PROCESOS PSICOLÓGICOS DE LA LECTURA

An advantage of Thematic Analysis (TA) for such research is that it allows the researcher to delve more deeply into the world of participants. In addition, it differs from other analytic methods that seek to describe patterns across qualitative data, in that they are bound by theory, whereas thematic analysis is less restrained (Braun and Clarke, 2006). As thematic analysis is about understanding people’s everyday experiences in great detail (McLeod, 2011), it is particularly well-suited to gaining insight into the motivations and developmental processes involved when an individual sexually exploits a child.

Thematic analysis offers two internal approaches to deriving themes from the data, namely an inductive/bottom-up analysis (Frith and Gleeson, 2004) and a theoretical/top-down examination (Hayes, 2013; Boyatzis, 1998). The former is a process of coding without the need for it to conform to the researcher’s pre- determined assumptions or a pre-existing coding framework (Braun and Clark, 2006). Nonetheless, researchers are not devoid of a framework but must consider their theoretical and epistemological commitments (Braun and Clark, 2006). In addition, thematic analysis is suitable when there is a dearth of literature and/or previous research on a topic (Strauss and Corbin, 1998) - as the world of the internet and its usage as a medium for sexually abusing children is a relatively new and unexplored one, this approach is applicable. Theoretical analysis is constrained by a more rigid framework and data that is influenced by pre-determined concepts (Braun and Clark, 2006), and was deemed too limited when a rich understanding of individual experience is required to begin to understand how sex offenders and producers of indecent images function. In addition, inductive thematic analysis is

pertinent when there is no explicit research hypothesis or fundamental question to be answered but rather allows themes to emerge from the data. This research was exploratory and wanted to focus on the individual experience and aetiology of sex offenders, rather than impose any form of pre-determined constraints on the data. In terms of inductive thematic analysis, Braun and Clarke (2006) recommend a six- phase approach which is outlined in Table 1 below:

Table 1: Key Phases of Thematic Analysis - Braun and Clarke (2006)

1. Familiarising yourself with your data:

Transcribing data (if necessary), reading and re-reading the data, noting down initial ideas.

2. Generating initial codes: Coding interesting features of the data in a systematic fashion across the entire data set, collating data relevant to each code.

3. Searching for themes: Collating codes into potential themes,

gathering all data relevant to each potential theme.

4. Reviewing themes: Checking if the themes work in relation to the

coded extracts (Level 1) and the entire data set (Level 2), generating a thematic ‘map’ of the analysis.

5. Defining and naming themes: Ongoing analysis to refine the specifics of each theme, and the overall story the analysis tells, generating clear definitions and names for each theme.

6. Producing the report: The final opportunity for analysis. Selection of vivid, compelling extract examples, final analysis of selected extracts relating back to the analysis of the research question and literature, producing a scholarly report of the analysis.

2.2.3 Pilot Study

Thematic analysis was ultimately chosen as the considered best option for this research. However, the pilot study (Sheehan and Sullivan, 2010) that was the starting point used Interpretative Phenomenological Analysis (IPA). The pilot study consisted of a small cohort of four participants. This was seen as an important part of the research process and a guide to subsequent refinement of interview structure and content, as well as a means of reviewing and analysing emerging themes as the cohort grew.

There is support within the qualitative methodology that allows for, and indeed encourages, the use of an initial phase of research as a guide for directing the main body of work, seeing it as exploratory and a useful basis (Turner et al., 2008).

Maxwell (2005) highlights the importance of how continually assessing the extent to which the design is working as the research proceeds, to factor in how it affects and is being affected by its context, and to make alterations and adaptations, is necessary. In qualitative analysis, maintaining your original data design can be seen as sacrificing thorough data analysis for consistency (Silverman, 2010).

IPA is an analytical tool developed by Smith (1996), with origins in health psychology. IPA views the participants as the experts in the field and the researcher’s task is to elicit their perspectives and make sense of their experiences, their worlds, their behaviour and the meanings they attribute to them (Smith, Jarman, and Osborn, 1999; Smith and Osborn, 2015). IPA is seen as particularly suitable for smaller samples and case studies, and so lent itself particularly to the pilot study

(Smith, Jarman and Osborn, 1999; Smith, 2004). However, this, in itself, was seen as a limitation when the cohort for this piece of research grew. IPA can influence and, in turn, limit analysis as it is considered restricted by an epistemological framework (Braun and Clarke, 2006). A less-rigid approach was considered more useful moving forward with the research cohort.

2.3 Design

2.3.1 Semi-Structured Interview

The semi-structured interview for this research was devised to facilitate the investigation of factors that might have contributed to the development and progression of this form of child sexual exploitation. The areas covered by the semi- structured interview are; life history; blocks to offending; role of fantasy; planning and preparation for offending; the abuse; and function of image/movie production. The design of the semi-structured interview was based on the conceptual framework

The Spiral of Sexual Abuse5 _{(Sullivan, 2002; Sullivan and Beech, 2004; Sheehan and}

Sullivan, 2010; Sullivan and Sheehan, 2016) and was further informed by other seminal works in respect of the etiology of sexual offending against children (Finkelhor, 1984; Marshall and Barbaree, 1990; Hall and Hirschman, 1992; Ward and Siegert, 2002).

A study of learning-disabled sexual offenders in the community noted that thematic analysis as a methodology and, more specifically, open-ended questions within semi-

5 _{The Spiral of Sexual Abuse is a framework that illustrates the development of sexual abuse that}

includes exploration of early life experiences, as well as the developmental stages in perpetrating sexual exploitation of a child. It allows for the individual story to unfold during interview whilst also exploring key identified areas in the aetiology of sexual offending.

structured interviews, allowed the greatest participant expression and yielded the best data (Hutchinson, Lovell and Mason, 2012). The interview questions for this study placed emphasis on open-ended questions, with the aim of trying to gain an understanding of the participants’ view of their process, as well as their attitudes and beliefs towards children and their own offending behaviour. Emphasis was placed upon questions that would lead the researcher into the data but not so broad as to lead to unlimited possibilities (Corbin and Strauss, 2008). Questions began with the general and moved to the specific (Stewart and Shamdasani, 1990).

While questions were prepared in advance against the broad backdrop of relevant literature, the interview content was also influenced by thoughts and hunches of the interviewee and, as such, was subject to change throughout the research process (Seale et al., 2007). However, it was recognised that while individuals will have a variety of responses on the same topic and that validity necessitates that each perception is acknowledged and valued, the limitation of any single viewpoint was also seen as an important factor (Porter, 2007). In addition, it was noted that interviewees may get different versions of the “probing” questions which, in turn, may produce different information (Higgins, 2009). Cognisance was also taken of the fact that excessive intervention from the interviewer can be counterproductive in the field of qualitative research (Guest, MacQueen and Namey, 2012). While researcher bias is inevitable in qualitative work and should be recognised and minimised, some bias will always exist (Yardley, 2008). The critical aspect is to acknowledge this fact (Hanson, 2009).

While all academic study is steered by research questions, these can be refined as the work progresses (Braun and Clarke, 2006). Following the pilot study, some changes were made to the semi–structured interview - in particular, with the inclusion of questions in relation to whether an individual had shared the images/videos they had taken or not. The pilot study identified that “sharing” and “non-sharing” producers had some differences in aspects of their offending behaviour, victim access and manipulation style. While the pilot was too small to be significant, it was nonetheless considered important to explore this issue more extensively with the full cohort. In addition, having been identified as themes during the pilot study, it was also decided to add questions in regard to the impact of the internet on offending behaviour and the relationship between viewing and the production of indecent images.

The semi- structured interview questions were as follows (see Table 2): Table 2: Semi-structured Interview Questions

Introductions _{1. Details of research project.}

2. Confirm they have read and understand confidentiality/data protection.

3. Emphasis that all information is ‘on the record’ and will be available to referring organisation.

4. Highlight the importance of not disclosing third party and/or incriminating information (reiterated throughout the interview).

Convictions 5. Summarise the offences for which you have been

convicted.

Personal history _{6. Tell me about your childhood and life growing up.}

7. Do you have a sexual interest in children? At what age did you become aware of this?

8. Were there any factors in your childhood which might have contributed to journey to producing IIOC?

Sexual exploitation of children-general

9. Have you engaged in other forms of child exploitation?

10. How did you target/access children for this?

11. Describe how you groomed children and others to this end?

12. Describe how you made it okay to engage in this behaviour-cognitive distortions /justifications used. 13. Have you downloaded IIOC/used the internet to

sexually exploit children?

14. Describe this process and your level of involvement.

Production of images _{15. How did the process of production begin?}

16. Tell me about the images you took?  Stills/video/both.

 Describe your production methodology- how you took the images.

17. Describe how you targeted/gained access to children.

18. Was your process covert or overt, or both? Describe how you did this.

19. Describe the grooming techniques/process you used with the child (children) and others that was directly linked to production.

 Were the images used as a threat to the child (children)?

 Did you involve the child in decisions about which images to take?

20. Describe how you made it okay to engage in the

production of IIOC-cognitive

distortions/justification.

21. Why were the images important to you? Their function(s)?

22. Did you share them with anyone?

 Why did you / did you not share?

 Describe you interactions with those whom you shared with/interacted with in relation to IIOC.

2.4 Participants

Participants were chosen from a range of individuals who were known to have produced child sexual exploitation images and, as such, met the criteria of appropriate (Yardley, 2008). Participants were initially contacted by a third party, either Law Enforcement, Probation, Prison Services or Local Authority personnel

who are involved in a professional capacity with these individuals. The researcher works regularly with such organisations and contacted them to explain the research and the wish to interview individuals who had produced IIOC. The organsiation the researcher works for is noted in the field of research for its involvement in studying individuals who have committed sexual crimes relating to children. As such, any approach was not atypical. As noted already, there is a dearth of information about producers and it is possibly an under detected crime. Accordingly, production was the crucial element in identifying individuals willing to be interviewed. In the main, the participants were incarcerated at the time of interview or were involved with law enforcement in the community. As a consequence, individuals who engage in this behaviour but have not been caught were therefore not included; neither were those who had been apprehended but refused to talk. All participants were male. The selection of only convicted offenders, apart from one, will have created some anomalies within the data.

The interviews relied upon self-report; however, case files were made available and allowed for independent verification of certain aspects of the accounts given. The aims and objectives of the study were explained, and written consent, in accordance with the British Psychological Society’s code of conduct for professional practice ‘Ethical Principles for Conducting Research with Human Participants’ (1993; 1994), was obtained from all participants. All participants were told that the interviews were being conducted for the purpose of research. A description of the project was provided to participants and only those who were content to sign a consent form were included in the study. The names of participants have been changed and any details of their offending or account that might identify either the participants or their

victims were removed. Strong emphasis was placed upon the fact that, if any participant disclosed an offence and/or incriminating information, this would be passed to the relevant authorities. However, it was also emphasised that participants were not to disclose any such information and/or third party details that would identify any victims.

Originally, the study group comprised 23 participants, however, one participant withdrew his consent for the interview to be used prior to the analysis of the content.

2.4.1 Inclusion Criteria

The cohort was defined by the requirement that they had created ‘produced’ images that were sexually exploitative of children. As can be the case with qualitative methods, there can be challenges in the area of sampling (Berg, 2007); however, there was diversity within the sample in terms of IQ, culture and professional backgrounds, as well as in respect of the specifics of their offending behaviour. In all cases, the following selection criteria were applied:

 Volunteers nominated by professionals working in the field;  Willing to give a recorded interview;

 IQ greater than 80;

 Acknowledged responsibility for the production of indecent images of children;

 Willing to take part in the research study/recorded interview;

 Willing to discuss their offence (therefore acknowledging some aspect of abusive behaviour).

2.5 Interviews

The interviews were carried out on a one-to-one basis over a minimum period of three hours. Some individuals were interviewed more than once - however, this was not always possible. All the interviews were video recorded onto a memory card using a Sony Digital camcorder. At the end of each interview, the file was transferred directly onto the encrypted hard-drive of a Sony Vaio laptop computer and the memory card was wiped using PGP wiping software. A back-up of the files was retained on a fully-encrypted, removable hard drive. The records of this study were anonymised and retained in password-protected Word Excel, SPSS and NVivo files on the encrypted laptop. No paper files were retained containing any personal details of the participants.

2.6 Analysis

As it has been suggested, that moving from the oral to the written text is a reconstruction as opposed to a direct copy (Steinar, 1996), the interviews were transcribed verbatim, retaining all grammar, pauses and unfinished sentences. As discussed above, thematic analysis, following Braun and Clarke’s (2006) six-phase approach, was then used to analyse the data.

Increasingly, researchers are utilising computer-assisted qualitative data analysis software to assist with the coding and cross-referencing of transcripts (Bringer, Johnston and Brackenridge, 2006). In this case, the QSR NVIVO (NVIVO) software package was used to store, manage and assist in the analysis of the transcripts. The benefit of the NVIVO software package is that it allows the researcher to code

passages in one document, while facilitating quick reference to similarly-coded passages from other transcripts. As themes, referred to as ‘nodes’, emerge from the data, they are stored in a manner that allows the creation, browsing and exploration of new themes and sub-themes as the ongoing analysis of the transcripts continues.

The first interview to be transcribed was read and re-read to familiarise the researcher with the data. The second stage involved the generation of initial codes, whereby interesting aspects were coded across the data. In stage three, following a broad grouping of codes, a preliminary list of themes was compiled. All subsequent transcripts of interviews were also reviewed and a revised list of themes was generated. Each revision required a review of the previously-coded data, in an attempt to ensure the reliability of the analysis and contribute to comprehensive data treatment (Silverman, 2000). In stage four, further refinements of themes were carried out, and dominant themes and sub-themes identified. Stage five involved defining themes and sub-themes. Stage six included the selection of extracts, final analysis and linking to literature.