Our project works best with at a range of decades so we have designated a decade to each person:
Ellie - 1960s
Jess - 1970s
Kate - 1980s
Liam - 1990s
Anna - 2000 - 2010
To make sure that we have a good range, with enough space between each paper, we have decided that we should all look at the 2nd and 7th years within our decade.
We will all then analyse our copies looking for the following features:
Formality
To make sure that we have a good range, with enough space between each paper, we have decided that we should all look at the 2nd and 7th years within our decade.
We will all then analyse our copies looking for the following features:
Formality
Tone
Colloquialisms
Archaic words
Layout
Purpose
When it comes to producing the final presentation, each person will feedback on their own decade.
When it comes to producing the final presentation, each person will feedback on their own decade.
Great stuff! I wonder how much data this will be? Don't give yourselves too much to cover. The sampling procedure you've come up with is a good one. You could go further and pick just one or two issues from the same time of year, and concentrate on items from each paper which are comparable (e.g. one editorial + one film review + one uni news story). Even this would give each person 6 short texts to analyse.
ReplyDeleteTry to agree an analytical framework that's a bit more specific than this list of features - e.g. you could code sentence length, lexical density etc., and you could specify more tightly how you will identify 'tone' and 'formality' (e.g. pronouns? abbreviations? colloquialisms are probably a sub-set of the tone/formality dimension, aren't they?)
Look at the v-lecture on data analysis; you could follow the suggestion that you each code a short stretch of your data independently and then compare results, to see if you're in agreement about how to classify the features.
One more suggestion: it may be easier to spot the contrasts if you compare similar parts of the paper across the decades. In that case, you might decide to divide the task not by decade but by text sub-type (Student A does film reviews; B does letters; C does news etc.)
Good luck - and do get in touch if you have questions etc.