Investigando como características desejáveis afetam o desempenho de perguntas sobre programação em sites de perguntas e respostas.

Abstract

The social query is a way to find information in social media age. It is the process of sharing a problem with the user of an online community and waiting for answers. Getting help demands that someone able and willing to help perceives the request. However, this is not always easy. In this context, the literature about this theme has focused in proposing question routing techniques, arguing that directing the problem to experts warns their attention and guarantees high quality answers. However, there are questions that are directed to nobody, but broadcasted, that also receive answers. In this research, we ask the community from Question and Answering Sites about Programming which characteristics they expect in questions they choose to respond. We gathered these answers and created a guideline with 16 suggestions on how to include “good” characteristics in the question. Next, we evaluated how the presence of more of these characteristics was correlated with the question performance (considering features like the number of views, number of answers and time for first response). We found very weak correlations only; thus, indicating that the presence of more of these characteristics has no meaningful impact, even if the community believes it does.