An Insider’s Look at Sentiment Analysis : What Works, and What They Don’t Tell You (watch on YouTube)
Instructor: Normand Péladeau, Provalis Research
Text analytics allows you to perform sentiment analysis that provides real value to help you shape and grow your business. That statement isn’t up for analysis. Text analytics works but it doesn’t necessarily work the same way for everyone. To make text analytics work for you, you need to know some of the pitfalls to avoid the pratfalls. We will show you sentiment analysis techniques and methods you can deploy, what’s behind them and what to watch out for.
Comparative Survey Research: Issues of Quality, Harmonization and Transparency (watch on YouTube)
Instructor: Irina Tomescu-Dubrow (Institute of Philosophy and Sociology, Polish Academy of Sciences, and CONSIRT at The Ohio State University and PAN)
Comparative survey data are at the heart of research that analyzes social phenomena for different populations (across countries, or social groups within country) and, frequently, over time (in cross-sectional or panel frameworks). To yield reliable results, such data need to be of good quality, and be used in informed way. This session focuses on the concept of quality as it applies to cross-national survey research, from the perspective of both data producers and users. First, we discuss the variability in survey quality that emerges through the design, collection, and documentation process. Here, we consider how ex-ante harmonization – that is, procedures that data producers apply during the design and implementation stages to improve comparability – can strengthen survey quality. Next, from the data user’s side, we discuss the problem of unequal quality within and between international survey projects. We consider the data quality challenges, and possible solutions, as they appear in ex-post harmonization of cross-national survey data. Ex-post harmonization is data reprocessing that facilitates the simultaneous use of diverse surveys conducted in multiple countries and across many time periods for comparative analyses. We conclude with a discussion of practical concerns regarding transparency of comparative survey research.