Table 8.

Summary of hypotheses and results

	Hypothesis	Result
H1	Sentiment of brand tweets positively impacts store visits	Brand sentiment has an inconclusive impact on foot traffic with no influence in the preferred model but a positive relationship in other models
H2	Disagreement of brand tweets negatively impacts store visits	More disagreement on social media leads to more foot traffic, indicating that greater disagreement represents higher brand engagement through diverse posts
H3	Subjectivity of brand tweets negatively impacts store visits	The subjectivity of social media posts has no measurable impact on retail foot traffic
H4	Popularity of a brand positively impacts store visits	More tweets about a brand leads to more store visits
H5	Number of likes for brand tweets positively impacts store visits	There is higher foot traffic when a brand get more likes; however, the effects are not as large compare to the per-tweet effects in H4
H6	Number of followers for a user tweeting about a brand positively impacts store visits	As the number of followers per tweet mentioning a brand increases, there is virtually no impact on retail visits
H7	Recency in days from brand tweets positively impacts store visits	Time has a nonlinear effect where social media activity is less impactful in the same day, more impactful after a few days, and insignificant within a week