The previous post looked to analyse what words were in context with Nick Clegg in a graphical manner. This analysis will take this one step further and show the actual scores for each word against Nick Clegg in our analysis, over the two TV debates and for the BBC Have Your Say discussion forum data and also the Telegraph online comments.
The first table below shows the analysis for the BBC Have your say data after the first TV debate. This mirrors the analysis seen previously with the word 'won' and 'well' appearing at the top of the list with high scores. The scores drop off fairly quickly however, indicating that the other words found have a low contextual score with Nick Clegg.
well | 120 |
came | 110 |
debate | 87 |
had | 80 |
won | 72 |
did | 70 |
very | 66 |
other | 64 |
has | 60 |
think | 54 |
will | 48 |
would | 44 |
good | 38 |
top | 34 |
night | 34 |
two | 32 |
people | 28 |
party | 22 |
can | 20 |
lose | 20 |
vote | 20 |
get | 18 |
far | 18 |
three | 16 |
thought | 16 |
The analysis for the Telegraph data shows a similar story:
won | 22 |
debate | 20 |
other | 14 |
talks | 8 |
then | 8 |
political | 8 |
reform | 8 |
easy | 8 |
candidates | 8 |
talk | 8 |
choice | 8 |
tory | 8 |
candidate | 8 |
walk | 8 |
incumbent | 8 |
relatively | 8 |
expose | 8 |
outmaneuver | 8 |
mere | 8 |
road | 8 |
blocks | 8 |
refuses | 8 |
This time the word 'won' is the first in the list, with a high score, and the majority of other words dropping off quickly.
The analysis for the data after the second TV debate is slightly different. Here the words that are most commonly in context with Nick Clegg are fairly neutral, indicating that there is no strong theme emerging, and the remaining words although a mixture of positive and negative concepts all have a relatively low context score.
debate | 44 |
last | 36 |
thought | 30 |
would | 28 |
were | 19 |
week | 17 |
other | 14 |
did | 13 |
make | 12 |
issues | 12 |
audience | 12 |
sky | 12 |
very | 10 |
common | 10 |
constantly | 9 |
leaders | 9 |
notice | 9 |
camera | 9 |
passion | 9 |
spoke | 9 |
serious | 9 |
bias | 9 |
This is repeated for the Telegraph data shown below:
debate | 38 |
stronger | 18 |
came | 18 |
people | 16 |
bad | 16 |
mind | 16 |
prime | 16 |
seeing | 16 |
tonight | 16 |
watch | 16 |
stage | 16 |
action | 16 |
minister | 16 |
2nd | 16 |
caring | 16 |
smiled | 16 |
figure | 16 |
weblink | 16 |
amazingly | 16 |
Here it can be seen that the scores are all relatively low and also that there are not many words coming through the analysis. This is due to the fact that the focus of comments on the Telegraph site in particular were on David Cameron and not Nick Clegg.
Trends can be seen to develop even in these two rather limited data sets. The word 'won' is at the top of both the BBC and Telegraph lists after the first debate with a score of 0.14 and 0.27 respectively. After the second debate this score drops to 0.03 on the BBC forum and does not even appear in the Telegraph data (effectively a score of 0).
As we collect more data we will be able to track and assess more emerging trends in public opinion.
We will continue this analysis and with more data feeds in the coming days and for the other party leaders.
Follow election trends on twitter http://twitter.com/electiontrends
Find out more about Election Trends http://electiontrends.blogspot.com/2010/04/election-trends-launches.html
Discover more about Emiore Trends http://electiontrends.blogspot.com/2010/04/emiore-trends-launches-digital-market.html
Discover more about Blueflow Technology http://www.blue-flow.com/
No comments:
Post a Comment