Surveillance capitalism and privacy. Knowledge and attitudes on surveillance capitalism and online institutional privacy protection practices among adolescents in Poland

. the purpose of the study was to determine the level of knowledge and attitudes towards surveillance capitalism and online institutional privacy protection practices among adolescents in Poland (aged 18–19), as well as to determine the relationships between these variables. Surveillance capitalism has emerged as a result of internet users’ activities and involves the collection of all data about these users by different entities for speciic beneits without letting them know about it. The dominant role in surveillance capitalism is played by hi-tech corporations. the aim of the study was to verify whether knowledge, and what kind of knowledge, on surveillance capitalism translates into practices related to the protection of online institutional privacy . the study was conducted on a sample of 177 adolescents in Poland. the main part of the questionnaire consisted of two scales: the scale of knowledge and attitudes on surveillance capitalism, and the scale of online institutional privacy protection practices. the results of the study, calculated by statistical methods, showed that although the majority of respondents had average knowledge and attitudes about surveillance capitalism, which may result from insuficient knowledge of the subject matter, this participation in specialized activities/workshops inluences the level of intensiication of online institutional privacy protection practices.


Introduction
In the digital era, digital data is a valuable resource. Companies collect, store, analyze and commodify it, oten without the approval of those from which the data is collected. he phenomenon is known as "traceability" (Latour 2007) and indicate that each of our online moves is tracked and enters a database to be analyzed in the

Surveillance capitalism in the age of Big Data
Modern technological infrastructure, from cookies through devices, online and mobile services to search engines and platforms, were designed to generate digital data that can be read algorithmically (Urichio 2011). As a consequence, the availability of a large amount of digital data combined with the improved computational technology of their processing (data mining), largely derived from research on artiicial intelligence, has enabled the efective dataication of social, cultural and political life on a wide scale and started a socio-cultural phenomenon called Big Data. hanks to various applications installed in smartphones, companies know how long we sleep, what activities we undertake during the day, how much time we devote to commuting to work, school and home, what we like, how we spend our free time, what interests we have, etc. Big Data provides a lot of diferent kinds of data that can be combined, compared and correlated in any way. At the same time this is both a signiicant value and weakness of Big Data. Since the majority of our professional and private activities are carried out using various digital devices, Big Data covers a broad range of our everyday activities. he more frequently we use digital devices, the more identiiable we become.
As Shoshana Zubof notes, Big Data is now the basis of a new logic of capital accumulation, which she calls "surveillance capitalism". Surveillance capitalism "determines what is measured, and what is passed over; how resources and people are allocated and organized; who is valued in what roles; what activities are undertaken -and for what purpose. he logic of accumulation produces its own social relations and with that its conceptions and uses of authority and power" (Zubof 2015, p. 77). Surveillance capitalism is a "new form of information capitalism [that] aims to predict and modify human behavior as a means to produce revenue and market control" (Zubof 2015, p. 75). As Zubof notes, surveillance capitalists quickly realized that the most valuable data is generated when consumers are pushed or forced into speciic behaviors that bring proit to companies and therefore their automated machine processes not only follow our behavior but also shape it (Zubof 2019, p. 18).
In case of surveillance capitalism, digital data is collected to not speciically deined goals that are unknown to a consumer (Lyon, Bauman 2013;van Dijck 2014, p. 205). As Debatin et al. notices what is visible (and what he is conscious) is only a small part (1/8) of the activities and network interactions. he whole rest (7/8) makes the sphere invisible, hidden, which consists of "data that trickle down from the interactions and self-descriptions of the users in the visible part. To maintain the separation (and the user's motivation to provide and constantly update his or her personal data), any marketing and advertising based on these data must be unobtrusive and subcutaneous" (Debatin et al. 2009, p. 88). he value of the data is not deined by their acquisition itself, but by the potential use in the future and the possibility of their unlimited use (Mayer-Schonberger, Cukier 2013). Moreover, data monitoring far exceeds any Pobrane z czasopisma Mediatizations Studies http://mediatization.umcs.pl Data: 14/09/2023 03:58:41 U M C S watching of individuals, because it penetrates deeply into social issues (Andrejevic 2012, p. 86) in order to predict complex phenomena such as consumers' decisions (Nissenbaum 2010, pp. 42-44;Citron, Pasquale 2014).
Surveillance capitalism is part of the struggles of new media corporations looking to achieve hegemony in the communication networks, and is a basic business model of not only such technology giants as Google or Facebook, but also of various start-ups ofering services and online applications. his hegemony is based on the asymmetry of relationship and information (Ramesh, Fish 2017;Nissenbaum 2010;Halavais 2009;Andrejevic 2014). Mark Andrejevic (2014) calls this asymmetry of the relation between those who collect, store and exploit large amounts of data, and those from whom this data is sourced, the big data divide. Each of these two groups has diferent options on the access and use of the data, which also determines the asymmetry of the relation (e.g. social sorters versus social "sortees", and forecasters versus those subordinating to the forecast). Moreover, Andrejevic notes that "the big data paradigm challenges the empowering promise of the Internet by proposing the superiority of post-explanatory pragmatics (available only to the few) to the forms of comprehension that digital media were supposed to make more accessible to the many" (Andrejevic 2014(Andrejevic , p. 1675. danah boyd and Kate Crawford (2011) speak in a similar way, and in their opinion the present Big Data ecosystem creates a new kind of digital divide between the Big Data rich and the Big Data poor (boyd, Crawford 2011). In addition, David Lyon raises: "the overall efects of consumer surveillance, especially through all kinds of internet use, are not only to cream of those contented consumers and promise further rewards and beneits, but also to cut of those who do not conform to expectations" (Lyon, Bauman 2013, p. 104-105). A new form of digital exclusion is revealed based on the interests of those technological corporations which ofer their services and products "for free". Users are aware of it, so they partly relinquish their privacy and agree to various forms of surveillance of their data, which simultaneously separates them from the threat of digital exclusion.
he asymmetry of information relies on controlling: • all information about the user's activities on the Internet, social networking sites or mobile devices, • information about the purposes and means of collecting, storing, analyzing and using the data and information that are collected about users, • available information, making its selection dependent on intelligent algorithms analyzing user behaviors and preferences. Surveillance capitalism should also be seen as a new form of "data colonialism" normalizing the exploitation of human beings through data and "paving the way for a new stage of capitalism whose outlines we only glimpse: the capitalization of life without limit" (Couldry, Mejias 2018). In a situation where our behavior as consumers of the digital products and services ofered by digital companies becomes what drives Pobrane z czasopisma Mediatizations Studies http://mediatization.umcs.pl Data: 14/09/2023 03:58:41 U M C S surveillance capitalism and provides proits only to these companies (Zubof 2019, p. 11), this phenomenon should be carefully analyzed in the context of institutional privacy and the protection practices against it.

Privacy in surveillance capitalism
In recent years, privacy in the digital era and in surveillance capitalism has been one of the most important issues. Surveillance of a user's online activity is very commonly identiied as a restriction or interference with the right to privacy (UNESCO 2015). Privacy is important not only because it is an inalienable human right to protect one's identity, but also because it is a part of the structure of social life (Blank, Bolsover, Dubois 2014, p. 4). According to Nissenbaum: "Activities online, mediated by the Net, are deeply integrated into social life: they may be continuous with brick-and-mortar correlates or, at the very least, have the power to afect communications, transactions, interactions, and activities in those realms (and vice versa)" (Nissenbaum 2011, p. 38).
Academics in previous studies focused more on social privacy protection practices (controlling access to personal information on social media platforms for the other users) rather than institutional privacy protection practices (controlling how digital companies and their partners might use personal and digital data) (Raynes-Goldie 2010). Many researchers have noted that users are concerned about their privacy on the web, but they are practically unable to take any action to protect it. his variance between attitudes and behavior was deined as the "privacy paradox" (Barnes 2006;Acquisti, Gross 2006;Taddicken 2014). Blank, Bolsover, and Dubois (2014, p. 15-16), however, point out that this was indeed oten the case at the beginning of SNS development, but now young users aged 14-24 are not apathetic toward online privacy. In turn, Taddicken (2014, p. 266) shows that there is no signiicant relationship between online privacy concerns on self-disclosure behavior, wherein she did not distinguish between public self-disclosure and self-disclosure behavior with clearly deined communities where users feel safe from privacy invasion. However, Trepte et al. (2014) argues that disparities between attitudes and behaviors might not indicate paradoxical and inconsistent behavior, but address diferent dimensions of privacy. His research shows that attitudes are not easy to use as predictors of privacy behavior, because there is more inluence for this behavior due to negative online experiences and internet skills 1 . Similar conclusions follow from the research of Büchi et al. conducted among Swiss university students, where it was observed that "attitudes are not the primary explanation for users' varying levels of privacy protection", but online privacy breaches and skills. Participants who have experienced privacy violation online will engage in more self-protective privacy U M C S behaviors (Büchi, Just, Latzer 2016, p. 16). It seems, however, that the situation in which there is a breach of privacy due to the action of a real person is more frequent and easier to observe than when users experience a breach of privacy due to the use of their data without their knowledge and approval, such as for sale to other entities or for proiling. Scandals such as that of Cambridge Analytica in 2018 occur extremely rarely, and, as demonstrated by the practice, did not translate into a massive outlow of Facebook users. What's more, proiling, and especially algorithmic personalization of content, can be perceived by users as beneicial, especially by those who do not have the motivation or the skills to personalize the content to their own preferences (Sundar, Marathe 2010). A factor that seems to play a signiicant role here involve the social beneits that users derive from disclosing information about themselves (Trepte et al. 2014, p. 8) which is rewarding for the user's brain (Tamir, Mitchell 2012).
Due to the debatable issue of the existence of the "privacy paradox" and the complexity and multi-dimensionality of online privacy behaviors, alternative proposals by researchers explain the discrepancy between privacy attitudes and real behavior as "privacy cynicism" (Hofmann, Lutz, Ranzini 2016) or the "privacy dilemma" (Matzner et al. 2016).
he distinction for social privacy and institutional privacy seems to be crucial to explain how users more likely to undertake protection practices against other users than against invisible and abstract web actors. According to the Eurobarometer e-Privacy survey carried out among the population of the European Union aged 15-24, only slightly negative attitudes to tracking online activities prevail. For example, only 54% respondents think that their online activities monitored in exchange for unrestricted access to a certain website is not acceptable and 58% say that is unacceptable for companies to share information about them without their approval, even if it helps these companies to provide new services they might like. What is interesting is that a signiicant number of respondents allow such activities, with 45% stating that having their online activities monitored in exchange for unrestricted access to a certain website is acceptable, while 42% think that is acceptable for companies to share information about them without their approval, even if it helps these companies to provide new services they might like (European Union, DG COMM 2016, p. 57-59). From the research by Young and Quan-Haase, students who use Facebook showed little or no concern about threats from Facebook or third party data companies, but they were concerned about how other people may use personal data and information (Young, Quan-Haase 2013, p. 493-494). Users of Tinder were more concerned about institutional privacy then social privacy (Lutz, Ranzini 2017).
Importantly, online privacy researches indicate that there is a signiicant link between internet and digital skills (boyd, Hargittai 2010; Litt 2013), knowledge (Matzner et al 2016, pp. 282-285, Nissenbaum 2011, and the intensity and the type of undertaken privacy practices. Some media scholars observe that the key to protecting our data is digital literacy. For example, Kennedy et  user' of digital media education is becoming more and more important nowadays (Kennedy, Poell, van Dijck 2015), which means that awareness of how much users know about data streams and their abilities to control them is really important to manage privacy. his is conirmed by the studies (Büchi et al.) showing that internet skills are an important complementary predictor of self-protective privacy behavior (Büchi, Just, Latzer 2016, p. 15-16). As Matzner et al. concludes, protection of social privacy requires diferent practices and activities then institutional privacy (Matzner et al. 2016).

Methodology and hypotheses
he study tests two research hypotheses that concern two factors (attitude and skills), as discussed above, and how they inluence the privacy behaviors: 1. Users with extensive knowledge and a negative attitude towards surveillance capitalism undertake online institutional privacy protection practices (IPPP). 2. Users participating in classes or workshops on internet safety and privacy (ISP) are much more likely to undertake online institutional privacy protection practices (IPPP).
In order to study how young Poles refer to surveillance capitalism, and whether they undertake related online institutional privacy protection strategies, an online questionnaire was created. he study was conducted in the period December 2017 -March 2018 among high school students and technical schools students aged 18-19 (N=177). Due to the selection of the sample, it was not representative.
he selection of the research group resulted from several factors. First of all, the 18-19 year-olds belonged to groups that use the Internet (54% daily or almost daily) and social media (40% daily or almost daily) most actively. Secondly, the 18-19 yearolds were in a formal stage of education.
he questionnaire consisted of 4 parts: 1) A statistical part/speciications (e.g. questions about age, gender, type of school, participation or non-participation in workshops on internet safety/digital literacy); 2) 5 questions regarding the use of digital devices and applications; 3) 33 statements verifying knowledge and attitude towards surveillance capitalism (KASC), e.g. "he websites I use have the right to collect private information about me", "Internet search engines, e.g. Google, Bing, Yahoo! Baidu records and stores the search terms", "Facebook is able to predict my emotions", "Based on online activity, someone can predict sexual orientation", "On the basis of online activity, companies create consumer proiles to which they direct personalized advertising", "If it is possible to sell private digital data in the future, I will be happy to use it", "In the information society, data collected by technology companies can only be used to improve the devices or applications we use"; 4) 30 statements verifying online institutional privacy protection practices (IPPP), e.g. "In an information society, data collected by technology companies can only be used to improve the devices or applications we use", "When I do not use the navigation or map function, I disable geolocation on a mobile device or social networking site", "I share my data on social networks with the number of steps or kilometers traveled and heart rate levels", "I use various search engines to protect my digital identity", "By logging into new websites using social media accounts, I check what permissions the page requests". In the case of conclusions verifying knowledge and attitude, the respondent had to indicate answers on a 5-point Likert scale from "strongly disagree" to "strongly agree". In the case of conclusions verifying practices, the respondent had to indicate answers on a 5-point Likert scale from "never" to "all the time".
Because the essence of surveillance capitalism is the collection and use of the users' data by technological corporations in order to achieve their business goals, parts 3 and 4 include in the questionnaire issues concerning: -Understanding how data is collected, -Attitudes towards data collecting, -Understanding the efects of surveillance capitalism, -Application and privacy policy, -Protecting against surveillance capitalism. Participation by the respondents was voluntary and everyone consented to the processing of collected data for scientiic purposes. he questionnaire was published online and a link to it was sent out along with an invitation to participate in the survey to teachers, based on the snowball method (subsequent teachers were to invite the next ones).
he study adopted explanatory variables: (1) participation in classes or workshops on ISP (yes/no), and variables explained: (1) knowledge and attitude on surveillance capitalism (KASC) 2 , (2) online institutional privacy protection practices (IPPP) 3 and co-occurring variables: (1) knowledge and attitude on surveillance capitalism, and (2) online institutional privacy protection strategies.
In the statistical analysis of the obtained results, descriptive methods and methods of statistical inference were used. To characterize the average value for quantitative traits, the arithmetic average (M) was calculated, and the standard deviation (SD) was assumed to be the scattering pattern, while the quartile measure (Q1, Mdn, Q3) was U M C S 57 Surveillance capitalism and privacy. knowledge and attitudes on surveillance… used to represent the position of the observation. he analysis of qualitative variables was performed by calculating the number and percentages of the occurrences of each value. he conformity of the distributions of quantitative traits with normal distribution was assessed using the Kolmogorov-Smirnov test (for N> 100) or Shapiro-Wilk (for N <100) and on the basis of analysis of skewness and kurtosis indices as well as standard errors, on the basis of a visual assessment of their histograms, and QQ charts.
Due to the lack of conformity of variable distributions coexisting with the normal distribution to the study of compounds, Spearman's rank correlation was applied. To compare the distributions of quantitative variables in two groups, a Mann-Whitney U test was used (for variables with a non-normal distribution) or a Student's t-test for independent samples (for variables with near-normal distribution). he Levene test was used to determine the equality of variance in the samples. For all analyses, the maximum admissible type I error α = 0.05 was assumed, while p ≤ 0.05 was considered statistically signiicant.

A. Characteristics of the studied group
A total of 177 people, high schools students and users of digital media, took part in the survey. he majority of respondents were 18 year-olds, at around 60%, while the remaining 40% were 19-year olds. Among the respondents, 51% attended a high school, while 49% attended a technical school. Almost two-thirds of the respondents (62%) were men, while over a third (38%) were women. he largest group of respondents involved people living in rural areas (38%).
Among the digital devices that the respondents indicated as the most frequently used, smartphones appeared in irst place (100% of respondents), laptops in second place (70%), and desktop computers in third place (62%). Tablets (20%) were the least frequently used devices.
In the case of a number of applications installed on a smartphone, 40% of respondents indicated the "6-10" range for the application, while 15% indicated the range "11-15" and "above 20". Due to the fact that a tablet is not one of the popular devices used by the respondents, the most common indication for the number of applications installed on this device was the range "0-5" (20%).
he surveyed group was diverse in terms of experience and participation in classes or workshops on internet safety and privacy (ISP) conducted outside school or at school by people unrelated to school. Among the respondents, 51% did not participate in classes/workshops ISP, while 49% conirmed their participation in such activities.
Since almost 50% of the respondents participated in classes or workshops on ISP, it should be assumed that their knowledge about surveillance capitalism is high. Of Pobrane z czasopisma Mediatizations Studies http://mediatization.umcs.pl Data: 14/09/2023 03:58:41 U M C S course, such an assumption has some limitations. Firstly because it is not known which issues were discussed during the classes/workshops, and secondly in what form they actually took place. hese two factors may have a signiicant impact on the way knowledge is acquired.

Knowledge and attitude on surveillance capitalism (KASC)
he lowest level of KASC of the respondents was 62 points, while the highest one was 154 points. A result on this scale in the range of 132-165 points means great knowledge and a negative attitude towards surveillance capitalism (Tab. 1). In "knowledge and attitudes" scale only 0.56% of respondents (one person) obtained a result showing very low knowledge and an extremely positive attitude toward surveillance capitalism, while about 40% of respondents (N = 71) obtained a score between 132-165 points, signifying a negative and deinitely negative attitude towards surveillance capitalism as well as a very large and extensive knowledge on this subject. he highest number of respondents, 59.32% (N = 105), gave a result between the extreme scale of values (M=126.85, SD= 15.15).
he analysis of individual answers, however, yields interesting results. Although the vast majority of respondents fairly and fully agree that the internet should not be controlled either by governments (66.10%) or corporations (68.36%), a large number of them having no opinion on these issues (about governments -17.51%, about corporations -20.34%). he same applies to the online privacy at stake. But, almost 50% respondents fairly and totally agree that data collected by tech companies can be used to improve the devices or application, and 30.51% have no opinion on this issue. A large group of respondents (41.78%) fairly and completely agree that the websites have a right to collect private information about them, but a similar group (38.42%) does not agree. A total of 25% likes the fact that websites suggest products that may be of interest. Some of them declared forgoing from online privacy in exchange for free access to digital content (16.38%), but the large group has no opinion on this issue (22.6%). However, the vast majority of respondents would not sell private digital data in the future (55.37%) or would not exchange it for cheaper or free services (54.24%). Knowledge and attitudes towards tracking and collecting user data are much more diverse than one might think, given the fact, that most adolescents strongly agree, and agree that each online activity or on a mobile device leaves a permanent footprint (87.5% of respondents).
When it comes to knowledge about the use of data by entities that collect and analyze them, that big group of adolescents strongly agree and agree, that companies collecting digital data can use programs to predict behaviors of online users (72%), but almost 21.5% adolescents neither agreed nor disagreed. he same number of survey participants strongly agreed, and agreed that mobile devices should collect data to determine the frequency, time, place and period of use by a user (72%), but 10% of adolescents had no opinion about this issue. 56% of the respondents thought that search engine results were complete and objective, although 25% neither agreed nor disagreed. In the case of knowledge by adolescents about the potential for using data for predicting personal trails or preferences, 38.41% strongly disagreed, and disagreed that Facebook was able to predict emotions, but a large group thought diferently (31.64%), while 29.94% neither agreed nor disagreed. Signiicantly difering points of view were presented by respondents in terms of predicting preferences based on online activity. Here, 76.84% of the respondents strongly agreed, and agreed that it was possible. Over half of the respondents also agreed that the following could be predicted based on digital data: conscientiousness and openness (53%), gender (67%), sexual orientation (67%), economic status (59%), and political preferences (70%). Still, a large number of respondents had no opinion about the possibility of predicting personal characteristics based on online activity, i.e. conscientiousness and openness (20%), gender (16%), sexual orientation (21%), economic status (25%), and political preferences (20%).
Due to the large number of responses indicating a lack of knowledge and opinions, people participating in extracurricular activities or workshops on online security and privacy conducted by non-school personnel were checked if they difered from those who did not participate in such classes, in terms of the level of knowledge and attitudes about the online collection of data. Table 2 shows the results obtained in the study.
he Mann-Whitney U test analysis did not show statistically signiicant diferences between the groups in the range of the tested variable: Z = 1.64, ni. his means that those who participated in extracurricular activities or workshops conducted by nonschool people on online safety and privacy did not difer from those who did not participate in such classes, in terms of level of knowledge and attitudes about online data collection. he lack of a clear diference between groups may result from the lack of the subject of data protection and practices related to their collection and use during classes about internet safety and privacy, which was not tested in the study. In summary, the knowledge and attitudes of users about surveillance capitalism are diverse. Although the vast majority of adolescents agrees with the statement that their online privacy is at risk and is aware of the fact that any online activity leaves footprints Pobrane z czasopisma Mediatizations Studies http://mediatization.umcs.pl Data: 14/09/2023 03:58:41 U M C S that can be used for diferent purposes, there is quite a large group of young people who do not have an opinion on such an important for their online privacy matters.

B. Online institutional privacy protection practices (IPPP)
he average level obtained by the respondents in the scope of IPPP was M=126.85 points with a standard deviation SD=15.15 point. On the other hand, the lowest level of IPPP among the respondents was 20 points, while the highest was 69 points; the average level obtained by the respondents in the scope of data protection practices was M=41.66 points with a standard deviation SD=9.27 points. he result on this scale in the range of 64-80 points means high and very high level of online privacy protection strategies.   Remark: n-number, r --average weight, R -total weight, Q 1 -1st quantile, Mdn -median, Q 3 -3rd quantile, M -average, SD -standard deviation, Min -lowest value, Max -highest value, Z -result of Mann-Whitney U test, p -signiicance level of Z-test As for the results on the "practice" scale, only 0.56% of the respondents (one person) obtained a result indicating a high degree of information privacy protection practices, 19.77% (N=35) presented a very low or low level; however, as much as 79.66% (N=141) obtained a result showing a moderate degree of privacy practices (between 33-64 points) (M=41.66, SD=9.27).
A signiicant number of respondents without a moderate level of intensiication of practices may result from a lack of knowledge about surveillance capitalism and competence in privacy management (Fig. 1), especially those purely technical, connected for example with the ability to change application settings, as well as lack of tools for obfuscation activity (Fig. 2). In the part of literature review it was indicated that digital competences are an important complementary predictor of self-protective privacy behavior (Litt 2013;Büchi, Just, Latzer 2016, p. 15-16).
Although the vast majority of the adolescents turn of geolocalisation when they do not use it (47%), a large group of them still does not modify default setting ater installation (17.51%) and 18.64% seldom modiies it. A large group also does not verify (12.53%) or seldom veriies (22.6%) the author and the source before uploading it. he respondents do not use or seldom use light obfuscation practices (Fig. 2). A large group of respondents does not use diferent browsers (41%) as well as does not install plug-ins against tracking, e.g. Ghostery (40.68%). here are still respondents who do not delete cookies browser (22.03%) or do it seldomly (25.99). Moreover, a large number of respondents do not use a private browser mode (22.6%) or do it rarely (23.16%). Besides, many users do not log in at all (22.03%) or do so rarely (28.25%) to new websites using a social media account ("linking accounts"). To sum up, adolescents in terms of their privacy protection practices against the use of their data by institutions for various purposes are characterized by large diversity. While some practices are undertaken oten or almost all the time by almost half of the adolescents (e.g. turn of geolocalisation when not used, checking of SNS settings and non-connecting of SNS accounts), others -equally non-engaging and not requiring specialized skills -are not taken at all or are rarely taken by more than half of the respondents (e.g. private mode in browser, use of diferent browsers, installing plug-ins against tracking or deleting cookies in the browser). his shows that privacy practices that can easily protect users from surveillance capitalism are not taken. C. Hypothesis 1 testing. Users with extensive knowledge and a negative attitude towards surveillance capitalism undertake IPPP.
In order to check the relationship between the level of knowledge and attitude on surveillance capitalism (KASC), and the level of online institutional privacy protection practices (IPPP), a rho-Spearman correlation analysis was carried out. Table 4 below shows the correlation coeicients from the analysis.
Correlation analysis did not show any statistically signiicant relationship between the level of KASC and the level of IPPP: rs = 0.10, ni. his means that in the surveyed group, both people with high and low knowledge about surveillance capitalism apply similar practices in the scope of online privacy protection. he level of knowledge and the negative attitude do not afect the online privacy protection practices. herefore, hypothesis 1 was not conirmed. In order to verify whether people participating in-school or out-of-school activities on ISP difered from those who did not participate in such classes, in terms of the level of IPPP, an analysis was carried out using Student's t-test for independent tests. Table 5 shows the results obtained in the study. Remark : n -number; M -average; SD -standard deviation; Min -lowest value; Max -highest value; Q 1 -irst quartile; Mdn -median; Q 3 -third quartile; t -student's t-test result for independent samples; p -signiicance level for the t test he Student's t-test for independent samples showed statistically signiicant differences between the groups in the range of the analyzed variable: t (170,72)=1.98; p <0.050; d=0.30 -this means that people participating in classes or workshops on ISP presented a higher level of IPPP than those who did not participate in such classes, so they used these practices more oten. Conclusions he purpose of the study was to determine the knowledge and attitudes on surveillance capitalism of adolescents in Poland aged of 18-19, and with what frequency they undertake institutional privacy protection practices. he study showed that adolescents present diverse knowledge and attitudes on surveillance capitalism as well as a various levels of online protection practices. However, the majority presented moderate knowledge and a lack of a clear attitude (59.32%), and a large group of adolescents presented a moderate level of protection privacy practices (79.66%). Such a large number of moderate indications on both scales may result from the issue being too abstract, unrelated to their experience as Internet and mobile devices users. Also, the fact that the questions mostly concerned general issues and less frequently referred to speciic social media, could also afect a large percentage of such responses.
he analyses did not show statistically signiicant relationships between the level of knowledge and attitudes of the respondents on surveillance capitalism and the level of their online institutional privacy practices. At this point, however, it should be noted that the majority of respondents (N=105) presented moderate knowledge and attitudes. he lack of a connection between knowledge and attitudes about surveillance capitalism and the practices of privacy protection may be explained by factors that inluence the shaping of practices, as mentioned in the theoretical part. As the researchers point out, one of the most important factors afecting privacy practices is the negative experience of privacy violations (Trepte et al. 2014;Büchi, Just, Latzer 2016). In this study, this relationship was not analyzed; however, a large number of "never" and "seldom" responses to such practices as turning of geolocalization when not used, modiication of default settings ater installation, or veriication of the author and source before uploading may indicate that the respondents do not see the risks associated with the violation of institutional privacy. he analysis of the correlation between participation or non-participation in classes or workshops on internet safety and privacy (ISP), and the level of intensiication of online institutional privacy protection practices (IPPP), has brought interesting results. Statistically signiicant diferences in the level of the IPPP were observed only in relation to the group of respondents participating in classes/workshops on ISP. hey concerned such practices as: veriication of the source and author of the application before downloading, checking what kind of access on a mobile device is requested for the installed application, disabling geolocation on a mobile device or social networking site when not using the navigation or a map function, modifying the supplied privacy settings on a mobile device, or checking the permissions the new page requests when the respondents log in using a social media account. Interestingly, in the case of such strategies as installing plug-ins for search engines that protect against tracking and private data collectors, using a mailbox that encrypts data, deleting cookies or using various search engines to protect digital identity, a similar Pobrane z czasopisma Mediatizations Studies http://mediatization.umcs.pl Data: 14/09/2023 03:58:41 U M C S level of intensiication was observed in both groups. Perhaps the diferences arise from the fact that certain privacy practices are more engaging and require more specialist knowledge (e.g. installing plug-ins in search engines against tracking and private data collectors or using a mailbox that encrypts data) and habitual repetition, which can be deined as habitual privacy protection strategies (e.g. deleting cookies or using various search engines).
Interpreting the diferences described above on online institutional privacy practices, it is worth referring to the concept of the types of privacy expressed by Katie Raynes-Goldie (2010). It distinguishes social privacy -when a user wants to protect their personal data against the misuse by other users, and institutional privacy -when a user wants to protect themself against data collection by companies and institutions. Each of these types requires diferent practices (Matzner et al. 2016). Institutional privacy protection involves the use of more advanced tools, which requires greater technical competence related to, for example, changing the application settings or installing an appropriate program. It also requires more active practices that "serve to build a protected sphere" (Matzner et al. 2016, p. 282).
he result of the study can be used to develop activity programs devoted to online privacy in the context of surveillance capitalism. hey also show that although the majority of respondents had moderate knowledge and attitudes about surveillance capitalism, which may result from insuicient knowledge of the subject matter, the participation in specialized activities/workshops can inluence the level of intensiication of online privacy protection strategies. However, the conclusions presented in the article contain some limitations. he study did not verify what kind of speciic contents were implemented within the framework of the classes and workshops on internet safety and privacy (ISP) in which the respondents declared participation. A detailed analysis of the issues carried out during classes and workshops could additionally show how efective those educational impacts were and what factors inluenced the respondents.