This dataset contains the the tweet ids of the tweets used in the article: 

Bovet, A., Morone, F. & Makse, H. A. Validation of Twitter opinion trends with national polling aggregates: Hillary Clinton vs Donald Trump. Sci. Rep. 8, 8673 (2018).

By using this dataset, you agree to the following points:

- The dataset is provided without any warranty or conditions of any kind. We assume no responsibility for errors or omissions in the results and interpretations following from usages of this dataset.

- You commit to cite our paper (above) in publications where you use this dataset.

- You commit to respect the Twitter Developer Agreement and Policy (https://developer.twitter.com/en/developer-terms/agreement-and-policy.html).

-----------------------------------------

The file tweet_ids.csv.gz contains the tweet ids of each tweet we collected from June 1st 2016 until November 9th 2016 (170696874 tweets) using the query "trump OR realdonaldtrump OR donaldtrump and hillary OR clinton OR hillaryclinton" (for tweets in english) repeatedly with the Twitter Search API.

Each line of the file has a tweet_id and its corresponding timestamp in EST timezone (comma separated).
           
The collection had some gaps (approximately 15 days are missing in total). See the article above for details.           
           
To download the corresponding tweets and recover the original user IDs of the retweet network, you may use the Twitter API using the corresponding tweet_IDs.