Íslenskt mál og almenn málfræði - 01.01.2023, Síða 147

The study was based on a twisted matched-guise test in the form of an online
survey. In this survey, 211 participants evaluated the “authors” (i.e. guises) of two
written texts by means of certain character traits on a 7-point Likert scale. Both
texts comprised the same content, namely a request for an apartment in the
Reykjavík capital area, but they differed in their writing style. The first text con-
tained features that are associated with informal digital writing including English
borrowings, spelling variations, emoticons, colloquialisms, etc. The second text,
on the other hand, corresponded with a more formal writing style as it did not
contain any informal features.
When I started this study, a housing request seemed to be a suitable topic for
a Facebook post as requests of this kind are very common. For example, numer-
ous Facebook groups are devoted to finding and advertising housing in Reykja -
vík and the rest of Iceland. Therefore, I expected the informants to be somewhat
familiar with rather informal requests like the one presented in the informal text
guise. However, the informants almost unanimously preferred the more formal
text guise. There were hardly any differences in the evaluations between men and
women or between younger and older participants. Only one research group
showed no clear preference for either text: informants with an elementary school
degree or an apprenticeship certificate. They were equally positive toward both
guises. Participants with a high school degree or any kind of university degree,
however, preferred the more formal text. In sum, the results suggest that even for
a topic that is typically discussed on Facebook, a more formal writing style is pre-
ferred.
The linguistic analyses of Icelandic Facebook practices, that were the primary
focus of my research project, partly contradict the results from the Dulin viðhorf
study. I analyzed actual Icelandic Facebook practices based on a research corpus,
which I compiled for the purpose of the project. The corpus consists of status
updates and comments of 28 Icelandic native speakers published between Sept -
ember 2012 and the end of October 2014. By the time that I started the linguistic
analyses, however, I realized that the data set was too big for me to handle as a sin-
gle researcher. Also, the material had become a bit outdated. Therefore, I decided
to consider only status updates and comments shared in 2014. This left me with a
research corpus of 8476 posts.
The quantitative analysis of this research corpus was concerned with three
research questions:
1. What are the formal characteristics of digitally written Icelandic?
2. What are the linguistic resources users draw upon?
3. To what extent and in what ways are features from these resources mixed
and combined?

Firstly, I analyzed the posts regarding the linguistic resources they entailed. For
this purpose, I coded the posts with respect to their participatory roles, status
Presentation of the thesis 147