Association between social media use (Twitter,Instagram,Facebook) and depressive symptoms: Are Twitter users at higher risk?

Abstract

Background:

The purpose of this study was to determine the association between social media dependence and depressive symptoms and also, to characterize the level of dependence. It was a transversal, analytical research.

Subjects and Methods:

The stratified sample was 212 students from a private university that used Facebook, Instagram and/or Twitter. To measure depressive symptoms, Beck Depression Inventory was used, and to measure the dependence to social media, the Social Media Addiction Test was used, adapted from the Internet Addiction Test of Echeburúa. The collected data were subjected for analysis by descriptive statistics where STATA12 was used.

Results:

The results show that there is an association between social media dependence and depressive symptoms (PR [Prevalence Ratio] = 2.87, CI [Confidence Interval] 2.03–4.07). It was also shown that preferring the use of Twitter (PR = 1.84, CI 1.21–2.82) over Instagram (PR = 1.61, CI 1.13–2.28) is associated with depressive symptoms when compared to the use of Facebook.

Conclusion:

Excessive social media use is associated with depressive symptoms in university students, being more prominent in those who prefer the use of Twitter over Facebook and Instagram.

Keywords

Depression social networking addictive behavior social network dependence

Introduction

Depression is a common recurrent disease that has a familial tendency and does not permit the affected person to reach full happiness; it causes a persistent feeling of sadness and loss of interest, as well as disturbance of sleep patterns and appetite (Mayo Clinic Staff, 2017). According to the World Health Organization (WHO), about 350 million people worldwide suffer from depression. In Peru, the prevalence of teenagers who suffer from this disease has increased from 5.4% to 8.3% in the last few years according to the Epidemiologic Study of Mental Health (EEMSMS) which took place in Lima and Callao. It was also found that there was a greater prevalence of depression in females (13.4%), than in males (3.9%). This evidence refutes the thought that teenagers are happier than adults and are now considered an at-risk population for depression (Vargas, Tovar, & Valverde, 2010).

Moreover, social media dependence is an alarming subject not only nationally but also internationally, and has been considered a public health problem in some countries. Social networks have always existed, but Randy Conrads created the first virtual social network in 1995. Nowadays, there are more than 200 virtual social networks, such as Facebook, Twitter and Instagram, which have been chosen to carry out this study (Dulworth, 2008). Demographically, teenagers are at more risk to suffer from social media dependence. This is because they are the ones who adapt faster to new technology and are constantly searching for new sensations.

In the past few years, there has been an increasing number of social media sites, Facebook, Twitter and Instagram being the most used. Facebook, according to Ellison and others, is a social network site where users create a profile to interact with people they know in real life or to meet new people over the Internet altogether. It enables the users to post pictures and status updates where they can express their feelings and/or what they are doing at the moment, accumulate and make new friends by sending a friend request and the other user has to accept it, like friends’ pictures and/or statuses, comment on each other’s pages publicly or privately, view each other’s profiles, and join groups based on their common likes and hobbies. Also, people can learn through each other’s profiles what their interests are, and the musical tastes and relationships between significant others or family (Ellison, Steinfield, & Lampe, 2007). On the contrary, Twitter is a micro blogging site whose purpose is to let users update their ‘tweet’ or a 140-character status. Twitter users follow others and are followed, making the tweets of the person you follow appear on your news updates. Unlike Facebook, the user does not have to be accepted as a friend to read other people’s tweets, they can follow a person and not be followed by back them. A certain characteristic that this site possesses is the retweet mechanism, that lets users spread information of their choice that other users have tweeted, so it reaches all the followers of the person who retweeted the tweet (Kwak, Lee, Park, & Moon, 2010). Finally, Instagram is the most recent social media site; users are able to capture and share life moments through their mobile phone only through photos and videos – with or without captions on them. It has the same follow method as Twitter, and followers can comment on the user’s photos and also like them (Hu, Manikonda, & Kambhampati, 2014).

The goal of this study is to determine the association between social media dependence and depressive symptoms. According to several studies that took place in different countries, such as the one done in Mexico, people who are dependent on social media have more difficulty establishing personal relations and as a consequence have a greater tendency of depression and suicide (Berenzon, Lara, & Robles, 2013). Due to an alarming increase of social media abuse and dependence by teenagers and the increase of depressive symptoms, it is fundamental to examine if there is an association between these two variables.

Subjects and methods

Subjects

The type of this study is transversal-analytic and took place in Lima, Perú, in a private university, in three different campuses. Participants were enrolled as students in the careers of Psychology, Architecture and Civil Engineering in their fifth term with a total of 397 students. To calculate sample size, a model-study was used where the authors searched for an association between Internet addiction and psychological problems in 10th graders, resulting in a sample size of 192 people (Evren, Dalbudak, Evren, & Ciftci, 2014). Ten percent of people were added for prevention due to loss of questionnaires or rejection, giving the final sample size of 212 students.

A stratified randomized sample with proportional fixation allocation was used, which selected 47 students from Psychology, 80 from Architecture and 85 from Civil Engineering.

Students who were chosen had to be enrolled in the university and in the fifth term of Psychology, Civil Engineering or Architecture and possess at least one social media (Facebook, Twitter or Instagram). The exclusion criterion was being clinically diagnosed with depression, people who did not want to answer the questionnaire and people who were younger than 18 years old.

The students were surveyed when they entered or exited their classroom. Before handing out the questionnaire, an informed consent was given, which they had to sign in order to complete the questionnaire. After this was taken care of, the self-questionnaire was started in a self-applicator mode. To determine if a questionnaire was valid, this had to be totally completed.

Measures

Addiction of Internet Test adapted by Enrique Echeburúa Questionnaire

The exposure variable was social media dependence on Facebook, Twitter or Instagram, which was detected with this questionnaire. It was modified to determine the dependence to social media only (Echeburúa & De Corral, 2010; Puerta-Cortésa, Carbonell, & Chamarro, 2012). It consisted of eight questions, five or more positive answers were needed in order to establish the presence of social media dependence.

Beck’s Inventory of Depression

The response variable was the presence of depressive symptoms. It was measured with Beck’s Inventory of Depression, which consisted of 21 questions, validated by Jesús Sanz and Caramelo Vázquez in the year 2005 (Melipillán, Cova, Rincón, & Valdivia, 2008; Sanz, García-Vera, Espinosa, Fortún, & Vázquez, 2005). A score of 17 or more was needed in order to establish the presence of depressive symptoms.

Data analysis plan

For data recollection, the results of the questionnaire were entered by double digitation in Microsoft Excel 2010 for Windows, and were then transferred to STATA 12 (STATA Corp, College Station, TX, US) for the result analysis. Percentages and statistic tests such as Shapiro–Wilk were used, in which normality was found in our only numeric variable to imply a median and an interquartile range to evaluate this variable. Also, the chi-square test was used to see if there is association between categorical variables. The crude prevalence range was calculated and adjusted through the simple and multiple regression of Poisson, all of this with a confidence level of 95%, using a p of < .05 as statistically significant.

Results

Two hundred forty students took the questionnaire and the amount of losses was 13. Applying the exclusion criteria, there were a total of 212 usable questionnaires, which are detailed in Figure 1.

Figure 1.

Diagram of sample extraction.

Table 1 shows five important variables. First of them, the sex of the surveyed students and, the median age, which was 20 years old. The careers of Architecture, Psychology and Civil Engineering were analyzed next, with Civil Engineering being the one with the most number of students surveyed, and Psychology with the least number of students surveyed. According to the statistics, Facebook was the most used social media, in second place of preference was Instagram and finally Twitter. The majority of people surveyed were not dependent on social media, but 38.7% of students were. Finally, it is important to address the percentage of people surveyed that showed depressive symptoms. According to the Beck questionnaire, 41% of students in the study showed depressive symptoms.

Table 1.

Socio-educative characteristics of the students surveyed.

Variable	n	%
Gender
Male	116	54.7
Female	96	45.3
Age (years)^a	20	18–35
Career
Architecture	80	37.7
Psychology	47	22.2
Civil Engineering	85	40.1
Priority of social media use
Instagram	37	17.5
Twitter	24	11.3
Facebook	151	71.2
Social media dependence^b
Not dependent	130	61.3
Dependent	82	38.7
Depressive symptoms^c
Not depressive	125	59.0
Depressive	87	41.0

Median and range.

Evaluated with test adapted from Echeburúa.

Evaluated with Becks Depression Inventory.

In the two-variable analysis, the relations between different variables and depressive symptoms or none depressive symptoms can be viewed. According to these results, there is an association between social media dependence and depressive symptoms. Out of the 82 students surveyed that presented social media dependence, 69.5% had depressive symptoms. On the contrary, we can state that two variables: social media dependence and career choice, are associated factors for developing depression, in contrast with variables such as age and sex. In addition, age and gender were also analyzed because several studies have found major association in the male sex than in females (Table 2).

Table 2.

Bivariate analysis of depressive symptoms versus social media dependence.

	Not depressive	Depressive	Total	p ^a

Social media dependence
Not dependent	100 (76.9)	30 (23.1)	130 (100)	<.001
Dependent	25 (30.5)	57(69.5)	82 (100)	<.001
Gender
Male	63 (54.3)	53 (45.7)	116 (100)	.130
Female	62 (64.6)	34 (35.4)	96 (100)	.130
Career
Psychology	30 (63.8)	17 (36.2)	47 (100)	.002
Architecture	57 (71.3)	23 (28.8)	80 (100)
Civil Engineering	38 (44.7)	47 (55.3)	85 (100)
Age
⩽20	69 (54.3)	58 (45.7)	127 (100)	.094
>20	56 (65.9)	29 (34.1)	85 (100)	.094

Chi-square test was used.

In the multivariable analysis, the association between developing depressive symptoms and the level of social media dependence was maintained. The career choice of Civil Engineering is highly associated with depressive symptoms in contrast with Architecture students (Table 3).

Table 3.

Multivariate analysis: Development of depressive symptoms and social media dependence.

Variable	p ^a	PR	CI95%

Social media dependence	<.001	2.87	2.03–4.07
Career
Civil Engineering	.007	1.66	1.14–2.40
Psychology	.108	1.51	0.91–2.51

Chi-square test was used.

In the analysis between depressive symptoms and type of social media used, it is shown that users with depressive symptoms prefer the use of Twitter over Instagram and Facebook. Furthermore, the use of Instagram is also associated with increased risk of suffering depressive symptoms (Tables 4 and 5).

Table 4.

Bivariate analysis between depressive symptoms and type of social media used.

Type of social media	Not depressive	Depressive	Total	p ^a
Facebook	100 (66.2)	51 (33.8)	151 (100)	.003
Instagram	16 (43.2)	21 (56.8)	37 (100)
Twitter	9 (37.5)	15 (62.5)	24 (100)

Chi-square test was used.

Table 5.

Multivariate analysis: Development of depressive symptoms and type of social media used.

Variable	p	PR	CI95%
Type of social media
Twitter	.005	1.84	1.21–2.82
Instagram	.008	1.61	1.13–2.28

Adjusted by age, sex and career. In the type of social media variable, Facebook is what it’s being compared to.

Discussion

The current study shows there is an association between social media use and depressive symptoms. The outcome obtained matches with various results found in different scientific studies in distinct age groups (Jelenchick, Eickhoff, & Moreno, 2013; Lin et al., 2016; Moreno, Jelenchick, Koff, & Eickhoff, 2012; Nesi & Prinstein, 2015).

A study showed that the more hours high school students spent on social media, the higher the score they obtain in Beck’s scale of depression (Jelenchick et al., 2013). In older adolescents, studies have also been done and it has been found that using social media for longer periods of time a day was significantly associated with depression; in a particular study in which the previous result was found, depression was measured with the Patient-Reported Outcomes Measurement Information System (PROMIS) survey and social media use was measured with different questions such as hours spent a day on social media (Lin et al., 2016). Likewise, a recent study found that the people dependent on social media search for acceptance and a higher ego in these sites leading to depressive symptoms by comparing themselves to others’ popularity (Nesi & Prinstein, 2015). In contrast, an investigation whose purpose was to evaluate the association between social media use and depression in older adolescents where the mean age was 18.9 years did not find evidence of a relationship between social media use and depression, but an important limitation was that the sample size of the study (N = 190) was small and only performed in a single university campus, in comparison to another article mentioned which included 1,787 people, therefore raised more accurate results (Moreno et al., 2012).

Other studies have also included adults in their population, such as the one performed in Mexico City in 2010 where it was observed that people who were dependent on social media had more difficulty establishing interpersonal relations and as a consequence had a higher tendency of depression and suicide. Even though these studies were performed in different countries, with different social, economic and educational realities, the results have been similar (Herrera et al., 2010; Labrague, 2014; Lin et al., 2016; Morrison & Gore, 2010; Pantic et al., 2012; Tandoc, Ferrucci, & Duffy, 2015; Wright et al., 2013). We can then conclude that this is an actual and alarming subject worldwide, and that there is indeed an association between these two variables.

There are sufficient studies to verify the association between excessive social media use and depression, but why may this be? Does dependence on social media cause depression or is it the other way around? It may be that people who have depressive symptoms seek attention through social media, for example, updating statuses on Facebook or writing about how they are feeling on Twitter. They may also be seeking for acceptance on a site like Instagram, where the audience posts photos of themselves on a daily and/or weekly manner.

An important finding of our study demonstrated that people who prioritized the use of Twitter over Facebook or Instagram had a higher tendency of suffering from depressive symptoms (62.5%). In contrast, the prioritized use of Facebook had a lower tendency of suffering from depressive symptoms with a percentage of 33.8%. These results show that being a Twitter user is an associated factor for the development of depressive symptoms and that being a frequent Facebook user is actually a protective factor against depressive symptoms (see Tables 4 and 5).

Twitter is the most prioritized social media site in people with depressive symptoms. The excessive use of this site acts as an associated factor for the development of depressive symptoms, or vice versa. This might be because of the exclusive purpose and use of this social media site. Users can go online and express their feelings in an unlimited amount of tweets. People who have depressive symptoms feel as if they are not heard, but by using Twitter, they feel like someone is listening to them, for example, their followers and different people all over the world not necessarily just real-life friends. Furthermore, the relationship between depressive symptoms and the use of Instagram could be because of two different scenarios. It is known that people with depressive symptoms seek attention and acceptance through sharing photos, especially image acceptance. While sharing photos on Instagram, people from all over the world can see and comment or like them, helping the one who posted the photo of themselves perceive image acceptance or elevate their low self-esteem. Another scenario could be that excessive Instagram use could lead to depression by setting body image standards since Instagram is filled with model-like pictures, leading the user to feel as if they are not good enough or perceive their life as boring since many photos are taken in different beautiful locations all over the world.

On the contrary, Facebook was the least preferred site by social media users with depressive symptoms and can act as a protective factor; this may be because Facebook is more personal, and more real-life friends based whereas Instagram or Twitter are not. People receive more feedback and support from friends than the other sites. A very recent study supports the conclusion that Facebook is not an associated factor for depression symptoms (Simonic, Kuhlman, Vargas, Houchins, & Lopez-Durán, 2014). This study measured Facebook use, the possible depressive symptoms that the adolescent may have developed, and personality domains, such as extroversion or emotional unstableness. The conclusion of this study was that there is not a direct association between Facebook use and depressive symptoms. In addition, this author also shares the idea that Facebook is a protective factor. From the 237 young adults who were studied, the females who most likely showed high neuroticism were found to be frequent Facebook users, and this was associated with lower depressive symptoms. Their research suggests that Facebook use may be protective against depression (at least for females with high levels of neuroticism) and may also be unrelated to depression per se.

Two main limitations should be considered. First, this study does not show causality, it can only demonstrate an association between the two variables studied. Therefore, it is recommendable to take into account longitudinal studies to obtain new information. It is important to show causality in further studies because it will help understand if depressive symptoms are caused by social media use or if excessive social media use is something depressive people seem to rely on. Second, the students surveyed were from only three different majors, which does not represent the entire university.

It is important to take into account the association between social media use and depressive symptoms. Therefore, the authors recommend a substantial screening of depression in social media users such as university students, which could be done by the university staff through a virtual or in-person setting. This setting could be an important opportunity for the prevention and early diagnosis and care of depressive patients and could also prevent the consequences of this disease such as suicidal ideations, suicidal attempts and poor sleeping habits (Wolniczak et al., 2013). A cohort study is also recommended to analyze the causality between these two variables.

Conclusion

Depression is a worldwide health problem with a rapidly increasing prevalence. Social media dependence has also raised its prevalence, being more prominent in teenagers and young adults. This study demonstrated the association between these two variables and also showed that people with depressive symptoms are more likely to prefer the use of Twitter over Instagram and Facebook. Also, it was found that the preferred use of Facebook over other social media sites such as Instagram and Twitter was a protective factor against depressive symptoms.

Footnotes

Acknowledgements

The authors thank all the students who were surveyed and made this study possible.

Ethical approval

All procedures were revised by the cathedra of the course Methodology Scientific Research of the Universidad Peruana de Ciencias Aplicadas.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

Informed consent

Informed consent was obtained from all individual participants in this study.

ORCID iD

Yhuri Carreazo

References

Berenzon

Lara

Robles

(2013). Depression: status of knowledge and the need for public politics and action plans in Mexico. Salud Pública de México, 55(1), 74–80.

Dulworth

(2008). The connect effect (1st ed.). San Francisco, CA: Berrett-Koehler Publishers.

Echeburúa

De Corral

(2010). Addiction to new technologies and social media in young adults: a new challenge. Adicciones, 22(2), 91–96.

Ellison

Steinfield

Lampe

(2007). The benefits of Facebook friends: Social capital and college students’ use of online social network sites. Journal of Computer-Mediated Communication, 12, 1143–1168.

Evren

Dalbudak

Evren

Ciftci

(2014). High risk of Internet addiction and its relationship with lifetime substance use, psychological and behavioral problems among 10th grade adolescents. Psychiatria Danubina, 26, 330–339.

Herrera

Pacheco

Palomar

Zavala

(2010). Facebook addiction and its relationship with low self-esteem, depression and lack of skills. Psicología Iberoamericana, 18(1), 6–18.

Manikonda

Kambhampati

(2014). What we Instagram: A first analysis of Instagram photo content and user types. In Proceedings of the Eighth International AAAI Conference on Weblogs and Social Media. Association for the Advancement of Artificial Intelligence. Retrieved from www.aaai.org/ocs/index.php/ICWSM/ICWSM14/paper/download/8118/8087

Jelenchick

Eickhoff

J. C.

Moreno

M. A.

(2013). ‘Facebook depression?’ Social networking site use and depression in older adolescents. Journal of Adolescent Health, 52, 128–130.

Kwak

Lee

Park

Moon

(2010). What is Twitter, a social network or a news media? Department of Computer Science, Korea Advanced Institute of Science & Technology. Retrieved from https://an.kaist.ac.kr/~haewoon/papers/2010-www-twitter.pdf

10.

Labrague

(2014). Facebook use and adolescents’ emotional states of depression, anxiety, and stress. Health Science Journal, 8, 80–89.

11.

Lin

L. Y.

Sidani

J. E.

Shensa

Radovic

Miller

Colditz

J. B.

(2016). Association between social media use and depression among U.S. young adults. Depression and Anxiety, 33, 323–331.

12.

Mayo Clinic Staff. (2017). Depression (major depressive disorder). In Diseases & conditions. Mayo Clinic. Retrieved from http://www.mayoclinic.org/diseases-conditions/depression/home/ovc-20321449

13.

Melipillán

Cova

Rincón

Valdivia

(2008). Psychometric properties of Beck II’s Inventory of Depression in Chilean adolescents. Terapia Psicológica, 26, 59–69.

14.

Moreno

Jelenchick

Koff

Eickhoff

(2012). Depression and Internet use among older adolescents: An experience sampling approach. Psychology, 3, 743–748.

15.

Morrison

Gore

(2010). The relationship between excessive Internet use and depression: A questionnaire-based study of 1,319 young people and adults. Psychopathology, 43, 121–126.

16.

Nesi

Prinstein

(2015). Using social media for social comparison and feedback-seeking: Gender and popularity moderate associations with depressive symptoms. Journal of Abnormal Child Psychology, 43, 1427–1438.

17.

Pantic

Damjanovic

Todorovic

Topalovic

Bojovic-Jovic

Ristic

Pantic

(2012). Association between online social networking and depression in high school students: Behavioral physiology viewpoint. Psychiatria Danubina, 24, 90–93.

18.

Puerta-Cortésa

Carbonell

Chamarro

(2012). Analysis of the psychometric properties in the Spanish version of the Internet Addiction Test. Trastornos Adictivos, 14, 99–104.

19.

Sanz

García-Vera

M. P.

Espinosa

Fortún

Vázquez

(2005). Spanish adaptation of Beck’s II Depression Inventory (BDI-II):3. Clínica y Salud, 16, 121–142.

20.

Simonic

Kuhlman

Vargas

Houchins

Lopez-Durán

(2014). Facebook use and depressive symptomatology: Investigating the role of neuroticism and extraversion in youth. Computers in Human Behavior, 40, 1–5.

21.

Tandoc

Ferrucci

Duffy

(2015). Facebook use, envy, and depression among college students: Is Facebooking depressing? Computers in Human Behavior, 43, 139–146.

22.

Vargas

Tovar

Valverde

(2010). Prevalence and associated factors of depressive episodes in adolescents of Lima Metropolitana and Callao. Revista Peruana de Epidemiología, 14, 91–98.

23.

Wolniczak

Cáceres-DelAguila

J. A.

Palma-Ardiles

Arroyo

K. J.

Solís-Visscher

Paredes-Yauri

… Bernabe-Ortiz

(2013). Association between Facebook dependence and poor sleep quality: A study in a sample of undergraduate students in Peru. PLoS ONE, 8(3), e59087.

24.

Wright

K. B.

Rosenberg

Egbert

Ploeger

N. A.

Bernard

D. R.

King

(2013). Communication competence, social support, and depression among college students: A model of Facebook and face-to-face support network influence. Journal of Health Communication, 18, 41–57.