The dataset is available on the Kaggle website here. It is a .csv file. The code I used to analyse the data is available here.
With Trump repetition is important, some examples:
How much? How much for my family?
And we have a Fed that's doing political things. This Janet Yellen of the Fed. The Fed is doing political -- by keeping the interest rates at this level.
No, you're wrong. You're wrong.
I've been all over the place. You decided to stay home, and that's OK. But I will tell you, I've been all over.
nobody was caring much about it. I figured you'd ask the question tonight, of course. But nobody was caring much about it.
I do want to say that I was just endorsed -- and more are coming next week -- it will be over 200 admirals, many of them here -- admirals and generals endorsed me to lead this country. That just happened, and many more are coming.
Trump is not a professional politician, and this came across during the debate. He did not have the same level of fluency as Clinton. He sometimes seems to use repetition rather than clarity or precision to get his message across. This is neither a criticism nor an endorsement of Trump - debating is a skill and a professional politician will have had much more experience in it than a business man.
Top ten words used by each candidate:
(would, 41), (think, 38), (people, 33), (know, 28), (going, 26), (Donald, 24), (need, 23), (us, 23), (one, 22), (that's, 21)
(going, 47), (country, 46), (look, 43), (they're, 41), (think, 38), (said, 35), (people, 35), (don't, 34), (say, 31), (I'm, 30)
Words common to both candidates = going, think and people
Clinton uses 'going to' and 'will' with almost equal frequency when talking about the future: (going to, 26), (will, 20)
Trump prefers 'going to' over 'will' when talking about the future: (going to, 40, will 21)
Trump uses 'think' mainly in the phrase 'I think...' (30 times) and 'I don't think...' (6 times) there was one occasion when he used the phrase 'if you think....' and once he used the phrase 'like you think...'
Clinton uses 'think' mainly in the phrase 'I think...' (35 times) and 'I don't think...' (3 times)
Clinton uses 'would' to express the conditional mood on 41 occasions, Trump uses it 21 times.
Clinton is fond of 'us' while Trump prefers 'I' and 'they'. When Trump uses 'they' he is talking about groups which he likes to demonize, for example: illegal immigrants, Mexicans, ISIS, Chinese, Iranians, member states of NATO (other than the US) and Saudi Arabians - so mostly foreigners, this is the xenophobic aspect of his campaign - the US is threatened by foreigners abroad and at home and he is the man who has the solutions.