“How Fear of Crime Affects Needs for Privacy & Safety”

Acceptance of Surveillance Technologies in Smart Cities

Julia van Heek, Katrin Arning and Martina Ziefle

Human-Computer Interaction Center, RWTH Aachen University, Germany

Keywords: Technology Acceptance, Crime Surveillance Technologies, Privacy, Safety, Smart Cities, Perceived Crime

Threat.

Abstract: These days, surveillance technologies are a key component of smart and networked cities preventing or

detecting crime and giving the residents a sense of safety. On the one hand, safety perceptions can be

supported by adequate surveillance technologies (e.g., cameras), however on the other hand, the systematic

use of surveillance technologies undermines individual privacy needs. In this empirical study, we explore

users’ perceptions on safety and privacy in the context of surveillance systems in urban environments.

Using an online survey, 119 users were requested to indicate their acceptance regarding different types of

surveillance technologies, differentiating perceived benefits and barriers as well as safety and privacy needs.

Also, we investigate acceptance differences towards surveillance technologies at various locations (private

and public). In this paper, we especially explore the impact of individual perceived crime threat on the

acceptance of surveillance technologies and on the needs for privacy and safety.

1 INTRODUCTION

One of the major challenges of modern societies is

to meet the complex demands of urbanization

processes and to maintain liveable, sustainable, and

smart cities. Up to 2030, more people will live in

cities than in other regions and this development is

forecasted to increase further. In line with these

fundamental urbanization processes, consecutive

challenges arise. Beyond issues of economy,

transportation or governance, nowadays’ major

keystones of urban planning are the broadly

accepted implementation of technical infra-

structures and (smart) city concepts (Ziefle et al.,

2014). All over the world, an increasing number of

surveillance technologies is used to prevent or time-

critically detect crime in order to improve safety (La

Vigne et al., 2011). Perceived safety represents an

essential prerequisite for the participation in social

and economic life and is a valuable good for cities.

However, the main drawback of surveillance

technologies is the perceived privacy violation by

the public through the recordings and processing of

data (Whitaker, 1999). Therefore, smart city

concepts must meet a wide range of residents´ needs,

including high comfort regarding safety,

sustainability, but also consider different levels of

perceived crime threat, and protection of privacy

(Ziefle and Wilkowska, 2015).

Facing the demographic change, smart city

concepts should also address the diversity of urban

residents. Although they are essential for all

dwellers, especially different ages of residents

should be taken into account. If individual needs

and wishes of both younger and older people are

considered, the fundament for liveable and safe

future cities is granted (Plouffe and Kalache, 2010).

2 ACCEPTANCE OF CRIME

SURVEILLANCE SYSTEMS

For a free, unrestricted and unworried life in urban

areas, people need to feel safe. In this context, crime

threat in cities is a central challenge (Smith and

Clarke, 2000; Marshall et al., 2007). The consequen-

ces of crime for urban safety and individual risk

perception are well described and represent a serious

barrier for many residents (e.g. Baumer, 1978;

Loewen et al., 1993). While it is undisputed that

safety and crime prevention are major goals for

urban development, the realization of effective

Heek, J., Arning, K. and Zieﬂe, M.

“How Fear of Crime Affects Needs for Privacy Safety” - Acceptance of Surveillance Technologies in Smart Cities.

In Proceedings of the 5th International Conference on Smart Cities and Green ICT Systems (SMARTGREENS 2016), pages 32-43

ISBN: 978-989-758-184-7

safety measures is controversially evaluated (Isnard,

2001/Wiecek and Seatnan, 2002/Sheldon, 2011).

Technically, surveillance technologies are at

hand and are already widely used in urban

environments to increase safety (Chattopadhyayr,

2013/ Song et al., 2013). Most of all city centers use

close-circuit television in public spaces and in public

transport systems. However, the acceptance of these

systems in general and, more specific, the individual

perception of safety does not necessarily rise, when

surveillance systems are installed (Lewis and

Maxfield, 1980). Instead, perceived fear of crime in

urban environments is rather shaped by physical

features, such as visibility or lighting, prospect such

as open spaces, opportunities to escape (Blöbaum

and Hunecke, 2005). But not only physical charac-

teristics of urban spaces, but also perceived incivi-

lities in surrounding areas strongly affect fear of

crime (Lewis and Maxfield, 1980). Therefore, the

installation of technical safety measure needs to

carefully address individual perceptions of safety at

different locations to support safe living in smart

cities.

Apart from the goal of enhanced safety,

surveillance systems also pose ethical concerns. In

terms of privacy, protection is one of the key human

rights. Technical monitoring of people in urban

environments for safety reasons conflict with

individual rights for privacy (Gumpert and Drucker,

2001), which – beyond legal concerns – might lead

to a public rejection of monitoring technologies in

city locations (Arning et al., 2013b). Accordingly,

the relationship between individual needs for safety

from crime and the individual need for protecting

one’s own privacy is complex and does not follow a

simple arithmetic, but rather varies with usage

context, individual characteristics and city needs

(Arning et al., 2013a). The safety-privacy-

relationship for crime surveillance technologies can

only be understood if the trade-off between both

basic motives is empirically addressed.

3 INDIVIDUAL FACTORS

The population in cities is characterized by a high

heterogeneity. Residents’ needs and wishes towards

quality of city life as well as related experiences and

attitudes are affected by a multitude of individual

factors. Though there are individual key

characteristics, which allow defining groups with

specific needs regarding safety and privacy. One

important factor is age, which becomes even more

relevant with the on-going demographic change. For

instance, age-related changes in health conditions

and changed leisure time activities after retirement

lead to specific mobility and accessibility needs

(Alsnih and Hensher, 2003). Especially for older

people, perceived safety in their living environment

is essential for maintaining social contacts

(Dickerson et al., 2007). Apart from age, gender is

another factor, which strongly affects needs for

safety and privacy. Elderly women, for example,

have higher needs for safety than men, reducing

their willingness to use public transport-tation or

carpooling (Arning et al., 2013b). Beyond age and

gender, but strongly interrelated, the perceived level

of crime fear is another important factor for the

acceptance of surveillance techno-logies in smart

cities. Fear of crime, defined as the emotional

response to possible violent crime and physical harm

(Covington and Taylor, 1991), has been intensively

researched in the last decades by various scientific

disciplines in the context of urban development.

Two central findings are specifically noteworthy in

this context: a) crime fear is an individual perception

not necessarily associated with objectively

measurable crime statistics. Thus, even when

persons live in a comparably safe residence, they

might perceive higher levels of crime fear; b)

individual factors, as age, gender, and experience

with crime, further affect fear of crime. A well-

replicated finding in this context is the inverse

relationship of victimization rate and crime fear: the

most fearful individuals (elderly women) have the

lowest victimization rate, the least fearful (young

men) have the highest victimization rate

(Scarborough et al., 2010). The strong interrelations

of age, gender and crime fear suggest, that age and

gender serve as “carrier variables” for different

levels of perceived crime fear. Accordingly, the

present study focuses on the inter-individually

different effects of crime fear on surveillance

technology acceptance.

The usage of crime surveillance technologies in

urban environments is one (technical) approach to

enhance perceived safety and to reduce crime rates.

Yet, only sparse knowledge is available about

acceptance patterns of residents towards benefits and

barriers of crime surveillance technologies, which

are assumed to increase safety perceptions in the

context of smart cities. The goal of the present study

is, thus, to understand peoples’ acceptance of crime

surveillance technologies in urban environments,

taking needs of safety and privacy as well as

individual factors such as perceived crime threat into

account.

“How Fear of Crime Affects Needs for Privacy Safety” - Acceptance of Surveillance Technologies in Smart Cities

4 METHODOLOGY

In the following section, the questionnaire, the

sample and the applied statistical procedures are

detailed.

4.1 Questionnaire

Questionnaire items were developed based on the

findings of focus group interviews carried out prior

to this study. The questionnaire was arranged in six

sections. The first part addressed demographic

characteristics of the participants, namely age,

gender, family status, children status, number of

persons living in a household together, type and

place of residence, housing status (homeowner or

tenant), educational level, (current or last) job sector

and current or last occupation.

The second part focused on the individual

perception of crime threat (PCT) and potential

experiences with crime. First, we asked for PCT at

different places by day and by night. For clarity

reasons, locations were arranged into four categories

(private (e.g., garden), semi-private (e.g., own

street), semi-public (e.g., shopping mall), and public

(e.g., train station) locations) based on results of

previous focus groups interviews. The question “to

what extent do you feel threatened by crime during

the day?” had to be evaluated for more than 20

different public and private locations (see Fig. 1).

Threat perceptions had to be rated on a six-point

Likert scale (1=not at all; 6=very strong PCT). In

addition, looking for possible differences of PCT

during day- and nighttime, participants had to

evaluate on a five-point scale (-2=much lower; -1

=lower; 0=no difference; 1=higher t; and 2=much

higher) if they would feel a different crime threat at

the same locations by night.

Based on PCT ratings for different locations the

between-factor “perceived crime threat” (PCT) was

calculated. Respondents’ PCT ratings were summed

up (max=156), transformed to a value of 100 and a

median split was conducted (cut-off=34.62), which

separated two groups with low PCT and high PCT.

Second, we asked for individual crime threat

concerning different crime offenses, which had to be

evaluated by the participants on a six point Likert

scale. The final aspect of this part of the

questionnaire focused on experiences with crime.

Here, participants indicated whether they, their

family, friends, relatives or acquaintances had

become victims of various crimes offenses

themselves, e.g. theft or bodily injury.

The third part assessed technologies and

traditional measures enhancing perceived safety in

private and public environments. Thus, different

technologies (e.g., camera surveillance, ambient

lighting, microphones) but also social measures

(e.g., police presence) had to be rated on a six-point

Likert scale (1=strongly disagree; 6=strongly agree)

for a private as well as a public context of use.

The fourth part of the questionnaire asked about

the acceptance of crime surveillance technologies at

different locations. First, the participants were asked

to evaluate to what extend they would accept

technologies like standard cameras, microphones,

cameras with face recognition and location

determination in their private living environment.

Then, participants had to do the same in the case of a

public environment. Further, we asked for

acceptance of surveillance cameras at different

private and public locations, which also had to be

evaluated on a six point Likert scale (1=strongly

disagree; 6=strongly agree). The next part of the

questionnaire asked about perceived benefits and

barriers of crime surveillance (also 6-point Likert

scale, see above). Benefits of crime surveillance

were examined in seven items, which referred to

safety aspects, e.g., prevention of crime, sense of

safety or the felt deterrent effect for potential

criminals. Barriers referred to eight items relating to

privacy aspects, e.g., protection of civil rights and

personal freedom, storage of recorded data or

inference of being under general suspicion.

The fifth and last part focused on the trade-off

between the need for safety, on the one hand, and

the need for individual privacy, on the other hand.

Participants were explicitly asked to trade-off

between their individual needs for safety and privacy

when considering the employment of crime

surveillance technologies at different locations on a

10-point scale (1=increase of safety; 10=protection

of privacy). Completing the questionnaire took about

20 minutes. Data was collected in an online survey

conducted in Germany. By using the online link, all

parts of Germany had been addressed, however,

participants predominately originated in North-

Rhine Westphalia. Overall, the questionnaire was

made available for about 8-10 weeks in the

beginning of 2013. In that time, there was no high

impact society event (e.g. terrorist attacks) and data

collection was also accomplished prior to the current

flow of refugees, which are moving to European

countries.

4.2 Sample

In total, 119 participants took part in the study. As

SMARTGREENS 2016 - 5th International Conference on Smart Cities and Green ICT Systems

only complete questionnaires (no missing answers)

could be used for further statistical analyses, 99 data

sets were analysed in the end. The mean age of the

participants was 37.8 (SD=15.5) with 58.6% females

and 41.4 % males. Asked for having children, the

majority of 65.7% answered to have no children.

Demanded for the number of persons living in their

household, 38.4% reported to live in pairs, 31.3%

live alone, 15.2% live in a threesome, 9.1% live with

four persons, 3.0% live with five persons and also

3.0% live with more than five persons in their

household. Asked for their residence, 23.2%

reported to live in a detached, 13.1% in a semi-

detached and also 13.1% in a townhouse. The

majority of 50.5% reported to live in an apartment

building. The participants were also asked for their

housing conditions: 45.5% reported to be the house

owner and 54.5% reported to rent. Regarding their

area of residence, 35.4% live in a city centre, 29.3%

in outskirts, 21.2% in suburban areas and 14.1% live

in rural areas. Participants volunteered to take part in

the study and were not gratified for their efforts.

5 RESULTS

The general results of the study concerning crime

surveillance acceptance have been published already

(van Heek et al., 2015). In this paper, the influence

of the user diversity factor perceived crime threat on

crime surveillance acceptance is focused.

First, the results of PCT regarding day- and

nighttime at various locations are described. In a

second step, the impact of individual perceived

crime threat on crime surveillance acceptance is

presented in detail. Initially, the segmentation of

two PCT groups is introduced. Afterwards the

influence of PCT on crime surveillance acceptance

in terms of technologies and traditional measures

enhancing perceived safety is shown. Further, the

results of surveillance technologies regarding

different locations as well as perceived benefits and

barriers of crime surveillance are presented

(depending on PCT). Finally, the results of the

trade-off between the needs for safety and privacy

are shown for both PCT groups. Data was analysed

descriptively and, with respect to the effects of user

diversity, by (M)ANOVA procedures (significance

level at 5%).

5.1 Perceived Crime Threat at Different

Public and Private Locations

Perceived crime threat at daytime: In total, i.e.

summed up for all locations, the PCT during

daytime was rather low (M=36.3 on a scale with

max = 100; SD=12.9).

The majority of private locations was perceived

as only lightly threatening, e.g., own garden (M=1.3;

SD=0.6) or own home (M=1.4; SD=0.8, see Figure

1). Semi-private locations were noticed as lightly

threatening, e.g., own street (M=1.8;SD=1) or hotel

(M=1.8;SD=0.9). Semi-public locations were

observed as slightly threatening, e.g., market

(M=2.4; SD=1.2) and public transport (M=2.6;

SD=1.3). Public locations were perceived as more

threatening, e.g., parks (M=2.8; SD=1.3), train

station (M=3.0; SD=1.4) or underground car park

(M=3.3; SD=1.6). Night time: In total, PCT

nighttime ratings were significantly higher (M=43.4;

SD=15.5) compared to daytime ratings

(F(1,97)=15.4,p<0.01). However, the PCT at night

did not vary strongly across the different locations.

Figure 1: Perceived crime threat by day and by night.

Private and semi-private locations were not

perceived differently by day or by night, except for

own street (M

Night

=2.1; SD=1.4; M

Day

=1.8; SD=1.0;

F(1,98)=18.7;p<0.01) and own house entry

Night

=2.1; SD=1.5;M

Day

=1.8;SD=1,1; F(1,98)=8.7;

p<0.01). Concerning semi-public locations a higher

123456

own garden

own home

own…

restaurants

own car

your street

hotel room

public…

hotels

departme…

own…

schools

cellar

main roads

shopping…

pubs

market

tourist…

public…

parking…

stations

parks

station…

train station

car park

u-car park

evaluation (min=1; max=6)

night day

private

semi-

private

semi-

public

not at

all

strongly

“How Fear of Crime Affects Needs for Privacy Safety” - Acceptance of Surveillance Technologies in Smart Cities

PCT was found, e.g., for market (M

Night

=2.9;

SD=1.4; M

Day

=2.4; SD=1.2; F(1,98)=35.7;p<0.01)

or public transport (M

Night

=3.2; SD=1.6; M

Day

=2.6;

SD=1.2; F(1,98)=51.7; p<0.01) by night. Regarding

public locations, nearly all locations were perceived

significantly more threatening by night, e.g., train

station (M

Night

=3.9; SD=1.8; M

Day

=3.0; SD=1.4;

F(1,98)=102.1;p<0.01) as well as parks (M

Night

=4.0;

SD=1.5; M

Day

=2.8;SD=1.2; F(1,98)=175.6; p<0.01).

5.2 Effects of Perceived Crime Threat

as User Diversity Factor

So far, the acceptance of crime surveillance

technologies with related benefits and barriers was

reported for the whole group of respondents (van

Heek et al., 2015). However, residents in urban

environments are highly heterogeneous. Since we

assumed that the perceived necessity and acceptance

of crime surveillance technologies is affected by

individual levels of crime fear, we systematically

included “perceived crime threat” as group splitting

variable in our analyses.

5.2.1 Segmentation of PCT Groups

Based on respondents’ ratings of crime threat at

different locations two groups with high and low

levels of perceived crime threat (high and low PCT,

cut-off=34.6 on scale with max=100) were formed

by median split. Below, groups are described by

socio-demographic factors. The group with high

PCT consisted of a higher proportion of women than

in the low PCT group (though not significant).

Concerning age there was a similar distribution in

groups 1 and 2 without significant differences.

Table 1: Segmentation of PCT Groups.

Group 1 (n=50)

„low PCT“

Group 2 (n=49)

„high PCT“

gender 52% female

48% male

65,3% female

34,7% male

n.s.

age M = 36.68

SD=14.42

M=38.88

SD=16.54

n.s.

familiy

status

single 60%

partner/married 38%

divorced 2%

single 36,7%

partner/married 59,2%

divorced 4,1%

.05

children

status

yes 24%

no 76%

yes 44,9%

no 55,1%

.05

type of

residence

detached house 16%

semi-detached house

14%

townhouse 10%

apartment building 60

detached house 30,6%

semi-detached house

12,2%

townhouse 16,3%

apartment building 40,8

n.s.

place of

residence

city centre 46%

outskirts 22%

suburban area 16%

rural area 16%

city centre 24,5%

outskirts 36,7%

suburban area 26,5%

rural area 12,2%

n.s.

Both groups differed in terms of family status and

children status significantly (p<0.05). Group 1 (low

PCT) consists mainly of singles (60%), while group

2 (high PCT) mainly consisted of married people or

people living with a partner. Regarding children

status there was a higher percentage of people with

children (44,9%) in the high PCT group than in the

low PCT group (24%). In terms of type and place of

residence there were in parts slightly different

distributions, which failed to meet significance level.

5.2.2 Fear of Crime Offenses

In a first step, we analysed to what extent people

with high and low PCT differ with regard to fear of

several crime offenses (see Figure 7). People with

high PCT reported to feel significantly more

threatened than those with low PCT

(F(1,97)=48.1;p<0.01), except for the item “bicycle

theft”. This result pattern applied for “light”

offenses, e.g. material damage (M

Low

=3.0; SD=1.3;

High

=4.2; SD=1.2; F(1,97)=22.9; p<0.01) or theft

(in/from house) (M

Low

=2.6; SD=1.4; M

High

=3.9;

SD=1.3; F(1,97)=20.9; p<0.01) as well as for

“serious” offenses, for example sexual crimes

Low

=1.7; SD=1.2; M

High

=3.4; SD=1.5; F(1,97)=

40.1;p<0.01), offenses against life (M

Low

=1.5;

SD=1.0; M

High

=3.2; SD=1.5; F(1,97)=44.4; p<0.01)

and terrorism (M

Low

=1.3; SD=0.8; M

High

=3.0;

SD=1.4; F(1,97)= 54.5; p<0.01).

Figure 2: Fear of crime offenses for low and high PCT

groups.

All PCT group differences were highly

1,7

1,3

1,5

1,7

2,4

2,9

2,7

3,5

2,6

2,2

3,0

2,7

3,0

3,2

3,4

3,7

3,8

3,9

4,0

4,2

123456

stalking

terrorism

offenses against

life

sexual crimes

(own) car theft

fraud

theft (in/from my

car)

bicycle theft

theft in/(from the

own house)

bodily injury

material damage

evaluation (min=1; max=6)

high

PCT

low

PCT

not at

all

strongly

SMARTGREENS 2016 - 5th International Conference on Smart Cities and Green ICT Systems

significant. However, for serious offenses (e.g.

offenses against life) the differences between fear of

crime ratings for people with low and high PCT

were stronger pronounced.

5.2.3 Acceptance of Crime Surveillance

Technologies in Private Environments

In a next step, we examined how PCT influences the

acceptance of crime surveillance technologies and

traditional measures in private environments (see

Figure 3). First of all, using visible and invisible

technologies in private environments was both

accepted by the high PCT group, while it was rather

rejected by the low PCT group (Visible: M

low

=3.0;

SD=1.7; M

High

=4.2, SD=1.4; F(1,98)=15.8; p<0.01;

Invisible: M

low

=2.7; SD=1.6; M

High

=4.0; SD=1.3;

F(1,98)=18.1; p< 0.01).



Figure 3: Acceptance of surveillance technologies and

traditional measures at private environments for low and

high PCT groups.

Concerning the technology type there was no

difference between the PCT groups for ambient

lighting, which was most accepted. All other types

of technology were significantly more accepted by

the high PCT group, e.g. cameras (M

Low

=2.9;

SD=1.7; M

High

=4.1; SD=1.4; F(1,98)=13.8; p<0.01)

or location determination (M

Low

=2.2; SD=1.2;

High

=3.2; SD=1.2; F(1,98)=24.3; p<0.01). The

group with low PCT even rejected the usage of

cameras, microphones, and location determination.

Regarding traditional measures more police

presence (M

Low

=3.7; SD=1.7; M

High

=4.6; SD=1.2;

F(1,98)=7.8;p<0.01) and more presence of private

safety services (M

Low

=3.1; SD=1.7; M

High

=4.1;

SD=1.3; F(1,98)=8.6; p<0.01) were more accepted

by the high PCT group than by the low PCT group.

Both groups did not differ in their evaluations of

other traditional measures like contact to

neighbours, present awareness of others and a dog.

5.2.4 Acceptance of Crime Surveillance

Technologies in Public Environments

To compare different contexts of application we also

asked for the acceptance of the same crime

surveillance technologies and nearly the same

traditional measures in public environments (see

Figure 4).



Figure 4: Acceptance of surveillance technologies and

traditional measures at public environments for low and

high PCT groups.

In general, use of surveillance technologies was

more accepted in public (M=71.5; SD=15.4) than in

4,1

5,0

4,8

3,1

3,7

2,2

2,1

2,9

3,9

4,0

4,9

2,7

3,0

4,4

5,2

4,9

4,1

4,6

3,4

3,2

4,1

4,7

4,8

5,1

4,0

4,2

123456

dog

present awareness of others

contact to neighbors

presence of private security

services

police presence

location determination

microphones

camera surveillance

alarm system

motion detectors

ambient lighting

invisible technologies

visible technologies

traditional measures type of technology technology

evaluaton (min=1; max=6)

high

PCT

low

PCT

rejection support

3,9

5,3

3,6

4,9

2,7

2,4

4,0

3,3

5,3

3,1

4,1

4,4

5,4

4,9

5,5

3,8

3,4

5,1

4,2

5,6

4,6

5,1

123456

to be only on the road by day

present awareness of others

presence of private security

services

police presence

location determination

microphones

camera surveillance

motion detectors

ambient lighting

invisible technologies

visible technologies

traditional measures type of technology technology

evaluation (min=1; max=6)

high

PCT

low

PCT

rejection support

“How Fear of Crime Affects Needs for Privacy Safety” - Acceptance of Surveillance Technologies in Smart Cities

private environments (M=66.0; SD=16.0; F(1,98)=

18.4; p<0.01). First, using visible and invisible

technologies in public environments was also

significantly more accepted by the high PCT group

than by the low PCT group (Visible: M

low

=4.2;

SD=1.6; M

High

=5.1; SD=1.1; F(1,98)=11.6; p<0.01;

Invisible: M

low

=3.2; SD=1.8; M

High

=4.6; SD=1.3;

F(1,98)=20.8; p< 0.01).

Concerning different technology types all

technologies are evaluated more positively by the

high PCT group. So there is an higher acceptance of

surveillance technologies by the high PCT group

than by the low PCT group, for example for cameras

low

=4.0; SD=1.5; M

High

=5.1; SD=1.1; F(1,98)=

13.5; p<0.01) and motion detectors (M

low

=3.3;

SD=1.7; M

High

=4.2; SD=1.4; F(1,98)=8.4; p<0.01).

Interestingly, the use of cameras was evaluated

positively by both groups for being used in public

environments. For people with low PCT cameras

were even the only accepted crime surveillance

technology. Regarding traditional measures

enhancing perceived safety more police presence

Low

=4.9; SD=1.2; M

High

=5.5; SD=0.6; F(1,98)=

9.8; p<0.01) and more presence of private safety

services (M

Low

=3.6; SD=1.7; M

High

=4.9; SD=1.1;

F(1,98)=18.7; p<0.01) were more accepted by the

high PCT group than by the low PCT group. Both

PCT groups did not differ in their ratings of present

awareness of others and to be travelling by day.

5.2.5 Perceived Benefits of Crime

Surveillance Technologies

In a next step we analysed to what extent perceived

benefits of crime surveillance were influenced by

PCT (see Figure 5).

Figure 5: Perceived benefits of crime surveillance

technologies for high and low PCT groups.

Nearly all benefits were significantly more

accepted by the high PCT group, except

investigation of crimes, which was most accepted,

but not evaluated differently by the two PCT groups.

Deterrent effect (M

Low

=4.2; SD=1.6; M

High

=4.8;

SD=1.1; F(1,98)=4.5; p<0.05), safer feeling in

darkness (M

Low

=3.6; SD=1.7; M

High

=4.8; SD=1.3;

F(1,98)=14.7; p<0.01), sense of safety (M

Low

=3.5;

SD=1.5; M

High

=4.8; SD=1.2; F(1,98)=20.0; p<0.01),

safer feeling when traveling alone (M

Low

=3.5;

SD=1.6; M

High

=4.7; SD=1.3; F(1,98)=18.8; p<0.01),

and measure against safety risks (M

Low

=3.6;

SD=1.6; M

High

=4.7; SD=1.3; F(1,98)=15.2; p<0.01)

were also accepted and favoured by the high PCT

group.

5.2.6 Perceived Barriers of Crime

Surveillance Technologies

We also examined, how perceived barriers of crime

surveillance technologies were influenced by PCT

(see Figure 6).



Figure 6: Perceived barriers of crime surveillance

technologies for high and low PCT groups.

The highest concern for both groups was the

protection of personal information. Interestingly,

both PCT groups did not differ significantly in this

concern. This also applied to the barriers reuse of the

stored data, implies being under general suspicion,

violation of personal data and protecting civil rights

and personal freedom. In contrast, group differences

were found for risk of violating personal rights

Low

=4.9; SD=1.3; M

High

=4.2; SD=1.3; F(1,98)=

3,7

3,6

3,5

3,6

4,2

4,7

4,5

4,7

4,8

5,0

123456

prevention of crimes

measure against security

risks

safer feeling when

traveling alone

sense of security

safer feeling in darkness

deterrent effect

investigation of crimes

evaluation (min=1; max=6)

high

PCT

low

PCT

rejection

support

3,6

4,3

4,4

4,9

5,0

3,3

3,9

4,3

4,2

4,7

123456

implies being under general

suspicion

privacy is violated

protect civil rights and

personal freedom

reuse of the stored data

storage of recorded data

continuous observation

risk of violating personal

rights

protection of personal

information

evaluation(min=1; max=6)

high

PCT

low

PCT

rejection support

SMARTGREENS 2016 - 5th International Conference on Smart Cities and Green ICT Systems

6.9; p<0.01), continuous observation (M

Low

=4.9;

SD=1.4; M

High

=4.3; SD=1.5; F(1,98)=4.4; p<0.05)

and storage of recorded data (M

Low

=4.4; SD=1.4;

High

=3.9; SD=1.3; F(1,98)=4.2; p<0.05). These

barriers were rated higher by the low PCT group

than by the high PCT group.

5.2.7 Acceptance of Crime Surveillance

Technologies at Different Locations

Further, we analysed to what extent PCT influences

the acceptance of crime surveillance technologies at

different locations. First of all, the usage of crime

surveillance technology was generally more

important for people with a high PCT (see Figure 7).

Figure 7: Influence of PCT on the acceptance of crime

surveillance at different locations.

The high PCT group evaluated the usage of

crime surveillance technologies significantly more

positively independent of different locations.

Although crime surveillance was not desired at

private locations, there was a broader acceptance for

it in the high PCT group, e.g. for living room

Low

=1.3; SD=0.6; M

High

=1.9; SD=1.3; F(1,98)=

10.0;p<0.01). At semi-private locations crime

surveillance technology acceptance was also rather

low, while at this point there were higher ratings of

the high PCT group as well, e.g. favourite pub

Low

=2.0; SD=1.2; M

High

=3.3; SD=1.4; F(1,98)=

22.1; p<0.01) or own house entry (M

Low

=2.7;

SD=1.8; M

High

=3.7; SD=1.6; F(1,98)=7.9; p<0.01).

At semi-public locations the low PCT group rather

rejected crime surveillance, while it was accepted by

the high PCT group, e.g. schools (M

Low

=3.3;

SD=1.7;M

High

=4.4,SD=1.5; F(1,98)=12.2; p<0.01)

or parks (M

Low

=3.3; SD=1.7; M

High

=4.7; SD=1.2;

F(1,98)=22.3; p<0.01). Finally, at public locations

crime surveillance technologies were rather accepted

by the low PCT group, while it was strongly desired

by the high PCT group, e.g. public transport

Low

=3.9; SD=1.7; M

High

=4.9; SD=1.2; F(1,98)=

11.4; p<0.01) or train station (M

Low

=4.3; SD=1.6;

High

=5.3; SD=1.0; F(1,98)=11.4; p<0.01).

5.2.8 Trade-off between Safety and Privacy

In a last step, we examined the effects of PCT on the

trade-off between looking for safety and protecting

one’s own privacy (see Figure 8).

Figure 8: Influence of PCT on the trade-off between need

for safety and privacy.

1,1

1,3

2,1

1,7

2,0

2,2

2,7

3,3

3,4

3,6

3,3

3,5

3,8

3,9

4,3

1,5

1,9

2,8

2,5

2,9

3,3

3,7

4,4

4,6

4,7

4,8

4,9

5,0

5,3

123456

bedroom

living room

garden

car

church

favourite pub

cellar

own house entry

schools

museum

main roads

shopping mall

park

market

public buildings

public transport

station

train station

evaluation (min=1; max=6)

high PCT

low PCT

public

semi-

public

semi-

private

rejection support

3,3

3,8

4,2

4,5

4,8

4,9

5,1

5,4

5,8

6,8

7,4

7,7

7,8

7,9

8,2

9,4

9,7

1,9

2,1

2,3

3,4

3,0

2,9

3,0

3,5

3,9

5,5

5,6

6,2

5,9

7,5

7,1

8,7

9,4

12345678910

need for safety need for privacy

(1=strong; 5=low) (6=low; 10=strong)

high PCT

low PCT

private

semi-

private

semi-

public

bedroom

living room

garden

car

cellar

church

favourite pub

ownhouse entry

museum

school

shopping mall

main road

market

park

public buildings

public transport

station

train station

“How Fear of Crime Affects Needs for Privacy Safety” - Acceptance of Surveillance Technologies in Smart Cities

All in all, there were significant differences in the

assessment of the relationship between safety and

privacy concerning both PCT groups. Concerning

private locations there were no differences between

both PCT groups, because both groups desired to

protect their own privacy at those locations.

Regarding semi-private locations, the low PCT

group had a significantly greater need for protecting

own privacy than the high PCT group for nearly all

semi-private locations, e.g. cellar (M

Low

=7.8;

SD=2.5; M

High

=5.9; SD=3.1; F(1,98)=10.7;p<0.01),

garden (M

Low

=8.2; SD=2.3; M

High

=7.1; SD=2.9;

F(1,98)= 4.7; p<0.05) or favourite pub (M

Low

=7.4;

SD=2.4; M

High

=5.6; SD=2.8; F(1,98)=11.8; p<0.01).

At semi-public locations there were significant

differences for all locations: for the low PCT group

privacy was more important, while the high PCT

group preferred safety, e.g. schools (M

Low

=5.4;

SD=3.2; M

High

=3.5; SD=2.6; F(1,98)=10.2; p<0.01),

main roads (M

Low

=4.9; SD=3.0; M

High

=2.9; SD=2.1;

F(1,98)=15.9; p<0.01). Concerning public locations,

the low PCT group had a significantly stronger need

for privacy, while the high PCT group strongly

favoured safety, e.g. train station (M

Low

=3.3;

SD=2.7; M

High

=1.9; SD=1.5; F(1,98)=10.3; p<0.01)

or public transport (M

Low

=4.2; SD=3.1; M

High

=2.3;

SD=1.9; F(1,98)=13.3;p<0.01).

6 DISCUSSION

This study revealed insights into acceptance patterns

regarding the use of crime surveillance technologies

in urban environments. In order to understand the

specific needs of a diverse resident population, we

examined the tolerance towards such technologies at

various public and private urban locations. The

results provide valuable insights for city planners

regarding an acceptable employment of crime

surveillance technologies at different locations in

urban environments, which consider individual

needs for privacy and safety.

6.1 Acceptance of City Surveillance

Surveillance technologies are accepted in those

locations in which crime threat is present. Crime

threat reports were higher in public spaces such as

train stations or parks, especially during nighttime.

Accordingly, conventional crime surveillance

technologies (i.e. CCTV systems), but also

conventional measures such as lighting are well

accepted - as long as they are visible and installed in

public spaces. Especially in urban transportation

hubs such as train stations, stations or main roads,

where a high number of people passes by,

surveillance technologies are strongly accepted.

Accordingly, a map or cartography of acceptable

locations for the acceptable installation of

surveillance technologies in urban environments can

be derived from our findings. A completely different

acceptance picture can be drawn for the acceptance

of surveillance technologies in private spaces. Here,

perceived crime threat is comparably low, and the

use of cameras or microphones for the surveillance

of private spaces is distinctly rejected. Instead,

lighting and motion detectors are the only accepted

measures. However, this finding does not allow

jumping to the conclusion that surveillance

technologies in private space are rejected in general.

Combined with different functionalities than crime-

stopping functions, surveillance technologies already

have entered private spaces, e.g. webcams for

medical monitoring or “nanny- or mummy-cams”

(Kientz et al., 2007). Moreover, the context-

specificity of technology-acceptance was already

shown for wireless technologies either used for ICT-

or for medical monitoring purposes (Himmel et al.,

2013). Future studies will have to investigate in

more detail the effects of usage context and

“monitoring target” (myself or others) on the

acceptance of surveillance technologies.

6.2 Influential Factors on Crime

Surveillance Technology

Acceptance

The assessment of individual privacy and safety

needs provides an explanation for the identified

acceptance patterns. In public spaces, people have a

higher need for safety, i.e. they “sacrifice” their

privacy rights for a higher safety from potential

crime assaults. In turn, in private spaces, where

perceived safety is higher, the need for privacy is

dominating. However, in the present study,

surveillance technology was operationalized as

“presence of a camera”, without giving information

about further processing or usage purposes of

recorded data. We assume, that this

operationalization is ecologically valid, since people

usually do not know, which of their actions are

monitored and how or for what purpose surveillance

data is further processed and used (Patton, 2000).

Accordingly, we doubt that people are fully aware of

potential privacy violations, which might occur

during the following data processing stages. A next

step of our research agenda is, to investigate the

effects of information about potential privacy

SMARTGREENS 2016 - 5th International Conference on Smart Cities and Green ICT Systems

violations of subsequent processing stages on

privacy perceptions and behaviour. Research on

privacy issues and user behaviour in social networks

showed, that people – although claiming to be aware

of their privacy rights – show a completely different

behaviour pattern, i.e. exposing huge amounts of

personal information (Debatin et al., 2009). Looking

at perceived barriers and benefits of crime

surveillance technologies, which might serve as

explanatory variables for acceptance, we were rather

surprised by the result pattern. Almost all benefit

items received comparable levels of affirmation,

which might be explained by a biased response

behaviour or by an insufficient item design. For

barrier-items we found a slightly more differentiated

result pattern. The barrier “being under general

suspicion” received the lowest affirmation in our

study. This is especially noteworthy, since the issue

of “general suspicion” is a widely used counter-

argument in research literature ethical implications

of surveillance technologies (e.g., Marx, 1998), but

is apparently not reflected in individual perceptions.

This result further indicates, that the ethical-

normative approach of technology acceptance

research needs to be complemented by a “user-

focused” perspective to derive implications and

design guidelines which meet public acceptance.

6.3 Effects of Perceived Crime Threat

User diversity in terms of different crime threat

levels is a crucial factor in the context of crime

surveillance acceptance. The contrast of people with

high and low crime fears shows – not surprisingly –

that crime surveillance measures and their related

benefits are more accepted by people with higher

fear levels. Interestingly, the two groups with high

and low crime fears did not differ in their age or

gender. There is not a hypothetically typical

distribution with mainly women and older people

who feel more threatened by crime than men and

younger people. The distribution of segmented PCT

groups indicates that nearly each city dweller could

be part of the group with a high PCT and that

perceived PCT should be the starting point for the

development of urban surveillance concepts.

Overall, the predominantly technology-centered

planning of infrastructural city concepts, without

integrating citizens into the decision-making

processes, seems not sufficient to cover persons’

attitudes regarding safety and privacy concerns in

the context of smart cities.

6.4 Limitations and Future Research

Our empirical research approach was provided

valuable insights into the acceptance of crime

surveillance technologies. Some methodological

issues should be taken into account, though.

First, some aspects have to be criticized in terms

of content. The very similar evaluation of perceived

benefits of crime surveillance showed that the item

content might have been too similar. For further

studies it would be desirable to use more specific

and tangible items concerning perceived safety

aspects, e.g. a quantifiable potential decrease in

criminality rates. The same applies for perceived

barriers of crime surveillance: participant’s feedback

showed that the queried items could be more

differentiated. In further studies more specifications

regarding privacy aspects will be examined

(different handling of recorded data, storage issues

or even face recognition). Concerning crime

surveillance technologies this study focuses on the

distinction between visible and invisible

technologies. Future studies should differentiate

between specific visible and invisible technology

types. Another note refers to the classification of

locations. Here, we assumed the classifications that

were made by the participants of previous focus

groups. The distinction between public and private

locations is comprehensive and uncontroversial,

whereas the difference between semi-public and

semi-private locations is rather small. Therefore, in

further studies a more precise definition of location

categories is necessary. Besides terms of content, for

further studies other methodological approaches

should be applied. Since four relevant attributes

(location types, safety aspects, privacy aspects and

technology type) were identified in this study, the

implementation of a conjoint analysis could be

useful to gain a deeper insight into the acceptance of

crime surveillance. This way, the relative

importance of different attributes could be

determined and the trade-off between safety and

privacy could be characterized precisely.

Also, some aspects concerning the sample could

be improved and continued in further studies: first,

the sample size of this study was rather small, so the

findings should be replicated in larger and more

representative samples, which contain a higher

number of men and a higher number of older

persons. To involve place of residence as a

hypothetically influencing variable, further samples

have to contain a higher number of people living in

rural areas. Finally, as this study only focuses

German city dwellers, our approach and findings

“How Fear of Crime Affects Needs for Privacy Safety” - Acceptance of Surveillance Technologies in Smart Cities

could be replicated in other countries to compare

crime surveillance needs and desires of city dwellers

of different countries and cultures.

A similar remark is directed to the flow of

refugees and emigrants from Arabic countries into

all parts of Europe. Under these conditions, where so

different cultural values and norms regarding

intimacy, protection needs as well as personal

nearness and distance meet if not clash, perceptions

of security might be different. Therefore, future

studies should replicate the findings.

A final note regards the development of

communal or political policies. Even though the

findings here do not allow the formulation of

concrete recommendations for the use of

surveillance technologies, still, the findings could be

integrated in the education of communal workers

which need to know both sides of the coin: security

for the individual and the commune as such but also

the respect of keeping privacy of the individual and

the commune.

REFERENCES

Arning, K., Ziefle, M. and Mühlhans, H., 2013a. Join the

ride! User requirements and interface design

guidelines for a commuter carpooling platform. In

Marcus, A. [ed.], Design, User Experience, and

Usability, Berlin Springer, pp. 10-19.

Arning, K., Kowalewski, S., and Ziefle, M., 2013b.

Modelling User Acceptance of Wireless Medical

Technologies. Wireless Mobile Communication and

Healthcare, 61, pp. 146–153.

Alsnih, R. and Hensher, D. A., 2003. The mobility and

accessibility expectations of seniors in an aging

population. Transportation Research Part A. Policy

and Practice, 37(10), pp. 903–916.

Baumer, T. L., 1978. Research on fear of crime in the

United States. Victimology, 3, pp. 254–264.

Blöbaum, A. and Hunecke, M., 2005. Perceived Danger in

Urban Public Space The Impacts of Physical Features

and Personal Factors. Environment and Behavior,

37(4), pp. 465–486.

Chattopadhyayr, D., Dasgupta, R., Banerjee, E, R. and

Chakraborty, A., 2013. Event Driven Video

Surveillance System using City Cloud. Proceedings of

the first International Conference on Intelligent

Infrastructure at the 47th Annual National Convention

Computer Society of India.

Covington, J. and Taylor, R. B., 1991. Fear of Crime in

Urban Residential Neighborhoods. Sociological

Quarterly, 32 (2), pp. 231–49.

Debatin, B., Lovejoy, J. P., Horn, A.-K., and Hughes, B.

N., 2009. Facebook and Online Privacy: Attitudes,

Behaviours, and Unintended Consequences. Journal of

Computer-Mediated Communication,15(1),pp.83–108.

Dickerson, A. E., Molnar, L. J., Eby, D. W., Adler G.,

Bédard, M., Berg-Weger, M., and Trujillo, L., 2007.

Transportation and Aging: A Research Agenda for

Advancing Safe Mobility. The Gerontologist, 47(5),

pp. 578–590.

Gumpert, G. and Drucker, S. J., 2001. Public boundaries:

Privacy and surveillance in a technological world.

Communication Quarterly, 49(2), pp. 115–129.

Himmel, S., Ziefle, M. and Arning, K., 2013. From Living

Space to Urban Quarter: Acceptance of ICT

Monitoring Solutions in an Ageing Society. In Kuroso,

M. (Ed.) Human-Computer Interaction. Users and

Contexts of Use, Berlin Springer, pp. 49-58.

Isnard, A., 2001. Can surveillance cameras be successful

in preventing crime and controlling anti-social

behaviours. Proceedings of the character, impact and

prevention of crime in regional Australia Conference,

Townsville, 2.-3.8.2001.

Kientz, J. A., Arriaga, R. I., Chetty, M., Hayes, G. R.,

Richardson, J., Patel, S.N., and Abowd, G.D., 2007.

Grow and Know: Understanding Record-keeping

Needs for Tracking the Development of Young

Children. Proceedings of the SIGCHI Conference on

Human Factors in Computing Systems, San José, CA,

30.4. - 3.5.2007, NY, USA, ACM, pp. 1351–1360.

La Vigne, N. G., Lowry, S. S., Markman, J. A., and

Dwyer, A. M., 2011. Evaluating the Use of Public

Surveillance Cameras for Crime Control and

Prevention. Final Technical Report. Washington, DC,

The Urban Institute.

Lewis, D. A. and Maxfield, M. G., 1980. Fear in the

Neighborhoods: an Investigation of the Impact of

Crime. Journal of Research in Crime and

Delinquency

, 17(2), pp. 160–189.

Loewen, L. J., Steel, G. D., and Suedfeld, P., 1993.

Perceived safety from crime in the urban environment.

Journal of environmental psychology,13(4), pp. 323-

331.

Marshall, R. D., Bryant, R. A., Amsel, L., Suh, E. J.,

Cook, J.M. and Neria, Y., 2007. The psychology of

ongoing threat: relative risk appraisal, the September

11 attacks, and terrorism-related fears. American

Psychologist, 62(4), p. 304.

Marx, G. T., 1998. Ethics for the new surveillance. The

Information Society, 14(3), pp. 171–185.

Patton, J. W., 2000. Protecting privacy in public?

Surveillance technologies and the value of public

places. Ethics and Information Technology, 2(3),

pp.181–187.

Plouffe, L., and Kalache, A, 2010. Towards Global Age-

Friendly Cities: Determining Urban Features that

Promote Active Aging. Journal of Urban Health,

87(5), pp. 733–739.

Scarborough, B. K., Like-Haislip, T. Z., Novak, K. J.,

Lucas, W. L., and Alarid, L. F., 2010. Assessing the

relationship between individual characteristics,

neighborhood context, and fear of crime. Journal of

Criminal Justice, 38(4), pp. 819–826.

Sheldon, B., 2011. Camera surveillance within the UK:

Enhancing public safety or a social threat?

SMARTGREENS 2016 - 5th International Conference on Smart Cities and Green ICT Systems

International Review of Law, Computers &

Technology, 25(3), pp. 193–203.

Smith, M. J. and Clarke, R. V., 2000. Crime and public

transport. Crime & Justice, 27, pp. 169-233.

Song, M., Tao, D., and Maybank, S.J., 2013. Sparse

Camera Network for Visual Surveillance – A

Comprehensive Survey. Cornell University.

Van Heek, J., Arning, K., and Ziefle, M., 2015. Safety and

privacy perceptions in public spaces: An empirical

study on user requirements for city mobility. In

Giaffreda, R., Caganova, D., Li, Y., Riggio, R., and

Voisard, A. (Eds.). Internet of Things 2014, LNICST

151, Springer Berlin Heidelberg.

Whitaker, R., 1999. The end of privacy: How total

surveillance is becoming a reality. Ney York, NY, The

New Press.

Wiecek, C. and Saetnan, A.R., 2002. Restrictive?

Permissive? The Contradictory Framing of Video

Survei-llance in Norway and Denmark, Norwegian

University of Science and Technology, Working Paper

Ziefle, M., Schneider, C., Valeé, D., Schnettler, A.,

Krempels K.-H. and Jarke, M., 2014. Urban Future

outline (UFO) A roadmap on research for livable

cities. ERCIM News (N. 98): http://ercim-

news.ercim.eu/en 98/keynote-smart-cities.

Ziefle, M., and Wilkowska, W., 2015. What makes people

change their preferences in public transportation –

opinions in different user groups. In Giaffreda, R.,

Caganova, D., Li, Y., Riggio, R., Voisard, A. (Eds.).

Internet of Things 2014, LNICST 151, Springer Berlin

Heidelberg, pp. 137-144.

“How Fear of Crime Affects Needs for Privacy Safety” - Acceptance of Surveillance Technologies in Smart Cities