The internet, the smartphone, and social media have changed the world. The resulting globalization, immediate gratification, and social networking influence many aspects of modernity. Social media’s purpose is to increase social networks through user-based content for social fulfilment and social identity. However, these technological advances also have their psychological and social disadvantages.
Today’s marketing adapts to the message flow and transmission route models of social media, employing the motives of social networking for branded speech. Henri Tajfel and John Turner (1979) proposed the theoretic framework for the social identity theorem (SIT) describing an individual’s self-concept as consisting of personal and of collective identity. The rise of social media, a tool designed to connect people, changes the communication behavior within social groups and that of the individual. Published in Psychology & Marketing, Vincent Dutot’s (2020) study, A social identity perspective of social media’s impact on satisfaction with life, investigates aspects of social media on identity, presenting a real-world laboratory for understanding the mechanics of socialization. A concerned public observes the negative experiences of social media and identifies possible negative effects, but how dark is this “dark side” of social media? Understanding the shadowy findings of social media presents marketing professionals with evidence-supported strategies for sharing branded messages. With descriptive statistics Dutot analyzes his sample for correlations and probabilities.
Focus of the Study:
By examining the constructs of social media addiction, fear of missing out (FoMO), and narcissism – social media’s “dark side” – Dutot measures their intersections with collective self-esteem and individual self-esteem and their effects on collective and individual satisfaction with life (SwL). In theory, these factors would increase a collectives’ self-esteem and an individuals’ self-esteem, their satisfaction with life, SwL, and that of their social groups (Dutot, 2020). Dutot focuses on measuring the connection between SIT and the dark side of social media.
Purpose of the Study:
The three constructs of addiction, FoMO, and narcissism affect collectives and individuals; but how much influence do they pose? Do the constructs affect collective self-esteem react and individual self-esteem similarly? Finally, do these correlations positively influence collective SwL or individual satisfaction? With his study, Dutot wishes to contribute to SIT in the specific context of social media.
Study Group:
The world’s population of adult social media users grows across demographics. Dutot conducts his study through a recruitment campaign consisting of a convenient online sample of (n = 260) of adult French speaking social media users, located within France. He accesses three separate social media platforms (LinkedIn, Instagram, and Facebook), chosen for their broad audiences, including sex and age segmentation within the sample.
To test his hypotheses evaluating the effects of social media on collectives, individuals, and SwL, Dutot pretested his survey with 10 social media professionals, correcting for mechanics and clarity. His analysis of predictors of influence is similar to multiple regression analysis and essential reliability analysis (Doane & Seward, 2024). Dutot uses hypothesis estimation (β coefficients) and multiple regression equations to estimate the effect of several independent variables (the predictors) of addiction, FoMO, and narcissism, on a dependent variable (outcome), SwL, of collective and individual self-esteem.
Research Methods:
Dutot’s study begins with a literature review, defining the characteristics and behaviors of the test population through SIT and social media. Next, he pre-tested his survey with 10 random social media professionals for accuracy and clarity. Through an online quantitative, cross-sectional survey,Dutot’s empirically measures the causal relations among the three disadvantages of social media variables, collective self-esteem, individual self-esteem, and SwL. Presented over five weeks, 260 individuals disclose the effects of the negative constructs of social media on their habits and behaviors. Subjective and self-reporting, the survey evaluates the effects of the dark side of social media on self-esteem and SwL by assigning value to a 5-point Likert survey and binary questionnaire.
Analysis Methods:
Using partial least square-based structural modeling (SEM), a more complex multivariate extension of multiple regression, Dutot analyzes survey responses. Dutot first validates his latent constructs using confirmatory factor analysis (CFA). This requires reliability, reporting composite reliability (CR), Cronbach’s α; convergent validity with the average variance extracted (AVE) ≥ 0.50; and the discriminate validity using Fornell-Larker Criterion ( > inter-construct correlation). These checks ensure the data quality before running any complex tests, aligning with the necessity of checking assumptions like normality and homogeneity of variance for simpler t-test and ANOVA to establish construct validity (Doane & Seward, 2024). Dutot’s model testing for overall fit (goodness-of-fit indices, comparative fit index (CFI) and root mean square error of approximation (RMSEA) go beyond the scope of Doane and Seward’s 2020 textbook.
Outcomes:
Dutot receives mixed results for rejecting his null hypothesis, which assumed that the dark side of social media does not influence SwL, accepting seven of his 10 hypotheses. A significant part of SwL (R2: 38.1%) provides support for the model and hypotheses (Dutot, 2020). When the probability (p-value, P) is < Cronbach’s α, Dutot’s observed relationships are highly unlikely to have occurred by chance, his null hypothesis is rejected, and his alternative hypotheses supporting the relationship. Additionally, one-tailed tests have a confidence level of σ/2. To determine the statistical significance of the beta coefficient (β), a TYPE II error, Dutot uses the t-statistic in his regression analysis. When: , or , or |t| > 1.96, Dutot accepts or rejects the alternative hypothesis due its standard deviation from the mean, determining statistical significance:
Where the standard deviation is less than 1.96 and the hypothesis is accepted, Dutot uses a one-tailed test for the skewed data. Overall, Dutot shows that it is probable that some of the dark side constructs of social media (addiction, FoMO, and narcissism) affect the collective self-esteem, some individual self-esteem, and SwL. Additionally, although the study rejects the hypothesis of individual self-esteem positively influencing SwL, evidence suggests the converse is true; and that individual SwL is actually the driver behind individual self-esteem.
Discussion:
Dutot contribution to SIT and social media evaluates the dark side of this emerging communication channel, in respect of individuals’ SwL, highlighting the intersection of addiction, FoMO, and narcissism with CSES, SSES, and SwL. SIT integrates self-esteem as a key factor for SwL and Dutot’s study validates these theories within social media. By analyzing the data, Dutot identified significant differences among the sexes and age groups and notes that “female and young users are not the only ones to be affected by social media use” (2020).
Implications of the Study:
By fostering genuine relationships and collective identity, brands can leverage social media and improve marketing efforts by extended reach and investing in collective SwL. Dutot’s study highlights the opportunities available to marketing teams creating social media marketing strategies designed to be shared and increase CSES and SSES. Additionally, his research suggests marketing professionals should personalize marketing strategies based on gender and generation.
Personal Reflection:
The strong correlation between narcissism positively influencing collective social-esteem, p < 0.001, is surprising. Dutot shows that this one seemingly dark side of social media positively influenced the collective self-concept. Conversely, it appears that the influence of narcissism on individual self-esteem was insignificant. I expected both the CSES and the SSES would not be significantly positively influenced. Unexpectedly, the assigned text does not cover the SEM analysis methods used by Dutot and Cronbach’s α were outside the scope of Doane and Seward’s text (2024). Separately, I couldn’t discern why some of Dutot’s t-statistics were below 2 SD. Additionally, Dutot’s tables do not align with the course material, making charting more challenging.
Next Steps:
After analysis, creating charts to visualize Dutot’s data would provide a cleaner representation. Then, a simple redundancy repeating the survey and calculating the probabilities will provide new sample means, validating the ANOVA. Using the same research framework but with different survey content could capture the probabilities for different social media constructs’ influence on other psychological factors, or within differently segmented samples. Additionally, it is five years after Dutot’s research and his data connecting social media to SIT. A longitudinal study could establish temporal precedence, proving that changes in the predictor variable (e.g., social identity) proceeds and therefore could potentially cause the outcome (e.g., self-esteem). Furthermore, he could also test how or when the relationship between social media constructs and psychological aspects intersect through experimental or quasi-experimental design. Using variable intervention would move the research past the survey’s limitations and prove a cause-and-effect relationship or causal link.
Conclusion:
Dutot’s (2020) social media study introduces the social science research behind quantifying qualitative data required for effective social media marketing. Representing foundational concepts of statistical analysis, Dutot evaluates the dark side of social media while introducing concepts for an intermediate statistics student. Of Dutot’s 10 hypotheses, only three pertaining to individual self-esteem or SwL failed to show significant influence. It might be that the experiment’s design does not accurately measure variables affecting individual self-esteem and SwL. Dutot’s experimentation reveals that social media’s dark side does influence collective self-esteem and SwL, and presents avenues to explore the effects of social media on social identity and marketing psychology.
References:
- Doane, D. P. & Seward, L. E. (2024). Applied statistics in business and economics. McGraw Hill. New York, NY. https://prod.reader-ui.prod.mheducation.com/epub/sn_358eb/data-uuid-597e228d8dcc4888b202f67f0f359078
- Dutot, V. (2020). A social identity perspective of social media’s impact on satisfaction with life. Psychology & Marketing, 37(6), 759–772. https://doi.org/10.1002/mar.21333
- Katz, E., Blumler, J. G. & Gurevitch, M. (1973, January 1). Use and gratification research. Public Opinion Quarterly, 37(4), pp. 509-523. https://doi.org/10.1086/268109
- Littlejohn, S. W., & Foss, K. A. (2009). Encyclopedia of communication theory. SAGE Publications. https://doi.org/10.4135/9781412959384
- Tajfel, H. & Turner, J. (1979). An integrative theory of intergroup conflict. In Austin & Worchel’s (Eds.), The Psychology of Intergroup Relations. Brooks/Cole.
Leave a Reply