(full disclosure, the title is such an attention grab, and really no one wins or loses IRL)

We all have a friend who can’t separate themselves from their mobile device for any given period of time, regardless of the activity. My friend, who here I will call Michael Jordan (MJ) drives and tweets, drinks and tweets, works and tweets, weddings and tweets, WiiU and Tweets: you get the picture. With this (obsession?) comes the assumption that while this is an over-exaggerated example, we are all online at certain times of the day, and more and more of that connection is not through a laptop (relatively fixed location), but on our smart devices(anywhere and everywhere we can get data).

Given this trend “it is important to understand how the rise of mobile computing has changed (if at all) social media content (Murthy, Bowman, Gross, & McGarry, 2015, p.1). An excellent article entitled Do We Tweet Differently From Our Mobile Devices? A Study of Language Differences on Mobile and Web-Based Twitter Platforms (Murthy et al., 2015) studied 6 weeks of Twitter spritzer stream data, containing 235 million tweets to better understand the way we use Twitter, and how our use of the app differs between our use of it on a PC vs. a mobile device.

The following questions were used to guide the research:

Research Question 1: Are mobile tweets more egocentric than web tweets?

Researcher’s Hypothesis:We hypothesize that tweets from mobile sources will contain more references to the self.This follows from Deuze’s (2012, p. 57) argument that mobile phones have facilitated “processes of personal transformation toward greater individualization.” (p.5)

My Hypothesis: Given my own use of Twitter (and I’m not a fiend), when I am out and about, my Tweets are often retweets or are sharing something interesting online that I figured would be worthwhile to my twittercrew (followers is such an uninspiring zombie like term. ‘crew’ isn’t that great, but I’ll use it until I find something better). After looking through my posts, I did notice that many of the ‘mobile tweets’ were in fact egocentric in nature (look at this picture I took of a toy my son has!, etc.), so the researchers may have a point.


Results: “Figure 1 illustrates the frequencies ofmobile self and other unigrams versus nonmobile self and other unigrams across an average day. The results demonstrate that both mobile and web have consistently higher frequencies for self than their counterpart. While self continues to increase steadily into the night for both mobile and web, other exhibits a shallower slope or even a decreasing slope at the same hour” (p.11).

Research Question 2: Do mobile tweets employ more “feminine” language than web tweets?

Before I present the researcher’s hypothesis, it’s important to give some more info regarding why they chose this as a research question (I’ll let them explain themselves):

“While the user population in Sloan et al.’s (2013) study represents near gender parity, women have been found to have different motivations for using social media (e.g., for keeping in touch with family and friends) compared to men (Smith, 2011). Mobile tweeting could be complementing these relationship maintenance behaviors. Additionally, different computer-mediated communication contexts have been found to have consequences for gendered language use (Huffaker & Calvert, 2005).” (p.5)

Researcher Hypothesis: “Mobile platforms may encourage feminine language through a perceived social connectedness.”

My hypothesis: I am not a woman, so I feel ill equipped to answer this. I don’t even understand what is ‘feminine language.’ So I looked it up. I found a peer-reviewed article by Liu and Mihalcea (2007) that researches the dichotomy between masculine and feminine language  which helped me to learn that studies have been done that dissect male and female writing to contrast and compare content, syntax, choice of vocabulary, etc. Still, after reading the article, I don’t really feel confident taking an educated stab at a hypothesis, and I really don’t understand why the question is relevant anyway, given that people choose to express themselves in different ways, regardless of sex.

Results: “For mobile and web-based tweets, unigrams associated with masculinity clearly dominate in typical usage. For mobile and web, “masculine” unigrams not only rise in frequency earlier in the morning than “feminine” unigrams, but also experience a dramatic rise and a sharp fall after midnight (see Figure 4). Furthermore, masculine unigrams exhibit a rise or plateau into the late evening, whereas feminine-associated unigrams demonstrate a general decline across mobile and web.”  


Research Question 3: Are mobile-based tweets more negative than web-based tweets?

Researcher Hypothesis: “Web tweets may encourage more careful reflections of the day’s events, leading to less sentiment-laden tweets.”

My hypothesis: Thinking of some of my friends, watching NFL on Sunday, needing to reach their team to share their disgust with them seems pretty common. That paired with my friends’ off-color comments about their commute in larger centres in the world seems to strike the point for me that Twitter (through mobile use) is for many a canyon to shout their obscenities and frustrations about the world around them.


Results: “Figure 7 illustrates that there are more occurrences of positive words than negative words for both mobile and web Twitter users. While the occurrence of positivity is higher than negativity, Figure 8 illustrates that we found a higher percent of negative tweets from mobile users over an average day in our sample. These data indicate that over an average week in our sample, there appears to be an unexpectedly large tendency toward negative tweets from both mobile and web sources on Saturday evening.” p.15


Research Question 4: Are mobile tweets more agentic than web tweets?

Researcher hypothesis: “We hypothesize that mobile tweets would also center more on individualistic attributes of the self. These traits tend to reflect bold, outspoken, and frank personalities as well as reflection and censorship. As a result, web-based tweets may be more representative of communal characteristics. These traits emphasize behavior that contributes to the health of the group as opposed to the individual.”(p.6)

My hypothesis: I have found that predominantly the tweets I am creating through a laptop are done in tandem with work or academic pursuits, and things shared are done so to share knowledge I believe is (at the time) interesting and relevant to those that are part of my crew. That said, I find I am using my phone more and more to read and research when not at my desk at work, or kitchen table at home. So, in effect, the tweets I am sending out on my phone are also communal in nature. However, given the fact that many of my friends’ use of the same technology are more sarcastic, or have links to Vines or some Meme, I’d be willing to bet that the agentic tweets are more prevalent from mobile devices.

Results: “When comparing the percent of agentic and communal tweets across source, we found that consistent patterns emerged, suggesting a greater number of communal tweets across both platforms”(p.18).


What can we take away from this? My fear was that due to the nature of people accessing the web and twitter through their mobile devices, Twitter would likely become more of a mouthpiece for narcissistic and emotional messages/rants that are agentic in nature and don’t serve any other purpose. Not surprisingly, while mobile tweets are still much more ecocentric in language than it’s pc tweet counterparts, there still remains a very strong communal aspect to Twitter, whether approached from the PC or an online device. While this is reassuring, negative tweets were far greater for mobile tweets than those created on a PC. What does that speak to? The article has brought up a lot of good starting points for future research, and helps us to understand that yes, there are differences in how we choose to interact with different technologies that are interacting with the same software.



  1. Jeremy,
    Great post. As a woman I am now wondering if I am using a more feminine syntax. I think I will go on wondering too what that means…. I can see why you skipped over that part.

    Anyways, I would have to agree that many of the tweets I post via my laptop are related to research. However, I have also noticed that if you are engaging in a live large group chat, ie: Saskedchat, that you need to use a laptop and use Tweetdeck or Hootesuite, if you truly want to keep up with the discussion. Although I agree with your arguments for the most part, this aspect might change the nature of the study.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s