The And Do Not Deal With Time Series Application

1861 Words8 Pages
cross- sectional and do not deal with time series application. The official model specification is as follows:
Socialhappy= βo +β1socialmedia+β2goodroomate+β3studyfrisat+β4academichappy+β5alc+β6tv+β7sports+β8outdoors+β9earnings
The main focus of this study was to determine how social media use impacts social happiness. The social happiness question is phrased as, “How happy are you with your social life at Colby? [with the following options] Very happy, Mostly happy, Indifferent, Not happy, Miserable” I transformed social happiness into a binary variable. If a participant selected “Very Happy” or “Mostly Happy” they are considered socially happy at Colby. One challenge I ran into with this study was the uneven categorical data responses from the survey. The uneven increment increases made analysis more difficult in that you could no longer describe how a one-unit increase in Bk affects social happiness. For example the survey asked, “How many hours each day do you typically spend on social media (Facebook, Snapchat, Twitter, etc.)” The listed responses were None, I don’t use social media, 1 to 2, 2 to 4, 4 to 6, 6 or more. I corrected this particular issue by recoding social media to grow by two-hour increments. Based upon how I observe my own peers relationship with social media, I predicted a negative relationship with social happiness. The added dictionary definition of FOMO also gives validity to this claim.
Other variables also suffered from irregular increments of
Open Document