close× Call Us
close×
Title Citizen Support for Government Surveillance: Evidence from a Survey Experiment in Kuwait
Post date 08/26/2019
C1 Background and Explanation of Rationale

There have been extensive revelations about government surveillance of ordinary citizens in the name of national security, even in the democratic United States. It is generally considered that people value their privacy. As privacy rights have become a key issue around the world, there has been a surge of studies examining the state's use of surveillance to monitor their citizens. Yet, we know little about how much security outweighs privacy to the masses: under which conditions are citizens willing to relinquish their privacy to the government?

In a survey experiment fielded in Kuwait, we demonstrate how exposure to different types of online threats and their targets affect public's risk perception and their support for government surveillance. Contrary to conventional wisdom, we expect citizens not to give up their privacy rights for the sake of national security. Instead, we hypothesize that individuals are more likely to accept government violation of their privacy when this is framed as necessary to protect citizens against threats to themselves (i.e. personal versus national) than when it is framed as necessary to protect the political order (national). Despite a low probability of terrorist acts, we expect citizens to be willing to sacrifice their privacy in the terrorism (versus cyber crime) condition due to devastating and physically evident consequences of these acts.

C2 What are the hypotheses to be tested?

This project argues that the effect of different types of threats and targets will produce divergent risk perceptions, resulting in divergent public support for surveillance. Using a 2 x 2 design, we directly manipulate a cyber threat's two key characteristics --- its type (criminal versus terrorist) and target (personal versus national) --- while keeping other threat's characteristics constant. We expect that risk perceptions of cyber threats that target participants directly (i.e criminal condition) will be relatively low, despite an increasingly growing evidence that such threats may present more risk to individual than other security threats, such as terrorism. Additionally, we hypothesize that individuals are more likely to accept government violation of their privacy when this is framed as necessary to protect citizens against threats to themselves (i.e. personal versus national) than when it is framed as necessary to protect the political order (national). Individual's cognitive biases might explain such a perception.

We include five categories of dependent variables: support for government surveillance, value of privacy, support for government spending on cyber capabilities, personal online behavior, and risk assessment.

C3 How will these hypotheses be tested? *

In our experimental design embedded in a survey fielded in Kuwait, we manipulate two variables of threats: type and target. This design allows us to systematically examine how the citizen responses vary in different plausible contexts. Subjects will be randomly assigned to one of four manipulation conditions or a control. The enumerators will read a corresponding short text.

Control: More than 90% of Kuwaitis today have one or more social media accounts

Treatment 1 (criminal / personal): More than 90% of Kuwaitis today have one or more social media accounts. Criminals have increasingly targeted Kuwaitis on social media to steal their private information.

Treatment 2 (terrorist / personal): More than 90% of Kuwaitis today have one or more social media accounts. Terrorists have increasingly targeted Kuwaitis on social media to steal their private information.

Treatment 3 (criminal / national): More than 90% of Kuwaitis today have one or more social media accounts. Criminals have increasingly targeted government entities in Kuwait
on social media to steal their secret information.

Treatment 4 (terrorist / national): More than 90% of Kuwaitis today have one or more social media accounts. Terrorists have increasingly targeted government entities in Kuwait
on social media to steal their secret information.

C4 Country Kuwait
C5 Scale (# of Units) 2,000
C6 Was a power analysis conducted prior to data collection? No
C7 Has this research received Insitutional Review Board (IRB) or ethics committee approval? Yes
C8 IRB Number IRB19-0686
C9 Date of IRB Approval May 30, 2019
C10 Will the intervention be implemented by the researcher or a third party? Ipsos Kuwait
C11 Did any of the research team receive remuneration from the implementing agency for taking part in this research? No
C12 If relevant, is there an advance agreement with the implementation group that all results can be published? not provided by authors
C13 JEL Classification(s) F50, H20