Young Saudis’ Evaluations and Perceptions of Privacy in Digital Communities: The Case of WhatsApp and Telegram

Abstract

Digital communities have gained increasing popularity in the past decade. However, they have put users at security risks, especially when they neglect to pay attention to the privacy statement and privacy settings. Therefore, digital community platforms must provide clear privacy statements and usable privacy settings. This research aims to evaluate the usability of privacy on WhatsApp and Telegram from the perspective of young Saudis. A total of 51 young Saudis participated in remote usability testing, followed by questionnaires and interviews. The results showed some privacy concerns in the two apps. Specifically, there are differences in the youths’ evaluation of the perceived privacy of the participants, with females seeing WhatsApp as more secure than the Telegram App. In the end, some recommendations for improving the privacy policy and settings for each app are discussed to ensure the safety and confidentiality of users’ information.

1. Introduction

Digital community platforms have become vital for daily communication [1]. People use digital communities for multiple purposes, such as chatting with friends and family or connecting to their colleagues at work. Most users of digital community platforms prefer accessing these platforms through their smartphone applications (apps) [2]. Almost half of the world’s population uses digital community platforms [3]. Statista reported that the number of social media users globally in 2021 was around 4.26 billion and is expected to reach 6 billion in 2027 [4]. The amount of personal information shared in digital communities has increased. In particular, sharing sensitive information, such as users’ names, dates of birth, locations, addresses, and device details [5], has raised privacy issues. Privacy has become more complex because the world has become more connected [6]. Furthermore, security breaches have increased because of digital communities.

Data protection laws allow individuals to control how their data are used. For instance, the EU’s General Data Protection Regulation (GDPR) sets a necessary global standard for privacy rights that allows users to control their data [7,8]. This control is managed through privacy settings. However, several reports have shown that most people tend not to pay careful attention to these settings [9,10]. Mostly, users are unaware of the option to change privacy settings or find them confusing and difficult to manage [11].

Studies that test the usability of privacy settings and policies in social media are limited. Albesher and Alhussain evaluated WhatsApp, Twitter, and Snapchat [2]. Alemerien tried to simplify the privacy settings related to sharing photos through a new visualization mechanism [12]. Dev et al. conducted a survey with Indian users about their privacy attitudes and behaviors and concluded with some recommendations to improve sensitive privacy defaults [13]. Vaziripour et al. carried out a survey with Iranian users to test their privacy preferences for Telegram [14]. Dias Canedo et al. administered a survey with Information and Communication Technology (ICT) practitioners to understand how these experts adopted data privacy concepts during software design [15]. As far as we know, most studies have evaluated the usability of privacy for one app [13,14,16,17,18], and only one study [2] has compared privacy settings and policies for several apps. However, the tested apps (Twitter, Snapchat, and WhatsApp) differ in the features they grant to users, which makes the privacy settings and policies hard to compare. Therefore, the current study compares the privacy settings and policies of apps with the same purpose (instant messaging and group chatting).

This study evaluates and compares the usability of the WhatsApp and Telegram apps’ privacy settings and policies. We intend to answer the following questions:

• What are young Saudis’ evaluations of usable privacy in WhatsApp and Telegram?

  o

  Are there statistically significant differences in young Saudis’ evaluations of usable privacy due to the app used (WhatsApp and Telegram)?

  o

  Are there statistically significant differences in young Saudis’ evaluations of usable privacy in the two apps (WhatsApp and Telegram) due to gender?

To answer these questions, we conducted experiments with 51 young Saudis. First, the participants completed specific tasks related to the privacy settings of one app. After that, they evaluated the app through questionnaires. The participants then performed the same procedures with the other app. At the end, the participants were met for an interview. All experiments were online, and notes were taken during the experiment and after watching every recording.

This research supports the design of more understandable and manageable privacy settings and policies in digital communities. In particular, this research improves the interface design for privacy settings and policies on instant messaging apps like WhatsApp and Telegram. The recommendations of this study can benefit the designers of privacy settings and policies in digital communities, the regulators who define the requirements of a usable consent interface, and the researchers who care about improving the interfaces designed for privacy and security purposes.

The remainder of this paper is organized as follows. Section 1.1 describes the link between sustainability and online communities and how privacy plays an essential factor in this relationship. Section 1.2 introduces privacy in digital communities, Section 1.3 describes the concept of usable privacy, and finally, Section 1.4 and Section 1.5 introduce the two tested apps (WhatsApp and Telegram, respectively). Section 2 includes the previous work on this study, while Section 3 explains the study’s methodology. Section 4 demonstrates the results for all the research questions. Section 5 presents a discussion associated with recommendations for improving the design of privacy settings and policies in social apps. Section 6 concludes the main findings of this research.

1.1. Sustainability in Digital Communities

The current study contributes to the field of sustainability by enhancing the sustainability of online communities. Particularly, it helps designers to create more usable privacy settings and statements. Sustainable online communities can increase civic engagement and public services. A community’s sustainable differentiation and long-term value-creating strategy may be significantly influenced by sustainable interaction, which also stabilizes and activates knowledge sharing, technology exchange, and integration for mutual benefit. Due to shifting user expectations, attention, and motivation, maintaining a sustained and engaged user base over the long term is challenging [19]. When users recognize that the privacy statements and settings of a specific online community are not helping them to protect their data, they will think to switch to another one. In January 2021, millions of users installed Telegram and Signal after WhatsApp announced changes in its privacy policy that sparked protests [20,21]. Moreover, Habibipour et al. [22] found that privacy affected sustained user engagement over time.

1.2. Privacy in Digital Communities

Privacy in digital communities focuses on a range of factors, including anonymity and dissemination of information, which can be defined here as “the claim of individuals to determine for themselves when, how and to what extent information about them is communicated to others” [23] (p. 24). This is also known in other terms, such as data privacy or data protection. Data privacy revolves around three key elements: the individual’s freedom to choose what personal data will be collected, how individual data are collected and shared, and compliance with data protection laws [24]. Moreover, privacy is a fundamental right of the individual that cannot be waived. In addition, users trust institutions and believe that laws can protect them from cyberattacks on their privacy. However, with the rapid development of social networks, dealing with these violations has become extremely complex, as these violations may amount to wiretapping, identity theft, and surveillance [25].

1.3. Usable Privacy

Saltzer and Schroeder [26] stated that “it is essential that the human interface be designed for ease of use so that users routinely and automatically apply the protection mechanisms correctly” (p. 9). Saltzer and Schroeder were arguably the first to note that computer systems must be secured. In addition, research on usable privacy and security has increased rapidly over the past two decades. Kirlappos and Sasse [27] used other terms with the same meaning, including usable security, which is defined as trusting in users and allowing them to participate. Another relevant term used is “privacy by design”, which means taking into consideration the privacy concerns at the design stage [28]. The usable privacy concept focuses on user interfaces designed with consideration for human factors, such as human–computer interaction (HCI); privacy and security aspects, such as user and email authentication; and web and social media privacy. These interfaces enable users to easily find information and privacy controls for the application and to understand and use them successfully [29].

1.4. WhatsApp

WhatsApp is a messaging app launched by Jan Koum and Brian Acton in 2009 that allows users to interact with family and friends worldwide, sharing messages, images, files, videos, and their geographical locations [30]. With more than 2 billion users, WhatsApp is one of the most popular messaging apps in the world [31]. One of its advantages is that it is free because of the use of free wireless access points, and its easy and ad-free interface is a distinct feature [32]. The privacy settings and policies are highly visible on WhatsApp, as they are all accessible in one place, and the sections are displayed with hints, like icons and tags. The lock sign indicates privacy within the account section, and the question mark denotes privacy policies within the help section [2].

1.5. Telegram

Telegram is a messaging app officially launched on 20 August 2013 by brothers Pavel and Nikolai Durov [33]. Nikolai took care of the technical side, and Pavel financially supported the endeavor. Telegram allows users to send instant messages and communicate in groups; it also has channels, to which users can subscribe. Telegram is considered a secure messaging app, but these features must be activated or selected by users instead of running by default. Privacy features include controlling who can see the last time users were online, who can call or add them to a group, and who can send them messages. Security features include a passcode lock to access the app [14]. Telegram has all of its privacy settings in the Privacy and Security section of the app, which is displayed with a lock. In addition, privacy policies are listed among the frequently asked questions (FAQs) and indicated by a question mark [34].

2. Background

2.1. Privacy Risks in Digital Communities

The privacy of digital communities is an emerging subject that has attracted increasing attention in recent years. Studies have examined privacy issues common in digital communities from different perspectives, such as identifying vulnerabilities, mitigating privacy risks, and suggesting ways to protect users from these issues. Ali et al. [9] discussed common privacy and security problems in social networking and data problems from service providers, as well as third-party “data collectors”. In a study aimed at educating online social network (OSN) users on how to protect themselves while using social media, a questionnaire was distributed randomly to undergraduate students to determine how they deal with privacy-related choices. The answers were disappointing because many users did not even use the privacy options provided by service providers; 23% of participants answered that they shared their personal information on OSNs. In contrast, 54% did not try to read their terms of use because they saw them as too long and difficult to read in a short time. In addition, 41% of users did not even protect their mobile devices with a password.

Schyff et al. [35] discussed how data monitoring has evolved and become more complex over several decades, especially since social media users are unaware of or are not interested in the risks they face. This, however, has enabled social media companies, such as Facebook, to collect data to predict and control user behavior. Thus, data monitoring represents risks that have profound implications for users. For example, people have historically been able to move to a new country and put their past behind them, but today, social media profiles remain on the Internet and are a stumbling block to development.

DeHart [36] investigated visual privacy in social communication. A total of 250 participants responded to two electronic surveys. Text analysis, variance analysis, and crowdsourcing were used to treat the sources. The results showed significant overlap in the privacy behaviors of participants by age and gender, a difference in the level of concern about identity theft and burglary, and increased fear of the dangers of social media sites. The study recommended increased awareness about the threats of visual content and the dangers of physical, mental, and emotional privacy, as well as recommending that app developers implement techniques that allow users to explore the trade-off between privacy and sharing.

2.2. Content Sharing in Digital Communities

One of the most important services offered by digital communities is content sharing. Users can employ this service to share media, like photos and videos, with other users. Despite the security and privacy concerns associated with content sharing, most users continue to share their content. In the following trials and surveys, usability, security, and privacy issues were investigated in the context of content sharing, with a focus on images.

Alemerien [12] introduced the visual privacy management policy (VPMP), a novel visualization tool for simplifying privacy settings when OSN users share their photographs with others. The authors conducted a comparison study of VPMP and the user interface of Facebook’s photo-sharing service in terms of usability and privacy by using five usability categories (intent to use, usefulness, ease of use, user satisfaction, and privacy concerns) and four privacy concern categories (trust, level of trust, and implications for privacy and content protection). A survey was conducted with 341 users, who performed 10 tasks using VPMP to ensure that the concepts used in VPMP are understandable and useful to OSN users. The results indicate that VPMP is effective and usable, and, based on the results, the study recommended the design of image-sharing mechanisms on OSNs carefully in terms of ease of use without compromising the privacy of personal information.

Hassan et al. [37] aimed to investigate the effects of applying image filters on the usefulness and aesthetics of images. By evaluating the effectiveness of 11 filters, they experimented with masking 20 different elements and features in images with consideration for two objectives: to protect privacy and to maintain the quality of images for viewers. Moreover, 570 people participated in the experiment on the effectiveness of blackout methods in a portion of the images. The results showed that, in some cases, privacy is clear versus utility, while in others, a high level of privacy can be achieved, while retaining utility. The study recommended investigating the relative trade-offs of image transformations to increase specificity without impairing the viewer’s experience.

2.3. Usable Privacy in Digital Communities

Kozlov et al. [38] presented a method for evaluating the effectiveness of experimental iterations of OSN counterfeit account verification systems: the Post-Authentication State (PAS) method. PAS was applied to the Facebook data. The results showed the ability of PAS to reduce classification time by 81%. Additionally, the PAS method reduces the volume of necessary human data by 70%. PAS also allows data scientists and engineers on Facebook to iterate evaluations more quickly by using new features.

Cohn-Gordon et al. [39] provided a framework for safeguarding the validity of deleting publications in OSN by controlling deletion and recovery capabilities. This framework requires and forces OSN developers to comment on all types of deleted data, report errors found, and suggest workable solutions. It was implemented on Facebook, and the results showed the ability of this framework to prevent and reduce the disruption of Facebook due to various errors in the deletion operations.

Minaei et al. [40] investigated the effectiveness of current deletion mechanisms on social networking sites. In this study, 191 participants were asked about their expectations and experiences with privacy using a questionnaire. The study showed that more than 80% of users deleted at least one post on social media, and 35% of the deleted posts occurred within a week of their publication. Participants expressed more concern about data collection companies than individuals on social networking sites. The participants recommended design guidelines that help improve deletion mechanisms.

Onaolapo et al. [41] created a tool to understand the effect of demographic characteristics on the behavior of the attacker in stolen accounts on social networking sites and observed the behavior of the attackers for six months. A total of 1008 real-life Facebook accounts were created, published, and exposed to criminals. The results confirmed that the demographic characteristics of victims affect the way criminals use victims’ accounts. Young people are more likely to receive online abuse and harassment and may fall victim to cybercrime more than others, while the elderly are more vulnerable to online fraud. The results also revealed that the gender factor has a significant impact, as female accounts received more friendship requests than male accounts.

Talukder [42] introduced AbuSniff as Android software to identify Facebook friends who are strangers, abusive, or banned. They used the AbuSniff system to explore user behavior in 7 trials with 263 individuals, as well as exploratory interviews and an online control study. The findings revealed that this program assists ONS users in protecting their personal information, such as images, posts, and status updates, from unauthorized access.

2.3.1. Usable Privacy of WhatsApp

Albesher and Alhussain compared the evaluation results of the privacy usability of WhatsApp, Twitter, and Snapchat [2]. Seven expert assessors applied the STRAP framework to assess the usability of privacy in all three applications. The results showed that Snapchat had the highest severity rating for usability issues, followed by Twitter and WhatsApp. In addition, a common severity rating among all apps was related to “the ability to revoke consent”. This means that there is no option to revoke the user agreement. The study recommended reconsidering how privacy policies and settings are displayed, such as locating them in the main section and attaching interactive visual signs and different colors to identify their differences.

Dev et al. [13], on Indian views of privacy and security for WhatsApp users, employed open and closed questions to inquire about the privacy attitudes and behaviors of 213 participants. A large percentage of participants said that they actively used privacy controls for different data types. The results also showed a high level of concern about privacy and site sharing via WhatsApp. The study recommendations included improved access control and settings that allow others to be invited to a group, rather than adding them directly. Additional recommendations were to make report risks clearer and more specific and to provide culturally sensitive privacy defaults.

Dev et al. [43] analyzed the impact of perceptions, behaviors, and personal experiences on the privacy of WhatsApp users in Saudi Arabia and India. Opinion polls were conducted among 820 participants from both countries who used WhatsApp daily. The results showed that individual perceptions and the origin and composition of the population influenced the use of WhatsApp’s privacy features. It was also strongly associated with personal experiences, such as contact with strangers. The study noted that cultural and social sites play a role in privacy behavior, as the Indian participants hid their profile and location information from colleagues and overlooked family and friends. Demographics affected each group differently; in terms of gender, women, for instance, tended to ban site sharing, and age had a significant impact on Indian users. The authors concluded that privacy concerns in all groups of participants were influenced by social motives and recommended more research on WhatsApp as a social network application, rather than a messaging app.

2.3.2. Usable Privacy of Telegram

Abu-Salma et al. [34] focused on two methods: a user lab study of 11 novices and 11 Telegram users and a hybrid analytic approach that combined cognitive mentoring and heuristic evaluation to assess Telegram. The evaluation revealed that users felt safe while using Telegram, although it had limited security features. In addition to the presence of problems in the user interface that affect the security of the program, such as making some features clear and others not, as well as the inconsistent use of terms, it does not support secret chat in group conversations. The study recommended that designers, developers, practitioners, and researchers consistently apply usability screening methods.

Vaziripour et al. [14] surveyed 392 users to determine the extent to which Telegram met privacy and security needs and explored the privacy preferences of Iranian users. They concluded that, for most users, privacy is important, and they believe that it is self-evident that messaging applications protect their privacy. Half of the survey respondents admitted to sending sensitive information while using the app, while 10% used end-to-end encrypted conversation, at least occasionally, in addition to their confidence in Telegram. The study included recommendations to make end-to-end encryption for chat messages the default and to use cookies to simplify privacy settings for users.

3. Methodology

The aim of this research is to look at young Saudi evaluations of usable privacy in WhatsApp and Telegram. The objectives of the study are:

• To explore young Saudis’ evaluation of the usability of WhatsApp and Telegram privacy settings and policies by analyzing the results of structured questionnaires.
• To present young Saudis’ perceptions and suggestions to improve the privacy settings and policies for WhatsApp and Telegram by analyzing the results of semi-structured interviews.
• To provide useful recommendations to improve the design of privacy settings and policies for WhatsApp and Telegram through the results of the experiment.

To answer the research questions, an experiment involving a mixed-methods (quantitative and qualitative) approach was adopted. For the quantitative method, the items of the structured analysis of privacy (STRAP) [44] were used in the form of a questionnaire. The questionnaire was based on the Likert scale (strongly disagree, disagree, neither agree nor disagree, agree, strongly agree). For the qualitative method, semi-structured interviews with open-ended questions were used. The STRAP framework combines elements of heuristic evaluation and goal-focused analysis to achieve viability, while reducing costs. This approach allows vulnerabilities to be identified, resulting in “privacy requirements” [15]. Researchers [44,45] have tested and analyzed the usefulness and efficiency of STRAP heuristics, concluding that it is a helpful tool for discovering security, privacy, and associated usability issues. In the social media domain, Albesher and Alhussain [2] used STRAP to evaluate and compare the usable privacy of Twitter, Snapchat, and WhatsApp. They concluded with several helpful recommendations for improving the design of privacy policies and settings in a way that makes them more readable and understandable. After the questionnaire evaluation, semi-structured interviews with open-ended questions were performed. This method allowed respondents to share more facts and opinions in their own words, encouraging the revelation of unexplored details resulting from the questionnaire evaluation. Semi-structured interviews were chosen for their flexibility and ability to generate new questions during the interview process [46]. Thematic analysis was chosen to analyze the qualitative results [47]. Specifically, MAXQDA analytics software was used [48,49], which was found helpful in other studies in the same field [50,51].

3.1. Participants and Requirements

The study was carried out in the Kingdom of Saudi Arabia, with a random selection of 51 young individuals between the ages of 18 and 35. The researchers prepared a message that asked for volunteers to participate in the study. This message was sent broadly to different groups in Telegram and WhatsApp. When we received requests for participation, we made sure that the participant had experienced the two programs. The study took place in a fully equipped online environment with a calm atmosphere and a high-speed Internet connection. The experiment lasted about a half an hour for each participant. To ensure unbiased results, each participant’s evaluation was conducted separately, and there was no direct contact between the participants. Participants needed a smartphone running iOS and the latest versions of WhatsApp and Telegram. Google Forms was used to collect the quantitative data from the participants. SPSS software was used for the data analysis. The reason for selecting iPhone is its popularity in Saudi Arabia. In fact, a recent study that investigated M-learning in Saudi Arabia through survey questions found that about 88.4% reported owning an iPhone, and 92.3% had the experience of using an iPhone [52].

3.2. Study Procedures

Before beginning the experiment, the research objective and brief were explained to the participants, the Internet connection speed was tested, and the quality of the audio, camera, and microphone was checked through an online meeting session using the desktop version of Skype. Then, participants were asked to read a consent form and sign it if they agreed to participate. During the experiment, participants were asked to open their cameras, and each participant was asked to perform four tasks relating to the case scenarios and to verbalize what they were doing while they were doing it. Tasks performed via the applications during this experiment included the following:

Task 1: Finding and locating “privacy settings” and “privacy policy” in WhatsApp and Telegram.

Task 2: Viewing the privacy policies of both apps.

Task 3: Changing the profile picture in both apps.

Task 4: Finding the last seen in both apps.

Performing all the tasks took approximately 15 min. Once the participants had completed the tasks, they were given a link to a questionnaire that used the STRAP framework to evaluate the privacy policy data and settings provided by WhatsApp and Telegram.

The participants were divided into two groups, and each group was given 10 min to examine the two applications. The first group started with the WhatsApp app and then moved on to the Telegram app. The second group started with the Telegram app and then moved on to the WhatsApp app. This procedure ensured that both groups had equal opportunities to examine both applications and complete the required tasks. After performing the task for the application, participants were given a link, which consisted of the STRAP framework questionnaire, to evaluate the privacy policy data and settings. After evaluating both apps, interviews were conducted with each participant. At the end of the experiment, the participants were thanked and dismissed.

4. Results

4.1. Demographic Results

A total of 51 participants evaluated the two mobile apps with different demographic characteristics, such as gender, age, city, university, and qualification. All participants had prior experience using the two mobile applications. Among the 51 participants, 61% (n = 31) were female, while only 39% (n = 20) were male. The age of the participants ranged from 18 to 35 years at the time of recruitment.

Table 1. Participant characteristics.

Category		Frequency	Percentage
Gender	Male	20	39%
	Female	31	61%
Age	18–24	35	69%
	24–35	16	31%
Academic level	High School	4	9%
	Diploma	0	0%
	Bachelor’s	43	84%
	Master’s	4	9%

and Figure 1 provide a summary of the participants’ characteristics.

4.2. Quantitative Result

4.2.1. What Are Young Saudis’ Evaluations of Usable Privacy in WhatsApp and Telegram?

To answer this question,

Table 2. Means, standard deviations, and the degree of youth evaluations of usable privacy in the two applications.

	Telegram			WhatsApp
Title	Arithmetic Mean	Standard Deviation	Rank	Arithmetic Mean	Standard Deviation	Rank
1. The information about app activities is always available in a way that is simple for me to access and understand.	3.90	1.025	1	3.86	1.059	1
2. Disclosure is complete, correct, and consistent in order for me to make informed decisions.	3.49	1.255	4	3.71	1.101	2
3. The relevant information was provided for each transaction to minimize the memory load and ensure that I’m aware of the consequences of the actions.	3.29	1.238	8	3.55	1.119	3
4. Disclosure takes into consideration limitations in my memory, ability, and attention and provides information that is brief and relevant.	3.37	1.183	6	3.53	1.206	4
5. Whenever possible, I was given real options rather than opt-in/opt-out choices to avoid coercion and maximize benefits.	3.35	1.278	7	3.33	1.260	7
6. The default settings reflected my privacy concerns and expectations.	3.45	1.006	5	3.53	1.084	5
7. I have avoided assuming consent whenever possible.	3.29	1.221	9	3.29	1.171	8
8. I have been provided with enough information to judge the security of the system and its information.	3.57	1.153	3	3.25	1.230	9
9. The app provides transparency in transactions and data use to build my confidence and trust.	3.63	1.019	2	3.39	1.060	6
10. I have access to all information the app has collected about me, regardless of source.	2.96	1.356	11	3.04	1.371	10
11. I have the ability to revoke consent.	3.24	1.210	10	2.80	1.400	11
Total mean youth rating for usable privacy	3.41	0.770		3.39	0.758

and Figure 2 summarize the statistical analysis of the STRAP heuristic questionnaire, which was used to evaluate the privacy policy data and settings of WhatsApp and Telegram. Table 2 includes the arithmetic mean, standard deviation, and rank after calculation for each question and, at its end, the sum of the arithmetic mean of both applications.

The results revealed that Saudi youth gave a medium evaluation score of 3.41 for the usability of privacy in Telegram and an average evaluation score of 3.39 for WhatsApp. This indicates that the usability of privacy for both applications is approximately equal in the perceptions of Saudi youth. The participants found that the information provided about the activities of both applications was easily accessible and understandable. However, they expressed dissatisfaction with the lack of access to the information collected about them in Telegram. In addition, the participants expressed annoyance with the periodic update of WhatsApp, which obliges them to agree and does not enable them to opt out. Overall, the study suggests that both Telegram and WhatsApp have similar usability of privacy in the perception of Saudi youth.

4.2.2. Are There Statistically Significant Differences in Young Saudis’ Evaluations of Usable Privacy Due to the App Used (WhatsApp and Telegram)?

To answer this question, the arithmetic means and standard deviations were extracted, and a t-test was conducted for independent samples to examine the significance of the differences in the evaluations of usable privacy between WhatsApp and Telegram from the perspective of youth in Saudi Arabia. The results are summarized in

Table 3. Means, standard deviations, and a t-test for independent samples to examine the significance of differences in youth evaluations of usable privacy for the two apps.

Scale	Application	Number of Participants	Arithmetic Mean	Standard Deviation	t-Value	Degrees of Freedom	Statistical Significance
Usable Privacy	Telegram	51	3.41	0.770	0.150	100	0.881
	WhatsApp	51	3.39	0.785

.

Table 3 presents the results of analyzing the data after utilizing the t-test. The t-value of youth evaluations in Saudi Arabia for usable privacy attributed to the application used (WhatsApp and Telegram) was 0.150, which is a non-statistically significant value at the significance level of 0.05, meaning that there are no statistically significant differences in the evaluations of young people in Saudi Arabia for usable privacy attributed to the app used (WhatsApp and Telegram), which indicates that the evaluations were similar for the two applications.

4.2.3. Are There Statistically Significant Differences in Young Saudis’ Evaluations of Privacy in the Two Apps (WhatsApp and Telegram) Due to Gender?

To answer this question, the arithmetic means and standard deviations were extracted, and a t-test was conducted for independent samples to examine the significance of the differences in the evaluations of youth in the Kingdom of Saudi Arabia of usable privacy in the two applications (WhatsApp and Telegram) due to gender.

Table 4. Mean, standard deviations, and a t-test for independent samples to examine the significance of differences in the evaluations of youth in Saudi Arabia of usable privacy due to gender.

Scale	Gender	Number of Participants	Arithmetic Mean	Standard Deviation	t-Value	Degrees of Freedom	Statistical Significance
Usable Privacy of Telegram	Male	31	3.54	0.731	1.472	49	0.147
	Female	20	3.22	0.806
Usable Privacy of WhatsApp	Male	31	3.61	0.676	2.669	49	0.010 *
	Female	20	3.05	0.833

* Statistically significant at the level of 0.05.

shows the results.

Table 4 indicates that the evaluations of young people in the Kingdom of Saudi Arabia regarding the usable privacy of the Telegram app showed no significant difference due to gender, as the t-value was 1.472, which is not statistically significant at the 0.05 significance level. This means that both male and female evaluations were similar. However, the t-value for youth evaluations of the usable privacy of the WhatsApp app due to gender was 2.669, which is statistically significant at the 0.05 significance level. This suggests that there are significant differences in youth evaluations of the usable privacy of the WhatsApp app due to gender, with females having a higher arithmetic mean than males. These results are important, as they shed light on how young people evaluate the privacy of different applications and how these evaluations may be influenced by gender.

4.3. Qualitative Results

Video recordings from the experiment activities and interviews were transcribed verbatim. The text was then independently reviewed and assigned codes using thematic analysis. The interview data were coded and analyzed using MAXQDA Analytics Pro 2022 software. Through thematic analysis, various opinions and perspectives were discovered and grouped into themes based on their similarities. From our content analysis of the interview data, we identified three significant categories that affect the privacy of apps: privacy concerns in two apps, the app with the highest perceived privacy by the participants, and recommendations for improving privacy.

Of the 51 participants, only 6 could complete all four tasks required for each application without any delay. Nine participants were not aware of the privacy policy in the two applications, mistakenly believing that it existed only during the first use of the app. In Telegram, two participants had trouble finding their privacy settings, and one participant was unaware of the last seen option. Three more participants were late in changing their last appearance. A total of 22 participants were unable to find the Telegram privacy policy, with 7 eventually finding it after some delay. Interestingly, all of the participants who faced this issue were using an iPhone. Among the 22 participants who were able to find the privacy policy quickly, 4 used Huawei devices, 5 had Galaxy devices, and the rest used iPhones. Android devices had an obvious privacy policy option, while iPhones had it embedded in the FAQ, and it was difficult to access. Interestingly, the “change display picture” task was the only task completed by all participants in both applications.

Privacy concerns were reported by participants regarding both WhatsApp and Telegram, which negatively impacted their overall experience. With WhatsApp, many users found it troubling that when a message was deleted, the sender was still able to see that it was deleted. In contrast, Telegram provides greater privacy in this regard. Furthermore, displaying a user’s connection status on both apps can also be a cause for concern. Another issue raised is that WhatsApp allows users to be added to groups without their permission, which can lead to strangers having access to their contact information. In contrast, Telegram hides the number of users and allows communication only via usernames. Finally, some participants reported that their contacts were able to communicate with them on Telegram without having access to their contact list, which is a privacy concern.

Privacy concerns were a significant issue when using Telegram and WhatsApp. The lengthy privacy policies for both apps made them challenging to read, and the language used in WhatsApp’s policy was unclear, lengthy, and not summarized, causing users to become distracted. The terminology used in the policies was difficult to understand, and the font colors were reported to be faint. WhatsApp’s policy tends to be legally oriented, making it challenging for the general public to understand, while Telegram’s policy is more technology-oriented, using technical terms, such as encryption and cookies, that only specialists would understand.

In addition, the participants pointed out that they were unable to access the privacy policy within the Telegram application and had to open it through a web link. Moreover, many participants raised concerns about the absence of an Arabic version of the Telegram privacy policy, which they considered a privacy issue.

From the survey conducted, it was found that 31 individuals believed that WhatsApp’s privacy policy was more user-friendly regarding privacy. They appreciated the clear presentation of the policy in bullet points, the use of bold text for important terms, and the implementation of three easy-to-spot colors. On the other hand, the remaining 20 participants found Telegram’s policy to be more comprehensive and detailed. They also found the application interface and colors to be clearer than WhatsApp’s. The policy was presented in sections and subsections, which made it easier for users to navigate and understand. Furthermore, Telegram’s policy allowed users to search within the document, enhancing overall accessibility and convenience.

Throughout the course of the discussion, the participants offered several ideas for improving the scalability and privacy of both applications. One of the most promising suggestions was to replace the current legal document, which is quite lengthy and packed with text, with illustrative icons and images that could more effectively convey the essence of the policy. This would make it easier for users to understand exactly what they are agreeing to when they use the app and could help improve transparency and trust overall. Additionally, the group recommended that clearer colors be used in the privacy policy in order to make it more visually appealing and user-friendly. Finally, there was a suggestion that the privacy policy should also be presented in Arabic in order to better serve the needs of Telegram users who speak this language. Overall, these suggestions were seen as highly valuable and could help improve the user experience of both applications significantly.

In addition, several participants offered their thoughts on how to improve the usability of privacy features. Among these proposals was the implementation of an alert system within both applications, which would promptly inform users if another individual was attempting to record the conversation. Moreover, it was suggested that deleted messages should be completely removed from WhatsApp to further safeguard user privacy. Finally, there was a recommendation to allow for communication via usernames rather than personal phone numbers, thus providing an extra layer of anonymity for those wishing to maintain their privacy.

5. Discussion and Recommendations

In this study, the focus was on scrutinizing the usability of the popular messaging platforms WhatsApp and Telegram with regard to privacy. To this end, the researchers utilized a combination of questionnaires and interviews. The outcomes of the study demonstrated that, despite some similarities, both platforms had distinct issues. Overall, the study was instrumental in providing significant insights into the usable privacy features of both WhatsApp and Telegram.

One of the areas in which the two apps excelled was providing information regarding their respective activities. Both WhatsApp and Telegram were rated highly for “availability of information with regard to the app’s activities”, which received the highest score ever in the two applications. Telegram stood out because of its easily accessible FAQ section, which was praised by many of the participants. On the other hand, WhatsApp’s help center was able to provide satisfactory answers to all user queries related to the app’s activities.

While both apps were appreciated by the users for their transparency, WhatsApp was preferred over Telegram when it came to the availability, accessibility, and clarity of privacy notices. This preference was understandable since Telegram’s privacy policy was found to be embedded within the FAQ section, making it confusing for most of the participants, especially on iOS devices.

Regarding the completeness, correctness, and consistency of disclosure, WhatsApp has a slight edge over Telegram. Based on feedback from participants, WhatsApp takes a user-friendly approach by dividing its privacy policy into different sections, which makes it easier to navigate and understand. In addition, WhatsApp considers the language barrier by providing its policy in Arabic, which is not available in Telegram. Furthermore, WhatsApp employs various formatting techniques, such as different colors and font sizes, to emphasize important sections and sentences, thus aiding users in identifying them. Overall, it appears that WhatsApp places a greater emphasis on providing a clear and transparent privacy policy to its users.

Participants gave both apps an average rating for their approach to “choice and consent” in terms of privacy settings. While most felt that the existing privacy settings were sufficient, some participants suggested that minor adjustments could be made to enhance protection. For instance, users requested options to modify their appearance status and hide personal information from strangers. However, the study revealed that users were concerned about sharing information with third parties, which was a recurring issue. In fact, one participant expressed worry that while the privacy policy of the applications could guard them against contacts within the app, it might not offer the same level of protection against third-party access. According to Telegram, neither the content of private chats nor that of group chats is made available to outside parties, and no requests pertaining to them are handled. Nevertheless, Telegram acknowledges that user data and account details remain susceptible to being accessed physically or remotely on users’ smartphones or computers [53]. There are examples of successful attempts to extract such data. Even though Telegram mentions collecting and storing metadata (IP address, devices used, history of username changes) in its privacy policy, it does not specify if it is encrypted when stored on its servers [53,54].

In addition, participants rated WhatsApp slightly higher than Telegram regarding their awareness of security mechanisms and transparency in transactions. The reason for this was that WhatsApp provided clearer explanations for why it collected data, while Telegram was seen as using technical language that non-professionals may not understand. However, both applications were urged to provide more compelling details about their use of user data and to increase transparency overall. Thus, it is important for apps to make privacy settings understandable and accessible to all users.

Based on the last group of questionnaires, it was found that WhatsApp outperformed Telegram when it came to users accessing their own records. This is because WhatsApp provides a clear option under the account section for users to request their account information, whereas Telegram does not offer such a straightforward method. However, Telegram received a higher rating than WhatsApp in terms of its ability to revoke consent. This is because WhatsApp periodically requires users to agree or withdraw from the application, which can be bothersome. In contrast, Telegram users are not subjected to such periodic updates, which alleviates their concerns. When WhatsApp sparked protests by announcing its privacy policy update in January 2021, millions of users installed other messaging apps, like Telegram and Signal [20,21]. One study [55] conducted a survey the following February with 1525 WhatsApp users from Mexico, Spain, South Africa, and the United Kingdom. The authors noticed that over 25% of them wanted to move at least some of their communication to other apps.

Due to many contacts, even strangers can post conversations, and a breach of privacy occurs without the person participating in the conversation knowing. Participants suggested giving the option to accept or decline before joining a group [13,43], setting a time for working groups and automatically deleting them when they did not participate or interact in the group for a certain period, and putting specific colors and icons on separate work or study groups from friends and family to avoid accidentally sending personal files that cause embarrassment.

Based on the results of the experiment comparing the privacy features of WhatsApp and Telegram, it was found that WhatsApp provides its users with the ability to initiate calls, add new contacts, and engage with others solely through their phone numbers. However, this presents a potential privacy concern when users are in a group with strangers, who can easily access contact information and reach out to other group members without consent [13,32,43]. On the other hand, Telegram offers the option of communicating using a username instead of a phone number, providing enhanced privacy and reducing the annoyance of receiving messages from unknown numbers.

To further improve both apps, users require better tools for overseeing group membership, developing group member trust, and filtering the information they share in various conversation contexts [56]. For improving WhatsApp’s privacy, participants in the study suggested adding an option to hide the phone number and communicate solely through a username. Furthermore, they proposed the idea of an alert feature that would notify users if someone in the conversation began filming, as many conversations could be shared without the consent of all parties involved.

In today’s world, it has become increasingly common for individuals to engage in conversations with both known and unknown acquaintances. However, this can lead to a potential invasion of privacy, without the knowledge of the person involved in the conversation. To combat this issue, several solutions have been proposed by participants. One such solution involves providing individuals with the option to accept or decline before joining a group, as suggested by Dev et al. [13,43]. Another solution involves setting a time limit for working groups and automatically deleting them if there is no participation or interaction within a certain period. Furthermore, some have suggested using specific colors and icons to distinguish work or study groups from friends and family to prevent the accidental sharing of personal information that can be embarrassing. By implementing these measures, individuals can feel more secure knowing that their privacy is respected and protected.

Overall, the results of the experiment have shown that messaging applications lack privacy features, which could potentially compromise the privacy of their users. Therefore, it is imperative that these applications improve their privacy measures to ensure the safety and confidentiality of their users’ information. The following summarizes the recommendations of this study:

• Telegram should not embed the privacy policy in the FAQ section.
• Telegram should add the Arabic language to their privacy policy.
• Telegram should divide its privacy policy into different sections.
• Telegram should employ various formatting techniques, such as different colors and font sizes, to emphasize important sections and sentences.
• Telegram should avoid using technical language.
• WhatsApp should not periodically require users to agree or withdraw from the app.
• WhatsApp should grant the ability to hide phone numbers and allow communicating by username. This becomes highly urgent now because users have become more engaged with strangers through “community”, a recently added feature by WhatsApp.
• WhatsApp should avoid auto-joining groups without permission.
• Both apps should provide more compelling details about their data usage.
• Both apps should grant users more options regarding sharing their data with third parties.
• Both apps should grant more users options to modify their appearance status and hide personal information from strangers.
• Both apps should put specific colors and icons on separate work or study groups from friends and family to avoid accidentally sending personal files.
• Both Apps should set a time limit for working groups and automatically delete the ones that have no interaction after alerting users.

6. Conclusions

After conducting an in-depth study on the privacy features of WhatsApp and Telegram, it was discovered that both platforms have distinct issues. However, both revealed approximately equal average scores for the usability of privacy. Saudi youth gave an average evaluation score of 3.41 for Telegram and 3.39 for WhatsApp. Additionally, there are no statistically significant differences in the evaluations of young people in Saudi Arabia for usable privacy attributed to the app used (WhatsApp and Telegram) since no significant value at the significance level of 0.05 was found. On the other hand, there are significant differences in youth evaluations of the usable privacy of the WhatsApp app due to gender, with females having a higher arithmetic mean than males.

Both apps stand out in their ability to provide comprehensive information on their activities, receiving the highest ratings in this category. Telegram’s FAQ section was particularly lauded by many participants for its ease of accessibility and usefulness, while WhatsApp’s help center was found to provide satisfactory answers to all user queries. Overall, this study was incredibly insightful in shedding light on the usable privacy features of two of the most widely used messaging platforms.