Get in Touch–Stay in Touch: Trust as a Core Issue For Success of Longitudinal Studies |
|
| Session Organisers |
Dr Götz Lechner (Leibniz Institute for Educational Trajectories (LIfBi)) Mr André Müller-Kuller (Leibniz Institute for Educational Trajectories (LIfBi)) Dr Roman Auriga (Leibniz Institute for Educational Trajectories (LIfBi)) |
| Time | Wednesday 17th July, 11:00 - 12:30 |
| Room | D25 |
This session–as part of a series established during the last ESRA–focuses on central methodological questions and advancements dealing with panel performance. The focus is on access to target persons and aspects of panel maintenance in the broadest sense.
Over the last decades, sophisticated survey-methodological possibilities for surveying have increasingly emerged alongside data-driven methods and digitization. On the other hand, public confidence in data-processing operations as well as in abstract systems in general (such as science, economy, politics, or technology) declined. Hence, there are challenges for survey processes that should be addressed in this session.
One of the main challenges for sustainable panel performance nowadays arises from structures that threaten the privacy or the expected outcome of action in participants’ everyday lives. Big data sampling structures with opaque and imponderable intentions and/or institutions such as Google and Facebook raise questions in public discourse about data confidentiality and security through their current ways of acting; thus endangering scientific surveying on a voluntary basis. In order to deal with these challenges, we propagate the visible establishment of a different model. Inspired by sociologists Luhmann and Giddens, we call it “Trust by Procedure”. Trust from this point of view arises mainly from transparency and continuity in procedure (and staff). For all the gatekeepers, respondents, and the researchers who develop instrumentation and/or work with the data later on, the study, their products, and the team behind it must be seen as trustworthy. In terms of scientific discourses, this means not to emphasize the “total survey error” idea only, but also the quality and trustworthiness in the survey process, from access to participants to data usage by the scientific community and the public in general. Trust needs time and proof by own experience. But how should trust be addressed and built up during fieldwork, between the researchers and the public? It requires a conviction that people behind the study represent competence, predictability, and reliability.
Thus, experiences, best-practices, and suggestions, as well as theoretical deliberations regarding fieldwork strategies, monitoring and evaluation systems, and sense-driven panel maintenance strategies to establish trust shall be the focus of this session.
Keywords: management and coordination efforts in large-scale studies, fieldwork strategies, panel maintenance
Dr Charlotte Herzmann (University of Bremen) - Presenting Author
Dr Tilman Reinelt (University of Bremen)
Dr Claudia Teickner (University of Bremen)
Professor Franz Petermann (University of Bremen)
Dr Hannes Kröger (German Institute for Economic Research (DIW))
Professor C. Katharina Spieß (German Institute for Economic Research (DIW))
Background. Early interventions fostering child development in families facing social or cultural challenges generally start during pregnancy or in the first year of life. However, both intervention programs as well as birth cohort studies often have difficulties recruiting these families. Reasons among others are language barriers or that these families are more likely to skip medical check-ups - a common pathway to recruit participants during pregnancy. Two strategies usually used in recruitment are (1) sending letters to households with potentially pregnant women or women, who have recently given birth, using register-based information and (2) building a network of people in the health care or social work sectors, who advertise the project to target families and might therefore function as gatekeepers. Yet, little is known about whether these two approaches lead to acquiring the same population. Methods. In an ongoing study, the Bremen Initiative to Foster Early Childhood Development (BRISE), both approaches are used. For about one year, information has been sent out to households (e.g., via letters, newspapers or flyers) and a network of gatekeepers (e.g., midwives, gynecologists, early prevention social workers) was established. So far, 161 families have been positively screened for inclusion in the study. Inclusion criteria comprise migration background, level of education and employment status. Results. Comparisons of families recruited via general household information and via gatekeepers revealed that families recruited by gatekeepers reported a significantly lower paternal education (U = 1682, p = .022) and by tendency lower maternal education (U = 2164.5, p = .084) than families recruited via general household education. No similar effects were found for migration background or employment status. Conclusion. Families facing social or cultural challenges are hard to reach. In order to reach these families and gain their trust, gatekeepers might be necessary, especially as these families might not be acquired via general study advertisements throughout letters or newspapers.
Ms Evanthia Leissou (University of Michigan ) - Presenting Author
Ms Terri Ridenour (University of Michigan )
Mr Ian Ogden (University of Michigan )
Dr Michael Elliott (Univerity of Michigan)
The Environmental Influences on Childhood Health Outcomes (ECHO) program is a seven-year National Institutes of Health initiative to study how environmental factors affect child health and development. The study design is based on its predecessor, the National Children’s Study (NCS), but unlike NCS, the ECHO design tried to maximize available resources by engaging 35 existing cohort studies, in addition to several new cohort recruitments (for a total of 50,000 children). The Michigan ECHO, a diverse team of researchers from five Michigan academic and health institutions (Michigan State University, University of Michigan, Wayne State University, Henry Ford Health System and Michigan Department of Health and Human Services), joined ECHO bringing onboard two sample cohorts (~ 2000 mother-child pairs): the Archive for Research in Child Health (ARCH), a Michigan State University cohort study collecting data annually since 2009, and MARCH, a new statewide probability sample of pregnant women. The MARCH sample is recruited in clinics, and during pregnancy, women complete surveys and provide bio-specimens. After the birth, mother and child are followed for up to 7 years, completing surveys, neurodevelopmental assessments, and providing bio-specimens.
Priorities for the project are to develop structures and technical systems to support the longitudinal design, new sample recruitment, panel maintenance, multi-mode data collection and data sources; to support multiple teams that implement the project design; to communicate and build trust with participants; and to harmonize data collected prior to ECHO with data collected under the ECHO design.
The focus of this paper is to describe the experience of building and maintaining a longitudinal sample in a community based project. We will discuss design challenges, resources engaged (human and technical), and communication protocols with stakeholders including research participants. More specifically we will discuss participant contact and panel maintenance strategies. Finally, we will discuss interviewer training and quality assurance protocols.
Mr Florian Bains (Leibniz Institute for Educational Trajectories) - Presenting Author
Miss Hanna-Rieke Baur (Leibniz Institute for Educational Trajectories)
Interviewers, especially their behavior during fieldwork, are crucial to ensure high quality data collection processes. From our point of view, here the issue of trust comes into effect in two different ways: On the one hand, researchers conducting a study need to trust interviewers that these implement the given instruments correctly via standardized behavior in the field. On the other hand, respondents need to trust their interviewers as competent representatives of a trustworthy study. Especially in the context of panel studies it is important that interviewers build a foundation of trust between them and respondents and as a result get respondents to participate repeatedly. Therefore the necessity of establishing a core staff of adequately trained interviewers that can be deployed repeatedly arises. Thus, intensive interviewer trainings are an essential element within the National Educational Panel Study (NEPS, Germany). Aiming to optimize the training, we currently extend our monitoring and evaluating measures by systematically integrating interviewers´ feedback to the training. We not only regard interviewers as a potential source of error but take their experiences from daily practice into account. For this reason, interviewers are asked about the usefulness of (certain) training (units) directly after the training (e.g. perceived preparation for contacting phase) and about possible consequences during fieldwork processes (e.g. experience of problematic situations concerning the contacting phase).
We would like to contribute to the session by presenting first results and discussing possible implications for the future (e.g. modification of the interviewer questionnaires or more individualized training according to specific needs)as well as influences on interviewer-respondent interactions. Taking interviewers´ input seriously or even implementing adjustments with regards to specific training units, should foster trust between interviewers and researchers (via mutual appreciation). It ideally pushes interviewers´ competence and self-efficacy and thereby has a positive effect on the interaction with respondents and on fieldwork in general.
Mrs Andrea Giersiefen (Frankfurt University of Applied Sciences - Forschungszentrum Demografischer Wandel) - Presenting Author
Mr Robert Lipp (Frankfurt University of Applied Sciences - Forschungszentrum Demografischer Wandel)
Dr Sven Stadtmüller (Frankfurt University of Applied Sciences - Forschungszentrum Demografischer Wandel)
Panel studies addressing children and adolescents at school face particular challenges. In addition to bureaucratic challenges in advance, the institution school but also the parents and not at least the children and young people themselves must be convinced to participate permanently.
The panel study "Health behavior and Injuries in School Age" (GUS) is intended to illustrate which measures can help to convince actors to participate in the study on a permanent basis.
The study started in grade 5 with about 10,000 students from about 150 schools. Currently, the number of participants is still around 10,000 students, but there are also a number of fluctuations behind it.
In the presentation, measures will be presented, which are helpful for a high participation rate: These are detailed and transparent descriptions of the study for all decision makers, individual result reports for the individual school and school certificates documenting the participation. Last but not least, recognition features such as a consistent process as well as a personal and continuous relationship to the scientists are of importance. Furthermore a newsletter and a homepage are used to report on current presentations and publications of project findings and to report on the general work in the project. These include e.g. interviews with students who do qualification thesis’s with the data, as well as reports from professional interviewers and interviews with the funding agency.