This Dataverse contains word embeddings, word embedding vocabulary files, and sentence files for sentiment analysis created from Daily Nation (a Kenyan newspaper) article text from 1998-2019. The uncleaned article text is also included for each year and section of the newspaper. There are embeddings created from text separated by individual years (e.g. 2005) and groups of three years (e.g. 1998-2000). There is a set of these word embeddings created using the GloVe algorithm, designated with "glove" at the beginning of the file, and the word2vec algorithm, designated with "daily nation" at the beginning of the file. See the Github repository https://github.com/emmapair/Kenyan-Embeddings-and-Sentiment for the analysis code.