Methodology

For my investigation, I collected twelve different peoples tweets, six of them being teenagers and six of them being adults. I collected them all over a similar period to make so it was a fair comparison. I chose six teenagers aged 17 and 18 and six adults in their 50's, both with similar social and ethnic backgrounds and from similar locations so their wouldn't be any dialect differences.

 I chose these ages as I feel a 17 year old will show the best representation of the young generation. I think if I were to use people aged in their 20's, they would still be more formal as they are a lot more mature and grown up to a teenager and would be more similar to the older generation in their tweets. Similarly I think if I used younger teenagers, they may either not be on social network sites yet, or would be a lot more immature and possibly tweet more like a child. I used adults in their 50's to analyse because its the kind of age they are very mature and if I used people younger, maybe in their 30's or 40's they may try to tweet like they're younger and act a bit younger. I think the age gap is big enough to get very different results so I will get very different reults to compare.

I collected the 20 different tweets from each person to get a representation of the kind of tweets they produce and the language they use so I could compare them for my investigation. I took every fifth tweet from their profile, because if I took the first 20, it could be unfair as it might be an unrealistic representation of their overall tweets. Also the first 20 tweets could be about the same or a similar subject matter and not explain what they talk about or how they talk at other times or most of the time.

It is a reliable data collection as it gives a similar and fair representations of each of the people as I used quite a large set of data per person. I also used a good number of different people to make my investigation more reliable as I am then able to compare more data and find more similarities to analyse.