Social Media, CMD, and Age: sociolinguistic variables change as social media users age.
Researcher/Author
Febuary - 2023
Social Media, CMD, and Age
A research essay on recent sociolinguistic research about age in relation to linguistic variables and discourse on a social media platform. It is about specific age groups (e.g. adolescents,…) and critiquing age as a social variable. It includes a literature review section, which summarizes some recent research on this subject.
Introduction
On entering the age of data and technological evolution, the introduction of media and social platforms such as Facebook and Twitter, blogs, forums, and instant messaging has made it possible for people of all ages to participate, share and connect. Social media offer the opportunity to collect large amounts of unprecedented sources of informal written language from many users who are collapsed into a single context: tweets/status/messages can be targeted to a person, a particular community, or the general public (Divita, 2012). As a result, language practices of different age groups of pre and post social media shows signicant variation, which leads to the growing body of sociolinguistic research focusing on Computer-mediated Discourse (CMD) and its linguistic connection to the diverse age range of internet users.
Within the scale of research of this essay, the relationship between age as a factor in linguistics variation and discourse on social media platforms will be drawn through some focal aspects: language style (words and phrases), topic of interest, and online behavior (using hashtags, links, tweet/status length, time of posts, number of friends) of different age groups represent different life stages, for example, under 20 - teenager, secondary - high school student, 20 to 40 - college student, 40+ - employee. Materials, research, and detailed analysis of recent expert studies will be provided to support and clarify that sociolinguistic variables change as social media users age.
Age-related variation and the invention of social media
Taking into account the stereotype perception and traits of each generation/age group towards the emergence of social media would aid linguistic examination on viewing age as a sociolinguistic variable.
The aging process correlates with higher levels of agreeableness, conscientiousness, and adherence to social norms (Hendrick, Knox, Gekoski, & Dyne 1988, cited in Cheshire 2006; Eckert, 1997; Pennebaker & Stone, 1999). Wisdom and commonsense are purported as people age, accumulating during middle adulthood and later years in life and based on the experience of variabilities of life, contexts exposure, and acquiring the basic issues of life; aging can reorganize the roles of cognition and that this process of reconstructing brings better cohesion, which produces greater regulation of emotion (Eckert, 1997). As a result, language practice shift as an individual grow older, more positive and fewer negative affect words are used (Pennebaker & Stone, 1999). As age proceeds, the language use of pronouns and determinations changes as there is a nding that social connection and social growth, in some cases, diminish because older adults are more satised with the sizes of their social networks than younger adults; emotional connection to relationships and friends is also be reported (Pennebaker & Stone, 2003; Rosenthal & McKeown, 2011). Susan Kemper et al. (1990; cited in Pennebaker & Stone, 2003) have found that over the life span, the level of complexity of narratives increases with gradual drops in assesses grammatical complexity and idea density over time. She also pointed out that the specic language patterns related to age and cognitive complexity will be curvilinear and peak in middle adulthood. Reviews above conclude that with the proliferation of age, more positive and less negative words are used, fewer self-references, and demonstrate a general pattern of increasing cognitive complexity.
Social media updates are self-constructed, personal, and involve emotional content (). Different discourses, social groups, and communities have their particular style of language, in the case of social media, users exchange information with one another via texts, messages, posts, comments, links, etc. with instant response; such environments emerge what is called ‘cyber language’. Ferrara, Brunner, and Whittemore (1991; cited in Peersman, C, Daelemans, W, Vandekerckhove, R, Vandekerckhove, B & Leona Van Vaerenbergh, 2016 ) have termed cyber language as “interactive written discourse” since it has the attributes of spoken language but is not spoken as written. Modications and adoptions are seen in capitalization multiple phonemes, conversion, initialism, clipping, blending, acronym, abbreviation, contraction, substitution, emoticon, nonstandard spelling, and punctuation. Social media afford unprecedented diverse sources of written language from diverse ages of users, which can be leveraged to study age as a factor of social variables.
Report and analysis
Recent research and studies on the relationship between age as a linguistic variable and discourse on social media platforms have mostly been analyzed through changes of language use of different people of different ages, not through changes of language use during the life span of an individual. Therefore, the resulting patterns are the reection of changes between different generations (Pennebaker & Stone, 1999).
1. Language style
Schwartz et al. (2013) conducted a study on the language use of posts and messages of 75 thousand authors on Facebook, which shows clear distinctions between different age groups of users, such as use of chat language (haha, xp), slang (ikr - I know right, idk - I don’t know), emoticon (:D, >:0) in the 13 to 18 age group. A similar pattern shared by the works of Cheshire (2006) and Nguyen et al. (2021), users age 30 and under are seen to have informal, unconventional style of language such as unconventional punctuation (e.g. ‘?!’, ‘!!!’, ‘???’, ‘!?!’) emoticons, ellipsis (...) and explicit stylistic modication such as alphabetical lengthening (e.g. ‘nooo waayyy’, ‘niiiiice’) capitalization (e.g ‘HAHA’, ‘pLeAsE’) more intensier (e.g ‘so’, ‘really’, ‘awful’). Whereas older individuals’ language use is more conservative, little use of lexical-stylistic features; words of positive emotion like ‘take care’, ‘hope’, ‘wish’, ‘thanks’ indicate support and wishing well, negation and negative emotion words also decrease with age. As the speaker enters the employment life stage - 23 and older, the pressure of using formal standard language to be taken seriously and get promotion rises (Eckert, 1997). Another report by Sap et al. 2014 pointed out that as age progress, the use of determiners like ‘the’, ‘an’, ‘a’ increase and serves as a representation for the intention of using more nouns; younger people are seen with characteristics such as low in agreeableness, conscientiousness, high in neuroticism, for example, people with high neuroticism tend to use phrases more frequently, ‘sick of’ is mentioned, not just ‘sick’.
How pronouns are used by people of different ages is one of the most studied aspects. According to Pennebaker and Stone (2003) and Barbieri (2008) (cited in Rosenthal & McKeown, 2011), younger internet users tend to use more rst-person (e.g. ‘I’) and second-person singular (e.g. ‘you’) pronouns, while older people prefer fewer self-references, rather refer themselves as ‘we’. The monodical decrease of ‘I’ can be explained as a proxy for social integration when a person realizes the increasing importance of relationships as they age (Rosenthal & McKeown, 2011).
2. Topic of interest
Several experts have found the subtle changes of topics of interest progressing from one age group to the next (Pennebaker & Stone, 2003; Schler, Koppel, Argamon, & Pennebaker, 2006; Giarla, 2019 ). Across one’s lifespan, the predominant topics change following the progression of school, college, work, and family, topics relating to relationships (e.g ‘brother’, ‘sister’, ‘daughter’, ‘son’, ‘mother’, ‘father’,...) continuously increase (ibid). Specically, studies of Facebook chat rooms have shown that to the group of 13 to 18 year-olds, school-related topics are shared (e.g. ‘school’, ‘homework’, ‘urr’, ‘exam’); while college-related topics, which includes ‘deadline’, ‘credit’, ‘semester’,... concerns college-age of 19 to 23. Those 23 and older appear to interest in work-related topics like job position/salary topic, ‘at work’, ‘unemployed’; nearly 2 million messages of 45 thousand authors aged 23 to 29 studied concern drunk and beer topics, for example, ‘throw up’, ‘hangover’, ‘wasted’, ‘drinking’, ‘ale’, ‘canned/fresh beer’ (Giarla, 2019; Khoo & Yang, 2020). Other studies of Huffaker and Calvert (2017) on interest and concern topic of generation Y and Z ( 35 and under) indicate that topics of environment, crime, business, health, and some future/current job-related social issues, generation X ( 35 to 55) search and write more about their local town, national politics, science and technology, and education, “baby boomers” (55 and above) are most interested in news related to topics similar to gen X with the addition of daily weather and health.
3. Online behavior
This section will investigate the online behavior of different age groups of users. As age proceeds, one’s depth of research in a certain piece of news increase, among thousands of studied users of Twitter, groups of 18 to 29 shows 59% of interest, 30 to 39, 40 to 59 share similar results of 75% and 77% respectively, those 60 and over take news 89% serious (Quinn, Chen & Mulvenna, 2011). From the data sets of two studied age cohorts on Facebook (younger: aged 15-30, older: aged 50 and above), younger ones have signicantly higher friend number at a ratio of 11 to 1 ( Rangel & Rosso, 2013). Ragel and Rosso also research whether the use of media features differs with age. Analysis of 500 proles of individual Facebook users from two age cohorts demonstrates that a statistical difference does exist between young and older users. Even though the main functions of both groups are commenting and replying, discrepancies are clear in the reply rate of both groups. Older users reply at almost half the rate of the younger ones. User comments and status comments are features that indicate young age, whereas the old user rate in replying to wall comments is over twice the rate of younger users, suggest the function is held in greater regard by older age. Hypermedia and media applications experience a greater degree of usage by people age 50 and above. Similar works and results of Huffaker and Calvert (2017) conrmed these ndings, with 92% of accuracy in applying mentioned age distinct behavior into predicting online user’s age. It is possible to conclude that the behavior of social media users can be differentiated by the age factor.
Conclusion
Studies and analysis covered in this paper aim to investigate how different age groups engage with online media platforms and clarify age as a sociolinguistic variable. Specically, through research and studies on each age group’s traits on preferring words and phrases, online behavior like the usage of hashtags, tweet/status length, and topic of interest. As users age, pronouns and negation are less used, while prepositions and determiners appear more often. Neologisms are clear indication of the youth, while the use of hyperlinks becomes more frequent with age. Content and topic of interest can be anticipated when it comes to a person’s age.
Bibliography
Cheshire, J 2005, ‘Age and Generation-specic use of language’, Sociolinguistics/Soziolinguistik: An International Handbook of the Science of Language and Society, vol. 2, pp.760-780, viewed on 15 September 2021,
Divita, D 2012, 'Online in later life: Age as a chronological fact and a dynamic social category in an Internet class for retirees', Journal of Sociolinguistics, vol. 16, no. 5, pp. 585-612, viewed on 15th September 2021.
Eckert, P, 1997, 'Chapter: 9, Age as a Sociolinguistic Variable ', The Handbook of Sociolinguistics, pp. 151-167, viewed on 15 September 2021.
Giarla, V 2019, 'Generational Social Media: How Social Media Inuences The Online And In-Person Relationships Of Gen X, Gen Y And Gen Z', Honors Theses, article 221 , viewed on 15 September 2021,
Huffaker, D & Calvert, S, 2017, ‘Gender, identity, and language use in teenage blogs’. Journal of Computer-Mediated Communication, vol.10, iss. 2, pp. 00 - 00, viewed on 15th September 2021.
Khoo, S.A.S & Yang, H 2020, 'Social media use improves executive functions in middle-aged e functions in middle-aged and older adults: A structural equation modeling analysis’, Computers in Human Behavior, Vol.111, p.106388, viewed on 15th September 2021.
Nguyen, D, Gravel, R, Trieschnigg, D & Meder, T, 2021. ‘How old do you think I am?: A study of language and age in Twitter’, Seventh International AAAI Conference on Weblogs and Social Media, vol. 7, no. 1, pp. 439-448, viewed on 15 September 2021.
Peersman, C, Daelemans, W, Vandekerckhove, R, Vandekerckhove, B & Leona Van Vaerenbergh, 2016, ‘The Effects of Age, Gender and Region on Non-standard Linguistic Variation in Online Social Networks’, Honors Theses, article 222, Viewed on 12th September 2021,
Pennebaker, W & Stone, D, 2003, 'Words of wisdom: Language use over the life span', Journal of Personality and Social Psychology, vol. 85, no. 2, pp. 291–301. Viewed on 12th September 2021,
Pennebaker, W & Stone, D, 1999, 'Linguistic styles: language use as an individual difference', Journal of Personality and Social Psychology, vol. 77, no. 6, pp. 1296-1312. Viewed on 12th September 2021,
Quinn, D, Chen, L & Mulvenna, M, 2011, ‘ Does Age Make a Difference in the Behaviour of Online Social Network Users?’, IEEE International Conferences on Internet of Things, and Cyber, Physical and Social Computing, p.266-272, viewed on 15th September 2020
Rangel, F & Rosso, P 2013, 'Use of Language and Author Proling: Identication of Gender and Age', Natural Language Processing and Cognitive Science, p. 177, viewed on 15th September 2021,
Rosenthal, S., & McKeown, K. 2011, 'Age prediction in blogs: a study of style, content, and online behavior in pre and post-social media generations', the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, pp. 763-772, viewed on 15th September 2021,
Sap, M, Park, G, Eichstaedt, C, Kern, L & Schwartz, A, 2014, ‘Developing Age and Gender Predictive Lexica over Social Media’, 2014 Conference on Empirical Methods in Natural Language Processing, pp. 1146-1151, viewed on 15th September 2021,
Schler, J, Koppel, M, Argamon, S & Pennebaker, J 2006, 'Effects of Age and Gender on Blogging', AAAI Spring Symposium: Computational Approaches to Analyzing Weblogs, pp. 199-205, viewed on 15th September 2021,
Schwartz, A, Eichstaedt, C, Kern, L, Dziurzynski, L, Ramones, M, et al. 2013, 'Personality, Gender, and Age in the Language of Social Media: The OpenVocabulary Approach', PLoS ONE, vol. 8, iss. 9. Viewed on 13th September 2021,
Seargeant, P & Tagg, C 2011, 'English on the internet and a ‘post-varieties’ approach to language', World English, vol. 30, no. 4, pp. 496–514. Viewed on 12th September 2021,
Wagner, S.E. 2012, 'Age Grading in Sociolinguistic Theory', Language and Linguistics Compass, vol. 6, Iss. 6, pp. 371-382. Viewed on 12th September 2021,