Joint Programming Initiative

More Years, Better Lives

The Potential and Challenges of Demographic Change

Panel Survey of Ageing and the Elderly (PSAE)
PSAE-Panel Survey of Ageing and the Elderly

Social Systems and Welfare
Relevance for this Topic
Country Sweden
More Topics


Contact information

Professor Mikael Stattin
Umeå University
901 87 Umeå
Phone: +46 90 786 57 14
Email: Mikael.stattin(at)

Timeliness, transparency

Around one year

Type of data

Registry + Survey

Type of Study

Longitude survey: long-term study of the same sample

Cross-section, regular

Data gathering method

Telephone interview (CATI)

Self-administered questionnaire

Access to data

On site access only data available only for scientific community, etc.

Conditions of access

Agreements with project leader

Few months.

Mainly anonymised microdata, aggregated tables, etc.

Dataset is compatible; mainly Excel, SAS, SPSS, STATA, Text, etc.

Data available in Swedish, but concepts/terms are translated


PSAE 2002/2003 • The annual cross-sectional ULF sample 2002-03 • Re-interview of the ULF panel going back to 1979, 1986-87, and 1994-95. • Re-interview of the PSAE extra sample first interviewed as part of the ULF 1994-95 cross-sectional sample PSAE 2010/2011 • The annual cross-sectional ULF sample 2010-11 • Re-interview of the ULF panel going back to 1979, 1986-87, 1994-95, and 2002-03 • Re-interview of the extra PSAE sample from 1994-95 and 2002-03 • Re-interview of additional PSAE extra sample first interviewed as part of PSAE 2002-02 cross-sectional sample


Yes, e.g. age, sex, region, etc.

Technically the PSAE is an integrated part of Statistics Sweden’s annual Survey of Living Conditions (ULF). Hence, the PSAE is a comprehensive survey of living conditions that covers the following arenas: household situation, housing and neighbourhood, employment, work life and work environment, education and training, incomes and economic standard, somatic health, psychosocial wellbeing, need of care, leisure activities, family- and social relations, political resources, and victimisation and fear of crime. Each of these arenas is covered by an extensive number of questions.

PSAE covers the whole country

The PSAE share sample with ULF which also means that ULF:s panel is integrated in the PSAE. Almost half of the annual ULF sample is part of panel that is re-interviewed every eight year. The other part of the ULF sample is made up of a cross-sectional sample. The ULF panel is integrated into the PSAE, which means that PSAE has a panel that were first interviewed in 1979, were re-interviewed a second time in 1986-87, a third time in 1994-95, a fourth time in 2002-03, and finally a fifth time in 2010-11. Thus, part of the PSAE provides detailed individual data that cover more than 30 years. In addition, compared to ULF the PSAE sample is boosted with an extra sample of 2,000 individuals in the 65+ age category. The extra sample is drawn from the cross-sectional part of the earlier ULF sample from 1994 and 1995, which turns this sample into a panel, and is included in both the 2002-03 and 2010-11 PSAE. In the 2010-11 round of PSAE we have, in order to compensate for attrition (mainly caused by mortality) filled up the extra sample with individuals from the 2002-03 ULF cross-sectional sample. The effective sample size for PSAE 2002-03 was 12,685 for the total population and 5,374 in the 55+ population. We expect approximately the same effective sample size for the 2010-11 wave (exact figures are not yet available). Finally register data from the Longitudinal Integration Database for Health Insurance and Labour Market Studies (LISA) have been added to the PSAE. These data, that gives detailed annual information about incomes and to some degree employment conditions, are currently covering the period 1991 to 2008 but the plan is to continuously update the register part of the PSAE.

Compared to a regular ULF survey the PSAE includes a substantial extension of the questionnaire for the age category 55+. The extension mainly concerned work related issues (attitudes towards work and retirement, work life history, and working conditions), health, psychosocial wellbeing, need of care, and social relations. Full information about the PSAE questionnaire is available at:

• Halleröd, B., Örestig, J., Stattin, M. “Leaving the labour market: the impact of exit routes from employment to retirement on health and wellbeing in old age”. European Journal of Ageing 10(1) (2013): 25-35. • Hult, C., Stattin, M. “Age, policy changes and work orientation: comparing changes in commitment to paid work in four European Countries”. Journal of Population Ageing, 2(3) (2009): 101-120. • Hult, C., Stattin, M., Janlert, U., et al. “Comparing mortality rates and recognizing health selection bias: A response to Wallman and Svärdsudd”. Social Science and Medicine 70(10) (2010): 1489-1491. • Hult, C., Stattin, M., Janlert, U., et al. “Timing of retirement and mortality: A cohort study of Swedish construction workers”. Social Science and Medicine 70(10) (2010): 1480-1486. • Nordenmark, M., Stattin, M. “Psychosocial wellbeing and reasons for retirement in Sweden”. Ageing & Society, 29(3) (2009): 413-430. • Örestig, J., Strandh, M., Stattin, M. “A wish come true?: A Longitudinal Analysis of the Relationship between Retirement Preferences and the Timing of Retirement”. Journal of Population Ageing (2013). • Stattin, M. “Pension preferences and work environment”. Promotion of Work Ability towards Productive Aging (2009): 143-154. • Stattin, M. ”Den äldre arbetskraften - deltagande, attityder och arbetsförhållanden: (Should I stay or should I go?)” 2009 • Stattin, M. ”Pensionspreferenser och arbetsvillkor bland den äldre arbetskraften”. 2006 • Stattin, M., Robroek, S., Schuring, M., et al. “Poor health, unhealthy, and unfavorable work characteristics unfluence on exit from paid employment among older workers in Europé: a four year follow-up study”. Scandinavian Journal of Work, Environment and Health (2012).


PSAE is linked to EU-SILC.

PSAE is built on Swedish Survey on Living Conditions and does also use data from LISA.

Data quality

Already at the start of the investigation the aim was to also follow up would be implemented and in the years 2010 and 2011, a replica of 2002/03-study in which most of the questions asked in 2002/03 was repeated. Between both survey years, however, changed data collection method when SCB was transferred from visits to telephone interviews. This has impacted on the way the questions and especially how answer options were designed. Most of the additional questions that were asked earlier respondents in 2002/02 has been retained in the data collection 2010/11.

The sample size has also changed so that the latter PSAE study holds fewer respondents than the former. This has implications primarily in terms to degrade performance reports in several subgroups. Both surveys have a certain loss. For ULF survey 2002/2003 the response rate was 74.9 percent and the survey 2010/2011, 60.2 percent. However, the response rate was somewhat higher among those who were part of the PSAE-specific selections. SCB has to obtain population representativeness calculated weights specific to each survey round. All results in this report are designed to weighted data.

No major changes in terminology during the project.


The PSAE was first launched in 2002 and 2003 (FAS Dnr 2001-1148) and recently a second wave of data collection has been finalized and the new data, collected in 2010 and 2011 (FAS Dnr 2009-1989), will be available at the beginning of June this year. The PSAE, managed by Björn Halleröd and Mikael Stattin (both participants in this program), was from the outset designed to provide the data necessary for the analysis specified in this program. Thus, it is not a coincident that the PSAE can be used in most of our WP:s it is a consequence of a long term investment in data collection. See above. Strength of PSAE is that it combines various registries and also focus on the work retirement transition which has high policy relevance in times of work prolongation and job longevity.

  • The information about this dataset was compiled by the author:
  • Kenneth Abrahamsson
  • (see Partners)