2. Materials and Methods
A representative sample of young people between 17 and 24 years of age in the Autonomous Region of Madrid was used to carry out this study. The Autonomous Community of Madrid is one of the seventeen autonomous regions that comprise the Spanish State, it has an population of 6,779,888 inhabitants (Instituto Nacional de Estadística 2020
) and includes the capital of Spain, Madrid. A two-stage sampling was applied, stratified by clusters, and selection of the primary sampling units (municipalities and census sections) was performed in a proportionate, random way, and the last units (individuals), by random routes and gender/age quotas. The sample base was designed using data corresponding to the population interpolated by age, with the reference date for each gender being 1 January 2019 for the Autonomous Region of Madrid. The sampling error was set at ±4 for overall data (p
= 50; confidence level of 95%).
The sample consisted of 533 respondents (51.2% boys and 48.7% girls; 36.6% were 17-19 years of age, and 63.4% were 20–24 years old). As a very slight deviation was observed between the distribution of the sample by age and gender with regard to what was presented by the statistical universe of young people in the Autonomous Region of Madrid, weights were included in the sample to adjust the results to the population data and maintain its representativeness.
2.2. Variables and Instruments
The sociodemographic variables considered were gender, age (17–19 and 20–24 years), and social class. The latter was developed using two variables: occupation and educational level of the father, assuming that he was the one who contributed the most income, except in those cases in which the father was unemployed, was a pensioner, etc. In those cases, the mother was considered the highest financial contributor.
A questionnaire created by the authors was used to collect information on the use of social networks and platforms. With the aim of learning more about the ways in which different platforms were used, and given the possibility of changing from one social network to another, leaving profiles open that were no longer in use, the young people were asked about an array of activities they are able to carry out on these platforms. The items included were based on a review of the state of the art, previous research and focus groups. By using this data, we have been able to identify the platforms on which young people carry out their activities, how these activities were distributed by platform, and the degree of activity developed on each one. The questions were closed, and the respondents could choose between different options.
The surveys were conducted between 17 June and 4 July 2019, after obtaining a favorable report from our institution’s ethics committee. The survey was implemented through the CAPI system (computer-assisted personal interview), and the duration of the questionnaire was approximately 25 min. The interviews were conducted personally in the homes of the respondents by a market research firm. The explicit consent of the respondent was requested after explaining the type of data that would be collected in the survey and what it would be used for, as well as informing them of its anonymous nature.
2.4. Data Analysis
The data was analysed using the statistical program SPSS, version 26, and the level of statistical validity was for the value of p < 0.05. The Chi-square test (χ2) was used to analyse the relationships (dependence–independence) between nominal variables (or ordinal variables with scant values), and qualitative variables, in which percentage data are described. For explanatory variables with several response categories, once the relationship was verified through χ2, the option of testing by pairs of the equality of column proportions was selected, offered by the SPSS custom table option.
The non-parametric Mann–Whitney U test have been applied to compare differences between two independent samples, when the dependent variable is either ordinal or continuous, but not normally distributed. According to Tomczak and Tomczak
), we use the Rosenthal correlation to measure how big the difference is between two groups (effect size). The Rosenthal correlation is a generic one that simply divides the standardized test statistics, by the square root of the sample size (Rosenthal 1991
). The interpretation various rules of thumb exist. The interpretation rule of thumb from Bartz
(1999, p. 184
) is: 0.00 < 0.20 = very low; 0.20 < 0.40 = low; 0.40 < 0.60 = moderate; 0.60 < 0.80 = strong; and 0.80 < 1.00 = very strong.
In the case of scale or continuous variables (with means), a Welch t-test was used for independent samples when the explanatory variable offered two partitions. When the explanatory variable offered several partitions (ordinal variable), a one-way ANOVA was performed, if the assumptions of normality and homogeneity of variances (homoscedasticity) were met. When these assumptions were not met, a Kruskal–Wallis non-parametric test and, then, a U-Mann–Whitney test (two partitions) were performed, and subsequently a post-hoc test using Dunn’s test with Bonferroni correction was carried out to compare the results ‘two at a time’ regarding the categories of the independent variable.
To facilitate understanding of the results, statistically significant differences were displayed in bold in the tables. Reference was made to these differences in the text only in those cases when they were statistically significant.
3.1. Use of Social Networks
The young people of Madrid have stated that they use social networks intensively, which is clearly shown by the fact that only eight of the 533 respondents have said they do not use social networks (1.5%), yet most (53%) have claimed they use these networks continuously, with higher percentages among women: 57.6% compared to 48.5% for men (χ2 = 4.48, df = 1, p = 0.034). Compared to this majority, 30.6% have stated that they use them several times a day; 10.7% use them a short time each day; and the rest (around 5%) use them less frequently. A U-Mann–Whitney test indicated that the differences by sex are significant for the whole population, but low, U (n1 = 273, n2 = 260) = 39,360, p < 0.016; r = 0.10), with a slightly higher scores of connection among women (Mdn = 7) versus men (Mdn = 6).
Regarding the differences by age group, a U-Mann–Whitney test indicated that the differences are significant but low for the whole population, U (n1 = 195, n2 = 338) = 28,603, p = 0.005, r = −0.12, with a slightly higher scores of connection among the younger age group (Mdn = 7) versus older age group (Mdn = 6). The youngest (17–19 years of age) said they use more these platforms continuously on a daily basis (60.5% compared to 48.4% of young people 20–24 years of age) (χ2 = 7.49, df = 1, p = 0.006).
Taking into account social class, young people at higher social levels said their use of social networks was highly sporadic, less than once a week. The Goodman–Kruskal gamma (γ = 0.015, p = 0.819) and independent-samples Kruskal–Wallis test showed no statistically significant differences, χ2 (4, N = 533) = 3.92, p < 0.417.
3.2. Use of Social Networks by Platform
Young people have not abandoned Facebook which continues to be one of the platforms on which most of the young people surveyed are active (77.5%), and where they carried out the highest number of activities (M = 8.5, SD = 5.18) of those users who carry out some form of activity of the twenty mentioned and mode of 5. The other platform that is the top leader in the highest amount of activity is Instagram (78.7% of users, M = 7.0, SD = 4.34, mode = 5). Thirdly, 74.3% of young people were active on YouTube, with an average number of activities that was lower (M =4.1, SD = 3.31, mode = 1) than that of Facebook and Instagram. Following: Twitter (39.2% of users, M = 4.4, SD = 5.18, mode = 5); Snapchat (16.4% of users, M = 2.3 activities, SD = 1.79, mode = 1); and others (81% of users, M = 7, SD = 4.33, mode = 2).
The most common tendency among young people was to be present on more than one platform. Along these same lines, of those who say they are active on at least one of the five platforms mentioned (Facebook, Twitter, Instagram, YouTube and Snapchat), only 9.9% are active on just one of them; 28% are active on two; 37.6% on three; and 24.5% are active on all of them. If we add the category of ’others’, only 3.7% have activities on a single platform, with the most common situation being that young people participate in at least four platforms (36.1%), considering that the category of “others” can gather several different platforms, or refer to only one. In short, in relation to the most heavily used platforms, it has been observed that 60% of all respondents are active on Facebook and Instagram simultaneously, and half are active on Facebook, Instagram and YouTube (51%).
Frequency of social network use is also related to the preference for different platforms. Those who said they continuously use social networks tend to be more active on Instagram and YouTube compared to those who did not engage in continuous use (60.1% used Instagram compared to 26.2 which didn’t use it, χ2 = 41.16, df = 1, p < 0.001; and 56.4% used YouTube compared to 42.9 which didn’t use it, χ2 = 7.1, df = 1, p = 0.007). In short, looking at it from another perspective, 89.4% of Instagram users were continuously connected to social networks compared to 79.6% on other platforms, 79% on YouTube, 77% on Facebook and 41.7% on Twitter.
Regarding the frequency of use of these activities, 31.8% of those who said they use Facebook carried out more than ten activities on this network; 8.4% on Twitter; 19% on Instagram; and 5.1% on YouTube. On the latter, the maximum number of activities performed was seventeen, while the maximum number of activities carried out on Snapchat was eight.
= 6, SD
= 4.5) were more active on Instagram than men (M
= 5.02, SD
= 5.1, t
(528) = −2.32, p
= 0.021) and on Snapchat (Mwomen
= 0.5, SD
= 1.17; Mmen =
= 1.1; t
(1, 521) = −2.22, p
= 0.027). The oldest group were also more active on Facebook (M
= 7.2, SD
= 5.72) than the youngest group (M =
= 5.78; t
(413) = −3.12, p
= 0.002. However, no significant differences were observed in the degree of activity that occurred on the rest of the social networking platforms, either by gender or age group (p
> 0.05) (see Table 1
The Goodman–Kruskal gamma showed that there was a significant weak positive association between social position and the tendency to participate in Facebook (γ = 0.143, p < 0.005). A Kruskal–Wallis test showed that social class had a modest significant effect on how different activities young people do on Facebook, χ2 (4, N = 533) = 15.85, p < 0.01, Ε2 = 0.044. A post-hoc test using Dunn’s test with Bonferroni correction showed the significant differences between: lower-middle classes (M = 7.8) and upper-middle classes (M = 8.7) (p < 0.01) and upper classes (p < 0.001) (M = 10.9); middle classes (M = 7.8) and upper-middle classes (M = 8.7) (p < 0.005) and upper classes (M = 10.9) (p < 0.001). Thus, upper class youth not only participated to a greater extent on Facebook or Twitter, they also maintained an average amount of activities that was higher on Facebook in comparison to those in the middle and lower-middle classes, and the lower class had less activity compared to the lower-middle, middle, and upper-middle classes.
3.3. Activities Carried Out on Social Network Platforms
Nearly all of the young people (99.3%) said they watched videos and music on social networks; those who watched videos or see photos of friends were 99%; those who talk to friends were 98%, while those who talk to family was 94.4%; and finally, the percentage of those who publish or post personal content on their profile were 95%. Moreover, 92% searched for entertaining content. In contrast, just over half indicated that they played online (53%), published opinions on social or political issues (56%), or posted criticism or complaints on a public profile occasionally (57%) (see Table 2
Facebook and Instagram are the two platforms on which most young people said they carried out the following activities: approximately three-quarters posted content of a personal nature on their profile; around two-thirds viewed videos or see photos of friends and family; a slightly lower percentage talked to friends; and about one-third of those who searched for information about celebrities or followed their profiles did so on both platforms
Although Twitter was used by the majority of participants to post personal content (51.7%), to publish criticism or complaints on a public profile (37%), and to give opinions on social or political issues (36%), this was not the platform where most network users claimed to carry out these activities, which mainly occurred on Facebook (52% and 53.1%, respectively), followed by Instagram (22.1% and 22.8%), and Twitter (18.6% and 15.3%).
Once again, Facebook was the preferred social network for talking to family members and sharing or recommending sites and links to others. It shared the leading position with YouTube in searching for entertaining content. In addition, YouTube was chosen more often for watching videos and music. Instagram was used more frequently to follow actors, singers, sports stars, etc.
On the other hand, those who play online, buy and sell, search for information regarding products and services, leisure, health or series on the networks, indicated they used other specific platforms more often for carrying out these activities.
Regarding gender differences (Table 3
), a higher percentage of young women indicated they preferred Instagram for activities such as posting personal content on their profile, watching videos, or looking at photos of friends and family, talking to friends, searching for information related to fashion and style, or following celebrities. Whereas young men most frequently used Instagram to search for sports content, buy and sell. On the other hand, more girls showed a preference for YouTube in seeking information about health, diet, nutrition, wellness, beauty, fashion and style, and they used Snapchat to post personal content, watch videos, look at photos of friends and family, and follow celebrities. Young men chose to search for information related to beauty, fashion and style on other social networks to a greater degree than women.
In addition, more young people between 20 and 24 years of age said they chose Facebook to carry out many of the activities compared to the youngest, which is a clear indication of their preference for this network. In contrast, the youngest people chose Twitter to publish personal content. As for Instagram, young people preferred this platform to search for sports information, seek audio-visual content, and follow professionals in the young people’s own fields of study. YouTube, however, was chosen most often by the 20–24 year age group to post personal content on their profile. Finally, the youngest people used other networks in higher numbers to talk to friends and family (see Table 4
In conclusion, it is important to note that social class influences the choice of platforms for carrying out some of the activities. It is worth noting that young people from the higher classes chose Facebook more often than those from other socioeconomic levels to publish personal content, search for information related to products, services, and audio-visual content, as well as to look for information about celebrities and follow their profiles. They also used this network to look for data regarding professionals in the young people’s own field of work or study, and to find content related to health, beauty, fashion, style, and entertainment. More than other groups, the lower class chose this platform to look for audio-visual information as well as data related to professionals in the young people’s own field of work or study. In addition, the latter chose YouTube in greater numbers to watch videos and music, and other platforms for buying and selling.
On the other hand, YouTube had a much greater number of users among young people from the upper-middle and middle class who claimed to have published personal content. There was a larger presence of the lower-middle class in watching videos and music, and more people from the lower-middle and middle class tended to look for information related to leisure. Those with middle and low social status chose Instagram to post content on their profile. We found that there was a greater preference among the middle classes for other social networking platforms to watch videos and music, search for information about audio-visual entertainment, celebrities, diet, nutrition, wellness, beauty, fashion, and style. The lower class tended to use them mainly for buying and selling (see Table 5
In general, young people in the Autonomous Region of Madrid appear to be highly intense users of social networks, with 84% accessing these platforms every day, and of these people, many remain connected continuously. The age and gender of the respondents seem to influence their frequency of use, while social class makes almost no difference. Women and younger people (17–19 years of age) were the ones who used social networks on a more continuous basis. This data contrasts with studies carried out by specialized research companies in Spain, which indicated that men were the most intense users. However, it should be noted that the comparison in that study was made with the general population, not just with young people (IAB 2019
In line with previous studies (Alhabash and Ma 2017
; Oh and Syn 2015
; Stanley 2015
), young people in Madrid showed themselves to be multiplatform users, with a quarter of them accessing all of the social networks mentioned: Facebook, Instagram, YouTube, Twitter and Snapchat, with the first three being the most heavily used sites. This contrasts with other countries such as the USA or Taiwan, where Snapchat was in second place among young people after Instagram (Knight-McCord et al. 2016
; Huang and Su 2018
), although it is in line with the data from specialized Spanish research companies (IAB 2019
). Thus, Instagram was the network where users spent the greatest amount of time in the context studied, with nearly 60% of its users being connected continuously. The use of YouTube also produced a greater tendency to be constantly connected.
On the other hand, a relationship between the frequency of social network use and the preference for different platforms has been observed. Those who said they continuously use social networks tended to be more active on Instagram and YouTube compared to those who did not engage in continuous use. In large part, these data are consistent with the work of Alhabash and Ma
), in which they pointed out that Instagram was the platform that stood out for its intensity of use.
Variables related to age, gender, and social class indicated some differences in the use of platforms, as reflected in the study, which mirrors the study carried out by Blank and Lutz
). Unlike these authors, who found no differences in Instagram use with regard to socio-demographic aspects, this platform was used more often by women and the younger group (17–19 years of age) in our context. Women used Snapchat largely, which is in line with the study by Vaterlaus et al.
). They especially used it to maintain contact in their most intimate and personal relationships, although it should be noted that young women preferred Instagram for this purpose. Regarding social class, the upper-middle and upper classes made more diverse and varied use in terms of activities on Facebook and Twitter, which is in agreement with a study carried out in Chile (Correa 2016
). Especially noteworthy is the preference of the upper class for Facebook in searching for different types of information, as well as information related to products and services, while a very high percentage of young people from the lower class chose other platforms to buy and sell.
Young people also make varied use of social media, as shown in previous studies (Alhabash and Ma 2017
; Alrajehi 2016
; Kaya and Bicen 2016
; Oh and Syn 2015
). While Alhabash and Ma
) found greater diversity in the reasons why young college students from Michigan use Instagram and Snapchat, this same diversity of motivation among young people in Madrid was found to be related to using Facebook and Instagram, with very similar levels.
Both platforms were used mainly for social and entertainment purposes. Facebook was used to a greater extent to talk to family members and share entertaining content. YouTube was used mainly for the latter purpose, and for watching videos and music. Instagram was largely used by those who followed famous people, such as actors, singers, sports stars, etc.
Even though Twitter stands out for offering a user profile capable of posting personal content, publishing criticism or complaints on a public profile, and communicating opinions on social or political issues (Boczkowski et al. 2018
), this is not the platform where most network users carried out these activities. For this purpose, young people reported using Facebook the most, which was also their preferred way to talk to family and friends. This may be in line with the study by Valenzuela et al.
), who found that Facebook was perceived to be more effective when protesting on personal networks with stronger links, while Twitter was more oriented toward criticizing people with whom the participants had weaker links. Consequently, this indicates that young people prefer civic participation (defined as a conversation on social and political issues) limited to their personal spaces that are closest to family and friends, which manifests itself via Facebook, as opposed to Twitter, which represents the institutionalized channel of political communication that allows direct contact with people’s representatives (parliamentarians, politicians, political parties, interest groups and institutions) (Campos-Domínguez 2017
On the other hand, those who play online, buy and sell, or seek information related to products and services, leisure, health, or series on the networks, mainly used other specific platforms for these purposes. The findings seem to confirm the conclusions of previous studies (Boczkowski et al. 2018
; Fergie et al. 2016
), which have suggested that user behavior on different platforms could be influenced by the perception of these platforms and their relevance for certain types of content.
Regarding the way that gender, age and social class are capable of influencing the activities carried out on different platforms, our results show that girls use Instagram in more diverse ways, specifically for social and entertainment purposes, with this being their favorite platform. The older group (20–24 years of age) show greater diversity in using Facebook for social and entertainment purposes as well. Moreover, upper class and upper-middle class people use Facebook more frequently, and in more diverse ways (motivations related to socialization, entertainment, professional information, products, and services) than those of the lower classes. In concordance with Blank and Lutz
), this allows us to conclude that sociodemographic characteristics, especially gender, age and social class, play an important role in defining the different ways in which young people use social media, and such influences vary depending on the platform under study, and even on the types of activities being carried out.
On the other hand, the first contribution of this work to the specialised literature is the differentiation of the use of different social networks by young people based on an exhaustive list of activities. In such a way that patterns of consumption and behaviour can be connected with the ability to distinguish consumption from different platforms for different purposes. Secondly, this work has been able to clarify the role played by sociodemographic factors such as gender, age and social class in creating these patterns of use. Thirdly, a comprehensive list of questions on social media actions is presented to the academic community for discussion and potential application on scales for further research.
Among the limitations of this study, we must point out the cross-sectional design of the information gathering process for a specific period of time. This has not allowed us to study changes in the trends of social network use that take place over longer periods of time. Likewise, the use of data exclusively from the Autonomous Region of Madrid might be a limitation that has prevented us from generalizing our results to young people in the rest of the country.
Another aspect to consider relates to the way in which social class has been recorded, using the father’s occupation and level of education as a reference, assuming that he was the main income earner in the household; and using the mother as a reference only if he was unemployed or a pensioner. Although this form of measurement may be questionable today, there is still a profound gender inequality in access to the labour market, in income inequalities, and in jobs, which particularly affects women. In fact, 54.7% of Spanish men indicated that they themselves were the person who contributed the most income to the household, while 27.8% of women said the same (CIS 2021
Furthermore, constant changes in the field of social media result in usage trends that vary from year to year. For future lines of research, we would like to mention the need to carry out longitudinal studies that would allow us to observe the changes in trends and analyze the role of socio-demographic factors in these tendencies.
On the other hand, among the main contributions of this study that bears mentioning is the use of a representative sample from the Autonomous Region of Madrid, as it has allowed us to gain deeper knowledge regarding the patterns of use by young people of the different platforms in this context, and to compare them with other regional and national studies, as well as to clarify the role played by sociodemographic factors such as gender, age and social class in creating these patterns of use.