Websites, blogs, and texts
In addition to forums and BBSs, researchers have begun to use the information available in various websites as a means to understand cybercrime and deviance using both qualitative and quantitative techniques. Websites include a range of content, such as social media feeds like Facebook and Twitter, as well blogs and simple text-based pages. The content of these sites provides direct information on the thoughts of an individual in theft own words through text, images, and video, as well as links to other groups and entities which share their interests (Hine, 2005). In turn, this information can be used to understand perceptions of, or attitudes toward, offending from a different perspective than the social exchanges observed in forums.
Criminological researchers have increasingly used content from webpages to understand aspects of cybercrime and offender behaviour. For instance, a study conducted by D’Ovidio and colleagues (2009) examined a sample of paedophile support forums and the content posted there to understand the ways these sites facilitate a social learning process of paedophilia (D’Ovidio, Mitman, El-Burki, & Shuman, 2009). Sociological and criminological researchers have also used data from dating sites and personal advertisements online to understand the ways that individuals solicit various partners for both traditional romantic entanglements and more deviant sexual practices (Frederick & Perrone, 2014; Grov, 2004; Tewksbury, 2003). For instance, a number of studies have explored the phenomenon of bug chasing, where HIV negative individuals actively seek out sex with HIV positive partners for the purposes of becoming infected (Grov, 2004; Tewksbury, 2003, 2006). Researchers utilised qualitative analyses of ads posted on dating sites and forums dedicated to this practice and found that individuals actively pursuing HIV infection utilised different languages to construct their ads compared to those who appeared to simply seek sexual partners.
A small number of studies have used data from blogs and social networking sites to understand self-disclosure of criminality and behaviour. Web logs, or blogs, also provide an important, though underutilised, data source in criminological research generally (Hookway, 2008). Blogs serve as a sort of electronic diary that documents individual experiences through user-generated text, video, and images which are typically listed in reverse chronological order that can be tracked over tune. Many blog sites operate to allow users to make their blog as a single web page, though increasingly social networking websites, like Facebook, serve the same purpose as a blog by documenting individuals’ thoughts and feelings (Hookway, 2008).
One of the key challenges evident in the use of blog and social networking profiles for research is the potential for identification of individual users’ real identities. Since these sites typically provide personally identifiable information or can be aggregated to document an individual’s offline identity, there is some degree of risk for researchers to use quotes or provide information in research articles as it may cause the subject personal harm. Social networking profiles may be publicly accessible and indexed via search engines, which would srrggest they can be treated as open source and unprotected data from an ethical standpoint. Individuals who have set their profile to private in order to limit outsider access present an ethical dilemma as researchers must consider whether they can use such information in research. Additionally, some individuals have begun to post disclaimers on thefr blogs or user profiles to dissuade third parties from using their information without thefr expressed consent.
Email as a data source
A small number of criminologists have begun to use email as a data source for both quantitative and qualitative scholarship as the sheer quantity of email accounts and messages sent every day constitute a massive potential mine of information.
Since email is essentially a free service, researchers can readily establish and maintain accounts over long periods of time to serve as data collection points. Such techniques afford opportunities to assess the methods of scammers and fraudsters, as well as malware writers and hackers who chose to use email as an attack platform (e.g., Bohrne & Holz, 2006; Frieder & Zittrain, 2007; Hanke & Hauser, 2006; Holt & Graves, 2007; King & Thomas, 2009; Niran et al., 2009; Rege, 2009; Wall, 2004).
To date, most researchers utilise small samples of several hundred messages as a basis for analysis of mass-distribution emails, or spam, to understand the methods of scammers (e.g.. Holt & Bossier, 2016). Several researchers have published studies examining advance fee fraud messages, or so-called Nigerian email scams, where the sender claims to need the help of the message recipient in order to transfer millions of dollars from one country to another (see Edelson, 2003; Holt & Graves, 2007; King & Thomas, 2009; Nhan et al., 2009). Should the recipient contact the sender to provide assistance, they will be entitled to receive a portion of the total funds. The recipient needs only provide a few hundred dollars in order to help pay attorney fees, taxes, or related costs. However, all funds sent will go to the scammer and will never lead to a good outcome for the message recipient.
Research on advance fee fraud messages illustrate that senders use one of a few different schemes, such as posing as public officials who have skimmed funds from business contracts or being a wealthy heiress with a large inheritance who needs assistance to move funds out of their country (see Edelson, 2003; Holt & Graves, 2007). The messages also include unique linguistic techniques in order to entice victims to respond, such as religious language regarding God or religious blessings, external web links to news stories to attempt to validate message claims, and emotional pleas for aid (Holt & Graves, 2007; King & Thomas, 2009; Niran et al., 2009). Senders also tend to request potential victims provide them with thefr contact information, home address, phone numbers, and sometimes banking details (see Holt & Graves, 2007). In the event that a sender receives this information, they can then solicit fluids from the potential recipient over time.
Kigerl’s (2012, 2015, 2018) research studies of spam email provide unique examples of potential big data mining and analysis techniques for criminological inquiry For instance, he published an analysis of the macro-level correlates of spam email distribution at the country level. The basis for this analysis was the 2008 spam archives of Untroubled Software, consisting of 725,037 messages which he coded through the use of a software script. Kigerl (2012) then extracted the originating IP address of the email and traced it to its country of origin. This data was then combined with other data sources to assess country-level economic factors and internet use variables to examine their relationship to the one-year rate of spam delivery for 132 total countries. Using regression analyses, he found that spam distribution rates were associated with a nation’s gross domestic product (GDP), the number of internet users per-capita, and higher unemployment rates.
