Perception of Information Sensitivity for Internet Users in Saudi Arabia

The rapid evolution of Internet use has led to the collection of big data about users, which has raised users’ privacy concerns about their personal information. This study adopts the hypothesis to evaluate the perception of the information sensitivity of Internet users in Saudi Arabia as research subjects. This study analyzes the sensitivity of 35 types of information through a questionnaire with answers from 508 participants from Saudi Arabia and estimates the cultural influence by comparing Saudi results with users from the USA, Brazil and Germany. Demographic information, such as age, gender, marital status and education, and attitude characteristics, such as disposition to privacy, propensity to take risks, privacy violation experience and trust in institutions, influence Saudi individuals’ perception of sensitivity towards various types of information. The survey results show slight differences in the sensitivity levels between Internet users in Saudi Arabia and their counterparts in other nations. Ultimately, this study contributes to improving the international model of information sensitivity perception between different nations.


Introduction
The Internet has completely changed our lives. It changes people's experience with shopping, learning, communication, social media, lifestyle, healthcare, and so on. In this regard, there has been a rapid increase in Internet users and connected devices, and more people are moving toward dealing with electronic services (Kapoor et al., 2018;Markos et al., 2017). The Internet dependency explosion has led to the creation and collection of large amounts of data called "big data." Big data is important for researchers and businesses in finding new knowledge and information and developing new products and services (Alashoor et al., 2017;van Zoonen, 2016). This big data includes valuable personal information about users. For various reasons, Internet users are forced to disclose their sensitive information, which creates serious security and privacy concerns (Alashoor et al., 2017;Fianu, Ofori, Boateng & Ampong, 2019;Kapoor et al., 2018;Markos et al., 2017;van Zoonen, 2016).
information about users to improve and manage services (Mutimukwe et al., 2019). Meanwhile, social media platforms and device operators collect different types of users' personal, biometric identification, browsing data, location, and daily processes data (Fianu et al., 2019;Hung & Cheng, 2009;Pires, Garcia, Pombo, & Flórez-Revuelta, 2016). That, in turn, has given rise to the concerns of Internet users about the information collected from the user in a way that threatens their privacy and security, or can be revealed to a third party (Mutimukwe et al., 2019;Xu et al., 2011). Therefore, the self-disclosure of information highlights inquiries about the importance and sensitivity degrees of information types being disclosed. However, the sensitivity varies based on many factors; it varies from person to person and from country to country (Bergström, 2015;Chang & Heo, 2014;Fianu et al., 2019;Krasnova & Veltri, 2010;Mutimukwe et al., 2019;Schomakers, Lidynia, Müllmann, & Ziefle, 2019;Stutzman et al., 2013).
The differences between nations' cultures is one of the most important factors that determine the degree of sensitivity of information types (Chang & Heo, 2014;Krasnova & Veltri, 2010;Markos et al., 2017;Schomakers et al., 2019). However, most previous studies targeted countries with a secular lifestyle, such as the USA, European countries, East Asia, and Latin America. Therefore, it was expected to find similar perceptions of information sensitivity (Bauer & Schiffinger, 2016;Hong & Thong, 2013;Markos et al., 2017;Pentina, Zhang, Bata, & Chen, 2016;Schomakers et al., 2019).
Article (Schomakers et al., 2019) claimed that studies on information sensitivity have focused on Western countries. For that reason, more studies about Asia, Africa, and Oceania are needed to shape the global model of information sensitivity among nations. In this regard, there exists an omission to analyze the information sensitivity of Internet users in the Middle East. Middle Eastern countries are significantly culturally different from the countries that were previously targeted. Middle Eastern countries are greatly affected by a conservative lifestyle that emerges from the interaction between Islamic rules and traditional extended family rules, which leads to a lifestyle that is somewhat different from Western countries.
This study aims to analyze and evaluate the perception of information sensitivity for people in the Middle East area, which is achieved by targeting Internet users in Saudi Arabia. The present study is developed according to the hypothesis of (Markos et al., 2017). We compare the outcomes of this study with the perception of information sensitivity of users from the USA, Brazil (Markos et al., 2017), andGermany (Schomakers et al., 2019). The results will contribute to clarifying the Saudi perception of information sensitivity in the international perception model. The rest of the paper is organized as follows. Section 2 presents the background and related works. Section 3 describes the adopted methodology and data. Section 4 shows the results and discussion, and Section 5 provides the conclusion.

Related Works
This section presents a review of literature about privacy concerns, information sensitivity, and the perception of information sensitivity between countries.

Background
In 1948, the United Nations Universal Declaration of Human Rights recognized privacy as a fundamental human right (Assembly, 1949). According to Westin (1967), privacy is people's desire to obtain the freedom of choice under any circumstances and choose to what extent they display their attitudes and behavior to others. With the rise of the Internet, information privacy has become an essential issue that interacts with several considerations (Bauer & Schiffinger, 2016;Bergström, 2015;Hong & Thong, 2013;Kapoor et al., 2018;Pentina et al., 2016;Xu et al., 2011). Pentina et al. (2016 pointed out that the rapid evolution of mobile technology and its apps, social networks, gaming, multimedia exchanges, and commercial transactions compel Internet service providers to collect many types of user information, such as gender, age, location, interests, personal preferences, browser history, calendar events, pictures, contacts, and so on. The main goal of collecting this data is to improve services for users. However, this data collection leads to security and privacy issues. Mutimukwe et al. (2019) claimed that with the increase of Internet use, creating a consistent basic definition of information privacy is important. They summarized the definitions of information privacy of many researchers by stating that it is the control of transactions between people to enhance autonomy and/or minimize risks. According to Hung & Cheng (2009), in the field of information systems, privacy is an individual's right to determine how, when, and to what extent information about the self will be released to another person or an organization. Based on (Öğütçü et al., 2016), users' self-disclosure highlights essential concepts, such as privacy concerns, self-disclosure behaviors, and information sensitivity for Internet users.

Privacy concerns
The number of social media users worldwide is expected to reach 3.02 billion by 2021 (Gilbert, 2019;Huo et al., 2019). Moreover, the number of devices connected to the Internet was expected to reach 50 billion different devices by 2020 (Mollah et al., 2019). This increase leads to disclosing more personal information by increasing the ability of organizations to collect, store, process, and exploit personal information from users, which reduces individuals' ability to control their information and increases their privacy risk (Mutimukwe et al., 2019). Milne et al. (2016) claimed that the most common worries of users are physical, monetary, social, and psychological risks.
Considering some privacy concerns of social networks as an example, (Fianu et al., 2019) mentioned that social media platforms primarily use users' data to make profits by selling user data to external parties, such as government and private organizations. Moreover, the default privacy setting allows unwanted people, such as friends of friends and group members, to obtain access to other users' profiles. This shares information with people who should not be able to access that information. In addition, Wirth et al. (2019) mentioned that friends share personal information, such as photos, only with their friends who trust them. However, one of these friends can then disclose the information to other people, even if the first user requested confidentiality for it.

Information sensitivity
Despite the privacy concerns of Internet users, the literature review shows various contradictions between privacy concerns and the behavior of information disclosure (Barth & de Jong, 2017;Roberts, 2012). For example, the most common password used is 12345, and most people use one password for multiple accounts. Moreover, people disclose their private information online, although they have security concerns about their privacy and data (van Zoonen, 2016). Kokolakis (2017) asked a question about this issue, why do users take so many risks? Pentina et al. (2016) claimed that Internet users make mindful trade-offs between benefits and privacy risks. Thus, most people are willing to share their data with the organization that requests it when these benefits are of direct personal relevance (Markos et al., 2017;Acquisti et al., 2013).
In this regard, van Zoonen (2016) claimed that users are interested in two specific points about their privacy. The first point is that people's privacy concerns vary according to the purpose for which the data is collected. The second represents the fact that people assume some data are more sensitive than others. Information type plays an important role in identifying the degree of privacy concerns. For example, personal data, such as nationality, gender, or age, are considered less sensitive information compared with medical and financial information that are considered highly sensitive for users. Users also differ in the perceived sensitivity of their information. For some people, this information is private, and for others, it is available to everyone without any concern (Robbins & Stylianou, 2002;van Zoonen, 2016).
Several works have studied the factors that influence the degree of information importance and sensitivity. Milne et al. (2016) performed a survey to analyze 52 types of information on the basis of expected risks. The results show a wide variation in the sensitivity of information types according to the perceived risk. Moreover, the current study demonstrates that cultural differences influence the assessment of perceived sensitivity to information types. Robbins and Stylianou (2002) examined the reflection of differences in national cultures on websites and on data sensitivity. The result shows that cultural considerations are an important factor in information systems and the degree of data sensitivity.
Veltri et al. (2011) compared the risk assessment of information disclosure on Facebook between American and Moroccan Facebook users. The results show that Moroccan users show greater privacy concerns than US users. Krasnova et al. (2012) analyzed the same point between German and American Facebook users. The results show that German users are more likely to self-disclose their private information than users from the USA. Li et al. (2018) studied the impact of culture to determine the intentions to share health information on social media. The case study on China and Italy shows a clear difference between the nations in that perception. Liu and Wang (2018) analyzed the influence of cultures in different countries on individual self-disclosure decisions. The comparisons between the USA and China show strong differences. Jiacheng et al. (2010) studied the mechanisms of knowledge exchange between individuals in China and the USA. The results show that the country's culture is an important factor in knowledge exchange. Chen et al. (2013) analyzed the respective levels of information privacy of mobile users in the USA and Korea. The results show that Koreans are more used to commercial activities via mobile phones, whereas Americans were more concerned about their information privacy. Trepte et al. (2017) studied the cultural factors that influence the perception of privacy risks in Germany, the Netherlands, the United Kingdom, the USA, and China. The results show important differences in user perception of privacy risks between nations. Markos et al. (2017) examined several factors, such as the country of origin, age, and type of information on data sensitivity and on users' willingness to disclose information on the Internet. The sample works comprised American and Brazilian users, and the results show significant differences according to the country of origin. Following the hypothesis of the previous study, Mutimukwe et al. (2019) added greater understanding of the cultural impact on the perception of data sensitivity. The German values concerning the disclosure of different types of information were compared with the results of American and Brazilian users in (Markos et al., 2017). The results show minor differences in sensitivity perception between the USA, Brazil, and Germany. The literature review above showed that the perceived levels of information sensitivity greatly vary between nations. Although previous works targeted certain countries (USA, China, and some European countries) that are considered secular and developed countries, unlike other nations that are considered conservative and traditional that have been ignored. Schomakers et al. (2019) claimed that the sensitivity of information has been studied in Western countries, and more information is needed about Asia, Africa, and Oceania. According to (Markos et al., 2017;Schomakers et al., 2019), further studies about different nations and countries are important to build a global model of perception of information sensitivity.

Differences of Saudi characteristics
The omission in the literature review to study the perception of information sensitivity in the Middle East is clear. This region includes many countries with similar cultural values from the interaction between Islamic rules and rules of traditional extended families. This leads to a significant difference in the lifestyle between the Middle East and Western countries, which entails differences in privacy concerns and perceptions of information sensitivity. Saudi Arabia is one of the key countries in the Middle East. Based on Alshahrani (2016) Figure 1 represents a set of cultural differences between Saudi Arabia, USA, Brazil, and Germany which cause differences in privacy concerns between Internet users. Saudi Arabia shows high degrees of large power distance and strong uncertainty avoidance results in strong religiosity, which indicates that the Saudi society follows the laws and administrative systems with high acceptance and accepts the belief that there is a single truth more than the USA, Brazil and Germany. Hoftede et al. (2010) claimed that people from cultures of high degrees of large power distance and strong uncertainty avoidance are careful when making information disclosure decisions. Saudi Arabia shows a low degree of Individualism. Strong religiosity and social norms display high collectivity. The Saudi society is a great example of tightly integrated relationships that bind extended families, such as the unquestioning loyalty and support for each other when a conflict arises with another group. Hoftede et al. (2010) unveiled that people from individual cultures are less likely to disclose information than those from collective cultures. Consequently, Posey et al. (2010) found that people from collective cultures are more concerned about the harm they may do to the privacy of their group.
According to the degree of masculinity versus femininity, Saudi Arabia has a value close to that of the USA and Germany, and it is more open than Brazil to accept differences. In terms of long-term orientation, the low degree indicates that US, Saudi Arabia, and Brazil honor and maintain traditions. Finally, the indulgence score shows that Saudi societal norms give average degrees of freedom to citizens in fulfilling their human desires, unlike Western countries that allow free gratification of human desires. With these terms, people tend to disclose more information with high scores (Hofstede, 2011).

Method and Data
This study adopted an experimental questionnaire approach to collect the data by using an online survey method. A total of 508 Internet users in the Kingdom of Saudi Arabia (KSA) were involved to estimate their perceived sensitivity of 35 different types of information on the Internet. This study analyzed the results statistically and compared them with the results of American, Brazilian, and German users according to experiments that have been conducted (Markos et al., 2017;Schomakers et al., 2019).

Selection of information types
This study adopts the hypothesis of (Markos et al., 2017) and the additions of (Schomakers et al., 2019). This study also uses the same information types as those studies to perform a descriptive comparison between Saudi users and users from the USA, Brazil, and Germany. Lastly, the researchers analyze the types of information suggested by (Markos et al., 2017), and modified by (Schomakers et al., 2019) to select suitable items for Saudi participants.
This study estimates which global types of information match the laws and culture in the Middle East, particularly in the KSA. We excluded certain information types from this study because of their seriousness in the Middle East. For example, certain information types are considered criminal behavior, such as homosexuality (sexual preference) and alcohol consumption. Others are rejected in conservative and religious societies, which are the majority of the population of the study area, such as online dating activities and women's hair color. In addition, this study drops DNA profile items because it is considered an unknown type of information for most people. Figure 2 displays the final list of 35 types of information.

Questionnaire and survey
This study presents the selected types of information as an online questionnaire. At the beginning, the topic was clearly presented to the participants. Then, the participants reported their demographic characteristics. Next, we asked the participants to evaluate the degree of perceived sensitivity of each type of information disclosed on the Internet or when they use smart devices. We also asked them to report any sensitive information that is not included in the questionnaire. Moreover, we instructed the participants to evaluate their privacy disposition, previous experiences of violation, a tendency to risk the disclosure of information, and their trust in institutions.
Given that we surveyed Internet users in the KSA, we asked a native speaker to formulate the questions in the Arabic language in a way suitable for the participants. We implemented and distributed the survey on an online platform called SmartSurvey. We divided our questionnaire into three parts. The participants provided their demographic information: age, gender, educational level, and marital status in the first part. Then, they evaluated the sensitivity of 35 types of information in the second part. We arranged these 35 types of information randomly to prevent sequence effects and asked the participants questions regarding their sensitivity toward the sample information types we provided for them (i.e., "How sensitive are the following information types for you?"). They rated their sensitivity to the given information types on a six-point Likert scale ranging from 1 "not sensitive at all" to 6 "very sensitive" (Markos et al., 2017;Schomakers et al., 2019). We used this type of scale to help participants distinguish the scores (Lozano et al., 2008;Markos et al., 2017). We also asked the participants to show their degree of agreement for eight points about their willingness, past violations, risk appetite, and trust in institutions in the third part of the questionnaire (Schomakers et al., 2019). They evaluated their agreement on a six-point Likert scale ranging from 1 "extremely disagree" to 6 "extremely agree." To compare our results with those of Markos et al. (2017) and Schomakers et al. (2019), we converted the scale to a 10-point scale.

The Sample
The researchers relied on various methods to distribute the questionnaire to reach a representative sample of Saudi Internet users. First, the questionnaire was distributed online to social groups from universities, employees, and communities. Also, face-to-face meetings were held with various members of the community to encourage them to participate as students and employees. To ensure adequate female participation, especially in the KSA, female students and employees were in buildings separate from males. Researchers contacted officials in the women's branches of educational and government institutions to encourage females and their families to participate. The previous process was applied in various areas of KSA.
In this survey, participants completed the questionnaire in January 2020. In total, 508 Saudi participants completed the questionnaire.

Measurements and analysis
To explore Saudi information sensitivity perception, we compared the outcomes with the available outcomes of Internet users in the USA, Brazil (Markos et al., 2017), and Germany (Schomakers et al., 2019). By imitating the methodology that was used by (Markos et al., 2017;Schomakers et al., 2019), we shaped the rank-order model of information sensitivity degrees on the basis of the average values according to the Saudi perspective. The resulting model includes the results of the average values of the USA, Brazil, and Germany to be compared with Saudi results. All 35 average values for perceptions are available for Germany, while only 29 are available for the USA and Brazil. Next, we categorized the variables into high, medium, and less sensitive data groups and analyzed them by using the threshold values between these three categories following the German experiments.

Results and Discussion
This section shows the descriptive results of the information sensitivity perception of Saudi Internet users. In addition, it shows a comparison between Saudi results and American, Brazilian, and German results on the basis of the outcomes of (Markos et al., 2017;Schomakers et al., 2019). Moreover, this section analyzes the sensitivity outcomes, highlights their characteristics, and finds factors that influence the sensitivity of Saudi Internet users.

Information sensitivity of Internet users in Saudi Arabia
On the basis of our survey, Figure 2 presents the rank-order of the average sensitivity ratings for each information type. The rank-order is sorted according to the average values (M) of Saudi Internet users from the highest sensitivity degrees to the lowest. As a result, the Password got the highest sensitivity by M = 9.64. Then, each of Fingerprint and Financial account information with some variance by M = 9.17 and M = 8.82, respectively. By contrast, Name of pet, Religion, and Occupation were less sensitive information for Saudi users in average about M = 2.4, M =3.08, and M =3.11, respectively.
Regarding our collected data, Internet users in the KSA showed some radical perception toward information sensitivity compared with their counterparts in other countries. From the 29 values common to all countries, Internet users in the KSA showed high degrees of sensitivity for each of passwords, fingerprint, digital signature, passport number, credit score, income level, and political affiliation. By contrast, they showed lesser degrees of sensitivity in seven items than their counterparts in the USA, Brazil, and Germany; these are license plate no., ZIP Code, mother's maiden name, weight, number of children, place of birth, and religion.
Saudi Arabia, Brazil, and Germany agree with the highest sensitivity for passwords by 9.6, 9.1, and 9.3, respectively. American users give the social security number as the most sensitive type (9.4). Saudi users clearly show a high sensitivity toward verification information types (passwords, fingerprint, digital signature, and passport number) than other countries. Moreover, political affiliation and law enforcement items show the most variance for Saudi users compared with other countries.
Among the participants, Internet users in KSA also showed lesser degrees of sensitivity in all the six values in common with Germany only, which are browsing history, online dating activities, medication, shopping behavior, sporting activities, and names of pets.
Based on the values of each country in common with the KSA, the correlation for information sensitivity shows high similarity between the KSA and Germany by r = 0.93, then r = 0.79 between KSA and USA, and r = 0.78 between KSA and Brazil. In detail, some differences are clear between our sample of Saudi users and other countries. The clearest difference between Saudi and Brazilian users was in law enforcement by 8.2 and 4.3, respectively. The clearest difference between Saudi and American users was in mother's maiden name by 4.1 and 7.2, respectively. The highest difference between Saudi and German users was in private phone numbers by 5.2 and 7.7, respectively.

Category of information and analysis
Similar to (Markos et al., 2017;Schomakers et al., 2019), we categorized information types into three groups of sensitivity value. We classified each type into low-sensitivity information, medium-sensitivity information, or high-sensitivity information classes as presented in Table 2.
To perform that, we scaled the average values of the sensitivity of each country between 0 and 10 by using the following formula: Where is the new scaled average value of sensitivity, and Sensitivity is the original average value of sensitivity. and are the minimum and maximum average values of the sensitivity of each country respectively. Then, we classified the variables by using the threshold levels from a previous study (Schomakers et al., 2019). The high-sensitivity items had values higher than 7.67, whereas low-sensitivity values had less than 4.7; in between was the medium sensitivity. We adopted these thresholds to obtain results comparable with the (Schomakers et al., 2019). The different numbers and scaling processes of used variables in this study could cause overlapping clustering when we use kmean clustering.
Comparing the high, medium, and low sensitivity information groups of Saudi users to the USA, Brazil (Markos et al., 2017), andGermany (Schomakers et al., 2019) shows some differences regarding the classification of the sensitivity of information types.
While the Credit score and Law enforcement were considered highly sensitive information for Saudi users, they were considered medium sensitive for German and American users, and some were low sensitivity for Brazilian users (Law enforcement). Moreover, Political affiliation was considered a medium-sensitivity information type for Saudi users, but it was considered low sensitivity for German, American, and Brazilian users.
Regarding the 29 items that are common to all the countries involved, we noticed various similarities between Internet users in the KSA and their counterparts in other countries. Internet users in the KSA were similar to those in Germany seven times, six times to those in the USA, and only twice to those in Brazil in the category with a high level of sensitivity. Internet users in the KSA matched seven times with those in Germany and five times with those in the USA and Brazil in the category with a medium level of sensitivity. Internet users in the KSA matched 11 times with those in Germany, eight times with those in the USA, and ten times with those in Brazil in the final category with the least level of sensitivity. Overall, Internet users in the KSA matched 25, 19, and 17 times with Internet users in Germany, the USA, and Brazil, respectively.
The rank distributions of Saudi Arabia, the USA, Brazil, and Germany were compared by using the average rank-order scores and their standard deviation analysis. First, we grouped and ranked all information from 1 to N, where N = 130. N is the total number of observations across all countries. Then, we extracted the mean (M) and standard division (SD). The USA had the highest average of ranks (M = 77.81, SD = 33.88), whereas Brazil had the lowest average of ranks (M = 49.75,SD = 35.91). The KSA and Germany showed average ranks (M = 61.79 and 68.4,SD = 40.94 and 32.5,respectively).
The outcomes of the last part of the questionnaire assess the attitude of Saudi users to a set of privacy concerns such as disposition to privacy, propensity to take risks, privacy violation experience, and trust in institutions. Participants demonstrate a high value of attention to their privacy (M = 4.97, SD = 1.38). In terms of propensity to take risks, they tend to risk disclosing their personal information online (M = 5.37, SD = 1.15). Regarding their experiences with online privacy violations, several participants lower than average had a bad experience, or misuse of information experiences (M = 2.57, SD = 1.86). Finally, participants demonstrate a discrepancy in the trust in institutions, depending on their type. They are highly trusting of government institutions (M = 4.76, SD = 1.66), whereas they show low trust levels in private companies (M = 2.15, SD = 1.48).

Discussion
This study focused on Internet users in Saudi Arabia as a case study aiming to evaluate the perceptions of information sensitivity in Middle Eastern and Islamic countries. We compared the outcomes of our collected sample of Internet users in the KSA with the outcomes of samples from different nations, namely, the US, Brazil, and Germany, as presented by Markos et al. (2017) and Schomakers et al. (2019). Based on the results, we deduced/found several similarities and differences between Saudi users compared with American, Brazilian, and German users.
The survey results showed that passwords, fingerprints, financial accounts, digital signatures, and passport numbers are the most sensitive information for Internet users in the KSA. In addition to the identifying information, we evaluated credit score, law enforcement, social security no., and IP address as highly sensitive information types for Internet users in the KSA. We considered them high risks because this information is used for security purposes in this country.
In contrast, the names of pets were the least sensitive type of information for Saudi users. It is reported as the lowest value of all information types for all nations. That is because Saudis, like most Muslims, are not interested in adopting many types of pets such as dogs and reptile due to Islamic rules that limit this behavior. In this regard, religion, occupation, information about sporting activities, and place of birth are the less sensitive types for Saudi users from low to high, respectively. Generally, Saudi users reported less averages in sensitivity about these types compared with other nations. It shows the cultural influence on the values of the sensitivity of information as claimed by Hofstede (2011). Moreover, these types that do not have any risk cases were taken into consideration; these information types are not used for any security purposes in Saudi Arabia.
The comparison between Saudi and American, Brazilian, and German users shows that the averages of Saudi users were higher than those of American, Brazilian, and German Internet users with information types that were classified as highly sensitive information. The averages were the lowest with information types that were classified as less sensitive information and were similar were classified as medium sensitivity information.
The rank-order of the perceptions of information sensitivity was almost similar across different countries. Internet users in the US, Germany, and the KSA showed the highest perception average of information sensitivity. By contrast, those in Brazil showed below average. Internet users in Germany and the KSA showed the closest average scores. However, the similar rank-order supports the guess-work of (Markos et al., 2017;Schomakers et al., 2019). They claimed that an international consensus exists about similar degrees of information sensitivity, regardless of the difference of cultures, nations, and religions.
The attitude of Saudi users towards their privacy concerns showed an interest in preserving their privacy online. Saudi users are not inclined to risk disclosing personal data online. In this regard, our survey shows that a low percentage of Saudis have a bad experience with their online privacy or have experienced the misuse of their friends' or family' data. In addition, Saudi users tend to trust government agencies to use their data. In contrast, they exhibit low levels of trust in private NGOs.
This study showed slight differences between Internet users in the KSA, the US, Brazil, and Germany concerning their perception of information sensitivity. Nevertheless, our results showed similarities concerning the global model of perception of information sensitivity. Internet users in the KSA were close in their perceptions with those in Germany, and they showed the most variance with those in Brazil. Except for personally identifiable information, Internet users in the KSA generally showed lower sensitivity levels than those in the US and Germany and higher sensitivity levels than those in Brazil.

Conclusion
This study aimed to evaluate the perception of information sensitivity of Internet users in Saudi Arabia as a case study to evaluate the perceptions of information sensitivity in Middle Eastern and Islamic countries. In addition, this study tested the factors influencing Saudi privacy perception. For that, we conducted an online questionnaire, analyzed and compared the results of 508 participants from Saudi Arabia with the information sensitivity perception of American, Brazilian, and German users. Based on our survey, results show slight differences between Internet users in Saudi Arabia, USA, Brazil, and Germany in the perception of information sensitivity.
Meanwhile, the rank-order of the sensitivity of information types shows similarity among nations. This study scoped an important area in the world. Generally, the Middle East and particularly Saudi Arabia have been ignored in these types of studies. Therefore, this study contributes to improving the understanding of the international perception of information sensitivity. This study adds views about the perception of a wide region different from the usual case studies of Western and Asian counterparts.