Web Survey Bibliography
This chapter discusses both the conceptual implications of applying sentiment to textual data as well as the operational steps to apply structure to the text under consideration. The primary result of sentiment analysis is a classification of each post in the dataset using a predefined set of categories. Depending on the unique attributes of the input text and the algorithms used to determine sentiment, different types of ordinal scales are used. Alternatively, coding text into nominal categories that define a specific emotional experience may be in some instances more applicable to the analysis. Finally, some posts or text express no sentiment at all. Several strategies improve accuracy when preparing data for analysis and processing the data to determine sentiment. These follow a specific process: first, a precursory exercise of understanding and planning the problem domain; second, harvesting the data; third, structuring and understanding the data; and fourth, analyzing the data.
Web survey bibliography - In Hill, C., Dean, E., Murphy, J. (eds.): Social Media, Sociality, and Survey Research. Wiley (6)
- The Future of Social Media, Sociality, and Survey Research; 2013; Hill, C., Dever, J. A.
- Collecting Diary Data on Twitter; 2013; Richards, A., Dean, E., Cook, S.
- Second Life as a Survey Lab: Exploring the Randomized Response Technique in a Virtual Setting; 2013; Richards, A., Dean, E.
- Virtual Cognitive Interviewing Using Skype and Second Life; 2013; Dean, E., Head, B., Swicegood, J. E.
- Sentiment Analysis: Providing Categorical Insight into Unstructured Textual Data; 2013; Haney, C.
- Decisions, Observations, and Considerations for Developing a Mobile Survey App and Panel; 2013; Roe, D. J., Zhang, Yu., Keating, M.