Methodology

I began my investigation by collecting my data. I transcribed five tweets from each participant. I chose five because I thought it would be enough to analise however to take my investigation further next time  I will tanscribe more tweets to get a more presise result. I collected random tweets from a certain month (July) this meant they were all writing things at the same time as one another and this made finding data easier for me. Also because they weren't all written on the same day it makes it more specific. 

The process, for my main data, I collected five tweets from each celebrity, copy and pasted them onto a word document. I then highlighted with different language techniques that I could compare and what were similar. For example I highlighted the use of swearwords, the formality of a sentence, accent, ellipsis, abstract nouns, and Internet slang.

This process was reliable as the tweets were directly from the celbrities twitter accounts . However there is false hope that a celbrity could of had somebody else writing there tweets for them this is called a ghost celbrity, somebody impersinating them because they are told what to write for them. In this situation I assume there was no observer parodox because that could change the reliability in my results. 

The participants for my data were well known rap artists and well known pop artists so all of the celebrities are widely known which makes the investigation more interesting to see the language they use to maintain their followers (fans) happy with the things they write. It was difficult and impossible to get results from celebrity's that were simular ages so there is a variety. Despite the age differences they are all in the same situation with a fan base and they all have the responsibility for their own actions.

I intended to anaylise my data on two frameworks, lexis and grammar. I have focused on politeness , swearwords, abstract nouns, Internet slang,ellipsis and socialect. These are going to help prove my hypothesis.