With the technology and communication tools available today, the boundary between work and life outside of work has become increasingly blurred, exacerbating issues around work-nonwork conflict. Can social media language be used to detect this conflict? Recently, Center for Working Well Faculty Associate Louis Tay and I conducted research investigating the potential of using social media text analysis to measure work-personal conflict, an increasingly important area for working well.
Previous studies found that 55% of workers send digital communication to colleagues after working hours, and 30% send digital communication on weekends — while expecting a same-day response. Work-nonwork conflict occurs in the other direction as well. Studies found that more than 60% of workers visit non-work-related websites at work, with more than half admitting to constantly surfing the Internet on the job. Additionally, 60% of online purchases and 70% of Internet porn traffic occur during the 9-to-5 workday.
While work-nonwork conflict has been studied by many researchers, most have considered family as the only or primary area of life outside of work. However, it is important to examine other nonwork areas as well, such as the personal area that includes friends, hobbies and travel. The personal area is becoming increasingly important as the number of single and/or childless workers, whose focus outside of work is likely on personal pursuits, continues to grow.
My colleagues and I set out to explore new ways to measure work-personal conflict. We were particularly interested in investigating social media language, as other scholars have used social media text analysis to measure constructs such as personality and job satisfaction. Additionally, social media text analysis is inexpensive and non-intrusive and is a great way to analyze large data sets with technologically advanced models.
We not only found that work-personal conflict was predictable from social media language, but we also found categories that predicted high and low conflict. We looked at both personal interfering with work (personal-to-work) conflict and work interfering with personal (work-to-personal) conflict. These include:
Netspeak. People with high personal-to-work conflict used more netspeak. For example, they used “u” instead of “you” and “coz” instead of “because,” and words like “lol,” “omg” and “welp.” This was an interesting discovery that would have been difficult to uncover using other, non-social media text methods. In fact, comparing people who use netspeak and those who do not may be a new direction for researchers looking to uncover how individual differences impact language.
Work. People with low personal-to-work conflict talked about work. This included language such as “working,” “overtime,” “shifts,” “clients” and “paycheck,” among other work-related words. Life satisfaction research shows that people are generally able to focus only on some (not all) areas of their life, which leads to deprioritizing the other areas. We theorize that people who post about work invest a lot of time in that area of their life, leading to less focus in the personal area. Subsequently, there is low conflict from the personal area to the work area for these people; in other words, low personal-to-work conflict.
Birthdays. People with low work-to-personal conflict talked about birthdays — a lot! From wishing others a happy birthday and thanking others for birthday wishes to posting about friends’ and family’s birthday parties, language about birthdays was the primary theme for people with low work-to-personal conflict. In line with life satisfaction research, one explanation is that birthdays are more personal-oriented (and not work-oriented). It is possible that people who post about birthdays invest more time in the personal area and less time in the work area, leading to low work-to-personal conflict.
Negative and Positive Language. We found that both high personal-to-work and high work-to-personal conflict were associated with negative language — swearing, negative tone and negative emotion. On the other hand, we found that both low personal-to-work and low work-to-personal conflict were associated with positive language — positive adjectives, excitement-oriented words and positive emotion.
What Does This Tell Us About Working Well?
Based on our research, work-personal conflict impacts people’s daily lives — so much so that it is related to what people talk about on social media. Work-personal conflict contributes to broader sets of well-being indices and ay impact economic indices as well. Establishing the ability to assess and track work-personal conflict — inexpensively and non-intrusively — through social media data will enable organizational researchers to contribute further to well-being discussions at organizational and societal levels. As communication technology continues to advance and more organizations move toward remote work and similar policies, the line between work and personal life may become even blurrier, making this line of research critical in the years to come.
Gloria Liou is a PhD student in the Department of Psychological Sciences in Purdue University’s College of Health and Human Sciences
