse for respondents and non-respondents can apply to Ipsos MORI for access. Ipsos MORI will consider all requests on a case by case basis.
The ADDResponse end of project workshop “Tackling survey nonresponse: The role of geocoded auxiliary data” took place at London Art House on on 26th May 2016.
Presentations from the event – and a summary of the expert panel discussion – are now available.
Members of the ADDResponse research team will be presenting at this event.
Location Matters: Techniques for exploring geographic variation in survey data
Monday 27th June 2016 5pm
Northampton Suite, City University London, EC1V 0HB
Geography – and particularly where people live – matters and can have a profound effect on people’s attitudes, behaviour and outcomes. However, all too often, analyses of survey data ignore this geographic variation and are restricted to ‘one size fits all’ global models. This event will provide an introduction to several techniques – including geographically weighted regression, small-area estimation and visualization – which can be used to explore low level geographic variation in survey data. Presentations will demonstrate how these techniques can be used to address a range of methodological and substantive research questions relevant to social scientists.
The event is jointly organised by the Royal Statistical Society Social Statistics Section and City University London.
The event is free to attend but please register via email@example.com
The end of project workshop “Tackling survey nonresponse: The role of geocoded auxiliary data” takes place at London Art House on Thursday 26th May 2016. Find out more.
Kaisa Lahtinen will present findings from the ADDResponse project at the GIS Research UK (GISRUK 2016) conference taking place at the University of Greenwich 30th March-1st April 2016.
Her presentation ‘Using geographically weighted regression to explore spatial variation in survey behaviour‘ will argue that the drivers of survey nonresponse vary geographically across the UK and that, therefore, relying on a “one size fits all” global model of response behaviour may be insufficient. Geographically weighted regression – which enables us to model geography as a continuous variable rather than being constrained by regional or other administrative boundaries – is one way to explore the spatial variation in nonrepsonse and provide new inisights into survey behaviour.
CALL FOR PAPERS
Special Issue of Survey Research Methods: Uses of Geographic Information Systems Tools in Survey Data Collection & Analysis
Guest Editor: Stephanie Eckman, RTI International
Guest Associate Editors:
Dr. Sarah Butt, City University London
Ned English, NORC
Dr. Robert Manchin, Consultant
(more to be named)
The application of Geographic Information Systems (GIS) and related tools such as remote sensing to survey data collection and analysis has dramatically increased in recent years. Traditionally, GIS tools have been applied primarily in the frame construction, sampling, and data collection phases of survey research. More recently, researchers have begun to use records of interviewer travel to detect falsification and to study how to make data collection more efficient. The techniques of geostatistics and geospatial models can provide new methods for studying and reducing nonresponse and measurement error. We also see surveys merging in additional variables via GIS tools: for example, data on air pollution, access to grocery stores and parks, and others can be merged in once we know respondents’ locations. However, the use of location data in surveys also raises new issues of respondents’ and interviewers’ privacy.
As these technologies become less expensive and easier to use, and geographic data becomes more widely available on the web, we expect survey researchers to find even more uses for these tools. While we embrace these tools, however, we should also maintain a healthy skepticism about their capabilities and limitations. We encourage papers on all these topics, as well as related issues.
Articles due by June 1, 2016.
To submit, go to the SRM website (http://ojs.ub.uni-konstanz.de/srm/) and upload the article as a PDF, just like a standard SRM article — be sure to mention the special issue in the field for “Author comments”.
We are presenting a poster on ‘Informing Non-Response Bias Model Creation in Social Surveys with Visualisation’ at the IEEE InfoVis conference in Chicago 25-30 October 2015. Our poster shows some of the visualisations we have done and illustrates how these might be useful in nonresponse modelling. You can have a look of our poster and abstract here.
We are taking part in the International Workshop on Household Survey Nonresponse in Leuven in September 2015. We will be presenting on the challenge of having a large pool of auxiliary data to use in nonresponse analysis, focusing on choosing between different model selection methods. The workshop will take place between 2nd and 4th September. More details on the workshop can be found from here.
We are presenting a paper at the European Survey Research Association (ESRA) conference in Reykjavik in July 2015. Our paper ‘Using geocoded auxiliary data to predict nonresponse in address-based samples: Are household- level commercial data any better than aggregate-level census data?‘ looks at using auxiliary data to predict nonresponse in social surveys and compares the usefulness of household level commercial data and aggregate level census data. We are trying to answer the question, is it worth spending money on commercial data when you could get census data for free? We are presenting in the morning of Wednesday 15th July as part of the ‘Enhancing survey data with geocoded auxiliary data 1’ session. The full conference programme can be accessed here.
We are conveying a session on Enhancing survey data with geocoded auxiliary data at the nest ESRA conference in Reykjavik in July 2015. The session invites studies that have combined survey data with geocoded auxiliary data to share their learning regarding the opportunities and challenges associated with this approach. We are interested in papers that provide insights into one or more of the following:
- The pros and cons of using different sources of geocoded auxiliary data
- Strategies for linking geocoded auxiliary data to survey data
- Modelling item or unit non-response using auxiliary data
- Combining auxiliary data and survey data cross-nationally
The deadline for paper submissions is 15th January 2015.
Find out more information on the next ESRA conference and how to submit papers here.