3.7 Data Analysis

As this study is a predominantly qualitative research, analyzing the research data did not start when I came back in the Netherlands, but already evolved during my research. The several methods I used alternated during my time in Costa Rica and influenced the way how I approached new (interview) situations. After several interviews I could see categories emerging within the fight against Costa Rica. This resulted in asking my interview questions in another way, as I knew more information. Also the documentary research, informal talks with Costa Rican people, the workshops and my observations in the field influenced my thoughts and subsequent questions in the process. In fact I have thus already been analyzing my data during the data collection. While referring to Miles and Huberman (1994), Jennings (2001: 195) comments on this as follows: “Qualitative data collection and analysis are not two discrete entities of the research process, as is often the case in quantitative analysis. The process of data collection and analysis is overlapping”.

During the three months I spent in Costa Rica I carried out a review of the five ‘steps’ (requirements) of the Code of Conduct, a quantitative research method. After collecting all the necessary information to do this; the annual reports, the signed Code of Conducts, the external symbols and letters, I started to analyze these as quantitative data by compiling them in an Excel spreadsheet, making a division in region and kind of tourism enterprise or institution, as requested by Paniamor. For each tourism enterprise or institution I filled in their accomplishments with each of the five steps of the Code of Conduct. In this way it became possible to review the progress of the Code of Conduct Project. One could call this analysis method quantitative content analysis.

The categories would then be: ‘did’ or ‘did not’ comply with a certain step of the Code of Conduct, which means that these categories comply with the requested ‘mutually exclusive’ aspect as well as the ‘exhaustive’ aspect (Jennings, 2001). The archival and documentary research was needed in order to place the findings from the other research methods into the context. For this reason I started analyzing these data before turning to the analysis of the final other data (except for the review of the Code of Conduct Project). I collected the archival and documentary data in Costa Rica at the several organizations, mainly at Paniamor, through the internet and in the library of Wageningen University. I used content analysis to analyze these data in order to collect quantitative and qualitative information about Costa Rica as a country as well as information about its child sex tourism situation and the fight against it.

When one looks at the types of content analysis I would argue that I used both summation and structuration [mainly for the empirical part on Costa Rica] (Jennings, 2001). Jennings (2001: 203) refers to Mayring (1983, 1985, 1988 in Sarantakos 1998) when explaining these types. In summation: “the data being analysed are reduced into categories that integrate and generalize the major themes of the documents” and in structuration: “the data are ordered according to a predetermined set of categories or an order determined through the texts themselves”. I recorded each interview with a tape-recorder and used pen and paper to make observation notes. Back in the Netherlands I transcribed these interviews and added my observation notes. For the workshops I used pen and paper to make observations myself, besides using the written comments from participants of these workshops. I did these observations without prior categorizations and collected them in one document.

I also collected all my written field observation notes and reflections in one document. These data consist of observations made in the field but also thoughts that popped up in my head wherever I was during the research process. Jennings (2001: 197) terms these ‘memos’: “messages researchers make to themselves during the course of research design, particularly during data collection and analysis”. Where applicable I connected observation notes to specific interviews or situations, for example in the case of visiting a beach resort for the weekend and conducting two interviews there. I analyzed all the data from these research methods, interviews, workshops and observations, by means of coding. According to Miles and Huberman (1994 in Jennings, 2001: 198) “codes are tags or labels for assigning units of meaning to the descriptive or inferential information compiled during study”. In this coding process I went through three different phases (Straus & Corbin, 1990 in Jennings, 2001: 200):

1. Open coding: The process of breaking down, examining, comparing, conceptualizing, and categorizing data.

2. Axial coding: A set of procedures whereby data are put back together in new ways after open coding, by making connections between categories.

3. Selective coding: The process of selecting the core category, systematically relating it to other categories, validating those relationships, and filling in categories that need further refinement and development.

After analyzing and coding my observation, workshop, interview and archival/documentary data during and after my research, I decided to use the same spreadsheet format for every research method in which I used the categories that emerged from the coding process and content analysis in order to structure my data (see Appendix VIII for an example). In this way I could clearly see the patterns and themes within the data which allowed me to document and summarize the outcomes in the results chapter of this report and draw the necessary conclusions. In the presentation of the data I will use the term participant researcher when I am referring to someone I have interviewed.

The issue of reflexivity plays a role in choosing this term instead of for example respondents or interviewees, as the person I have interviewed has ‘co-created’ my research. The people I have interviewed are not just subjects to be used, but they play an active role in the creation of data. I will not use any names when I am referring to certain participant researchers. I will call them representatives of a certain type of organization because of anonymity issues. When necessary I will use the name of the concerning company (with permission).