Concept Paper

Effect of Social Media Use On Self-Esteem In Adolescent Girls

Andreia Tonia Da Costa

School of Psychology, Capella University

PSY5201: Integrative Project

1
 Effect of Social Media Use On Self-Esteem In Adolescent Girls

Background

Scholars have become more interested in studying the effects of social media on teenage

mental health during the past 20 years, with a particular emphasis on body image and self-

esteem. Peer pressure in real-world settings, traditional media representations, and in-person

encounters have historically had a significant impact on teenagers' psychological development.

Teenagers were exposed to ideas of beauty through periodicals, television, and peer comparison.

However, the teenage experience has seen a significant transition into the digital realm due to the

rapid advancement of technology and the rise of social networking sites like Instagram, TikTok,

and Snapchat. These platforms foster unattainable beauty standards and exacerbate appearance-

based comparisons since they are visually dominant, algorithmically driven, and dependent on

social feedback mechanisms. Particularly susceptible to these comparisons are teenage girls,

who frequently gauge their value by looking at well chosen, filtered photos and receiving likes

and comments from the public (Farooq et al., 2023; Steinsbekk et al., 2021). Because of this, the

dynamics of teenage self-esteem are now closely linked to perceived attractiveness and online

visibility.

Higher social media use has been repeatedly linked to decreased self-esteem, particularly

in girls, according to research. However, correlation rather than causality is the main emphasis

of a large portion of the material now in publication. Although social media use has been linked

in numerous studies to poorer self-image and lower body satisfaction, the underlying

2
mechanisms underpinning this relationship have not been properly investigated. For example,

the mediating role of body image—how one's view of their physical appearance affects the

relationship between social media and self-esteem—has been recognized for a long time but has

rarely been examined in reliable experimental settings. Additionally, cross-sectional self-report

designs, which might limit causal inference and add reporting bias, are a major component of the

majority of current research. Few studies assess digital interventions using rigorous longitudinal

or quasi-experimental methodologies, despite the fact that they have demonstrated some promise

in enhancing body satisfaction and resilience (Gordon et al., 2021; Mahon & Seekis, 2022).

Examples of these interventions include social media literacy programs and time-use restriction

strategies. Furthermore, despite mounting data that suggests private peer interactions, including

group chats or direct messaging on Instagram, may perpetuate body shaming and appearance-

related comparisons, their impact is still poorly understood (Alluhidan et al., 2025).

In light of these drawbacks, this study offers a fresh approach for investigating the ways

in which body image influences the connection between teenage girls' self-esteem and social

media use. Additionally, it aims to determine whether a structured digital media literacy

intervention can effectively guard against low self-esteem and a poor body image. This study

fills important gaps in causation, platform-specific analysis, and the real-world implementation

of interventions by using a quantitative, quasi-experimental methodology. By offering scalable

methods to assist teenage females in navigating digital environments, the results are meant to

influence public health policy, mental health services, and educational curricula. In addition to

addressing actual gaps in the body of literature, this study intends to provide workable solutions

that can be applied in communities and schools to encourage better digital habits and enhanced

mental health among teenagers.

3
 Research Problem

Adolescent girls are especially vulnerable to the detrimental psychological effects of

social media use, particularly with regard to self-esteem, according to the corpus of current

research. Online exposure to unrealistic beauty standards and comparisons based on appearance

are closely linked to this sensitivity. Several research have shown that girls who spend more

time on image-centric platforms are more likely to be dissatisfied with their physical appearance

and have poorer levels of self-esteem (Farooq et al., 2023; Colak et al., 2023). Notwithstanding

the strength of these results, the great majority of current research is correlational in character

and does not thoroughly investigate the ways in which social media affects self-esteem.

Therefore, whether body image serves as a mediating element that aids in explaining this

relationship is yet unknown.

Furthermore, not much research has looked into how interventions, such social media

literacy courses, could buck this trend and encourage better results for self-esteem and body

image. The scant literature on interventions is mostly exploratory or focused on short-term

results, paying little attention to longitudinal changes or platform-specific impacts. This research

gap makes it evident that a focused, empirically supported investigation with a strict quantitative

design is required. Such a study must assess whether structured digital interventions may

successfully enhance psychological outcomes for adolescent girls in addition to investigating the

mediating impact of body image. By providing an empirically supported examination of the

underlying psychological processes and intervention outcomes, this proposed study directly

addresses these shortcomings.

Research Question

4
Primary Research Question:

To what extent does body image mediate the relationship between social media use and self-

esteem in adolescent girls, and can a digital media literacy intervention significantly improve

these outcomes?

Null Hypothesis (H₀):

There is no statistically significant mediating effect of body image on the relationship between

social media usage and self-esteem among adolescent girls, and the digital media literacy

intervention does not significantly improve self-esteem or body image.

Alternative Hypothesis (H₁):

Body image significantly mediates the relationship between social media usage and self-esteem

among adolescent girls, and a digital media literacy intervention significantly improves self-

esteem and body image.

Goals and Objectives

The main purpose of this study is to learn more about how body image affects the

relationship between social media use and self-esteem in teenage females, and to see if a digital

media literacy program can help with these problems. This study is especially noteworthy since

it throws light on a major public health issue: the mental health of teens in the digital age. The

study's goal is to help educational programs, community education, and legislative initiatives that

promote digital well-being by finding effective interventions.

Objectives:

5
To find out how social media use affects the self-esteem of teenage girls in numbers.

To find out what function body image plays in the link between self-esteem and using social

media.

To find out if a structured media literacy program may help people feel better about their bodies

and their self-esteem.

To find out how different social media sites, such Instagram and TikTok, affect body image and

self-esteem in different ways.

To give parents, teachers, and mental health experts recommendations based on the evidence.

Population and Sample

This study's general demographic is teenage girls between the ages of 13 and 19. This is a

time when they are more self-conscious, aware of their bodies, and influenced by their peers.

We chose this age group because studies in developmental psychology shows that adolescence is

a key time for forming an identity and being open to social comparison (Valkenburg et al.,

2021).

The target group will be teenage females who go to public high schools in a mid-sized

city in the United States. The schools chosen have students from a wide range of ethnicities,

socio-economic backgrounds, and access to technology, which makes the results more

generalizable. All of the schools that took part in the research had internet connection, and

almost all of the kids have cellphones and social media accounts, which are both necessary for

the study.

6
 There will be 150 teenage girls in the study, and they will come from five different high

schools. We will use a stratified random sample method to make sure that people from all age

groups, school years, and social and economic backgrounds are represented. A power study for

mediation models with a medium effect size, α = 0.05, and power = 0.80 shows that at least 114

people are needed. The sample size has been raised to 150 to make up for possible dropouts and

incomplete answers. People will only be able to take part if they want to, and both parents and

participants will have to agree. An Institutional Review Board (IRB) will review and approve

the study protocol to make sure it follows ethical standards.

Methodology and Procedures

Quantitative Research Approach

This study uses a quantitative research design, which is the best way to test hypotheses

and look at measurable variables using systematic statistical analysis. Quantitative research is all

about being objective, being able to repeat the results, and being able to trust them. This study

uses a quantitative technique, which uses numbers to find connections between factors, unlike

qualitative research, which looks at how people feel about things. This makes it especially good

for looking into the link between body image, self-esteem, and social media use, as these ideas

can be put into action and measured with reliable psychometric tools.

More specifically, the study uses a quasi-experimental design with a pretest–posttest

control group structure. This is a strong design that is widely employed in behavioural and

educational research when full randomization isn't possible because of ethical or logistical

reasons (Campbell & Stanley, 1963). To keep contamination between conditions to a minimum,

7
participants will be randomly put into either the intervention or control group within their school.

However, the assignment won't be completely blind because of the nature of the intervention.

This design lets you compare groups over time, both within and between them. This gives you

information about how effectively the intervention worked and how strong and directional the

relationships between variables were.

There are a number of reasons why a quasi-experimental design is a good choice. First, it

is more ecologically valid since participants learn in their natural environments (public schools)

instead of in fake labs. Second, this design helps with ethical issues by giving the control group

the intervention after the trial is over, making sure that everyone who took part benefits in the

end. Lastly, using more than one measurement point makes the results more reliable because it

lets the research uncover both short-term and long-term effects of the intervention.

Instrumentation

The study uses three standardized and psychometrically validated tools to measure the

important constructs: social media use, body image, and self-esteem. This is done to make sure

that the methods are sound. We chose each of these tools because they are reliable, valid,

appropriate for teens, and have been used in a lot of previous empirical investigations.

1. Social Media Usage Questionnaire (SMUQ)

The SMUQ is meant to find out both the good and bad things about how teens use social

media. This program keeps track of how often, how long, what platforms people prefer, and what

kinds of engagement they have (such passive scrolling vs. active posting). There are 20 questions

on the questionnaire, and they are all on a 5-point Likert scale that goes from "Never" to

"Always." Some examples are: "I check social media as soon as I wake up" or "I compare how I

8
look to people I follow on social media."

Farooq et al. (2023) checked the scale and found that it had a Cronbach's alpha of 0.89,

which means that it was quite consistent within itself. We made some small changes to the

questionnaire for this study so that it includes questions on popular platforms for teenage girls,

such TikTok, Instagram, and Snapchat. This makes sure that the questions are culturally and

contextually appropriate. Researchers can look at both the amount of use and the kind of

interactions that happen, both of which may have an effect on psychological outcomes.

2. Body Esteem Scale for Adolescents and Adults (BESAA)

The BESAA is a tool that looks at how people see themselves in terms of their physical

appearance from many angles. It has three parts: (1) Appearance, (2) Weight Satisfaction, and

(3) Attribution (how other people see your body). The tool has 23 items, and each one is graded

on a 5-point Likert scale that goes from "Never" to "Always." "I like what I see when I look in

the mirror" and "I am proud of my body" are two of the statements.

Mendelson et al. (1997) first created the BESAA, and it has since been tested and found

to be reliable for teens. Cronbach's alpha values are usually above 0.90. In this study, it is an

important mediating variable that shows how social media affects body image and, in turn, how

it affects self-esteem.

3. Rosenberg Self-Esteem Scale (RSES)

The RSES is a well-known tool for measuring self-esteem around the world. There are 10

items in it, some of which are positive and some of which are negative. For example, "I feel that

I have a number of good qualities" and "I feel I do not have much to be proud of." Responses are

written down on a 4-point Likert scale that goes from "Strongly Agree" to "Strongly Disagree."

This tool has been tested on people of all ages and cultures, including teens, and it has a

9
Cronbach's alpha of 0.85 in teen samples (Colak et al., 2023). The RSES was chosen because it

is short, easy to use, and has a good ability to predict mental health outcomes. In this study, the

dependent variable is the main psychological result that we are interested in.

Data Collection Procedures

The study will gather data at three important times to make sure that the data collection is

systematic and consistent. This will allow for both short-term and medium-term evaluations of

how well the intervention worked:

Time 1: Pretest:

All participants, both in the intervention and control groups, will fill out the SMUQ, BESAA,

and RSES to get baseline measures. This phase is very important to make sure that any variations

between the groups in the beginning are taken into account during the analysis.

Time 2—Posttest (Right After the Intervention):

After the four-week intervention is over, individuals will again fill out all three instruments. This

assessment will show the immediate impact of the intervention and be used to compare groups.

Time 3: Follow-Up (Four Weeks After Posttest):

Four weeks after the intervention, there will be a third round of tests to see if the effects last.

This step is very important for finding out if any improvements in self-esteem and body image

last over time.

School-issued digital tablets will be used to collect data in the classroom, with

experienced facilitators watching over the process. The digital format makes it less likely that

people will make mistakes while entering data and makes it easier to automatically encrypt and

10
store responses safely. To keep the data private, each participant will get a unique anonymous

code. This will make it easier to match data from all three time points. Parents must give written

authorization, and students must agree to participate. A certified Institutional Review Board

(IRB) will give its clearance for the project.

Intervention Design

The intervention group will take part in a four-week media literacy program based on the

SoMe Social Media Literacy Curriculum created by Gordon et al. (2021). This curriculum is

founded on critical media theory and cognitive behavioural principles. Its goals are to help teens

understand how the media may be used to manipulate people, improve their ability to think

critically, and lower the number of social comparisons based on looks.

 Each week, there will be a session that lasts about 60 to 75 minutes.

 Breaking Down Altered Images: Participants will look at digitally changed photographs

and talk about how seeing perfect bodies online affects their mental health.

Peer-Led talks: These are group talks concerning identity, social validation, and the stress

of being online. People will be asked to think about how social media affects how they

feel about themselves.

 Reflective Journaling: Students will fill out guided journal prompts that will help them

think about how they feel about their body image and their internet experiences. This

metacognitive activity helps you remember important ideas.

 Role-Playing and Scenario Analysis: Participants will act out situations where they have

to deal with unfavourable internet comments or peer pressure to look a certain way.

11
The control group won't get the intervention during the active study time, but they will be able to

access it after data collection is done to keep things fair.

Data Analysis Plan

A mix of descriptive, inferential, and multivariate statistical methods will be used to

make sure that the data analysis is thorough and strong. We will use SPSS (v.28) and AMOS (for

structural modelling and mediation testing) to do the analysis.

1. Descriptive Statistics

Demographic factors like age, grade level, and socioeconomic position, as well as pretest and

posttest results on all three instruments, will have basic descriptive statistics (means, standard

deviations, and frequencies) made for them. This will give you a quick look at the sample and

help you find any differences between the groups at the start.

2. Inferential Statistics

To make sure that randomization worked, independent-samples t-tests will look at the pretest

results of the intervention and control groups.

We will utilize ANCOVA (Analysis of Covariance) to compare posttest scores while keeping

baseline values the same. This method makes the statistics more powerful and takes into account

any differences between the groups from the start.

Repeated Measures ANOVA will look at how things changed within each group across the three

time periods: before the test, after the test, and after the follow-up. This will assist figure out if

the effects of the intervention last over time.

3. Analysis of Mediation

The main premise of the research is that body image mediates the link between using social

12
media and self-esteem. Hayes (2017) created the PROCESS macro (Model 4), which is a

common tool for doing mediation, moderation, and conditional process analysis. We will utilize

this tool to do mediation analysis.

Important factors for mediation are:

 The total effect (c path) is the effect of using social media on self-esteem.

 The direct influence (c' path): How using social media affects self-esteem while taking

body image into account.

 The indirect effect (a*b path) is the result of the effect of body image on self-esteem (b

path) and the effect of social media on body image (a path).

 We will utilize bootstrapping (5000 resamples) to figure out the confidence intervals for

indirect effects. If the confidence interval for the indirect effect does not contain 0,

mediation will be supported.

4. Effect Size Estimation

We will publish effect sizes to show how important the results are in real life:

 Cohen's d (for t-tests and ANOVA): 0.2, 0.5, and 0.8 mean minor, medium, and large

effects, respectively.

 Partial eta squared (η²) for ANCOVA and repeated measurements ANOVA shows

how much of the variance is explained by the intervention after taking other factors

into account.

13
Validity, Reliability, and Limitations

Reliability

All of the chosen instruments have shown good internal consistency, with Cronbach's

alpha ratings for adolescent samples ranging from 0.85 to 0.91. The repeated-measures design

makes things more stable over time, which makes the changes that are seen more reliable.

Construct Validity Using tools that are generally regarded and based on theory helps to prove

validity. The control group and pretest–posttest comparisons help improve internal validity.

Using a broad sample from many schools helps External Validity and makes the results more

generalizable.

Limitations

Self-report bias can make answers less accurate, especially when it comes to sensitive subjects

like body image and self-esteem.

 There could be attrition between the posttest and the follow-up, which could affect the

results.

 Uncontrolled variance may come from things like peer pressure outside of the

intervention.

 These problems will be fixed by careful teaching, supervision, and by adding control

variables to statistical models.

Conclusion

This concept paper looks at how social media affects the self-esteem of teenage girls,

with a focus on how body image might operate as a mediator and how digital interventions might

14
help. The research dilemma comes from solid evidence that using social media can impair self-

esteem, especially when people compare themselves to others visually and are unhappy with

their bodies. But we don't know much about how body image works as a mediator or if media

literacy programs can really help with these problems.

The proposed study, based on Objectification Theory and Social Comparison Theory,

will apply a strict quantitative method to look at these connections. The project aims to give

teachers, mental health experts, and policymakers useful information by using stratified sample,

validated measures, and a structured media literacy intervention. In the end, this study could add

to the body of knowledge in the subject of teenage digital mental health, which is developing. It

could help create better support systems for young girls who are dealing with the complicated

social constraints of online life.

References

Alluhidan, A., Park, J. K., Akter, M., Rodgers, R., Razi, A., & Wisniewski, P. J. (2025).

Unfiltered: How teens engage in body image and shaming discussions via Instagram

direct messages (DMs). arXiv preprint arXiv:2504.02176.

http://dx.doi.org/10.48550/arXiv.2504.02176

15
Colak, M., Sireli Bingol, Ö., & Dayi, A. (2023). Self-esteem and social media addiction level in

adolescents: The mediating role of body image. Indian Journal of Psychiatry, 65(5), 595–

600. https://doi.org/10.4103/indianjpsychiatry.indianjpsychiatry_306_22

Farooq, H. O., Farrukh, H., & Khan, Z. (2023). The influence of social media on adolescents’

self-esteem. Qlantic Journal of Social Sciences & Humanities, 4(3), 45–60.

http://dx.doi.org/10.55737/qjssh.413440589

Gordon, C. S., Jarman, H. K., Rodgers, R. F., McLean, S. A., Slater, A., Fuller-Tyszkiewicz, M.,

& Paxton, S. J. (2021). Outcomes of a cluster randomized controlled trial of the SoMe

social media literacy program for improving body image-related outcomes in adolescent

boys and girls. Nutrients, 13(11), 3825. https://doi.org/10.3390/nu13113825

Mahon, C., & Seekis, V. (2022). Digital interventions for adolescent women’s body image: A

systematic review. Frontiers in Global Women’s Health, 3, 832805.

https://doi.org/10.3389/fgwh.2022.832805

Prince, T., Mulgrew, K. E., Driver, C., Mills, L., Loza, J., & Hermens, D. F. (2024). Appearance-

related cyberbullying and its association with the desire to alter physical appearance

among adolescent females. Journal of Eating Disorders, 12(1), 125.

http://dx.doi.org/10.1186/s40337-024-01083-z

Ra Valkenburg, P. M., Pouwels, J. L., Beyens, I., van Driel, I. I., & Keijsers, L. (2021).

Adolescents’ social media experiences and their self-esteem: A person-specific

susceptibility perspective. Technology, Mind, and Behavior, 2(2).

https://doi.org/10.21942/uva.14095971

Steinsbekk, S., Wichstrøm, L., Stenseng, F., Nesi, J., Hygen, B. W., & Skalická, V. (2021). The

impact of social media use on appearance self-esteem from childhood to adolescence–A

16
 3-wave community study. Computers in Human Behavior, 114, 106528.

https://doi.org/10.1016/j.chb.2020.106528

Vincente-Benito, I., & Ramírez-Durán, M. D. V. (2023). Influence of social media use on body

image and well-being among adolescents and young adults: A systematic review. Journal

of psychosocial nursing and mental health services, 61(12), 11-18.

https://doi.org/10.3928/02793695-20230524-02

Voggenreiter, A., Brandt, S., Putterer, F., Frings, A., & Pfeffer, J. (2024, May). The Role of

Likes: How Online Feedback Impacts Users' Mental Health. In Proceedings of the 16th

ACM Web Science Conference (pp. 302-310). https://doi.org/10.1145/3614419.3643995

17