Data Replication for Sentiment and Emotion Network Analysis [SENA]
Description
Textual database to replicate procedures showcase in SENA, an analytic framework and software that conducts Sentiment and Emotion Network Analysis of textual data. The compressed folder contains two main forms of textual data. The folders called: "Essays_not_labeled" and "Essays_named_labeled" contain the same 11 essays on the reasons to participate in a data science seminar. The main difference of these files is that there in the "Essays_not_labeled" folder, the essays show the original name of the participants. In the "Essays_named_labeled" folder, these names have been replaced by an attribute of the respondent. In our case, this attribute is the gender of each participant. Based on this process, the analyses will conduct analyses based on the categories represented in these file names. When no labels or categories are added to file names, as in the case of the files contained in the folder "Essays_not_labeled" only aggregated SENA worldcloud analyses will be deployed. Finally, both folders contain Microsoft Word files saved as "*.doc" files. If your texts are in ".*docx" format, you need to save them as "*.doc" before uploading them to the software SENA. The second type file is a "*.csv" database with the same textual content as the two folders but stored in rows and columns. Each of these documents were decomposed into sentences and then the sentences were classified into topics following LACOID (Gonzalez Canche, 2023, reference below). To demonstrate the use of SENA, users may decide to include the topic identified with LACOID as an attribute of the text. If this is done, the resulting SENA analyses will yield comparative results. If users decide to ignore these attributes, only aggregate analyses will be displayed. González Canché, M. S. (2023). Latent Code Identification (LACOID): A Machine Learning-Based Integrative Framework [and Open-Source Software] to Classify Big Textual Data, Rebuild Contextualized/Unaltered Meanings, and Avoid Aggregation Bias. International Journal of Qualitative Methods, 22. https://doi.org/10.1177/16094069221144940
Files
Steps to reproduce
To reproduce unzip the folder. Once uzipped you may select to load the csv file or the word documents in the SENA application. To access the SENA software application follow the links described below. Mac users can access SENA here: https://cutt.ly/QwhYruBr Windows users can access SENA here: https://cutt.ly/YwhJJKvO