Computers & Education 43 (2004) 193204

www.elsevier.com/locate/compedu

Internet safety in emerging educational contexts

Jocelyn Wishart
*

Department of Information Science, Loughborough University, Loughborough LE11 3TU, UK

Abstract Concern has arisen for the safety of children using the Internet to support their education outside the school context. Inappropriate material such as pornography, inammatory and racist writings can be accessed both by accident and with deliberate intent to view. Children are also perceived to be at risk from approaches by strangers, particularly in web-based chat rooms. A survey of the international literature in this area highlighted worldwide concern for the safety of young Internet users and it was generally agreed that schools have a fundamental role in ensuring their safety. The survey ndings indicated that a thorough audit of Internet Safety teaching and practices was a vital stage in examining these practices and informing future planning. This paper reports on the consequent audit of Internet Safety practices in over 500 schools from 27 Local Education Authorities (LEAs) across England, commissioned by Becta and conducted during the Summer term 2002. Independent, state and special schools were included in the survey at both primary and secondary levels. ICT advisors from the sampled LEAs and representatives of Internet Safety organisations were also invited to complete a linked questionnaire. A number of recommendations for Internet Safety teaching, in particular, ensuring that children are aware of safe practices for surng the Internet in less regulated contexts outside the school, will be presented based on the evidence gathered in the survey. Additionally emerging concerns for Internet Safety practices in schools such as the technical and moral diculties of ltering Internet access via mobile technologies will be highlighted. 2003 Published by Elsevier Ltd.
Keywords: Improving classroom teaching; Media in education; Internet safety

Tel.: +44-1509-223082; fax: +44-1509-223053912. E-mail address: j.m.wishart@lboro.ac.uk (J. Wishart).

0360-1315/$ - see front matter 2003 Published by Elsevier Ltd. doi:10.1016/j.compedu.2003.12.013

194

J. Wishart / Computers & Education 43 (2004) 193204

1. Introduction Childrens use of computers in contexts outside the school has increased dramatically in the past few years. The Department for Education and Skills (DfES) report Young People and ICT (DfES, 2002) found that 75% of children use computers at home with Internet access being available in 64% of households. Concern has therefore arisen for unsupervised childrens safety when surng the web as the Internet both contains material that parents may not wish their children to view and allows children to chat with people they have not met and who may not be as they seem. Several recent UK research programmes raise the issue of Internet Safety. For example, the Cyberspace Research Unit examined young peoples use of chat rooms (OConnell, Sange, & Barrow, 2002); the Screen Play Project assessed childrens use of computers (Economic & Social Research Council (ESRC), 2000) and the Cyberkids project examined virtual communities and online spaces (ESRC, 1999). The DfES report highlights the current concern for the safety of young Internet users nding that Three-quarters of parents said they were concerned about Internet Safety issues. The percentage was similar across all child age groups and by social grade... ( DfES, 2002, p. 36). So, although a recent report examining young peoples experiences reported that Children are missing out on the real gains of the Internet due to parents fears of dangers in cyberspace (Institute for Public Policy Research (IPPR), 2001), parents remain fearful, and reportage of survey ndings in the press tend to exacerbate this worry. There is also a conict between perceived and actual risks. One of the largest perceived problems is accessing unsuitable material, however, OConnell et al. (2002) found that Accidentally going on these sites [adult sites] often is very low but does seem to increase with age. The key problem she found was children giving out their personal details over the Internet. Children start giving personal information such as rst name (5%), last name (4%), e-mail address (3%), photograph (2%), phone number (1%) and home address (0.7%) in chat conversations from the age of 9. By the time the children reach the age of 16, they seem to be divulging a higher percentage of information at a rapid rate. Furthermore, OConnells ndings suggest that 1 in 10 children who use chat rooms have attended a face-to-face meeting with someone they met online. Additionally there is the problem of online victimisation with it being reported that 1 in 4 children have experienced online bullying via mobile phone text messages, email or chat rooms (NCH, 2002). Whilst no comprehensive survey of Internet Safety practices in English schools was found, the literature itself highlighted that schools (both in England and internationally) were perceived to have a vital role in promoting and ensuring Internet Safety. For instance, a survey undertaken in Ireland revealed that 49% of parents thought that schools should provide online safety information (Amarach, 2001) and a Canadian survey showed that 86% of parents thought it very important that schools improve the online safety of children using school computers (Media Awareness Network, 2001). In addition to parents, the Childrens Charities Coalition supported the notion that schools had a fundamental role to play in delivering Internet Safety measures Clearer guidance should be oered to schools on the safe use of Internet...emails...Chat rooms...school web sites...ltering and blocking software (Childrens Charities Coalition for Internet Safety, 2001).

J. Wishart / Computers & Education 43 (2004) 193204

195

OConnell et al.s (2002) ndings indicate a perceived shortfall in Internet Safety training in UK schools which arguably results in children not being adequately equipped to safely deal with the challenging circumstances they may encounter in an on-line situation. Her reports recommendations were important in supporting the need for the kind of comprehensive information gathering conducted for the project reported in this paper. Also OConnell recommended that Schools ought to be the main point of delivery (2002, p. 10) in providing a program of education for Internet Safety guidance, and that they should foster a synergy between home and school so that young peoples two main sources of advice work together to impart the same messages. (2002, p. 10). This point is also made by Livingstone (Institute for Public Policy Research (IPPR), 2001, p. 17), who recommends A co-ordinated response across school, community and home is essential for safe and fair use of the Internet by children. Thus the literature survey ndings indicated that a thorough audit of Internet Safety concerns and practices in schools was a vital stage in examining these practices and informing future planning by education professionals. The objectives of this audit were: To identify which schools teach Internet Safety, in what ways, with which age groups and in what areas of the curriculum. To identify which are the current Internet Safety issues for schools, which are emerging and the overall importance schools assign to the topic. To identify where schools currently get advice from on Internet Safety and how they respond to that advice. To identify what breaches of Internet Safety have taken place within the school and how this has impacted upon their teaching of the subject.

2. Method 2.1. Sample In order to gain a suitable number of schools that formed a representative cross section of schools across England, three Local Education Authorities (LEAs), one city based and two suburban or rural, were randomly selected from each of the nine areas in England dened by the National Grid for Learning (NGfL) Website (2002). Forty-three schools from each of these LEAs were then randomly selected from the lists of Primary, Secondary, Independent and Special schools displayed by Schoolsnet (2002) and the Times Educational Supplements (2002) Schools Directory. This made a total of 1160 schools however, 60 of the selected schools were dropped from consideration after it was found that they had recently taken part in another Becta project. 2.2. Research instrument Respondents were given the option of completing a questionnaire or responding to the same questions as part of a telephone survey. Burnard (1994) describes the advantages of using structured rather than unstructured telephone interviews, in particular it makes for ease of data analysis. Therefore, it was decided to design a schedule of questions for a telephone interview.

196

J. Wishart / Computers & Education 43 (2004) 193204

This schedule was then sent in a questionnaire format to the prospective interviewees beforehand to brief them fully about the proposed research in order to gain their informed consent. It was particularly important that the participants had the opportunity to see the questions rst as the survey concerned Internet Safety issues, which are a sensitive subject for schools allowing Internet access to children. A second revised version of the questionnaire was sent by e-mail to the ICT advisors in the 27 LEAs selected for the schools survey and a third version questionnaire was tailored for and sent by email to representatives of the following English Internet Safety organisations: Childnet International, NCH, Internet Watch Foundation, National Society for the Prevention of Cruelty to Children, Childline, dot.safe and the Parents Information Network. 2.3. Data collection Telephone interviews were chosen as the preferred data collection method. They have been found by McCann, Clark, Taylor, and Morrice (1984) to gain a better response rate, better answers to complex and sensitive questions and more complete recall of information than postal questionnaires. Also the data had to be collected before the end of the summer term making for a very short time scale and it had to be collected from a wide geographical range of schools right across England. Schools were contacted by a member of the project team at a time convenient to the school and structured telephone interviews using the questions from the questionnaire were conducted between a member of the project team and the schools ICT co-ordinator. On a few occasions the network manager completed the survey in place of the ICT co-ordinator. 2.4. Analysis Data collected from the three questionnaire surveys (both written and telephone) was input into an Excel spreadsheet, collated and transferred to SPSS (Statistical Package for the Social Sciences) for analysis. For questions collecting quantitative data, frequencies of the responses to each question were calculated and cross-tabulations of the results were made. Where appropriate the v2 statistic was used to test for statistically signicant associations within the cross-tabulation. For responses to the open-ended questions seeking qualitative data a thematic content analysis was carried out by identifying common themes in the respondents answers and reporting on the frequency of their occurrence. Where correlation between variables was calculated Spearmans rank order correlation coecient was preferred as the data did not follow parametric rules.

3. Results and discussion Data was obtained from 577 schools mostly by telephone interview. Questionnaires were also received in the post and by e-mail (see Table 1). As shown in Tables 24, the schools contacted as part of the audit represented a wide crosssection representative of the dierent types of school across England. Responses were also received from 18 LEA representatives and from representatives of 3 of the Internet Safety organisations.

J. Wishart / Computers & Education 43 (2004) 193204 Table 1 Method of data collection Phone Email Post Face to face Number received 367 12 197 1

197

Table 2 School type Community Voluntary aided or controlled Foundation Special Independent No data provided Number 332 125 23 27 58 12

Table 3 Age phase Middle (age approx 813) Prep (age approx 512) Primary (age up to 11) Secondary (age 1116 or 18) Through (age 3 or 518) Number 15 8 319 192 40

Table 4 Area East Midlands East of England North East North West Outer London South East South West West Midlands Yorkshire and Humberside No data provided Number 76 67 58 66 56 60 73 48 62 11

Only twenty schools (3.5% of those approached) did not have Internet access and these were excluded from the rest of the analysis. Of these twenty, fourteen were Independent schools. 3.1. Internet safety teaching Teaching Internet Safety was reported in 85% of the schools. It is most likely to take place solely within the subject area of ICT (66% of schools teaching Internet Safety) and, as shown in

198 Table 5

J. Wishart / Computers & Education 43 (2004) 193204

Types of schools and teaching methods (% of schools in each type shown) Primary Secondary Middle Prep All through

Induction Programme 39 73 67 75 41

Policy

Posters

Whole class

Worksheets

Discussion

50 64 47 38 29

45 47 33 38 24

67 59 53 63 29

17 22 7 0 9

49 37 20 38 26

Table 5, is more likely to be delivered via an Internet induction programme or whole class teaching than through a specic scheme of Internet Safety work. Making use of the schools acceptable Internet use policy as a teaching vehicle is also popular. Primary schools are more likely than schools with other age groups to use discussion activities and Secondary schools are more likely to use their Internet acceptable use policy as a teaching vehicle. The observed dierences between primary and secondary schools use of induction, acceptable use policy, whole class teaching and discussion for Internet Safety teaching were tested for statistical signicance using v2 and were found to be statistically signicant at p < 0:001 (v2 416:97, df 3). Use of worksheets was mentioned by 17% of Primary and 22% of Secondary schools but when teaching materials sent to the project team were looked at this was found to mean more general Internet worksheets than specic ones on Internet Safety. Additionally whole class reminders are common when using the Internet for research, especially in Primary schools. All through schools which tend to be Special or Independent are less likely to be teaching Internet Safety at all.

4. Breaches of internet safety in school As shown in the chart (Fig. 1) breaches of Internet Safety reported by schools in the main study were most likely to be pupils accidentally accessing inappropriate material. The LEA advisors also reported accidental access of inappropriate material as the most frequent breach of Internet Safety heard of by them with 5 of the 18 hearing about it more than once a term followed closely by deliberate access (4 of the 18 hearing about it more than once a term). It was found that the frequency of breaches of Internet Safety in school increased with the number of pupils attending the school. The strength of this relationship was calculated using Spearmans rho as the correlation co-ecient. The correlation calculated was equal to 0.6 indicating a moderately strong relationship which was found to be statistically signicant with the probability of the relationship occurring by chance being less than 0.001 (p < 0:001). Accidental access of inappropriate material happens most often in larger Secondary schools with over a quarter seeing accidental breaches more than once a term. This is despite 99% of these schools having ltering systems in place. In smaller Primary, Middle and Prep schools accidental breaches at this frequency are not seen at all and only 13% of these schools see inappropriate material accidentally more than once a year.

J. Wishart / Computers & Education 43 (2004) 193204

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

199

Never < once a term > Once a term

Downloading without permission

Access of SMS or chat

Accidental access

Fig. 1. Frequency of dierent breaches of Internet Safety in Main Study Schools.

5. Internet safety concerns In fact, as shown in Table 6, accessing inappropriate material is the ICT Co-ordinators single most important Internet Safety concern and the ICT Co-ordinators reported that accidental access is slightly more of a worry than deliberate access.
Table 6 Ten most frequently cited single most important Internet Safety issues Accessing inappropriate material Filtering Email Ensuring safe Internet access at school Pupils giving out personal details Supervising pupils (No issues currently) Safe Internet access at home Chat rooms Monitoring Internet use Current awareness of Internet safety issues No. of times suggested 142 61 39 36 35 27 25 18 14 13 9

Deliberate access

Unsolicited (junk) e-mails

Bullying e-mails

Inappropriate e-mails

200

J. Wishart / Computers & Education 43 (2004) 193204

There was much less agreement amongst the LEA advisors as to what was their single most important Internet Safety concern with fourteen dierent themes emerging. The most commonly reported concern from four LEAs referred to childrens abilities to interfere with the ltering systems, a further two reported potential for contact between children and strangers as their single most important Internet Safety concern and another two, Internet Safety awareness in schools. Contrastingly, all three the representatives of Internet Safety organisations reported childrens Net literacy as their single most important Internet Safety concern. It appears that teachers feel a responsibility to act in loco parentis and are worried about pupils viewing something that is out of their control. Deliberate access of inappropriate material is slightly less of a concern for schools who report knowing their pupils, their abilities and their motivations. The LEA advisors also considered deliberate and accidental access of inappropriate material to be their most serious Internet Safety concerns though for them deliberate access was slightly more of a concern than accidental. However, the breaches of Internet Safety that were of most concern to all of the representatives of Internet Safety Organisations were inappropriate access of chat or instant messaging, bullying via e-mail and receiving inappropriate e-mails. Thus education professionals concerns over Internet Safety can be seen to stem from the perspective of their role, with teachers and LEA advisors focusing on within school experiences, whereas those of the Internet Safety professionals focus on the potential for harm from others and the need to know how to use the net safely. In response to ICT Co-ordinators concerns over accidental viewing of inappropriate material and time limited teaching sessions schools tended to rely heavily on supervised Internet access, often ensuring that pupils only visited web sites recommended by the teacher. For instance, as shown in Table 7, 62% of schools supervise all Internet access at all times. For Secondary schools this is more than is recommended by the LEAs with 33% of the LEA advisors recommending supervision of all Internet access and 56% of schools reporting it. The opposite is occurring in Primary schools with 78% of the LEA advisors recommending supervision of all Internet access and 66% of schools reporting it. This emphasis on supervision may lead to both a lack of awareness of good Internet Safety practice amongst children when surng the Internet outside of school and a lack of emphasis in school on developing independent search and evaluation skills. This is a particular issue with regard to learning to use chat and instant messaging which as shown in Table 8 are largely unavailable to pupils in schools. Only 32 schools (5%) allowed their use for school work and fewer, 24 (4%), allowed pupils to use them recreationally.
Table 7 Supervising Internet access Always by teacher/sta/helpers In classroom or computer suite only Usually/mostly Younger pupils only Remote monitoring used Some pupils only Internet not used No data Number of schools 347 26 15 9 3 5 83 65

J. Wishart / Computers & Education 43 (2004) 193204 Table 8 Numbers of schools allowing chat/instant messaging for All pupils All in class/under supervision In named site only (MSN messenger, think.com, Not.OP, Sourceror) In internal chat site only Restricted range of pupils only None No data/do not know Allowed for school work 19 2 3 2 6 509 16 Allowed for recreation 16 3 0 0 5 517 16

201

This is viewed to be a particular problem by OConnell et al. (2002) and the Childrens Charities Coalition for Internet Safety (2001) who, like the Internet Safety organisations consulted in this study, argue that schools can better enable childrens safety by providing them with the knowledge and skills to allow them to deal safely with chat room situations rather than by restricting their access.

6. Emerging internet safety issues A particular focus of this study was to nd out what teachers foresaw as being concerns for childrens Internet Safety in the future. Fig. 2 shows the most commonly reported ICT Co-ordinators concerns for emerging Internet Safety issues. Schools expressed concern about increasing use of e-mail with e-mail being new to the school and its use required by pupils following the national Qualications and Curriculum Authority (QCA) Scheme of Work (QCA, 2000). ICT Co-ordinators worried in particular over how to
Internet Access via mobile devices 2% Increased use of E-mail 30%

Downloading 3% Keeping up to date 4% Giving out details 4% Effective policy 6% Safety in Chatrooms 5% Safe use at home 5%

Increased Internet access Pupils increased IT 13% Proliferation of capability 7% unsuitable material 9% Effectiveness of filter 8%

Fig. 2. Emerging Internet Safety Issues for ICT Co-ordinators.

202

J. Wishart / Computers & Education 43 (2004) 193204

monitor and control use and the ethics of doing so. They were also concerned over the time and network resources e-mail used. Emerging Internet Safety issues for LEA Advisors followed a similar pattern as shown in Fig. 3. However, the emerging concerns of Internet Safety Organisations, shown in Fig. 4, dier signicantly from those of the education professionals with all the representatives citing worries over Internet access from mobile phones and two of the three raising potential abuse of peer to peer networks and bullying via e-mail. In the UK mobile phones are owned by 34% of 1114 year olds rising to 70% of 1618 year olds (DfES, 2002) and currently schools vary in their policy of whether to allow them in schools or not. It is quite possible that children are bringing WAP enabled phones into school making a mockery of

Access to videoconferencing 6% Hacking 19%

Increased use of E-mail 26%

Downloading 6% Keeping up to date 6% Safety in Chat rooms 6% Safe Internet access at home 6%

Increased Internet access 6% Proliferation of unsuitable material 13% Effectiveness of filter 6%

Fig. 3. Emerging Internet Safety Issues for LEA ICT advisors.

Raising awareness/ Net literacy 13%

Abuse of peer2peer networking 25%

Internet Access via mobile devices 37%

Bullying emails 25%

Fig. 4. Emerging Concerns of Internet Safety Organisations.

J. Wishart / Computers & Education 43 (2004) 193204

203

the schools eorts to lter Internet access to provide a safe surng environment within school. Filtering for these phones is both technologically dicult as they will be operating on a variety of networks and fraught with moral dilemmas such as who owns the phone in question (parent as purchaser or child carrying it) and whether teachers should be looking at private phone messages. Peer to peer networks such as those set up by using Kazaa or Filetopia enable le sharing, usually music, amongst computers connected to the Internet. Some are moderated or supervised but others are not. By downloading peer to peer software the user is allowing strangers to share their les and to chat. Also once peer to peer networking is installed it can run undetected by Internet logging software as a direct http connection can be made, though both parents and schools can set ltering software to prevent download of the software client.

7. Conclusions and recommendations It is concluded that whilst the vast majority of schools are teaching Internet safety they are doing so in a restricted environment with safety conscious supervision preventing exploration and in particular, the use of chat. This will cause problems when children are surng the web and using chat and instant messaging at home which may well be an unsupervised and an unltered environment. It is recommended that stakeholders such as government organisations and childrens charities provide: Advice for LEAs on enabling chat in schools and support for schools aimed at teaching children about the use of chat rooms and instant messaging safely. An updating service to alert schools and LEAs to developments in technologies and new guidance on their use in school. Teaching materials for schools to use with pupils aimed at developing Net literacy and safe surng practices that enable pupils to use the Internet responsibly and usefully both in and outside school. In fact since this study was carried out the an Internet Prociency Scheme aimed at 711 year olds has been set up by Becta, the DfES, Gridclub and the QCA. The aim of the scheme is to provide teachers with easily accessible support materials to help their pupils develop a set of safe and discriminating behaviours to adopt when using the Internet and help pupils demonstrate what they know. The scheme itself is based around a CyberCafe where the caf owner and his customers assist the children to consider the correct behaviour e to follow when using the Internet and other communications technologies, using fun interactive activities. It is intended that children will learn to: manage e-mails; respond to chat; evaluate web sites and manage text/instant messages. More information is available at www.gridclub.com/grown_ups/cont_ict_cyb.shtml. Additionally it is recommended that means of ltering or monitoring Internet access for children using mobile technologies to surf the web need to be investigated through negotiation with Internet Service Providers (ISPs). Their help will also need to be sought on monitoring peer to peer network use by children.

204

J. Wishart / Computers & Education 43 (2004) 193204

References
Amarach Consulting (2001). Irish parents fear for children online. Report, 5 October 2001. Available: www.nua.ie/ surveys/index.cgi?f VS&art_id 905357266&rel true [Accessed 7.8.02]. Burnard, P. (1994). The telephone interview as a data collection method. Nurse Education Today, 14, 6772. Childrens Charities Coalition for Internet Safety (2001). Working to make the Internet a safer place for kids. Available: wwwcommunicationswhitepaper.gov.uk/pdf/responses/ccc_internet_safety.PDF [Accessed 7.8.02]. Department for Education and Skills (DfES) (2002). Young people and ICT: ndings from a survey conducted Autumn 2001. NGfL Research and Evaluation Series. No. 5. London: Becta. Economic & Social Research Council (ESRC), Bristol University, Cardi University & University of Wales College Newport. (2000). The Screen Play Project. (ESRC 19982000). Available: www.bris.ac.uk/Depts/Education/ fullresultsabbreviated.doc [Accessed 7.8.02]. ESRC (Economic & Social Research Council), University of Loughborough, University of Sheeld. (1999). Cyberkids: Childrens Social Networks, Virtual Communities, and On Line Spaces. April 1997March 1999. Available: www.hull.ac.uk/children5to16programme/details/valentine.htm [Accessed 7.8.02]. Institute for Public Policy Research (IPPR), Citizens Online, Livingstone, S. (2001). Online freedom & safety for children. IPPR/Citizens Online Research Publication No. 3 November 2001. Norwich: The Stationery Oce. Available: www.citizensonline.org.uk/pdf/pub3.pdf [Accessed 7.8.02]. McCann, K., Clark, D., Taylor, R., & Morrice, K. (1984). Telephone screening as a research technique. Sociology, 18, 393402. Media Awareness Network (2001). Young Canadians in a wired world: the students view. October 2001. Available: www.connect.gc.ca/cyberwise/pdf/wired_e.pdf [Accessed 7.8.02]. National Grid for Learning (2002). School Web Sites Available: www.ng.gov.uk/directories.jsp?sec 31&cat 601&clear y [Accessed 23.5.02]. NCH (2002)News Page. 1 in 4 children are the victims of on-line bullying says childrens charity. Report. 15 April 2002. Available: www.nchafc.org/uk/news/index1.asp?auto 194 [Accessed 7.8.02]. OConnell, R., Sange, S., Barrow, C. (2002). Young peoples use of chat rooms. Implications for policy strategies and programs for education. Preston: University of Central Lancashire. Draft report, July 2002. Available: www.fkbko.net/launch.htm [Accessed 7.8.02]. QCA (2000). Schemes of Work: ICT at Key Stage 3. London: QCA. Available: www.standards.dfee.gov.uk/schemes2/ secondary_ICT/?view get [Accessed 30.4.03]. Schoolsnet (2002). Schools Guide. Available: www.schoolsnet.com [Accessed 23.05.02]. Times Educational Supplement (2002). Schools Directory. Available: www.schoolsdirectory.com [Accessed 23.05.02].