CoChat
Sign Up
Cited by 4Privacy, Security, and Data ProtectionTechnology Adoption and User BehaviourInnovative Human-Technology Interaction

The privacy and control paradoxes in the context of smartphone apps

Open access
Ayres-Pereira, Vanessa, Pirrone, Angelo, Korbmacher, Max, Tjostheim, Ingvar, Böhm, Gisela·2022·Frontiers in Computer Science
DOI 10.3389/fcomp.2022.986138
Open this paper in CoChat — free accountOpen PDF

Abstract

This research examines how various factors, such as the degree of e-privacy concerns and control over data access permissions, can influence a user's intention to install a smartphone app. We conducted two survey-based experiments with 441 participants. In each experiment, we manipulated the degree of control over the number and type of data access permissions granted to different fictional apps. In Study 1, participants were informed about the set of permissions the apps required. In Study 2, participants indicated which individual permissions they were willing to grant to the apps. In both experiments, we assessed the level of e-privacy concerns, perceived app importance, and the intention to install the apps. The results suggest that the type of app plays a central role in determining both the perceived benefit of installing the app and the level of e-privacy concerns. The intention to install an app is more strongly associated with perceived app importance than with e-privacy concerns (especially when app importance is high, and users have explicit control over which specific data access permissions they want to grant). The implications of these results are discussed regarding psychological factors involved in app installation decision-making process and the importance of promoting data protection by design.

AI snapshot

Written from the paper’s full text.

TL;DR

Two survey experiments with Norwegian university participants found that perceived app importance was the strongest predictor of intention to install fictional smartphone apps, while e-privacy concerns mattered mainly for low/moderate-importance apps and when users lacked fine-grained permission control. Increased control over permissions produced only limited evidence of a control paradox and did not consistently raise installation intention.

Key findings

  • Across both studies, app type drove perceived importance and installation intention: medians for Messaging vs Weather vs Puzzle were 9.0/7.0/1.0 for importance and 6.0/5.0/2.0 for installation intention (0–10 and 1–7 scales). App importance was the only significant predictor in Study 2 for all three apps (Puzzle B=0.454, Weather B=0.246, Messaging B=0.498, all p=0.001) and significant in Study 1 (Puzzle B=0.438, Weather B=0.390, Messaging B=0.522, all p=0.001).
  • Privacy concerns were multidimensional and context-dependent: smartphone privacy concern (Mdn=7.0) never predicted installation intention (refuting H1), but in Study 1 app-specific concerns did: discomfort over data access reduced Puzzle installation intention (B=-0.037, p=0.002), and data-security importance reduced Weather installation intention (B=-0.113, p=0.002).
  • Study 2 permission control weakly supported the control paradox: Weather installation intention was higher in Study 2 (M=5.89, SD=0.87) than Study 1 (M=4.52, SD=1.98), p<0.001, small effect r=-0.195; Puzzle and Messaging intentions did not differ significantly. When choosing, participants granted few function-consistent permissions: Puzzle M=0.62, Weather M=1.03 (mostly location), Messaging M=2.43.
  • Data sensitivity: internal storage was more discomforting than contacts, microphone, and location (Wilcoxon p=0.001, 0.001, 0.002; small effects r=-0.161, -0.154, -0.149).

Methods

  • Two survey-based experiments (Study 1 N=227; Study 2 N=214) recruited at a Norwegian university psychology faculty in Oct–Nov 2019; mostly 18–30 years old (95.2% Study 1; 96.7% Study 2), with more women than men.
  • Participants rated three fictional apps (Puzzle Game, Messaging, Weather Forecast) in counterbalanced order. Study 1 manipulated number/type of required permissions (0–4 data types) as an all-or-nothing decision; Study 2 asked participants which permissions they would grant.
  • Measures included perceived app importance, installation intention, perceived data-security importance, discomfort over data access permissions, and smartphone privacy concern. Analyses used nonparametric tests (Kruskal-Wallis, Mann-Whitney, Friedman, Wilcoxon) and bootstrapped multiple regressions with Bonferroni-adjusted alpha.

Limitations

  • Stated: the sample came from one Norwegian psychology faculty and likely had high education/privacy literacy, limiting generalizability; future work should include diverse samples and privacy-literacy scales.
  • Stated: app importance and data requirements were confounded (low-importance apps needed few permissions; high-importance Messaging needed multiple), so future studies should vary these orthogonally; fictitious apps in questionnaires may not emulate real smartphone permission decisions.
  • Evident: self-report intentions rather than observed installation behavior; convenience sample; no actual app use, so findings may not generalize beyond young, educated Norwegian students.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3389/fcomp.2022.986138

Cites

43 of 43 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

10 of 71 cited sources checked · 7 supported · 0 flagged · accuracy not yet measured

  • Premisecitation

    Some app developers have collected excessive amounts of data from users for non-functional purposes, such as selling the data as a commodity.

    “One issue of concern is that some app developers have collected excessive amounts of data from users for non-functional purposes, that is, for so-called secondary uses such as selling the data as a commodity (e.g., Spiekermann et al., 2015;Zang et al., 2015).”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites Spiekermann et al., 2015 source

      The opening passage of “Personal data markets” (labeled Abstract; page unavailable) describes firms auctioning personal information and a large third-party data marketplace, supporting the claim that user data is traded. It does not say that app developers collected excessive data or that they collected it for purposes unrelated to app functionality. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites Zang et al., 2015

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Premisecitation

    Liu et al. (2015), Gu et al. (2022), and Shahidi et al. (2022) produced evidence in favor of the privacy calculus theory by demonstrating that whether a user chooses to use an app is a result of the trade-off between the app's functionality or usefulness and the user's privacy concerns.

    “Liu et al. (2015), Gu et al. (2022), and Shahidi et al. (2022), for example, produced evidence in favor of the privacy calculus theory by demonstrating that whether a user chooses to use an app is a result of the trade-off between the app's functionality (or usefulness) and the user's privacy concerns.”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites Liu et al. (2015) source
    • cites Gu et al. (2022)

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Shahidi et al. (2022) source
  • Interpretationown data

    The results from Study 1 suggest that the decision to install an app (and grant data access permissions as required) results from a trade-off between the app's functionality importance and the user's privacy concerns.

    “The results from Study 1 suggest that the decision to install an app (and grant data access permissions as required) results from a trade-off between the app's functionality importance and the user's privacy concerns (e.g., Dinev and Hart, 2006;Liu et al., 2015;Shahidi et al., 2022).”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites Dinev and Hart, 2006 source
    • cites Liu et al., 2015 source
    • cites Shahidi et al., 2022 source

      Section 5.2 (page unavailable) reports that perceived usefulness increased intention to use StopCovid, while privacy concerns decreased it; perceived permission sensitivity also increased privacy concerns. This supports the suggested benefit–privacy trade-off, but the study measured intention to use a specific contact-tracing app, not actual installation or granting of data-access permissions. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
  • Interpretationown data

    The results support the existence of the privacy paradox in the context of apps.

    “These results support the existence of the privacy paradox (Barnes, 2006) in the context of apps, extending and generalizing the results reported by other studies about privacy decisions (e.g., Hann et al., 2002;Hui et al., 2007;Acquisti, 2009).”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites Barnes, 2006 source
    • cites Hann et al., 2002
    • cites Hui et al., 2007 source
  • Findingown data

    In Study 1, respondents reported a moderate to high median level of smartphone privacy concern (Mdn = 7.0 on a 0-10 scale).

    “In general, respondents reported a moderate to high median (Mdn) level of smartphone privacy concern (Mdn = 7.0; on a scale from 0 to 10) and a moderate to high degree of discomfort from granting data access permission to all types of data (Mdn = 7.00; on a scale from 0 to 10).”
    Quote in the paperNot a substantial claim
  • Classified as probably not a substantial claim. 28
  • May combine two claims. 44

  • Classified as probably not a substantial claim.

    Premisecitation

    E-privacy concern is a multidimensional concept.

    “Moreover, e-privacy concern is a multidimensional concept.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Buck et al., 2018

  • Classified as probably not a substantial claim.

    Premisecitation

    Decisions to disclose data can be based on a trade-off where individuals choose to exchange their private data for benefits or services that they consider important.

    “suggest that decisions to disclose data can be based on a trade-off where individuals choose to exchange their private data for benefits or services that they consider important (e.g., Hann et al., 2002;Hui et al., 2007;Acquisti, 2009).”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites Hann et al., 2002
    • cites Hui et al., 2007 source
    • cites Acquisti, 2009 source

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Premisecitation

    There are many facets of e-privacy that people may be concerned about.

    “That is to say, there are many facets that people may be concerned about (e.g., Buck et al., 2018).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Buck et al., 2018

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Location is an important aspect of e-privacy.

    “Location is an important aspect of eprivacy (see Rowe, 2020, for an interesting discussion)”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Rowe, 2020 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Several studies reported the privacy paradox in the context of social network sites and e-commerce.

    “Several studies reported the privacy paradox in the context of SNSs and e-commerce (for reviews, see Barth and de Jong, 2017;Kokolakis, 2017).”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites Barth and de Jong, 2017 source
    • cites Kokolakis, 2017 source

  • May combine two claims.

    Interpretationown data

    The results of Studies 1 and 2 confirm the hypothesis that app perceived importance has a positive effect on app installation intention (H4), replicating previous studies.

    “Therefore, the results of Studies 1 and 2 confirm the hypothesis that app perceived importance has a positive effect on app installation intention (H 4 ), replicating previous studies (e.g., Egelman et al., 2013;Shahidi et al., 2022).”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Egelman et al., 2013 source
    • cites Shahidi et al., 2022 source

      In §5.2 (page unavailable), the cited paper reports that perceived usefulness positively influenced intention to use StopCovid. Section 3.2 defines that adoption intention to include downloading the app, so the finding is relevant to installation intention. The paper tests perceived usefulness, however, not the citing study’s distinct construct of app perceived importance. (LLM, conf=0.83)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingown data

    In Study 1, the Messaging app was perceived as more important than both the Puzzle and Weather Forecast apps, and the Weather app was more important than the Puzzle app (Mdn Messaging = 9.0; Mdn Weather = 7.0; Mdn Puzzle = 1.0 on a 0-10 scale).

    “As shown in Figure 1, the Messaging app was perceived as more important than both, Puzzle and Weather Forecast apps, and the Weather app was more important than the Puzzle (Mdn Messaging = 9.0; Mdn Weather = 7.0; Mdn Puzzle = 1.0; on a scale from 0 to 10).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In Study 1, intention to install the Messaging app was greater than intention to install the Puzzle and Weather apps, and intention to install the Weather app was greater than the Puzzle app (Mdn Messaging = 6.0; Mdn Weather = 5.0; Mdn Puzzle = 2.0 on a 0-7 scale).

    “The intention to install the Messaging app was greater than both the intention to install the Puzzle and Weather apps, while the intention to install the Weather app was greater than the Puzzle (Mdn Messaging = 6.0;Mdn Weather = 5.0; Mdn Puzzle = 2.0; on a scale from 0 to 7).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In Study 1, data security was perceived as significantly more important in the Messaging than in the Puzzle and Weather apps, without significant difference between the Puzzle and Weather apps.

    “In Study 1, post-hoc analyses indicated that data security was perceived as significantly more important in the Messaging than in the Puzzle (Z = -8.046,p < 0.001, r = -0.542)and Weather apps (Z = -8.380,p < 0.001, r = -0.562),without significant difference between the Puzzle and Weather apps.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Although data security was not a significant predictor of the Puzzle Game installation intention, only in this app data security importance was positively related to the number of requested permissions.

    “Although data security was not a significant predictor of the Puzzle Game installation intention, only in this app data security importance was positively related to the number of requested permissions.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    App importance (p = 0.001, B = 0.390, 95% CI [0.296, 0.470]) and app data security importance (p = 0.002, B = -0.113, 95% CI [-0.189, -0.040]) were both significant predictors of the intention to install the Weather app.

    “App importance, p = 0.001, B = 0.390, 95% CI [0.296, 0.470], and app data security importance, p = 0.002, B = -0.113,95% CI [-0.189, -0.040] were both significant predictors of the intention to install the Weather app.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In Study 2, data security was significantly more important in the Messaging than in the Puzzle and Weather apps, without significant difference between the Puzzle and Weather apps.

    “As in Study 1, the analysis of Study 2 also indicated that data security was significantly more important in the Messaging than in the Puzzle (Z = -8.407,p < 0.001, r = -0.584)and Weather app (Z = -7.942,p < 0.001, r = -0.556),without significant difference between the Puzzle and Weather apps.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Although participants reported a moderate to high degree of discomfort from granting data access permission to all types of data (Mdn = 7.00), post-hoc analysis with a Wilcoxon Signed-Rank test indicated that discomfort from granting access to internal storage was significantly higher than to contacts, microphone, and location, with small effect sizes.

    “Although participants reported a moderate to high degree of discomfort from granting data access permission to all types of data (Mdn = 7.00), post-hoc analysis with a Wilcoxon Signed-Rank test indicated that discomfort from granting access to internal storage was significantly higher than to contacts (Z = -3.35,p = 0.001), microphone (Z = -3.18,p = 0.001), and location (Z = -3.07,p = 0.002), with”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The Messaging app was perceived to be significantly more important than both the Puzzle and Weather Forecast apps.

    “Pairwise posthoc analysis with a Wilcoxon Signed-Ranks test indicated that the Messaging app was perceived to be significantly more important than both the Puzzle (Z = -17.218,p < 0.001, r = -0.840),and Weather Forecast apps (Z = -10.349,p = 0.000, r = -0.503).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In both studies, the intention to install the Messaging app was significantly greater than both the intention to install the Puzzle and the Weather app.

    “In both studies, post-hoc analysis with a Wilcoxon Signed-Ranks test indicated that the intention to install the Messaging app was significantly greater than both the intention to install the Puzzle (Z study1 = -11.677,p study1 <0.001, r study1 = -0.780;Z study2 = -11.397,p study2 <0.001, r study2 = -0.788)and the Weather app (Z study1 = -7.032,<0.001, r study1 = -0.470;Z study2 = -3.877,p study2”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In general, respondents reported a moderate to high level of smartphone privacy concern (Mdn study1 = 7.0; Mdn study2 = 7.0; on a scale from 0 to 10), without statistically significant differences between studies, U(N study1 =223; N study2 =210) = 22478.50, Z = -0.73, p = 0.467.

    “In general, respondents reported a moderate to high level of smartphone privacy concern (Mdn study1 = 7.0; Mdn study2 = 7.0; on a scale from 0 to 10), without statistically significant differences between studies, U(N study1 =223; N study2 =210) = 22478.50,Z = -0.73,p = 0.467.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The Friedman test with Study 2 data replicated Study 1 and rendered χ2(N = 208) = 226.694, p < 0.001.

    “Results of the Friedman test, with Study 2 data, replicated Study 1 and rendered χ 2 (N = 208) = 226.694,p < 0.001.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The intention to install the Puzzle and Messaging apps were not significantly different between studies.

    “Results of Mann-Whitney U tests (with Bonferroni adjusted p-value = 0.017) indicated that the intention to install the Puzzle, U(N study1 =226; N study2 =210) = 22192.50,Z = -1.214,p = 0.225, and the Messaging apps, U(N study1 =224; N study2 =210) = 22873.00,Z = -0.522,p = 0.602, were not significantly different between studies.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Participants differed in their understanding of what the Messaging app would require; the number of granted permissions was about equally distributed from one to four.

    “Participants differed yet in their understanding of what the Messaging app would require; the number of granted permissions was about equally distributed from one to four.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The level of importance given to data security varied depending on the number of data accessed and the type of app: the greater the number of data accessed, the greater the importance given to security in the Puzzle app.

    “One dimension, the level of importance given to data security, varied depending on the number of data accessed and the type of app: the greater the number of data accessed, the greater the importance given to security in the Puzzle app.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    There was no significant difference between studies, except for a higher intention to install the Weather app in Study 2.

    “As shown, there was no significant difference between studies, except for a higher intention to install the Weather app in Study 2.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In Study 1, the discomfort produced by the number of permissions requested increased privacy concerns and decreased the intention to install an app considered to be of little importance.

    “For example, in Study 1 we found that the discomfort produced by the number of permissions requested increased privacy concerns and decreased the intention to install an app considered to be of little importance”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    A condition of greater control did not affect the respondents' privacy concerns when using smartphones or the importance they gave to data security.

    “However, a condition of greater control did not affect the other aspects, the respondents' privacy concerns when using smartphones or the importance they gave to data security.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In Study 2, a Game app correlated to granting access to no data, while a Weather app correlated to the disclosure of location data only.

    “In Study 2, a Game app correlated to granting access to no data, while a Weather app correlated to the disclosure of location data only.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The type of app plays a central role in determining both the perceived benefit of installing an app and the level of concern for data privacy.

    “The results indicate that the type of app plays a central role in determining both the perceived benefit of installing an app and the level of concern for data privacy, supporting recent studies (Gu et al., 2022).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Gu et al., 2022

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingown data

    In Study 2, although there was no reduction in the level of importance attributed to data security, no correlation was observed between this dimension of privacy concerns and the intention to install any app, unlike Study 1.

    “An interesting finding was that in Study 2-although there was no reduction in the level of importance attributed to data security-we no longer observed a correlation between this dimension of privacy concerns and the intention to install any app (unlike Study 1).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductown data

    In Study 1, a mixed within-and-between-subjects design controlled for the number and type of data access permissions described as required by the application.

    “A mixed (within-and between-subjects) design controlled for the number and type of data access permissions that were described as being required by the application.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    The results suggest that the type of app plays a central role in determining both the perceived benefit of installing the app and the level of e-privacy concerns.

    “The results suggest that the type of app plays a central role in determining both the perceived benefit of installing the app and the level of e-privacy concerns.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    Privacy calculus theory postulates that the decision to disclose personal data is a function of the discount of the expected losses from the gains promised by the decision to disclose data, where individuals decide to provide information when potential gains surpass potential losses.

    “Privacy calculus theory postulates that the decision to disclose personal data is a function of the discount of the expected losses from the gains promised by the decision to disclose data (Dinev and Hart, 2006), where individuals decide to provide information when potential gains surpass potential losses (producing the privacy paradox).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Dinev and Hart, 2006 source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Previous research suggests that users are concerned with their privacy and even willing to pay premiums for apps that are less likely to request access to personal data.

    “Besides, previous research suggests that users are concerned with their privacy and even willing to pay premiums for apps that are less likely to request access to personal data (Egelman et al., 2013).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Egelman et al., 2013 source

  • May combine two claims.

    Premisecitation

    Polls conducted in the US found that Americans report avoiding apps due to privacy concerns; 54% of respondents reported that they decided not to install an app due to the amount of personal data needed to use it.

    “For example, polls conducted in the US found that Americans report avoiding apps due to privacy concerns; 54% of respondents reported that they decided not to install an app due to the amount of personal data needed to use it (Boyles et al., 2012).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Boyles et al., 2012

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    In Study 1, an experimental error led to a small number of respondents in the condition where the app requested access to internal storage, location, and contacts.

    “An experimental error led to the small number of respondents in the condition where the app requested access to internal storage, location, and contacts.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Greater control individuals have over the release of personal information is associated with a greater likelihood that they engage in risky behavior, that is, intentionally disclose the data.

    “Brandimarte et al. (2012) observed that the greater the control individuals have over the release of personal information, the greater the likelihood that they engage in risky behavior, that is, that they intentionally disclose the data.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Brandimarte et al. (2012) source

  • May combine two claims.

    Premisecitation

    Users are usually unaware of the amount of data collected and with whom the data are traded.

    “This practice is ethically questionable because, usually, users are unaware of the amount of data collected and with whom the data are traded (e.g., Acquisti and Grossklags, 2007;Felt et al., 2012).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Acquisti and Grossklags, 2007

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Felt et al., 2012

  • May combine two claims.

    Premisecitation

    Although users report increasing levels of concern about e-privacy, they tend to engage little in behaviors to protect their data.

    “Another problem is that research has shown that, although users report increasing levels of concern about e-privacy, they tend to engage little in behaviors to protect their data.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Barnes, 2006 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    In Study 1, the apps were presented in counterbalanced order across participants to minimize carryover effects.

    “The apps were presented in counterbalanced order across participants to minimize carryover effects.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Users want applications to perform specific tasks.

    “Previous research shows that users want applications to perform specific tasks (Egelman et al., 2013)”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Egelman et al., 2013 source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Data have been used in cybercrimes, to charge different prices for the same product depending on consumer characteristics, and to micro-regulate and persecute citizens in some countries.

    “Such data have been used, for example, in cybercrimes, to charge different prices for the same product depending on the characteristics of the consumer (price discrimination) and to micro-regulate and persecute citizens in some countries (e.g., Acquisti and Varian, 2005;Zuboff, 2015).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Acquisti and Varian, 2005 source
    • cites Zuboff, 2015 source

  • Classified as probably not a substantial claim.

    Premisecitation

    The literature indicates several factors that may influence e-privacy decisions, such as the degree of control over the release of personal information, benefits from data disclosure, and e-privacy concerns.

    “The literature indicates several factors that may influence eprivacy decisions, such as the degree of control over the release of personal information, benefits from data disclosure, and eprivacy concerns (e.g., Brandimarte et al., 2012;Felt et al., 2012;Egelman et al., 2013;Zafeiropoulou et al., 2013;Buck et al., 2014).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Brandimarte et al., 2012 source
    • cites Felt et al., 2012
    • cites Egelman et al., 2013 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    The variability across studies can be due to several factors, including that research on social network sites and e-commerce deals with decisions other than the decision to install apps.

    “The variability across studies can be due to several factors, including the fact that research on SNSs and e-commerce deals with decisions other than the decision to install apps.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Previous research found that most participants do not pay attention to permission during installation.

    “Previous research found that most participants do not pay attention to permission during installation (Egelman et al., 2013).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Egelman et al., 2013 source

  • May combine two claims.

    Premisecitation

    Factors influencing e-privacy decisions have usually been studied in isolation, and most studies focused on the context of social network sites and e-commerce.

    “So far, such factors have usually been studied in isolation and most studies focused on the context of social network sites (SNSs) and e-commerce (cf.Kokolakis, 2017).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Kokolakis, 2017 source

  • May combine two claims.

    Interpretationown data

    For apps with lower importance (Puzzle and Weather Forecast), privacy concerns were more likely to take precedence and predicted a decreased intention to install the app.

    “In contrast, for apps with lower importance (Puzzle and Weather Forecast), privacy concerns were more likely to take precedence and predicted a decreased intention to install the app.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    Results concerning the Messaging app were obtained after removing 14 outliers; removal did not change significant p-values and allowed meeting assumptions of multiple regression (absence of significant outliers and homoscedasticity).

    “Results concerning the Messaging app were obtained after removing 14 outliers (case IDs 7 , 26, 46, 60, 77, 81, 97, 150, 159, 170, 173, 183, 190, 210).The removal of outliers did not change the results (i.e., significant p values), however, it allowed to meet the assumptions of multiple regression analysis-absence of significant outliers and homoscedasticity.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Gu et al. (2022) assume that hedonic apps elicit imagery processing, in which attention is governed only by obvious (or explicitly stated) privacy cues.

    “Hedonic apps, in contrast, are assumed to elicit imagery processing, in which attention is governed only by obvious (or explicitly stated) privacy cues.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Gu et al., 2022

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim.

    Premisecitation

    A study published recently (Gu et al., 2022) argued that not only the degree of benefits-but the nature of the benefits-produced by an app would influence the installation decision-making process (or the information processing mode).

    “However, a study published recently (Gu et al., 2022) argued that not only the degree of benefits-but the nature of the benefits-produced by an app would influence the installation decision-making process (or the information processing mode, in the author's terminology).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Gu et al., 2022

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim.

    Premisecitation

    Gu et al. (2022) argue that utilitarian apps trigger an analytical information processing mode in which individuals examine all available clues to form a privacy concern before balancing risks and benefits.

    “They argue that utilitarian apps trigger an analytical information processing mode in which individuals examine all available clues to form a privacy concern before balancing risks and benefits.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Gu et al., 2022

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    Results concerning the Weather App in Study 2 were obtained after removing 18 outliers; removal did not change significant p-values and allowed meeting assumptions of multiple regression.

    “Results concerning the Weather App were obtained after removing 18 outliers (case IDs 266 , 276, 302, 306, 321, 328, 342, 363, 379, 382, 441, 264, 281, 286, 307, 308, 320, 332, 380, 392, 438).The removal of outliers did not change significant p-values, however, it allowed meeting the assumptions of multiple regression analysis-absence of significant outliers and homoscedasticity.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Privacy concerns tend to be less influential on users' decisions about whether to install an app than benefits, especially when individuals have greater control over which data can be accessed by the app.

    “However, privacy concerns tend to be particularly less influential on users' decisions about whether or not to install an app, compared to benefits (Liu et al., 2015)-especially when individuals have greater control over which data can be accessed by the app.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Liu et al., 2015

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Interpretationown data

    Despite being statistically significant, the difference in Weather app installation intention between studies was small and did not alter data trends across apps.

    “Despite statistically significant, the difference was small and did not alter data trends across apps.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Smartphone literacy is a type of privacy literacy that includes, among others, knowledge about the type of data needed for an app to function.

    “Smartphone literacy is a type of privacy literacy that includes, among others, knowledge about the type of data needed for an app to function (Ketelaar and van Balen, 2018).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Ketelaar and van Balen, 2018 source

  • May combine two claims.

    Interpretationown data

    The present study demonstrates that e-privacy concerns are a sensitive measure and that both the levels of privacy concerns and their correlation with the decision to install an app vary depending on various parameters.

    “Second, the present study demonstrates that e-privacy concerns are a sensitive measure and that both the levels of privacy concerns and their correlation with the decision to install an app vary depending on various parameters.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    The variation in correlations regarding the different facets suggests that the use of different instruments to measure privacy concerns can contribute to variations in terms of conclusions across studies, sometimes showing the so-called privacy paradox, sometimes contradicting its existence.

    “The variation in correlations regarding the different facets suggests that the use of different instruments to measure privacy concerns can contribute to variations in terms of conclusions across studies, sometimes showing the so-called privacy paradox, sometimes contradicting its existence (e.g., Acquisti and Grossklags, 2003;Egelman et al., 2013).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Acquisti and Grossklags, 2003

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Egelman et al., 2013 source

  • May combine two claims.

    Interpretationown data

    These results supported hypotheses 2 and 3 (H2 e H3) that these other dimensions of e-privacy concerns may reduce installation intention.

    “These results supported hypotheses 2 and 3 (H 2 e H 3 ) that these other dimensions of e-privacy concerns may reduce installation intention.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    Our results contrast with the literature and suggest that our respondents had partial knowledge about the type of data needed for an app to function.

    “Our results contrast with the literature and suggest that our respondents had partial knowledge about the type of data needed for an app to function.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    Overall, the results suggested that app-specific privacy concerns (not general concerns regarding smartphone use) can have a negative effect on app installation intention under two particular conditions: when the user does not have fine-grained explicit control over which data access permissions they may grant (Study 1) and second, when the app is considered of low to moderate importance (Puzzle и

    “Overall, the results suggested that app-specific privacy concerns (not general concerns regarding smartphone use) can have a negative effect on app installation intention under two particular conditions: when the user does not have fine-grained explicit control over which data access permissions they may grant (Study 1) and second, when the app is considered of low to moderate importance (Puzzle”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    Results concerning the Messaging App in Study 2 were obtained after removing 4 outliers; removal did not change significant p-values but allowed meeting assumptions required by multiple regression analysis.

    “Results concerning the Messaging App were obtained after removing 4 outliers (case IDs 379, 290, 363, and 388).The removal of outliers did not change significant p-values, but it allowed meeting the assumptions required by multiple regression analysis.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    From our data from Study 2, it can be concluded that granting access to the location might be seen by the user as taking advantage of desirable location-sensitive features of a Weather app, as well as photo storage can be an advantage of a Messaging app.

    “For example, from our data from Study 2, it can be concluded that granting access to the location might be seen by the user as taking advantage of desirable location-sensitive features of a Weather app (Egelman et al., 2013;Zafeiropoulou et al., 2013), as well as photo storage can be an advantage of a Messaging app.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Egelman et al., 2013 source
    • cites Zafeiropoulou et al., 2013 source

  • Classified as probably not a substantial claim.

    Premisecitation

    Previous research suggested that most people are unaware of which type of data is often needed for specific apps to work.

    “Previous research suggested that most people are unaware of which type of data is often needed for specific apps to work (e.g., Felt et al., 2012;Buck et al., 2014).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Felt et al., 2012
    • cites Buck et al., 2014

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Participants provided their written informed consent to participate in this study.

    “The patients/participants provided their written informed consent to participate in this study.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Consumers do not make choices as entirely free agents; their choices are highly influenced by contextual factors.

    “In general, our results suggest that it is important for policymakers to consider that consumers do not make choices as entirely free agents; their choices are highly influenced by contextual factors.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    The chosen high-importance app had functionalities that might require access to multiple data for functional purposes.

    “On the other hand, the chosen high-importance app was such with functionalities that might require access to multiple data for functional purposes.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    The study used fictitious apps and presented them in a questionnaire.

    “Finally, we used fictitious apps and presented them in a questionnaire.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationcitation

    Higher control can reduce certain types of privacy concerns, partially corroborating hypothesis 6.

    “These results support that higher control can reduce certain types of privacy concerns (Hoadley et al., 2010)-partially corroborating hypothesis 6 (H 6 ).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Hoadley et al., 2010 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The methods were acknowledged by the Norwegian Center for Research Data (NSD; notification form 695287).

    “The methods were acknowledged by the Norwegian Center for Research Data (NSD; notification form 695287).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The study was registered in the university system of risk and compliance in the processing of personal data in research projects (RETTE).

    “The study was registered in the university system of risk and compliance in the processing of personal data in research projects (RETTE).”
    Quote in the paperNot a substantial claim
66 of 66

  • The authors state this as an assumption.

    The four data sensitivity items reflect a common latent discomfort factor.

    “It was presumed that the four items reflected a common latent discomfort factor.”

  • The authors state this as an assumption.

    The question of how the degree of control over the release of personal information affects the intention to install a smartphone app has not received sufficient attention in the e-privacy literature.

    “To our knowledge, the question of how the degree of control over the release of personal information affects the intention to install a smartphone app has not received sufficient attention in the e-privacy literature.”

  • The authors state this as an assumption.

    The means by which increased control would increase the intention of data disclosure remains to be further explained.

    “The means by which increased control would increase the intention of data disclosure remains to be further explained.”

  • The authors state this as an assumption.

    The number of studies on the interactions between control and privacy concerns in digital contexts is still limited.

    “However, the number of studies on the interactions between control and privacy concerns in digital contexts is still limited.”

  • The paper relies on this without support.

    Some studies about the decision to install apps and allow them to access data produced mixed results regarding the privacy paradox.

    “However, some studies about the decision to install apps and allow them to access data produced mixed results regarding the privacy paradox.”

  • The paper relies on this without support.

    Most applications require for their installation some extent of self-disclosure of private information.

    “Most applications (also referred to as apps) require for their installation some extent of self-disclosure of private information.”

  • The paper relies on this without support.

    The three fictional apps (Puzzle Game, Messaging, and Weather Forecast) are very common, generally installed by most people, and have different levels of functionality importance.

    “These apps were chosen because they are very common, generally installed by most people, and we expected they would have different levels of functionality importance.”

  • The paper relies on this without support.

    The fictional apps Puzzle Game and Weather Forecast can be defined as hedonic and utilitarian, respectively.

    “In our study, the fictional apps Puzzle Game and Weather Forecast could be defined as hedonic and utilitarian, respectively.”

  • The paper relies on this without support.

    Messaging apps can hold both hedonic and utilitarian functions.

    “Messaging apps can hold both hedonic and utilitarian functions.”

  • The paper relies on this without support.

    The psychological mechanisms through which a condition of increased control affects the decision-making process when installing apps remain to be explained.

    “The psychological mechanisms through which a condition of increased control affects the decision-making process when installing apps remain to be explained.”

  • The authors state this as an assumption.

    Benefits tend to dominate installation decisions.

    “As previously discussed, we presume that this is because the benefits tend to dominate installation decisions.”

  • The paper relies on this without support.

    In the real environment, apps with more functionalities tend to be more popular but also need more data to function.

    “In practice, this correspondence between the nature of the benefits, the degree of importance, and data requirements is also observed in the real environment: Apps with more functionalities tend to be more popular, but also need more data to function.”
12 of 12

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Consistentz = -2.857reported p = .004 · recomputed p = .004

    “,Z = -2.857,p = .004”

  • Consistentz = -3.09reported p = .002 · recomputed p = .002

    “,Z = -3.090,p = .002”

  • Consistentz = -0.73reported p = .467 · recomputed p = .465

    “,Z = -0.73,p = .467”

  • Consistentz = -0.35reported p = .721 · recomputed p = .726

    “,Z = -0.35,p = .721”

    The paper states an exact rank test

  • Consistentz = -0.03reported p = .979 · recomputed p = .976

    “,Z = -0.03,p = .979”

  • Consistentz = -2.09reported p = .037 · recomputed p = .037

    “,Z = -2.09,p = .037”

  • Consistentz = -2.43reported p = .015 · recomputed p = .015

    “,Z = -2.43,p = .015”

  • Consistentz = -3.35reported p = .001 · recomputed p < .001

    “(Z = -3.35,p = .001”

  • Consistentz = -3.18reported p = .001 · recomputed p = .001

    “(Z = -3.18,p = .001”

  • Consistentz = -3.07reported p = .002 · recomputed p = .002

    “(Z = -3.07,p = .002”

  • Consistentz = -0.204reported p = .839 · recomputed p = .838

    “,Z = -0.204,p = .839”

  • Consistentz = -0.315reported p = .753 · recomputed p = .753

    “,Z = -0.315,p = .753”

  • Consistentz = -1.249reported p = .212 · recomputed p = .212

    “,Z = -1.249,p = .212”

  • Consistentz = -17.218reported p < .001 · recomputed p < .001

    “(Z = -17.218,p < .001”

  • Consistentz = -10.349reported p = 0 · recomputed p < .001

    “(Z = -10.349,p = .000”

  • Consistentz = -16.283reported p < .001 · recomputed p < .001

    “(Z = -16.283,p < .001”

  • Consistentz = -8.046reported p < .001 · recomputed p < .001

    “(Z = -8.046,p < .001”

  • Consistentz = -8.38reported p < .001 · recomputed p < .001

    “(Z = -8.380,p < .001”

  • Consistentz = -8.407reported p < .001 · recomputed p < .001

    “(Z = -8.407,p < .001”

  • Consistentz = -7.942reported p < .001 · recomputed p < .001

    “(Z = -7.942,p < .001”

  • Consistentz = -1.214reported p = .225 · recomputed p = .225

    “,Z = -1.214,p = .225”

  • Consistentz = -0.522reported p = .602 · recomputed p = .602

    “,Z = -0.522,p = .602”

  • Consistentz = -4.074reported p < .001 · recomputed p < .001

    “,Z = -4.074,p < .001”

  • Consistentz = -0.052reported p = .959 · recomputed p = .959

    “,Z = -0.052,p = .959”

  • Consistentz = -0.545reported p = .586 · recomputed p = .586

    “,Z = -0.545,p = .586”

  • Consistentz = -2.945reported p = .003 · recomputed p = .003

    “,Z = -2.945,p = .003”

26 of 26

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Messaging data-security median differs between reports

    Major

    The results give the Study 1 Messaging app a median data-security importance of 10.0, while the later figure text gives it a median of 9.0.

  • Friedman test has the wrong degrees of freedom for four data types

    Minor

    The sensitivity comparison covers four data types, for which a Friedman test has three degrees of freedom, not the reported two.

  • Puzzle regression coefficient falls outside its confidence interval

    Minor

    The reported smartphone-privacy-concern coefficient is −1.101, but its reported 95% confidence interval runs from −0.209 to 0.019. The point estimate cannot fall outside its own interval.

  • Weather regression footnote lists 21 IDs, not 18

    Minor

    The footnote says 18 outliers were removed, but its parenthesized list contains 21 case IDs.

  • Installation-intention scale has two reported lower bounds

    Minor

    The measure is defined as running from 1 to 7, but the Study 1 results describe it as running from 0 to 7.

5 of 5

Cites

Flagged first, then by citations

Research works printed but not linked: 2

  1. Acquisti, 2003, Losses, gains, and hyperbolic discounting: An experimental approach to information security attitudes and behavior,, 2nd Annual Workshop on Economics and Information Security, 1
  2. Boyles, 2012, Privacy and Data Management on Mobile Devices

Cited by

Sorted by citations

Related

Authors
Ayres-Pereira, V., Pirrone, A., Korbmacher, M., Tjostheim, I., & Böhm, G.
Affiliations
  • University of Bergen (NO)
  • University of Inland Norway (NO)
ORCID
Venue
Frontiers in Computer Science
Publisher
Frontiers Media SA
Published
2022-09-09
Volume
4
ISSN
2624-9898 (ISSN-L)
DOI
10.3389/fcomp.2022.986138
OpenAlex
W4295008965
Semantic Scholar
3b0fc42e5e22e52cf136a991bb08243ddb0b5dec
Cited by
4
References
39
Citations by year
2026: 1 · 2025: 1 · 2024: 1 · 2023: 1
FWCI
1.06
Type
journal article
Language
en
Open access
Licence
Locations
  • Frontiers in Computer Science (publishedVersion, cc-by)
  • London School of Economics and Political Science Research Online (London School of Economics and Political Science) (acceptedVersion)
  • London School of Economics and Political Science Theses Online (London School of Economics and Political Science) (acceptedVersion)
  • NORA - Norwegian Open Research Archives (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Norges Forskningsråd (270969)
Topics
  • Social Sciences › Sociology and Political Science › Privacy, Security, and Data Protection
  • Social Sciences › Decision Sciences › Information Systems and Management › Technology Adoption and User Behaviour
  • Physical Sciences › Computer Science › Human-Computer Interaction › Innovative Human-Technology Interaction
Keywords
mobile application, data access permissions, user control, privacy paradox, data protection by design

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-19.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-19.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review