Subscribe to DSC Newsletter

Is the economic crisis over? What is the sentiment of people regarding US Economy and the future? These are some of the questions that many people ask these days and the signs are somewhat mixed. Dow Jones is close to the 10000 mark and some US Economy Indices show that the worse is behind. But do people feel the same?

To answer these questions 10000 Tweets containing the word economy were collected with the purpose of finding out what people think and how they feel about the US Economy and the economic crisis. The following web chart shows some of the results :

PositiveSentiment is an annotation type that includes all words that suggest positivity such as good, better,advances while the opposite annotation (NegativeSentiment) exists for all keywords that suggest negativity.

The bolder the lines between words the heavier the association. To get an idea of how people feel, look at the line that connects NegativeSentiment and the word still which implies that the strongest sentiment is that US Economy is still under big problems.

Some other findings :

- US President tells that the economy gets better but people don't feel the same.

- Economy cannot be getting better while at the same time there are layoffs.

- People expressing very negative feelings after losing their jobs.

Full story at:

The sentiment on US Economy from Twitter
Monday, November 02, 2009 1:32 AM
By Themos Kalafatis

Views: 200


You need to be a member of AnalyticBridge to add comments!

Join AnalyticBridge

Comment by Themos Kalafatis on February 9, 2010 at 10:53am
Hello Andrzej,

Just now i saw your comment so i am sorry for the late reply.

It is true that analyzing information in that way has several problems, for example :

1) It is taken from a specific population such as Americans that *also* use Internet that *also* use Twitter.

2) The techniques that are used to capture sentiment may be biased because the programmer -me in this example- could be also biased to capture more negative sentiment than positive through the coding that i do.

3) See also the discussion started by myself on "Research Methods" group of LinkedIn for potential problems and pitfalls of Social CRM Analytics.

On the other side :

1) Increasing the sample size to 20K made no difference in the (specific) results that i was able to disclose (not all results were shown).

2) Americans using Twitter were the target of the analysis and not the Polish. If that was the case the results would have been different most likely. That fact however does not cancel the insights extracted

3) My experience has shown that despite the sampling / programming issues of analyzing text information from Social Media, the insights have proven themselves over and over again. A statement which -unfortunately- cannot be elaborated further.

Best Regards,

Comment by Andrzej Góralczyk on December 29, 2009 at 3:52pm
I like this excellent study, but have also some doubts.

1. We cannot compare positive and negative sentiment directly by simple comparison of the count of associations (co-occurences) due to some psycho-cultural bias. For example, in Poland people tend to be over-pessimistic, and there is no reason for them to express sentiment if something is OK. So, there are much more negative expressions than positive ones for almost every subject. In US opposite, people tend to be optimistic. Clear example is overall positive sentiment towards financial institutions presented in the figure; in Poland sentiment towards financial institutions is very negative, despite these institutions were much more "honest" (in practice we had no financial crisis in Poland).
So, the count of associations should be "normalized" in some way before putting those of positive and negative sentiment into one figure.

1a. If Americans are over-optimistic (I don't know), the overall picture should be worse than that presented in the figure.

2. There are only 10,000 tweets annotated and obviously some particular associations counts are very low. For such cases there is a problem of geometry that should be resolved, again as a unique problem of normalization, usually difficult to tackle. I wonder if a lonely NoCharacterization association to Debt is the kind of geometry effect...

On Data Science Central

© 2021   TechTarget, Inc.   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service