Differences in Attitudes towards Theft between British and Romanian Students

Fear of crime as a subject has received increasing levels of attention from scholars over the years. The phenomenon has been explored across diverse crimes and cultures with varying results, the better part of the literature revolving around general fear of overall crime, serious offences and Western, English-speaking countries. Adopting a more focused approach to the subject, the present study considers a lower-level crime, namely personal theft, and its effects on students in Romania as well as the UK. The aims of the project were to explore the differences in fear of crime characteristics such as sensitivity to risk and perceived likelihood of victimisation at a country level, to compare physical and psychological responses to theft, and to assess the influence of external factors such as the media. It was found that Romanian students experience significantly more fear and awareness of possible threats in almost any context. Trust in the police and reporting attitudes were highly similar between the two countries. Romanian students’ attitudes towards theft were affected by both media influences and exposure to theft information through the social network, while British students were only sensitive to the latter. Several other relationships between variables are explored throughout the study. Possibilities for future research as an expansion of this topic are proposed.


Introduction
Reported crime levels have been steadily decreasing across the world in recent times.
England and Wales have witnessed a continuous reduction in crime rates since 2002 (Office of National Statistics, 2018) and a similar pattern is occurring throughout the EU (Vieno, Roccato & Russo, 2013;Torrente, Gallo & Oltra, 2016;van Dijk, Tseloni & Farrell, 2012). However, a side-effect that does not seem to be decreasing in the general population is fear of victimisation. Despite the apparent sudden drop in offending, research suggests that the degree to which people report fear of crime is not always balanced against police reported trends and rates of crime and violence (Vauclair & Bratanova, 2017). Fear of crime has received increased interest from scholars over the years. Yet, despite the ever-expanding body of research, it is still not fully understood as a phenomenon that -it is suggested -has become a more significant issue than crime itself (Hanslmaier, 2013). Fear of crime is a multifaceted construct that encompasses a vast range of social and psychological reactions to the prospect of victimisation (Collins, 2016). The personal consequences of it include negative effects on mental healthranging from mild anxiety to depression and even PTSD (Lorenc et al., 2012) -lower quality of life due to reduced physical activities, constrained behaviours and less social interaction (Hanslmaier, 2013). Often, several negative outcomes occur simultaneously in high-risk individuals, who have difficulty coping under such circumstances (Ellis & [3] Renouf, 2018). Therefore, the underlying concepts and ramifications of fear of crime need to be well understood in order for its effects to be combated effectively.
The phenomenon of fear of crime has been explored across diverse crimes and cultures with varying results (Xiong, Nyland, Sue Fisher & Smyrnios, 2017;Brown, 2016;Vieno et al., 2013). Findings such as the surprising paradox of the least victimised being the most afraid (Ferraro, 1996;Ugwu & Britto, 2015), the media influence on feelings of vulnerability (Fox, Nobles & Piquero, 2009;Callanan & Rosenberger, 2015), and the importance of the environment and social cohesion in individual perceptions of one's community (Kriegler & Shaw, 2016;Roberts & Gordon, 2016), stand out. However, the better part of the literature revolves around general fear of crime, serious offences and Western, English-speaking countries (Shapland and Hall, 2007). Empirical research that aims to explain the differences in fear of crime between countries is surprisingly limited (Vauclair & Bratanova, 2017). Adopting a more focused approach to the subject, the present study considers a lower-level crime, namely theft, and its effects on students in Romania as well as the UK. Romania was chosen as a counterpart to the UK in an attempt to contribute to the underdeveloped body of research surrounding this country, particularly in the area of criminology. Moreover, a recurring element in fear of crime research is the importance of the socio-economic context (Alda, Bennett & Morabito, 2017;Hirtenlehner and Farrall, 2013). In particular, ex-communist countries are known to display both higher anxiety levels when confronted with the subject of victimisation, as well as an inclination towards police distrust (Hollis, Downey, del Carmen & Dobbs, 2017;Stickley, Koyanagi, Roberts, Goryakin & McKee, 2015). It was therefore reasoned that a substantial contrast will be found between the areas of study, given the difference in [4] socio-economic backgrounds. Little academic literature surrounding this topic has emerged from within Romania (Andreescu, 2010;Ghitescu & Banciu, 2010;Gruia & Gruia, 2014), and fear of crime studies targeting several nations, even within Eastern Europe, have overlooked Romania (Barni, Vieno, Rocatto & Russo, 2016;Stickley et. al., 2015). Thus, both the research subject and area have under-representation as a common element. Lastly, comparisons of this nature are somewhat rare and could potentially provide further insight into the role of culture in fear of crime.

Aims
The primary aim of this project was to explore whether there are any differences between British and Romanian students in the ways they understand and react to theft, or rather the possibility of it. This aim was further split into three main objectives, which targeted certain areas of interest derived from the literature. To address them, a comparison of the levels of fear of theft criteria such as severity, risk and perceived likelihood of becoming a victim was attempted, in accordance to the 'sensitivity to risk' model developed by Warr (1987). Following these criteria, assessing the differences between British and Romanian students constituted the first objective. The second objective was to examine whether the two groups would deal with the possibility and outcome of theft differently, in terms of precautions taken against it as well as physical and emotional responses. Finally, the project aimed to look at the possible roots of any such differences, as well as to place them in a cultural context, unpicking the influences behind this fear. Media influences and indirect encounters with theft were particularly scrutinised as potential underlying reasons. [5] The study focuses solely on 'theft from the person', which covers 'theft (including attempts) of item or items directly from the victim, but without the use of physical force against the victim, or the threat of it ' (Home Office, 2013). The data is based on responses about snatch or stealth theft -i.e. pickpocketing and similar forms -and it is with this meaning that the word 'theft' is used throughout the entire study.

Methodology and methods
A quantitative methodology was adopted in this study. The procedure was appealing for its deductive approach and positivistic theoretical perspective, as the main goal of the present piece of research was to produce 'results that are objective, valid and replicable' (Gray, 2014, p. 128). The empirical findings presented in the following section are based on data from an internet-based questionnaire study, which took place from February to April 2018. The survey was initially designed in English, then translated into Romanian and distributed in each country. The first page of the survey consisted of a participant information section, followed by 17 questions and ended with a debriefing page. The first three questions were reserved for the collection of demographic data such as gender, age and country of birth. The remaining 14 questions were split as follows: one asked participants about their feelings towards theft, four asked them to assess their own levels of fear towards the subject, three to evaluate their possible reactions to hypothetical situations, and the remaining six focused on the way they viewed theft in their respective countries. Responses were measured on a 6-point Likert scale, 1 generally corresponding to 'not at all' and 6 to 'extremely so'.
In the first stage, respondents were recruited through social media, while the rest of the data was gathered through snowball sampling techniques (Gray, 2014;Goodman, 2011).
The only requirements for participation were that the respondents must be over 18 years old, born in one of the target countries and currently at University -regardless of level of study. The survey was originally sent to 30 students from each country, with the request to pass it on to others who fit the profile and would be interested in participating. This method was selected as the most accessible and affordable way to reach a relatively large sample from the target audience (Wharton, Hampl, Hall & Winham, 2003). In the first instance, the surveys were meant only to be distributed to university students in the cities of Huddersfield, UK and Sibiu, Romania. The two had been chosen as they have a similar size, population levels and composition, and living environment. However, limiting the research in this way yielded very few responses, thus the areas were expanded to students in the whole country, in both instances. While this may be seen as a disadvantage, it could be countered with the notion that all students must have lived in a town large enough to have a university, which they can likely afford to attend, at the point of questionnaire completion. It may therefore be suggested that their attitudes are not remarkably different from respondents belonging to a single university. A total of 140 students attempted the survey (N=79 in the UK; N=61 in Romania). However, after controlling for missing answers and respondents native to countries outside the two targeted, an evenly split sample of 112 remained. The data was analysed using the statistical package IBM SPSS Statistics, version 24.

Findings and analysis
The findings were split into three sections, each addressing the differences in levels of fear of theft components, physical and emotional responses to the possibility of victimisation, and factors that may have influenced these attitudes, respectively. It was [7] found that Romanian students displayed higher levels of fear and awareness of the offence in nearly every instance or scenario, compared to their counterparts. Emotionally, however, while students in the UK mainly reported anxiety as the first response to theft victimisation, Romanians were slightly more inclined towards annoyance. In terms of physical responses, here measured through the likelihood of reporting the theft to the police, the two countries showed similar inclinations. The Romanian students suggested that they were more likely to report the crime than those in the UK sample, but attitudes were largely positive in both samples. In terms of influences, respondents in the UK showed scepticism towards the media representation of theft, while exposure to the phenomenon through the peer network seemed to strongly affect their attitudes. In Romania, both factors had similarly substantial effects on the way respondents viewed theft and their own victimisation.

Levels of fear
To begin, a first comparison was made between all the study variables. This was presented in a visual format to facilitate the interpretation of the differences in levels of fear of theft, and factors surrounding it, between the two countries. In Figure  reporting, beliefs about the prevalence of theft, media coverage and public opinion on the extent of theft. Overall, 21 variables were analysed, comprising all the survey questions that were based on a numeric response. The average response for each question was used for the comparison. It is promptly noticeable that, except for one corner, the UK area is fully contained inside the one corresponding to Romania. Thus, Romanian students [8] have scored higher in nearly all instances, showing higher fear, awareness and generally stronger attitudes towards theft. The only spot where British students surpassed their counterparts corresponds to 'Likelihood of victimisation', meaning the UK sample was slightly more likely to believe the possibility of becoming a victim of theft was high.
Likelihood of reporting and general awareness of belongings reached similar numbers in the two countries. The largest differences were recorded in the 'Theft exposure' area, where participants were asked how often they were confronted with the subject of theft both from media sources and outside of it. Whether they believed that degree to be representative of the true extent of the offence was also questioned. In summary, British students believed themselves to be more likely to be victimised and showed less trust in the media coverage and public opinion of theft, but showed less fear and alertness than their counterparts in all situations. *FoT stands for 'fear of theft' *AoB stands for 'awareness of belongings'

Physical and emotional responses
This section refers to the physical and emotional responses that participants reported to the possibility of theft victimisation. The former was measured by their likelihood of seeking legal action in such an event -i.e. reporting it to the police -while the latter analysed their feelings towards the offence. In Figures B1 and B2, the main emotions participants felt while considering the possibility of becoming a victim of theft are represented. In addition to the selection of arguably negative emotions displayed in the legend, which largely related to fear and frustration, respondents also had the option of choosing indifference, sympathy, happiness and relief. In the UK, feelings of anxiety were mainly reported, as the choice of 32% of the sample. Anger came second with 21%, followed by irritation in third place with 14%. The rest of the options enjoyed less attention overall. Interestingly, 2% of the sample reported feelings of sympathy. In Romania, the situation was somewhat different. The surprising main emotion reported by 21% of students in response to the possibility of theft was annoyance. Nonetheless, anxiety followed close behind, chosen by 18% of the sample. Only a few steps below were fright at 16% and irritation at 13%. Sympathy was not reported by any of the Romanian participants. Neither country showed indifference towards the possibility of theft victimisation, nor a positive emotion.
[10] The likelihood of reporting a theft incident to the police of each country is represented below, in Figure C. The horizontal axis represents the level of agreement towards reporting -1 corresponding to 'would definitely not report' and 6 to 'would definitely report' -while the vertical axis shows which percentage of the sample chose a particular option.
In this graph the two lines follow a similar route, both countries selecting 'definitely would report' as their leading choice, by 36% in Romania and 27% in the UK. Curiously, the line corresponding to the British population peaks once more at 3, with 25%. This divide suggests that the students are equally likely to be fully positive and somewhat reticent towards reporting. In Romania, however, the line climbs steadily with each option, soaring at 'definitely would report', where the percentage of the previous choice doubles. scale. There appears to be a great deal of consensus between the two countries. Both British and Romanian students seem to have quite similar opinions about the police performance in theft situations; in fact, out of all the study variables, the least differences were found here. An overwhelming majority (65%) in both countries stated that they do not believe the police would bother for such a minor event. A much smaller percentage declared that reporting is too troublesome (8-9%) or the incident would not matter enough to them (4% each). It should be safe to assume that, from the point of view of the respondents, it is the police who believe theft to be too minor an issue to address it. The rest of the distribution between the two countries was similar, with two minor exceptions.
First, Romanian students are up to four times more likely to believe that the police would be unable to solve the incident than British students. This may either translate to a lack of confidence in the competency of the law enforcement or an acknowledgement of the difficulties of tracking down petty thieves who may operate in crowded areas and leave no evidence behind. Second, a small percentage (9%) of the UK sample stated that their decision to report depends on the stolen goods, perhaps based on the financial or emotional value of the item. It should be noted, however, that this was not a predetermined option; the question gave participants the ability to add any other reasons for not reporting, which only a number of British participants chose to do. If it had been a predetermined option, this might have been selected by Romanian participants as well, or the UK percentage might have been higher.
[13] The top question refers to the frequency of theft incidents seen by respondents in the media, specifically in news reports. The average response from the Romanian sample was 4.48, a rather high score on a scale of 1 to 6 -the latter corresponding to 'extremely [14] often'. In the UK sample, the mean fell to 3.16, suggesting that theft representations are somewhat infrequent in the British media. As a follow-up, participants were then asked whether they believed the amount of theft presented in the media to be representative of the extent to which it occurs in reality. The average Romanian response fell in the middle of the scale (M = 3.39), which would translate to 'neither accurate nor inaccurate'.
British students, however, showed less confidence in media accuracy, reporting a mean of 2.04. It can only be speculated as to whether they believed the news regarding their portrayal of the extent of theft to be too positive or too negative. Moving on from media coverage, outside influences were measured. The students were tasked to consider the degree to which they are informed of theft risks by sources outside the press. Specifically, they were asked how often theft is brought into conversation by their peers, either to discuss a theft occurrence or the general phenomenon. The reasoning behind this question was to gain the ability to explore whether indirect crime exposure was more likely to affect attitudes towards an offence when real-life people were involved, as opposed to experiencing it more distantly through the media. Alas, participants from both countries seem to believe the subject was not brought into conversation particularly often (M = 3.27 in Romania, M = 2.86 in the UK). Finally, regarding the accuracy of public perception regarding the extent of theft in their country, the answers varied between 'somewhat inaccurate' and 'somewhat accurate'. The British sample showed slightly more scepticism with a mean of 3.05, while the Eastern European students averaged 3.8 for this question. [15]

Figure E. Public opinion on exposure to theft, split by country
Multiple linear regressions were performed to assess the degree to which two different sources of theft exposure -the media and the social network -are able to predict different attitudes. The levels of perceived likelihood of victimisation, how common respondents believed theft was, and how much their daily routine was affected by the threat of theft, were tested against these predictors. Table 1, displayed below, shows the situation in Romania. Exposure from the media had a significant effect on 'likelihood of victimisation' (F (1, 53) = 9.123, p = 004) and perceived 'frequency of theft' (F (1, 54) = 12.974, p = .001), but did not manage to explain more than 1% of 'routine disturbance', with a high pvalue of .610. High levels of media coverage most effectively affected students' perceptions of how frequently theft occurs, the Beta coefficient being moderately high at [16] .440 on a scale of 0 to 1, and the adjusted R squared indicating that nearly 18% of the variance in the dependent variable is explained by the predictor. On the other hand, exposure to theft through discussion with peers predicted all three variables at a significant level; it is noteworthy that 31% of the perceived likelihood of victimisation reported by Romanian students was explained by this variable (F (1, 53) = 25.418, p = .000, β = .596). 'Routine disturbance' showed a weak (adj. R2 = .060), albeit significant (p =.040) correlation with 'social exposure', while 'frequency of theft' was predicted slightly better (F (1, 54) = 8.265, p = .006). This analysis suggests that Romanian students' beliefs and reactions towards crime are largely influenced both by media coverage and exposure to the subject from the public. The predictors behaved similarly when measured against the dependent variables, although the latter showed stronger explanatory powers.
Controlling for outliers revealed that the predictions are rather precise and thus reliable, as there were no more than 3 (often as little as 1, in a sample of 56) respondents outside the 2+ standard deviations line in any given analysis.
[17] The same analysis presents a different picture in Table 2, where multiple regressions were calculated to assess the situation in the UK. In this case, media exposure did not significantly predict any of the dependant variables, the p-value ranging from .105 to .782, extremely high numbers when compared to the results from the Romanian sample.
'Frequency of theft' showed the strongest association with media coverage, which was also the case in the Eastern European country, yet only 3% of the variance (adj. R2 = .030) was explained by the predictor. 'Social exposure', on the other hand, was a much stronger predictor. Nearly 30% of the variance in perceived likelihood of victimisation was was not predicted at the same level, R squared showing only 8%, yet the relationship is still significant on its own (p = .017) and more significant than in Romania (p = .040).
Outliers were once again very few (up to 4 at 2+ standard deviations away), suggesting a consistent and valid prediction. Therefore, it may be concluded that students in the UK are much more influenced in their attitudes towards theft by their social network than by the media, perhaps due to the fact that the former both bears more credibility and provides a more direct link to crime by removing the element of detachment. This may also be explained by the finding that British students believe the media representation of theft to be highly inaccurate. [19]

Discussion
Three main research questions guided the current study. The first objective was to discover whether Romanian or British students reported higher levels of fear of theft criteria -such as sensitivity to risk and perceived likelihood of victimisation -than their counterparts. The most consistent finding throughout the analysis was that Romanian students displayed a significantly greater degree of fear of theft characteristics than the British students. This was the case, both in terms of fear of victimisation and awareness .240 .085 [20] of victimisation possibilities, in all situations. This finding satisfies the hypothesis that Romanians are more sensitive to risk in their daily lives, consistent with similar research on the subject and in countries with a similar background (Stickley et al., 2015;Andreescu, 2010;Hollis et al., 2017). In terms of perceived likelihood of victimisation, however, the results were unusual. While it was not a substantial difference, students in the UK scored higher than those in Romania in this scenario. Considering the fact that this was the only point where British students reported overall higher numbers than Romania, it may warrant further examination.
The second objective was to learn whether students in the two countries react to the possibility of theft differently, in terms of both feelings and behaviours. According to Shapland and Hall (2007), individuals reacting to crime tend to fall into one of two categories: either a generally emotionally distressed group or an action-oriented group.
The former experience high anxiety levels that they carry throughout their lives and have difficulty finding coping mechanisms, while the latter focuses on taking precautions or organising crime prevention measures in their community. In the UK sample, emotional responses were measured to reveal anxiety, anger and irritation as the prevalent psychological reactions to theft, consistent with a large body of literature from the country (Rühs, Greve & Kappes, 2017;Shapland & Hall, 2007). The most common reaction to theft in Romania was annoyance, however, followed by anxiety, fear and irritation. While the element of fear was certainly present, feelings related to frustration took the lead.
Scholars have indicated that, while the psychological effects of crime tend to decrease over time, the opposite is true for preventive behavioural changes (Shapland & Hall, 2007). In terms of practical responses, the likelihood of reporting a criminal incident was [21] assessed, alongside attitudes towards the police. Students in both countries displayed a high likelihood of reporting, more so in Romania than in the UK. It is unclear whether this translates to more police trust in Romania, or if the loss of property would simply mean more to an Eastern European for socio-economic reasons. High levels of frustration towards the incident may also be a reason for higher reporting, as annoyance may motivate a reaction more than anxiety. Reasons for not reporting were similar between countries, with a strong consensus towards the idea that the police would not bother for a theft incident.
Lastly, the project aimed to examine the influences behind fear of theft and whether they differed by country. Media coverage and indirect exposure through the social network were the chosen possible predictors. British students encountered significantly less theft in the media than their counterparts, but they believed that image to be an inaccurate representation of the actual nature and level of theft happening in their nation. On the other hand, they felt theft was a rather common topic of conversation in their daily lives.
Some scholars suggest that social interaction may lessen fear of crime through the element of support (Scarborogh, Like-Haislip, Novak, Lucas & Alarid, 2010; Gibson, Zhao, Lovrich & Gaffney, 2002), while others argue that anxiety may actually increase through prolonged social interaction due to the increased communication on the subject of crime (Lorenc et al., 2012). The effect of both media coverage and social interaction on theft and victimisation opinions were assessed. It was found that in Romania both elements had a similar impact on students' attitudes, while in the UK the media element did not successfully predict any beliefs or inclinations. It was therefore concluded that the [22] social network has a much stronger impact on British students' attitudes towards theft than sources such as the news.

Limitations
The piece of research presented here faces several limitations. To begin with the methodology, the fact that data was gathered from two different countries via online surveys opened more than one possibility for information to be lost in translation.
Alongside this initial hurdle, the general limitations of unsupervised survey completion were present, such as inability to clarify a question or the issue of subjective interpretations of question or response meaning, which may vary depending on the past experiences and personality of the respondent (Ellis & Renouf, 2018). There is also no certainty that the survey was taken seriously or the answers were honest (Bryman, 2016).
Attempts to tackle both these uncertainties included the use of few, straight-forward and non-invasive questions. In terms of sample, the size was relatively small in each country due to lack of time and resources, which may reduce generalisability (Bryman, 2016). For similar reasons, the survey was confined to a university level and only targeted students, yet the attitudes of the general population may be quite different. It is likely that the socioeconomic background of participants also played a role in the distribution of responses, because respondents' living environment and circumstances have been found to greatly affect dependent variables such as fear of crime.
In terms of analysis, an unexpected obstacle was working with Likert-type scale data, the efficacy of which is widely debated to this day (Cheng & Yui, 2016). For this reason, any results should be interpreted with caution as they may not be fully valid or reliable. One [23] other source of error may have been the fact that large amounts of data were introduced manually into SPSS and modified several times, a process during which errors may have occurred. A final limitation was the result of ethical restraints; whether the participants had ever been victims of theft, or known people who have, was not a possible question.
Therefore, relying only on hypothetical scenarios, it cannot be predicted to what degree this information would have affected the results. A comparison with national statistics was attempted, however no such data was found in Romania regarding theft and using only UK statistics would have caused disproportionality in the study.

Conclusion and recommendations
To conclude, the present research is an exploratory study that aimed to find differences in attitudes towards theft between British and Romanian students. Such differences were found consistently throughout the analysis. Theft as a stand-alone offence remains underanalysed as it does not produce strong enough reactions in comparison with serious crimes to warrant further research. However, this study revealed that it is a substantial worry in the lives of many individuals. As expected, all aspects of theft were a significantly more dominant factor from the perspective of the Romanian students. As a country severely lacking in criminological research, this is a call for and one step towards such undertakings in Romania, as well as in other parts of the world that social research has yet to reach. In the future, similar studies may benefit from larger sample sizes, comparisons with national statistics of the offence in question, ideally from victimisation surveys.
In terms of practicality, research on fear of theft may help combat theft and improve public [24] satisfaction with the law enforcement system and people's own lives. A gap in the literature was addressed in this piece of research. This new knowledge would ideally be disseminated, expanded into a more comprehensive and generalisable project and eventually acted upon to further minimise fear of crime. Future comparative research of a related nature may benefit from a media content analysis at the time of the study, to verify for instance the frequency at which an offence is mentioned as well as the context and manner of presentation. Such information would be usefully linked to participants' responses surrounding media coverage, as well as provide better insight into the ways culture and mediatisation shape one another. Finally, to delve deeper into the social and cultural aspects that underpin such differences in attitudes between countries, follow-up qualitative research may yield valuable results. Alternatively, the continuing on the quantitative route may also provide further insight into the topic, by recreating the research in other countries and comparing the overall results.
In typical social research fashion, more questions arise from the questions answered, each offering the possibility to uncover more knowledge about social phenomena. Fear of crime has been and will continue to be a topic of interest for many years to come, and the research presented here hopes to be a modest contribution to the subject.