Social Comparison and Kids' Self-Esteem: What the Research Shows

Leon Festinger proposed social comparison theory in 1954 to explain a fundamental human drive: we evaluate our own abilities, opinions, and worth by comparing ourselves to other people. In environments where objective standards are unavailable — and most things children care about deeply, like attractiveness, popularity, and social belonging, have no objective standard — comparison with others becomes the primary measurement tool. What Festinger didn’t know, and couldn’t have, is what happens to that comparison process when children and adolescents are placed inside platforms algorithmically optimized to surface the most engaging, aspirational, and emotionally activating content available — which is overwhelmingly content showing peers and strangers at their most polished, successful, and desirable.

The result is not simply more comparison. It is a systematic bias toward upward comparison with curated targets — a situation with no real precedent in human development research and real, measurable consequences for self-esteem in children who are still forming their sense of who they are.

Key Takeaways

• Festinger’s social comparison theory predicts that upward comparison (comparing to those who seem better off) reduces self-esteem when the gap feels permanent or unchangeable — exactly the condition created by comparing to curated social media profiles.
• Appearance-based comparison is the most consistently harmful type, with Fardouly and Vartanian’s research showing it produces body dissatisfaction independent of the platform.
• Children ages 8–12 are in a developmental phase of particularly intense social comparison, coinciding with the age at which most platform use begins.
• Downward comparison (comparing to those who seem worse off) temporarily raises self-esteem but is associated with decreased empathy and is not a reliable self-esteem protection strategy.
• Secure attachment relationships with parents are the most robustly evidenced protective factor against comparison-driven self-esteem damage.
• Domain-specific self-concept — a child’s sense of competence in a particular area — provides insulation from general comparison effects.

Festinger’s Framework in a Digital Context

Festinger’s original theory made two key distinctions that are directly relevant to social media. First, he distinguished between upward comparison (comparing to people who seem to outperform us) and downward comparison (comparing to people who seem to underperform). Second, he argued that comparison targets tend to be similar — we compare ourselves to people who seem reasonably close to us in relevant dimensions, not to those who are wildly different.

Social media platforms systematically violate this second principle. The people most prominently featured in children’s feeds are not similar peers — they are algorithmically selected because they generate high engagement, which typically means they are unusually attractive, successful, funny, or talented. The comparison is structurally upward and structurally distorted in ways that Festinger’s theory predicts would be harmful.

A 2015 paper by Vogel, Rose, Miller, and Bhattacharjee published in the Journal of Social and Clinical Psychology directly tested this dynamic in adult populations: participants who viewed profiles of highly attractive, successful peers showed significant decreases in self-evaluations of attractiveness and success compared to control conditions. The effect was strongest for participants who were already lower in self-esteem — a finding with direct implications for children, whose self-esteem is still being formed and is therefore more volatile than adults’.

The developmental timing matters enormously. Children in middle childhood (roughly ages 8–12) are in what psychologist Susan Harter has called a period of “social comparison saturation” — they are developmentally oriented toward peers as their primary self-evaluative reference group in a way that is more intense than either early childhood or late adolescence. Placing children in this developmental window into a highly curated social comparison environment is, in terms of self-esteem theory, a worst-case scenario for timing.

Appearance Comparison: The Most Consistent Harm

Of all comparison types studied in the social media and self-esteem literature, appearance comparison has the most consistent and replicable harmful effects. Jasmine Fardouly and Lenny Vartanian’s 2015 review of social comparison and body image, published in Psychology of Women Quarterly, analyzed studies across multiple platforms and social contexts and concluded that appearance-focused social comparison is robustly associated with body dissatisfaction, negative affect, and disordered eating attitudes in women and girls — and that digital environments intensify these effects by dramatically increasing comparison exposure and by enabling comparison to artificially curated images.

The curated nature of social media images is a specific amplifier that older comparison research didn’t account for. Instagram users in their own surveys report that roughly 70% of images they post to the platform have been filtered, edited, or otherwise altered before posting. Children viewing peers’ social media posts are frequently comparing themselves to images that have been substantially modified — a comparison they often know is distorted but that their emotional systems respond to as genuine anyway.

Fardouly’s subsequent research examined the mechanism: it’s not simply viewing attractive images that harms self-esteem, but specifically appearance comparison — the cognitive act of measuring oneself against the image. Passive scrolling that doesn’t activate direct comparison produces weaker effects than the same scrolling with active comparison thinking. This distinction has intervention implications: teaching children to notice when they’re engaging in comparison rather than simply viewing — a metacognitive skill — has some evidence of effectiveness.

The gender differences in appearance comparison effects are real but not absolute. Research consistently finds girls and young women show larger appearance comparison effects than boys, likely reflecting greater cultural emphasis on female appearance. Boys’ social comparison harms tend to cluster more around status, athleticism, and dominance dimensions, with appearance comparison still present but producing smaller average effect sizes.

What Platform Features Amplify Comparison — and Which Reduce It

Not all social media platforms produce equivalent comparison effects. Platform architecture shapes comparison processes in measurable ways.

Platforms with quantified social metrics — visible like counts, follower counts, view counts — provide numerical comparison targets that intensify the process. A study by Burrow and Rainone (2017) in the Journal of Experimental Social Psychology found that visible like counts directly influenced users’ state self-esteem, with posts receiving fewer likes than expected producing self-esteem dips. Children who associate their social worth with measurable social media metrics are particularly vulnerable.

Platform Feature	Comparison Effect	Self-Esteem Impact	Evidence Quality
Visible like/follower counts	Amplifies quantified comparison	Negative — especially for low performers	Moderate (experimental)
Stories/ephemeral content	Lower curating pressure, less comparison	Smaller negative effects	Preliminary
Close friends / private sharing features	Reduces audience size, lowers stakes	Smaller comparison effects	Limited studies
Algorithm-curated feeds (vs. chronological)	Surfaces highest-performing content, maximizes upward comparison	More negative than chronological	Growing evidence
Creator content (professional influencers)	Extreme upward comparison targets	Strongest negative body image effects	Robust
Peer-generated content	Similar-to-self comparison, more realistic	Mixed — depends on peer selection	Moderate

Platforms that rely on algorithm-curated feeds rather than chronological timelines systematically increase the proportion of high-performing, aspirational content users see. This is not a coincidence — it is how these platforms maximize engagement time, which is their commercial objective. The result, from a social comparison standpoint, is that children on algorithmic platforms are systematically exposed to more upward comparison targets than they would encounter in organic, chronological peer content.

Research by Twenge, Joiner, Rogers, and Martin, published in Clinical Psychological Science in 2018, analyzed a nationally representative dataset of over 500,000 adolescents and found that time spent on social media was significantly associated with depressive symptoms — and that the association was stronger for girls than boys, and stronger for platforms with higher image-sharing (Instagram) than those with lower (Twitter/text-based). The mechanism isn’t perfectly established, but social comparison is the most parsimonious explanation for the gender and platform differences.

Downward Comparison: Not the Answer

A reasonable-seeming intervention strategy is to encourage children to notice people they compare favorably to — downward comparison. Feeling better about yourself by recognizing that others are less well-off is a real psychological mechanism, and research confirms it temporarily raises self-esteem. But it has significant costs and limitations.

Downward comparison is associated with decreased empathy and increased in-group/out-group thinking. Children who regulate their self-esteem primarily by finding others to compare down to are developing a fragile self-esteem structure that requires other people to be worse off. This is not a sustainable or socially healthy self-concept strategy. Research by Gibbons and McCoy (1991) in Journal of Personality and Social Psychology showed that chronic downward comparers were less likely to help those they had compared downward to — a direct empathy cost.

Encouraging children to seek downward comparison targets on social media — to find accounts of people struggling, failing, or looking worse — also introduces its own content-quality concerns and doesn’t address the underlying comparison drive.

The protective factors that research actually supports are more structural.

What Protects Children: The Research on Protective Factors

Secure attachment is the most robustly evidenced protective factor. Research by Valkenburg, Patti, and Patti (2021) in Developmental Psychology found that adolescents with secure parental attachment showed significantly smaller self-esteem fluctuations in response to social media feedback than those with insecure attachment. Attachment security doesn’t eliminate comparison effects; it provides a stable self-esteem foundation that is less disrupted by negative comparison outcomes.

What this means practically: the quality of a child’s relationship with parents is more predictive of their social media self-esteem vulnerability than the number of hours they spend on platforms. This doesn’t mean limits are irrelevant — it means that strong, secure parent-child relationships are a more powerful protective investment than usage restriction alone.

Domain-specific self-concept is the second major protective factor. Research by Susan Harter at the University of Denver on self-concept development in children found that children who have strong, clear, positive self-concepts in domains that matter to them are relatively insulated from general comparison effects. A child who identifies as a builder, a coder, a musician, or an athlete has a self-concept anchor that social comparison in other domains can’t easily destabilize. The key is that the domain has to genuinely matter to the child — adult-assigned achievement (good grades) provides less protection than self-chosen mastery.

This is part of why supporting children’s development of genuine competence in areas they care about — rather than generic self-esteem boosting — is what the research recommends. Engineering and making projects that produce concrete, visible results create the kind of competence-based self-concept that comparison to filtered Instagram images doesn’t undermine.

Age-Differentiated Parental Responses

For children ages 8–10: the research supports conversations about curation before platform use begins. Children in this age range can understand the concept that social media shows people’s highlight reels — that they’re seeing the best moments, not typical ones. Use specific examples. Research by Zosuls, Miller, Ruble, and Levin on media literacy interventions suggests that even brief, concrete explanations of how media is constructed reduce children’s automatic comparison responses.

For children ages 11–13: media literacy moves toward the algorithm. Explaining that the platform is showing you these images because they generate engagement — not because they represent what real people actually look like or how their lives actually go — gives children a frame that disrupts automatic comparison. This is distinct from “those images are fake”; it’s explaining the selection mechanism.

For teenagers: research by Nesi and Prinstein (2015) in Journal of Abnormal Child Psychology found that adolescents’ social media comparison effects are moderated by their existing peer status concerns — adolescents already anxious about social standing show stronger comparison effects. For this group, the parental response that has evidence behind it is supporting contexts of offline peer connection that provide actual social information rather than curated content.

What to Watch for Over the Next 3 Months

Watch for comparison-triggered self-deprecating language after social media use: “I’m so ugly,” “everyone else is having more fun,” “I never look like that.” These are comparison statements in the moment. They’re worth noting without amplifying — a matter-of-fact “what made you think of that?” opens a conversation better than either reassurance or dismissal.

Notice whether your child is avoiding activities or situations because they feel they don’t measure up — a real behavioral consequence of comparison-driven self-esteem damage that goes beyond mood and into decision-making. A child who stops wanting to post photos because they don’t look good enough, or who avoids situations where their appearance or performance will be visible, is showing comparison effects that have moved into behavioral avoidance.

Track whether your child has domains of genuine competence and investment outside of social appearance and popularity. If social comparison online is the primary arena where your child measures their worth, the protective factor of domain-specific self-concept is absent. Supporting the development of any area of genuine skill and mastery — not just praising effort — is the most durable response the research supports.

Frequently Asked Questions

At what age does social comparison become harmful?

Social comparison processes are present from middle childhood (roughly age 7–8), when children begin using peer comparison as a primary self-evaluation tool. The most acute period is early adolescence (11–14), when identity formation, peer importance, and social media use typically coincide. But research documents comparison effects on self-esteem in children as young as 8–9 who are using image-sharing platforms — which is why platform age restrictions have developmental rather than arbitrary logic.

Is following celebrities worse than following peers?

Generally yes, for self-esteem. Fardouly and Vartanian’s research and subsequent studies consistently find that comparison to professional influencers and celebrities — who are selected precisely for aspirational qualities and whose images are professionally produced — generates stronger negative self-esteem effects than peer comparison. Following peers has comparison costs, but those targets are at least within a realistic range. Influencer feeds are structurally designed for maximum upward comparison.

Can complimenting my child’s appearance help buffer social media effects?

Appearance-focused compliments from parents are less protective than competence-focused attention. Research by Carol Dweck and others on praise and self-concept suggests that praising appearance (“you look so beautiful”) ties self-worth to an appearance standard, which intensifies rather than reduces comparison sensitivity. Commenting on what your child has done, built, or figured out — competence domains — builds a self-concept less vulnerable to appearance comparison.

Should I look at my child’s social media feed to understand what they’re seeing?

Yes — not to surveil their social activity but to understand the comparison targets they’re encountering. A parent who looks at their 12-year-old’s Instagram explore page for five minutes has much more concrete information for a conversation than one who talks abstractly about “social media.” The specific content — the particular kind of upward comparison your child is encountering — makes the conversation more grounded and more useful.

Does limiting time on social media improve self-esteem?

Research by Hunt and colleagues (2018) in Journal of Social and Clinical Psychology found that limiting social media use to 30 minutes daily for three weeks produced significant reductions in loneliness and depressive symptoms in college students. Equivalent studies in children are limited, but the mechanism (less comparison exposure) is plausible at any age. Time limits are not the only or the most powerful intervention, but they reduce exposure, which reduces comparison frequency, which reduces comparison effects.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

Sources

1. Festinger, L. (1954). “A theory of social comparison processes.” Human Relations, 7(2), 117–140.
2. Fardouly, J., & Vartanian, L. R. (2015). “Negative comparisons about one’s appearance mediate the relationship between Facebook usage and body image concerns.” Body Image, 12, 82–88.
3. Vogel, E. A., Rose, J. P., Miller, L. R., & Bhattacharjee, K. (2014). “Social comparison, social media, and self-evaluation.” Psychology of Popular Media Culture, 3(4), 206–222.
4. Twenge, J. M., Joiner, T. E., Rogers, M. L., & Martin, G. N. (2018). “Increases in depressive symptoms, suicide-related outcomes, and suicide rates among U.S. adolescents after 2010 and links to increased new media screen time.” Clinical Psychological Science, 6(1), 3–17.
5. Nesi, J., & Prinstein, M. J. (2015). “Using social media for social comparison and feedback-seeking: Gender and popularity moderate associations with depressive symptoms.” Journal of Abnormal Child Psychology, 43(8), 1427–1438.
6. Harter, S. (1999). The Construction of the Self: A Developmental Perspective. Guilford Press.
7. Valkenburg, P. M., Patti, M., & Patti, M. (2021). “Social media and adolescent self-esteem.” Developmental Psychology. https://www.apa.org
8. Hunt, M. G., Marx, R., Lipson, C., & Young, J. (2018). “No more FOMO: Limiting social media decreases loneliness and depression.” Journal of Social and Clinical Psychology, 37(10), 751–768.
9. Burrow, A. L., & Rainone, N. (2017). “How many likes did I get?: Purpose moderates links between positive social media feedback and self-esteem.” Journal of Experimental Social Psychology, 69, 232–236.