You are on the Natcen site

Click here for Scotcen

natcen map

You are on the Natcen site

Click here for Scotcen

natcen map

How to design a bespoke survey about pensions

Posted on 18 February 2019 by Mari Toomse-Smith, Director
Tags: methodology, pensions, surveys

The UK has an ageing population, and ageing workforce, and has introduced a range of policy changes to both State Pension and private pensions over recent years. It has never been more important to make sure that policymakers have reliable information about how the public is planning and preparing for later life. Department for Work and Pensions (DWP) recently commissioned NatCen to explore the feasibility of a survey in this area. See the report.

Firstly, we worked with the DWP researchers and stakeholders from across the government to understand what it is that they really need to know about how people are preparing and planning for later life. Our partners on this feasibility study, Daniela Silcock from the Pensions Policy Institute and Rosie Gloster from the Institute for Employment Studies, have written a useful blog about what the DWP needs to know about pensions. In summary, it is important to understand:

  • How people make decisions about retirement;
  • How they access their private pensions (there’s now more flexibility around this); and
  • How these two things interact.

Our next step was to work out the best way to obtain this information. We spent a lot of time looking at existing surveys and other data sources to see whether this information already exists. Findings showed there are some large gaps in the data that is already available or could be created by linking existing sources together. We also explored modifying some existing large-scale surveys, but the gaps were too large for that to work. It was only after this that we concluded that a new survey is indeed needed.

How should people be interviewed?

DWP needs to know a lot about one person, so we knew that the questionnaire had to be long. We also know that pensions isn’t the most exciting topic for a survey interview, so getting people to take part is unlikely to be easy. Given this, we concluded that the new survey should ideally be conducted face-to-face, as this is the way we get the highest response rates and can field longer questionnaires.

Where would the sample come from?

While retirement will at some point affect almost everybody, we concluded that the DWP would be better off focussing on those who are aged between 40 to 75. This is because people in this age range are more likely to be actively planning and preparing for later life. This makes things more difficult for a survey, however, as it means that we will have to ‘screen’ for people in the right age range, which can be costly. To avoid this, we recommended to the DWP that the new survey follows up respondents from the Family Resources Survey (FRS), face-to-face survey about household finances. This has the double benefit of allowing us to ask fewer questions and already knowing the age of the people.

How many people should the DWP talk to?

The size of the FRS sample has another benefit: it will give us a large follow up sample – up to 9,000 if we combine two years. This may seem extravagant, but is essential to ensure that we capture enough people in rarer groups, who we currently lack reliable data for. Two groups that the DWP is particularly interested in are the self-employed and carers, both of whom are at the risk of not having enough savings for later life.

Should the DWP follow up the same people?

Time was spent pondering over whether to follow up the same people. Information about the same people over time is something that the DWP needs. However, things don’t usually move very fast in people’s pension plans, so we wouldn’t need to collect information from them more frequently than in every 3-5 years. That is not frequent enough for a longitudinal survey about a topic that’s not very interesting (or salient as we say in survey world), because we would lose too many people between waves. Our recommendation was to keep the face-to-face survey cross-sectional, meaning that each face-to-face survey would start with a new sample. For gathering information about changes over time we proposed two approaches: annual or even twice-yearly web-telephone follow up surveys between large face-to-face waves to address burning policy questions and working with existing longitudinal surveys (such as the Understanding Society and the English Longitudinal Study for Ageing) to make sure those capture the most important longitudinal information about retirement.

To sum it all up

Our analysis of what information is available shows that the best way to address information needs around planning and preparing for later life is through a bespoke new survey. The baseline of findings from a survey will enable the development of policy and practice to support an ageing workforce.

comments powered by Disqus
Blog filters
  • 2022
  • 2021
  • 2020
  • 2019
  • 2018
  • 2017
  • 2016
  • 2015
  • 2014
  • 2013
  • 2012
  • 2011
  • 2010
  • Jan
  • Feb
  • Mar
  • Apr
  • May
  • Jun
  • Jul
  • Aug
  • Sep
  • Oct
  • Nov
  • Dec
Clear Filters

Subscribe to the RSS Feed: