Discourse analysis on the UEF website

As a starting point of my research, I examined the UEF’s international website. The collected data was analysed by the method of discourse analysis. Fairclough (2012: 9) characterizes discourse as (a) an element of the social process that creates meaning, (b) a language that is regularly associated with certain social fields or practices (e.g. ‘academic discourse’) and (c) a way in which different aspects of the world are associated with certain social perspectives.

According to Kress (2012: 35), discourse analysis has been shadowed by the vagueness around the homonym ‘discourse’, and in my opinion, it is why there are numerous different approaches to conducting this form of analysis. Hyland (2012: 412–423) combines discourse analysis with EAP, and highlights that by doing so, researchers can emphasize concrete texts rather than institutional social practices. Researching EAP usually means analysing specific academic

30

genres (e.g. the research article, the conference presentation or the student essay) (Hyland 2012:

141).

When examining the UEF’s international website, I kept in mind the method carried out by Jenkins in her research on various other universities’ websites: searching for any written material that refers to language policy matters. I made notions whenever there seemed to be a clear omission of content in English whilst the information is otherwise provided in the Finnish version of the website. However, since I did not intend to make a thorough comparison of the two language versions, I was more interested to hear whether the international students have come across instances where there is less information provided in the English version of the website.

Jenkins (2013: 81) assumes that most of the content (especially if it relates to policies found in mission statements etc.) on the universities’ webpages would be under the control of senior management and faculty, but does recognize the marketing/branding forces behind the institutions’ website. Nevertheless, she also presumes that all findings about English language policies do reflect the ways in which English language issues at universities are regarded (Jenkins 2013: 81). Jenkins (2013: 84) and her research partner Panero explored the international universities’ websites, their programme documents, entry forms, videos of classes etc. To be more exact, they searched for information of the following kind:

- the institution’s stated/implied conceptualization of ‘international’ HE - the institution’s stated/implied approach to ‘EMI’

- range of courses/subjects/levels offered in EMI

- language requirements (entrance/exit examinations, mandatory language courses, differentiation between groups in terms of requirements)

- type of EAP support offered, if any

- teaching staff on EMI courses (local/non-local, NES/NNES) - resources used on EMI courses (based on Anglophone or own)

- visuals (videos, photos, logos, images, graphs, hyperlinks, typography, etc.) - testimonials, discussion boards, interactive features

- other (anything of interest and relevance not covered by the above).

Jenkins also refers to Pauwels (2012: 252), who has developed a model to help researchers analyse websites as social and cultural data sources. His model includes six phases that advance from looking at the more apparent features and collecting data to more in-depth interpretations of all elements and their complex relations. All of the phases are displayed in Appendix 1. Some of the elements in Pauwel’s framework are outside the scope of my Master’s level study, which

31

is why I considered all the phases of Pauwel’s model before concluding to utilize only a few of them. Since I have already become accustomed to using the UEF website, albeit in Finnish, it was impossible for me to conduct the first phase of the framework: Preservation of First Impressions and Reactions. The first phase has to precede the actual analysis and record the affective reactions of the user, which is why I was not able to include it in my analysis.

However, it was vital to include the second phase of the framework: Inventory of Salient Features and Topics. At first, it was difficult to conduct this phase, because I had not decided on the scope of which I was to collect data from the English version of the UEF website. After spending 3 hours collecting material from the website, my data already amounted to 140 pages of text, and there seemed to be no end in sight. The scope of the material had to be narrowed down, which is discussed more thoroughly in Section 3.2.1.

It was important to carry out parts of phase three of Pauwel’s model: In-depth analysis of content and stylistic features. As Pauwels (2012: 253) explains, this phase “proposes to first look at the potential information that resides in the separate modes (intra-modal analysis) and then to look at the complex forms of interplay between the different modes (cross-modal analysis).” Performing intra-modal analysis meant that I observed the impression provided by the UEF website at the level of lexicography and syntax. I decided to leave out the typographic signifiers, visual representation signifiers as well as sonic types and signifiers from my data, since I assumed that the faculty members who produce the information for study programmes do not have much influence on these matters compared to the marketing team and communications department. Moreover, I correctly presumed that the most recent update on the outlook of the UEF website (published in the spring of 2015) has been designed by a marketing agency.

In terms of observing the Layout and design signifiers, I followed Knox’s (2007) explanation of the multimodal nature of websites. Knox (2007: 19–53) describes this feature by categorizing websites into three main dimensions. First, webpages usually recognize the head-tail opposition, and thus place the most important content on the ‘first screen’ (Knox 2007: 38).

Basically, the longer it takes to scroll down the page to reach the sought information, the lesser its importance (Jenkins 2013: 86). Secondly, by separating the columns of a webpage into the widest centre column (which displays the primary information) and the narrower side columns (displaying the less important information), information value of primary-secondary is realized

32

(Knox 2007: 39). Lastly, webpages make use of the ‘guaranteed viewing area’ in the top-left corner of the screen by positioning the content with particular significance in there – thus utilizing the MacroTheme-Rheme dimension (Knox 2007: 46).

Since I left out a number of parts from the Intra-Modal Analysis phase, I was not able to carry out the Analysis of cross-modal interplay, which would compare the relationship between linguistic, visual, auditory, spatial and time-based elements (Pauwels 2012: 256). However, I managed to make notions related to In-depth ‘negative’ analysis, which analyses aspects and issues that are, for some reason, not present in the collected data. According to Jenkins (2013:

85), these aspects seem to become significant for their absence alone. Analysis of this kind was also used by Saarinen and Nikula (2013: 133), who mention that noting the absence of references to language use is important, because the decision to discard them from a website reflects the general ideologies of the HE institution.

In the fourth phase of the framework – Embedded point(s) of view or “voice” and implied audience(s) and purposes – research moves from observing what is being said or implied to who says it to whom and why (Pauwels 2012: 256–257). This phase, while complex, was an important part for my discourse analysis. In the fifth phase – Analysis of dynamic information organization and spatial priming strategies – the focus is on the structural and navigational options and constraints of websites (Pauwels 2012: 258). I think this phase might be useful when the in-depth ‘negative’ analysis is under observation, which is why I do not intend to cut it out entirely. The last phase – Contextual analysis, provenance and inference – looks at the larger context in which the website is produced, and it was something I considered during the analysis of my thesis.