All time references are in CEST
Surveying Older People in Times of Crisis: Methodological Issues 3
| Dr Michael Weinhardt (DZA German Centre of Gerontology)
Dr Jan-Lucas Schanze (GESIS – Leibniz-Institute for the Social Sciences)
Dr Michael Bergmann (Survey of Health, Ageing and Retirement in Europe (SHARE))
Dr Annette Scherpenzeel (Netherlands Institute for Health Services Research (Nivel))
Dr Stefan Stuth (DZA German Centre of Gerontology)
Dr Julia Simonson (DZA German Centre of Gerontology)
|Tuesday 18 July, 16:00 - 17:00
European societies are undergoing difficult times, where the aftermath of the COVID-19-pandemic and the current energy crisis add to already existing challenges, such as drought and heat waves due to climate change. Many of these challenges impact the ageing populations in Europe in particular ways and surveys are essential to assess the life situations of the older population. National and international surveys that target and cover the ageing populations can deliver this important knowledge but, at the same time, face a range of very specific methodological issues. These include, but are not limited to:
- fieldwork management, contact strategies and participant engagement for older target groups
- age-related coverage errors of different sampling frames
- the institutionalised older population: undercoverage and interview challenges
- nonresponse bias and measurement error due to declining cognitive and physical (listening, reading) abilities
- restricted use of modes (online, mobile) and mixed-mode surveys for older people
- age-related interview challenges and the need for specific interviewer skills and training
- the use of proxy interviews or 'triad interviews' and consequences for data quality
- conversion of gate keepers who refuse / hesitate to let an older person be interviewed
- tracking /tracing of panel members, e.g., for people moving to institutions or the deceased
- data linkage: developments and new sources, for example to public health / death records
- age differences in the quality of data in face-to-face or self-administered interview modes
- the question of “vulnerability” of older respondents and ethical issues in general
- new survey tools to improve data collection and quality among older people
We invite papers addressing these or related issues of surveying the older population. We also invite papers on the adjustments necessary to survey older respondents during the recent pandemic and the consequences they had on the various aspects of total survey error.
Professor Emanuela Sala (University of Milano Bicocca)
Dr Giulia Melis (University of Cagliari) - Presenting Author
Dr Daniele Zaccaria (SUPSI)
One of the main challenges of longitudinal research is panel attrition, i.e. the reduction in wave-on-wave survey response. Developing effective sample maintenance strategies is therefore key to maintaining both sample representativeness and sample size. This is especially the case in surveys of older people, which are also more likely to be affected by participants' “natural” death. Despite the relevance of the topics, there is a shortage of research on participant engagement for older target groups. Against this background, this paper provides an evaluation of the sample maintenance strategies implemented in the course of wave 1-3 of the Italian (qualitative) Longitudinal Study on Older People’s Quality of Life during the Covid-19 pandemic (ILQA-19). Specifically, the sample maintenance strategy we have developed is based on 1. the production of short videos with the team members that are sent to participants during the festive seasons 2. sending personalized text messages, via WhatsApp, and 3. the organisation of local meetings with the participants. To evaluate the effectiveness of the sample maintenance strategy, we shall look at the qualitative feedback we received from the study participants (via WhatsApp) after sending the video materials and the personalised text messages and at the wave-on-wave response rate. We shall evaluate the lessons learnt from this qualitative longitudinal study in the context of large scale surveys of older people.
Professor Patrick Kutschar (Paracelsus Medical University) - Presenting Author
Professor Martin Weichbold (Paris Lodron University Salzburg)
Professor Juergen Osterbrink (Paracelsus Medical University)
Background and Aim: Face-to-face surveys are applied frequently for nursing home residents (NHR) with cognitive impairment (CI)—a vulnerable population in which care decisions should build upon high-quality empirical evidence. Interviewers (INT) are of particular consideration for data quality and Gender-of-Interviewer (GOI) may affect item nonresponse (INR) or validity of answers. It was examined whether gender-based interviewer-respondent-dyads lead to a) different INR, and b) differences in each items’ response distribution (RD).
Methods: Data come from the registered cRCT 'PIASMA’ (2016-2018): 598 NHR with no/mild (MMSE 18-30) and moderate CI (MMSE 10-17) of 15 German NH were interviewed about pain, depression, and quality of life. ANCOVA was applied to test effects of interviewer-respondent-dyads on INR; differences in RD were transferred into comparable effect sizes (Cohen d,w,f).
Results: INR were highest in case female NHR were interviewed by female INT (INR=5.7%) and fewest in male dyads (INR=3.3%). ANCOVA shows weak effects of GOI on INR (eta-square=0.9%, p<0.05). NHR’ responses significantly vary by dyads. Pain frequency reports are higher when NHR were interviewed by female INT (p<0.010, w=0.33). Female NHR surveyed by male INT report a poorer quality of life (p<0.05, d=0.23). In male NHR, RD show less pain (p<0.05, d=0.20) and less frequent anxiety/depression states (p<0.001, d=0.42) if questioned by male INT.
Discussion: INR and RD vary by gender-dyads. These effects seem to be moderated by questions’ saliency and residents’ cognitive impairment. GOI appear to initiate gender-specific question-answer-processes, represent a considerable source of error, and should be accounted for in future studies with older populations.
Dr Nicole Watson (University of Melbourne)
Dr Mossament Nesa (University of Melbourne) - Presenting Author
Tracking people through various life stages is an important part of longitudinal studies. One such stage is the transition to retirement and old age. We use household panel survey data from Australia (that is, the Household, Income and Labour Dynamics in Australia (HILDA) Survey) collected over the last 21 years to examine the representativeness and data quality of sample members across three age groups: 55-64, 65-74, and 75 and over. We compare population estimates for health, labour market and demographic variables from the HILDA Survey to those from a range of official sources. We also examine various aspects of data quality over time, such as re-interview rates, response to the self-completion questionnaire (which is in addition to the individual interview), item non-response, straight lining and rounding. In addition, we track the success of following and interviewing people in nursing homes and the like and flagging deaths over time.
Mr Tim Kuttig (Robert Koch-Institute) - Presenting Author
Dr Judith Fuchs (Robert Koch-Institute)
Ms Hanna Perlitz (Robert Koch-Institute)
Mrs Christin Wolff (Robert Koch-Institute)
Mr Hans Butschalowsky (Robert Koch-Institute)
Dr Christa Scheidt-Nave (Robert Koch-Institute)
Dr Beate Gaertner (Robert Koch-Institute)
Old and impaired persons are often excluded from or underrepresented in health surveys. Health 65+ is a national health interview and examination survey in Germany with a focus on the COVID-19 pandemic. It is tailored to the needs of the population 65 years and older. We assess the effects of a mixed-mode design on contact and response rates and sample composition for the baseline interviews.
12248 individuals 65+ years were drawn from 128 local population registries through a two-stage stratified random sampling procedure and invited to take part in the study. A previously tested three-step procedure (letter, telephone, home visits) was applied to contact the study population between June 2021 and April 2022. Baseline participation comprised answering a self-completed questionnaire or an interview. Proxy-interviews and written consent from legal guardians were also possible. Contact and response rates were calculated for the adjusted gross sample. Sample composition was analyzed according to data available for the total sample; register-based information (e.g. age, German citizenship) and further information (e.g. nursing home resident and purchasing power according to postal address).
Contact and response rates after the initial invitation letter were 10.8% and 8.8%, respectively. Both rates increased with each additional contact step, to finally 84.5% and 30.9%. With each contact step, deviations in the sample composition of participants (e.g. in terms of age, German citizenship, purchasing power) relative to the composition of the adjusted gross sample decreased. Home visits were especially effective at including nursing home residents.
A sequential mixed-mode approach is effective at including hard-to-reach groups such as older and impaired individuals in population-based health studies. Furthermore, it reduces deviations in the sample composition of participants relative to the gross sample.