BOCSAR web survey �€“ meeting our clients�€™ needs (Jun 2011)

Background


From January to December 2010, there were an estimated 333,883 user sessions on the BOCSAR website. Between July 2010 and May 2011, we responded to more than 600 requests for crime and court information through our information service. While the number of requests for information in 2010/2011 was lower than in previous years, BOCSAR implemented significant reforms to the website in 2010/2001. Anecdotal feedback suggests that it was these improvements that led to a reduction in requests for information as clients were much better able to find the information they required directly from the website.

As part of a determined attempt to improve our information service and to asses the reforms made to our website, we conducted a survey of our website users during April and May 2011. The key aims of the survey were to assess 1) whether clients were able to easily access the information they required and 2) the functionality of our current online data tools.


Method


Participants were recruited to the survey using several methods:


1) A link from our home page

2) An internal email to all of the Department of Attorney General and Justice

3) A link from the email signature of all BOCSAR staff

4) An email alert from bulk mail out lists (i.e. Crimnet, BOCSAR seminar email list)

5) Other forms (Twitter, word of mouth)


Results

Between April 15thand May 30th, 701 people completed the BOCSAR website survey. In what follows we summarise the results and highlight some of the comments made by respondents.

Respondent type
Seventy eight per cent of the survey respondents were government employees, with 66 per cent of these coming from the state government (Table 1).

Table 1. Type of survey respondent
Respondent type
Per cent of respondents
State government employee
65.8%
Local government employee
9.8%
Private citizen
5.6%
Other (please specify)
4.5%
University student
3.3%
Commonwealth government employee
2.0%
Research organisation employee
2.0%
University faculty or staff
1.7%
Media representative
1.2%
Private industry employee
1.2%
Non-profit organisation representative
1.1%
School teacher
0.6%
High school or primary school student or teacher
0.5%
Advocacy organisation representative
0.5%


Assessing client needs
Clients accessed the BOCSAR website for a variety of purposes. Figure 1 shows that 44 per cent of respondents were looking for general crime information, followed by 40 per cent of respondents looking for crime statistics for a local area.

Web survey figure 1

When asked whether they had found what they were looking for, 82 per cent responded �€œYes�€, 14 per cent of respondents said �€œI found some things but not others�€ and only 3.5 per cent responded �€œNo�€. Of those who responded �€œI found some things but not others�€ and �€œNo�€, open text answers show that 18 per cent of were looking for lower level crime data than what is currently available on our website.Some if these comments were:

�€œLocalised/suburb based crime statistics�€

�€œUp to date information; information broken down by police command rather than local government area.�€

�€œSuburb specific crime statistics.�€

�€œIt would be great to have access to postcode stats - at least in the city centre�€

Other key themes emerging from the open text answers show that participants were looking for but couldn�€™t find data on victim and offender characteristics or ethnicity/Indigenous data. Comments regarding victim and offender characteristics included:

�€œCrime data by victim and offender and age breakdown�€

�€œSpecific victim data�€

Comments regarding ethnicity/Indigenous status data included:

�€œEthnicity�€

�€œSpecific statistics relating to Aboriginal people and court appearances [sic].�€

�€œEthnic identity�€

�€œSpecific information regarding Aboriginal statistics�€.

Assessing our current online data tools
Of those surveyed, 38 per cent had used the online data tools. Seventy per cent of these had used the Crime Trends tool, 69 per cent had used the Crime by Type of Place tool and 50 per cent had used the Ranking of Regions tool.

Figure 2 shows that 86 per cent of these users either agreed or strongly agreed that the tools were easy to use.


Web Survery Figure 2

When asked �€œis there anything else you would like us to include in tools?�€ open text answers showed that, as above, many users wanted specific crime data at a lower level than is currently available. Some of these comments were:

�€œPerhaps the tools can be modified to allow narrowing down the stats by suburb? For example it'd be useful for me to know how often a certain type of crime occurs such as a 'break in' so I can add more security to my home�€

�€œPostcode data�€

�€œTo be able to generate reports at postcode level to get more local information rather than just LGA�€

�€œLocalised suburb based statistics�€

While not specifically asked, feedback regarding BOCSAR�€™s Crime Maps highlighted areas for improvement. Comments included:

�€œMore explanatory notes on the maps�€

�€œAccess to the maps is a little difficult - i.e. you need to bring up the whole crime map report which is slow to load. Access to the maps individually would be helpful�€

�€œLGA hotspot maps should be updated more frequently�€

General Feedback
Other comments were generally positive. For example:

�€There is always room to improve websites but this is a good one that clearly has a development pathway. As a member of another government agency working extensively with data, I am very impressed with the range and presentation of the data - exemplary practice. There are many clever ideas and ways of communicating information that could easily overwhelm the non-expert. I am very impressed with the Hot Maps as a useful and useable communication tool. Congratulations on your strong client focus and thoughtful presentation of data of interest to the public.�€


However, the ability to search for data was an area that was highlighted as needing improvement. For example:

�€œEasier ways of searching for specific things, throughout the website�€

�€œA search facility...�€

�€œI found the A-Z listing frustrating as when I clicked on a topic, it didn't really take me to anything that was relevant. If there is nothing available on the topic, it shouldn't be there.�€

�€œInclude a search function to make it easier to find certain subjects/crimes/statistics�€.

In addition, numerous users identified the lack of victim, offender and time of data on our website. As this data is currently available online improving access and navigation to these pages is a priority.


Conclusion


The BOCSAR website survey identified that people are generally very happy with our website and the services available online. However, the survey also identified some areas where we can improve our client services. The primary area where our services appear to be falling short of meeting clients�€™ needs is in the level of aggregation of our crime data. BOCSAR�€™s primary unit of measurement is Local Government Area yet it appears that many clients are seeking postcode or suburb level data. Other key reforms that may be required are to make victim and offender more easily accessible online, to increase the amount of Indigenous crime data online and to introduce a search facility on our website.

Some of these reforms have already been delivered. We have made reforms to our information services policy to make postcode data freely available. It is hoped that the Department�€™s new web platform will enable a search function to be embedded on our site. BOCSAR�€™s proposal to reform our online data tools, if progressed, will enable much greater access to low-level crime data, as well as more easily accessible victim and offender data.