The data set used and analysed consisted of results from residents (N=300) who participated in the 2014, Gold Coast Community Survey on fear of crime and the factors that are associated with individual perceptions of what contributes to their fear. The data gathered from the survey analyses groups of categorical variables including fear, demographic characteristics, news and information, as well as community characteristics. Fear and News and information are categorised into their own variables with multiple values, whereas demographic and community characteristics are grouped represented by individual variables and further represented through multiple values. Demographic characteristics include; gender, age, income and education level. Community characteristics include; collective efficiency and Social cohesion. A detailed description of the data set including values is shown in Table 1. In this analysis the primary focus is to determine the association between fear and various other factors, thus determining that fear is the categorical dependant variable and the subsequent variables are independent variables.
Sub-sample size and Frequencies of variables. (N=300)
|Variable||n||% of variable|
15 – 24
25 – 64
55 – 65
|Highest Level of Education Completed
Year 11 or 12 or equivalent
|Primary Source of News and Information
To determine whether there was a connection between fear of crime and various factors that could possibly influence or are associated with each individual’s perceptions, a chi-square r x c test for independence was conducted on the assembled data. This test was chosen to be conducted for this analysis due to all the variables being used are categorical with multiple values. Therefore meeting two assumptions for the chi-square test for independence; all categorical variables (Nominal or Ordinal) and should consist of two or more categorical variables. The other primary assumption of the chi-test for independence, which is the expected frequency should not drop below five in more than 25% of the cells in a contingency table was also met. The results displayed only two (4.55%) cells falling below the expected frequency count of five, with the minimum being 2.08, therefore not contributing to more than 25% cells of the contingency tables.
A chi-square r x c test for independence was performed to examine the relationship or association between fear of crime and various factors that contributed to each participants perceptions. Within this analysis there were multiple variables to be examined to determine the association with fear of crime, the significant findings will be discussed prior to results table 2. Within the age of participant variable, 48% of participants over the age of 65 were fearful of crime, compared to 2.3% of participants aged between 55 and 64, 4.7% of participants between the ages of 25 and 54 years, and 3.3% of participants aged between 15 and 24 years. The relation between the dependant variable fearful / not fearful and the variable age of participant, showed that there was a significant association (X² (3, N=300) = 106.59, p ≤ .001). The Cramer’s V was 0.59, thus resulting in approximately 35% variance of frequencies of fear can be explained by the variance of age. Within the variable news and information, 46.7% of participants perceived the television increased fear of crime compared to 3.3% due to the radio, 7.0% due to print, 1.3% due to the internet and 0% due to other sources. The relation between the dependant variable and the variable of news and information, showed that there was also a significant association (X² (4, N=300) = 59.39, p ≤ .001). The Cramer’s V was .445, thus resulting in approximately 20% variance of frequencies of fear can be explained by the variable of news and information. Both the Age variable and the news and information variable showed statistically higher associations with fear of crime, representing factors from demographic characteristics and news and information; compared to alternate variables, particularly community characteristics. Further detailed results of variables shown in table 2.
Results of chi-square test on variables associated with fear on crime
|Variable||df||χ²||p value||V||Variance %|
|News and Information||4||59.39||≤.001||.445||19%|
The variables age and news and information both have a significant association with the fear of crime within the Gold coast community. Addressing the research questions, the preceding data demonstrates that demographic characteristics and news and information both are related to residents fear of crime thus, concluding that the answers to research question one and two are, true, there is a relation The third research question enquiring the relationship between community characteristics and residents fear of crime, although the data concluded there is a slight relationship, it is not as significant as the other variables. Therefore it is suggested that strategies address the residents fear on crime by focusing on the factor of age and the production of news and information of crime, to alter the perceptions.