Relevant FactorsThe article evaluates the following extrinsic factors that may influence the response rate in social question and answering (social Q&A). The numbers in brackets indicate the correlation coefficients obtained.
- Number of followers (0.21)
- Frequency of posting (very high and very low frequencies reduce the number of answers; -0.14)
- Number of @mentions (which is also considered a strong predictor of information diffusion and a significant factor for increasing a post's visibility; 0.20).
- Number of @mentions to general/verified accounts within the micro blogging site (-0.33/0.03)
- Use of hashtags (-0.18)
- Emoticons (0.10)
- Repeated punctuation and interjections (not significant)
- Expression of gratitude (not significant)
- Topic (depends on the category - between -0.08 for electronics and 0.63 for entertainment)
- Posting time (the time period between 0:00 and 6:00 am yielded the lowest number of responses, the differences between the other time periods are not significant).
EvaluationThe evaluation draws upon Sina Weibo, China's largest micro blogging site, analyzing questions and answers within a 10-day period and considering only distinct answer. The authors use multiple linear regression analysis to verify the impact of these predictors yielding a model which accounts for roughly 19% of the total variance. Multicollinearity was checked by examining the bivariate correlations across all predictor variables.