Undermining our data: implications for trust in the population census

Killick, Lynn; Duff, Alistair S; Deakin, Mark; Hall, Hazel

Abstract

This paper draws on empirical work conducted as part of a multi-method research study funded by the Arts and Humanities Research Council (AHRC). It is concerned with public perceptions of an online population census and adds to prior work exploring the perceptions of policymakers of the utility of census outputs(Killick, Hall, Duff, & Deakin, 2016).
This paper makes particular reference to the findings of a survey conducted as part of the wider AHRC funded study. The author used NoviSurvey software to support the design and distribution of the survey. The survey is concerned with the motivation of the public to participate in the population census of 2021 and attitudes about accessibility, data security and data handling. This findings presented in this abstract relate to 330 responses received in 2017.
Governments understand that information about the population of their nation can inform policy decisions, with particular regard to policy decisions that could improve health and social outcomes for the population (White, 2009). Accordingly, the government compels citizens to participate in a count so that it can make policy decisions for the benefit of society. Enumeration of the population by way of a confidential census continues to be the preferred option to gather population information in the UK. However, public sector reform combined with the availability of new technologies has led to a review of the traditional census (Boyle & Dorling, 2004; Coleman, 2013; Dugmore, Furness, Leventhal, & Moy, 2011; White, 2009). Following a detailed research exercise and consultation conducted by the census offices of the UK home nations, the 2021 census in Scotland will require the public to provide household information using online platforms.
The census office also intends to augment census data with information extracted from administrative records. This is to plug data gaps in the census and meet requests of data users for more frequent output reports. The issue of data sharing between government sector organisations raises questions regarding informed consent. For example, with particular reference to data sharing, albeit in a US census taking context, McMillen (2004) argues that government does not properly inform the public about the use of personal data held by its institutions (McMillen, 2004). Others note that the public place trust in the ‘census office’ but express misgivings about the actual use of census data (Baffour, King, & Valente, 2013; Courtland, 1985; Singer & Neugebauer, 2003).
Finally, the complex and costly data systems required to collect census data is likely to result in the outsourcing of elements of the online census. Based on procurement protocols, this creates the conditions for the private sector and non-UK based companies to have full access to sensitive census data. The literature suggests that the involvement of private organisations in processing census data could harm the levels of public trust in the census (Cullen & Reilly, 2008; Cullen, 2009).
This study found that survey respondents are not content with private sector companies handling census data. When asked specifically about UK-based private sector firms 47% stated that they were not comfortable with such companies processing their data. When asked about their attitudes regarding non-UK based companies handling the data there was a marked shift 77% stating that they would not be comfortable with foreign companies handling their census data. The respondent's strength of objection to private sector companies handling census data is consistent with the work of Cullen (2008) and suggests that outsourcing of data processing is a risk to public participation in the population census.
The literature suggests that public participation in the (UK) census could be adversely affected by the potential for data breaches or the perceived misuse of census data (Heeney, 2012). There are no reported instances of census data abuses in the UK, and the Scottish census office assures the public of the right to confidentiality. Furthermore, a privacy impact assessment of the census highlights its confidential nature and the robust data control measures in place. However, The Statistics and Registration Service Act 2007 provides an exception, allowing prior access to statistics by other government agencies (Holt, 2007). This is an important provision, particularly in the context of policy responses to the current heightened terror threats in the UK (e.g. Zedner, 2016) and historical examples of misuse of census information (Aly & Roth, 2004; Seltzer & Anderson, 2001).
This issue of anonymity appears to be a major factor for the respondents to the survey discussed in this paper. Sixteen percent of survey respondents were happy for the census office to share anonymised census data with UK intelligence agencies. However, 60% of respondents reported opposition to the sharing of identifiable information. Respondents also noted a greater level of trust in the census office than ‘government’ to treat their personal data with respect for their privacy (73% v 66%).
Concerns regarding improper use of census data by other areas of government prompted scores of people to avoid the population count in 1991 (Simpson & Dorling, 1994). The findings from this study suggest these concerns are still present and the shift to a predominantly online census may increase such concerns.
In summary, this paper examines attitudes to sharing personal information with e-government; it also explores issues of accessibility and readiness for a transition to online data collection to understand if the proposed new format of the census will have an impact on the quality of the census outputs. Thus it aligns with the key conference theme of patterns of information behaviour in different contexts and the social, cultural and economic impacts of engagement with information. i3 delegates with interests in how the public respond to data gathering by e-government and shifts to online channels will find this paper of relevance.

Citation

Killick, L., Duff, A. S., Deakin, M., & Hall, H. (2017, June). Undermining our data: implications for trust in the population census. Paper presented at Information: interactions and impact (i3)

• Presentation Conference Type: Conference Paper (unpublished)
• Conference Name: Information: interactions and impact (i3)
• Start Date: Jun 27, 2017
• End Date: Jun 30, 2017
• Acceptance Date: Apr 7, 2017
• Deposit Date: Apr 24, 2017
• Publicly Available Date: Apr 24, 2017
• Pages: 1-4
• Chapter Number: unkown
• ISBN: unkown
• Keywords: Census, information security, perception, trust,
• Public URL: http://researchrepository.napier.ac.uk/Output/830689
• Contract Date: Apr 24, 2017