Social Isolation

Theories of social isolation (or “social location”) have been used to explain lower cooperation in responding to surveys among certain subgroups in society, such as the elderly, minority racial and ethnic groups, and lower socioeconomic status groups. The social isolation theory of unit nonresponse states that subgroups in a society that are less connected to the dominant culture of the society—that is, those who do not feel part of the larger society or bound by its norms—will be less likely to cooperate with a survey request that represents the interests of the dominant society. According to the leverage-saliency theory, respondents decide to participate in a survey depending on survey attributes such as how long the interview might take, the presence of an incentive, and what the data might be used for. They may also make this decision depending on who is sponsoring the survey and what the topic of the survey is. The hypothesis of social isolation comes into play in these last two aspects.

Survey researchers have noted that sometimes sentiments of “civic duty” prompt survey participation and that these feelings of civic duty are associated with survey participation, especially when a properly constituted authority is requesting the participation of the sampled respondent. In theory, organizations with greater legitimacy, for example, those representing federal government agencies, are more likely to positively influence a respondent's decision to participate in a survey than, for example, a commercial survey research firm with less authority. However, according to some researchers, individuals who are alienated or isolated from the broader society feel less attachment to that society and thus have a lower sense of civic duty, which in turn will lead to a higher refusal rate in surveys conducted by and for those perceived as being in authority. Similar reasoning is often advanced regarding interest in and knowledge of the survey topic, in particular when focusing on political and election surveys. There are a number of studies that link political participation with survey participation. Results from these studies indicate that individuals who feel more isolated from the dominant society may also lack interest in politics and may feel less willing to respond to a political poll. In addition, nonresponse among members of socially isolated groups may be due to self-disqualification. That is, socially isolated individuals may feel they are not qualified to give an opinion on the topic or that their opinion is not valued.

Aspects of social isolation are both structural and social-psychological. For this reason, multiple indicators of social isolation should be used to study its relationship with survey cooperation. With demographic indicators (e.g. age) as a proxy for social isolation, some studies have found that elderly people are less likely to cooperate than are other subgroups. Elderly people often have fewer ongoing relationships with the larger society and thus are more isolated from the larger society. However, the effect of older age also has been found to be mitigated by living conditions and household composition. Single-person households tend to be less cooperative, whereas households with young children tend to be more cooperative. It is hypothesized that parents of young children are more likely to interact with strangers in the community on a daily basis, thereby increasing the level of connection with the society and thus also their likelihood of cooperating in surveys when a stranger (the interviewer) contacts them to seek cooperation with the survey.

...