PDF Publication Title:
Text from PDF Page: 032
2.2.2 Methods of data analysis Quantitative data analysis To avoid confirmation bias as much as possible,14 and for the sake of discovery of unforeseen patterns, the academic corpus was, firstly, compared with the Google corpus on word frequency to discover differences in extensive use of certain words, or lack of use of words.15 Since the process of collection, interpretation, and analysis is non-linear the results from this qualitative analysis were analysed, before going further to the next step in analysis, since the findings possibly influenced the direction of focus in the qualitative analysis. Only the 5000 most frequent words were showed in NVivo (NVivo Qualitative Data Analysis Software 2014)16. In this list, I looked for words (and their synonyms) that the academic literature had considered important, such as opacity, transparency, ethics, responsibility, bias and discrimination (see Table 2 in the next chapter).17 Some differences between the corpuses stood out (see Table 2, Figure 5, and Figure 6 in chapter 3: “Analysis/ interpretation”). Consequently, a qualitative approach was performed to understand how these words were used. While doing the qualitative analysis and the interpretation of the findings, the findings of the quantitative analysis were used to compare quantitative and qualitative findings: whether they reinforced or contradicted each other (see chapter 3: “Analysis/ interpretation”) In chapter 4.2, “Using mixed methods”, I will reflect on using both quantitative and qualitative analysis. Qualitative analysis: close reading and semiotic analysis The texts were subjected to multiple close readings and coding. The codes were distilled from Baker and Ellece’s Key Terms in Discourse Analysis (2011) and include elements such as modality, the use of metaphors, and mitigating words (See Appendix A for the list with codes). The elements in this list were selected based on the criterion that these elements are often studied in Critical Discourse Analysis, since they may be indicators or unequal power relations, or rhetorical techniques. 14 Combining different methods, such as quantitative analysis of corpuses and close qualitative analysis is a form of triangulation, which minimizes the risk of bias — Reisigl and Wodak propose that this multi- methodological approach should be done in combination with “a variety of different empirical data as well as background information” (Reisigl and Wodak 2001, 35). 15 My approach could be regarded as a mix between a corpus-driven and corpus-based approach: in a corpus-driven approach minimal theoretical assumptions are used. It relies for its direction of analysis on “frequency and other statistical analysis”, while the later uses “the corpus as a source of examples to check researcher intuition” (Baker and Ellece 2011, 29). This study seems somewhere in between: on the basis of researcher intuition, a corpus is put together and this is firstly subjected to an exploratory data analysis (corpus-driven), which directs in consequence the researcher’s intuition (Baker and Ellece 2011, 29). 16 While Nvivo is qualitatitive data analysis software, I have only used the quantitative features such as word frequency queries and lists. 17 Baker and Ellece would possibly consider this approach to be subsumed under corpus-assisted discourse studies (CADS) (2011, 24–25). 32PDF Image | peek into the discursive construction of the Google Search Algorithm: A critical discourse analysis
PDF Search Title:
peek into the discursive construction of the Google Search Algorithm: A critical discourse analysisOriginal File Name Searched:
thesis-google-search-algotithm.pdfDIY PDF Search: Google It | Yahoo | Bing
Cruise Ship Reviews | Luxury Resort | Jet | Yacht | and Travel Tech More Info
Cruising Review Topics and Articles More Info
Software based on Filemaker for the travel industry More Info
The Burgenstock Resort: Reviews on CruisingReview website... More Info
Resort Reviews: World Class resorts... More Info
The Riffelalp Resort: Reviews on CruisingReview website... More Info
| CONTACT TEL: 608-238-6001 Email: greg@cruisingreview.com | RSS | AMP |