Green Capitals "in the Hearts and Minds of the People"
Data Analysis and Interpretation
Download 0.67 Mb. Pdf ko'rish
|
GreenCapitalsintheHeartsandMindsofthePeople
Data Analysis and Interpretation
Analysis and interpretation are two interlinked but distinct processes when carrying out qualitative research. In the Oxford Handbook of Qualitative Research, analysis is defined as being concerned with organising the data by dividing and re-assembling them in a consistent way (Trent & Cho, 2014). Interpretation is a parallel step in which the researcher finds the meanings data express and compares them to other sources, literature or common sense in order to identify patterns of meaning (Trent & Cho, 2014; Patten & Newhart, 2018). The two processes occur in parallel in an iterative process where complexity is alternatively reduced and enlarged until a narrative of what the data reveal of the phenomenon emerges (Saldaña, 2009; Patten & Newhart, 2018). 33 When analysing the data, one can summarise and categorise them by using graphic maps, critical questions, memos, but the most widespread tool, and the one adopted in this research, is coding (Trent & Cho, 2014). A code is “a word or short phrase that symbolically assigns a summative, salient, essence-capturing, or evocative attribute for a portion of language-based or visual data” (Saldaña, 2009; p.3). When coding a piece of qualitative data, such as the transcriptions of interviews of this research, codes are associated to parts of the text in order to summarise what they deal with and access more easily to them when interpreting the data (Saldaña, 2009, 2014; Trent & Cho, 2014; Patten & Newhart, 2018). Such a functional role becomes particularly important during interpretation. Codes get intertwined with the variables of the research and can represent (or neglect) aspects of the data that are fundamental to find patterns of meaning (Saldaña, 2014; Patten & Newhart, 2018). Because of that, the data are re-coded throughout the analysis and interpretation of data in order to drop those codes that are less relevant and create new ones for the parts that are related to the emerging meaning pattern (Saldaña, 2014, Trent & Cho, 2014; Patten & Newhart, 2018). Codes can be created before coding starts and then be applied to the data, or they can be created while reading the data (Patten & Newhart, 2018). There are more specific coding strategies dealt with in the literature, most notably the categorisation by Saldaña (2009, 2014). As Trent & Cho (2014) summarise it: ● process coding (uses gerunds, “-ing” words to capture action); ● in vivo coding (uses the actual words of the research participants/subjects); ● descriptive coding (uses nouns to summarise the data topics); ● versus coding (uses “vs” to identify conflicts and power issues); ● values coding (identifies participants’ values, attitudes, and/or beliefs)” (in Trent & Cho, 2014; p.647). These strategies are not mutually exclusive or exhaustive of all the possible ways of coding, but they offer a comprehensive overview of the multiple ways coding can be carried out. By using this terminology, it can be said that this research has made use mostly of descriptive coding, and occasionally of process and versus coding. Even if an analytical framework was drawn from existing literature already before coding, the fact that the interview questionnaire was based on that already meant that most of the questions were centred on the framework variables. As a consequence, most of the codes were created in an inductive way starting from the expressions interviewees used when talking about the 34 different variables of the questionnaire, i.e. reasons for applying, benefits, costs and political considerations. A complete table of codes can be found in Appendix B. In a way, then, coding and theory met rather than proceeding from one to the other, as the interviewees were left free to give whatever impression they wanted as to how the variables manifested in their city. It was up to the interpretation phase then to find common patterns in the variety of answers officials gave for every single variable. As mentioned earlier in this section, interpretation is a process taking place throughout the phases of the research, from data collection through their representation, analysis and discussion (Trent & Cho, 2014). Interpretation involves many non-mutually exclusive techniques as well, including but not limited to: asking critical questions, seeking advice from other researchers, connecting to the literature or existing theories, accounting for power dynamics or reconnecting to one’s personal experience as a researcher (in Trent & Cho, 2014). In this research, interpretation has taken place mostly by asking critical questions and connecting to existing literature. When carrying out the interviews, transcribing and coding, some themes emerged from what the interviewees had to say. Interpreting these accounts in the light of existing theories helps to understand how the EGCA was managed and why it is said to have had some precise impacts on the behaviour of actors. Other methods, for example seeking advice, were discarded because the high amount of available data would make it difficult for any colleague or professor to offer a sound opinion on the analysis. Power dynamics, as it has been discussed in the previous section, have influenced the way the research has been carried out. As to their impact on the answers, though, the atmosphere of the interviews was so informal and the topic so little sensitive that it can be presumed to have been quite weak. Lastly, as this is only the second research involving interviews the author carries out, it is unlikely that any recollection of past experience may reveal something relevant during interpretation. Now that the main methods adopted in this research have been defined, the next chapter is going to relate how analysis and interpretation took place and discuss their results more in detail. 35 |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling