Tuesday 16th July
Wednesday 17th July
Thursday 18th July
Friday 19th July
Download the conference book
Not able to participate: a neglected cause of nonresponse 1 |
|
| Convenor | Dr Ineke Stoop (SCP) |
| Coordinator 1 | Mr Achim Koch (GESIS) |
| Coordinator 2 | Dr Hideko Matsuo (KU Leuven) |
There are two main causes of survey nonresponse: refusal and noncontact. It can also happen that a person is not able to participate because of language problems, or because someone is mentally or physically not able to participate. Sometimes "being unable" is the consequence of design aspects (someone who cannot hear cannot participate in a telephone survey, and someone who is blind cannot complete a paper questionnaire). In these cases practical solutions can be found. In other cases someone may not be able to answer survey questions for more general reasons: persons with severe learning disabilities may not be able to participate in a general survey whatever the mode. They may however be able to participate in a survey that has especially been designed to accommodate these groups.
Not being able to participate is usually assumed to be a minor factor in survey nonresponse and the bias caused by this is usually assumed to be small. In the European Social Survey, however, the share of persons not being able to participate cannot be ignored in a number of countries.
We invite submissions that address : a) the measurement of reasons for nonparticipation, b) the relationship with design issues (survey mode, sampling frame) and reasons for nonparticipation, c) ways to minimise nonparticipation due to not being able (and potential effects on measurement error and comparability), and d) effects of "not able" on nonresponse bias.
AAPOR (2011) subdivides eligible nonrespondents into refusals, non-contacts and “other” cases: instances in which the respondent is/was eligible and did not refuse the interview, but no interview is obtainable because of: a) death, b) the respondent is physically and/or mentally unable to do an interview; c) language problems; and d) miscellaneous other reasons. “Respondents who are physically or mentally unable to participate in an interview would include both permanent conditions (e.g., senility, blindness or deafness) and temporary conditions (e.g., pneumonia or drunkenness).”
In practice, causes of nonresponse may be hard to determine and hard to distinguish. Is it easier to pretend not being able to speak the survey language than to refuse? How to distinguish between permanent and temporary conditions? How to complete the screening process when the person opening the door does not speak the language? When interviewers are involved in selecting respondents within households, it may be tempting to ignore those household members who cannot participate and select someone who is able to do an interview. And when time is running short it might be tempting to classify a temporary condition as being permanent.
These problems will be exacerbated in cross-national studies in different countries with different survey traditions, sampling frames, interviewer remuneration schemes and population composition. The presentation will focus on the European Social Survey and will show how difficult it is to achieve comparable final disposition codes across countries.
The starting point of this presentation is to emphasize the absence of sensory disabled people in quantitative surveys, among which the European Social Survey which will be used as an example. We will focus our examples on the case of the Deaf in the Belgium ESS round 6.
Sensory disabled people represent a rather big population (around 4% of Belgian population) but they are nearly absent in general quantitative surveys. The aim of this paper is to present and discuss some mixed-mode techniques to reach sensory disabled people and try to include them in general opinion surveys.
Because every deaf cannot read texts, we couldn't opt for a "classic" web based questionnaire, so an adapted version of the questionnaire has been set up. It is a web-based version that integrated videos of the translation of every question in sign language.
We will show some results of the round 6 experiments that we did and show how our specific mixed-mode adapted procedures tend to prove that, with reasonable costs, it is possible to include deaf people in the general quantitative surveys and to reduce one of the specific non-response bias. The ethical issue is also at the center of our concern and the proposed techniques should contribute to a fairer use of the quantitative surveys and of inference.
We will also focus on the translation procedures (from French into Sign Language) and on the associated risk of measurement error.
As a survey designed to explore and investigate the impact of disability over time it was imperative that the development of the Life Opportunities Survey included consultation with people with differing disabilities. A reference group was established to advise and guide the development of the questionnaire, field and interview methods. Awareness of respondent requirements during the survey development and testing among a sample of respondents selected to reflect the diverse disabled population proved invaluable to the ongoing success of this longitudinal survey, now in its third wave.
The lessons learnt in the development of the Life Opportunities Survey are applicable across all social surveys. In particular they have led us to develop guidance for interviewers on providing an 'equal chance' for all respondents and accessible means of communicating with respondents selected to take part in our surveys. This is important so that our surveys are truly representative, covering all groups of people, including those with disabilities and those whose language is not English.
Group Quarters (GQs) population in the US Census refers to institutionalized population in correctional and juvenile facilities and nursing facilities, as well as non-institutionalized population in college/university resident halls and military barracks. Not all of GQ persons are necessarily able to participate in the US Census unless they are accommodated with measures tailored to the needs of particular GQ persons. In this paper we examine the best practices of enumerating in the US Census the conventional not-able-to-participate GQ population and discuss their implication for controlling coverage errors and associated costs. We apply a survey error and cost model (e.g., Groves, 1989) to assess GQ enumeration tools tailored to controlling for the not-able GQ persons. We present case studies to illustrate merits of the best GQ practice to deal with the not-able population and remaining challenges to address in the 2020 Census.