Source-linked AI summary
Mobile social media usage and academic performance
Fausto Giunchiglia, Mattia Zeni, Elisa Gobbi, Enrico Bignotti, Ivano Bison
TL;DR
Existing research links social-media and smartphone use with academic performance, but prior evidence often relies on surveys or focuses on addictedness rather than logged activity. This paper defines usage and performance measures, combines smartphone logs with time diaries, and compares students’ social-media behavior with GPA and credits. Results show negative associations that vary by academic activity, while the authors identify limits from the short observation window and sample size.
Problem
Prior studies either focus on social-media or smartphone addictedness or use surveys that provide only approximate usage estimates when relating behavior to academic performance.
Method
The study defines social-media usage and academic-performance measures, then combines smartphone app logs with time diaries to compare activity-specific use with GPA and CFU.
Results
The study finds negative correlations between social-media use and academic performance, with different patterns depending on the activity.
Takeaways & Limitations
Controlling social-media use during studying or lessons is more informative for its association with academic performance than considering general use without activity distinctions.
Takeaways & Limitations
The study is limited by a two-week observation window and a relatively small student sample, with 67 students included in the final analysis.
Abstract
from arXiv · showhide
Among the general population, students are especially sensitive to social media and smartphones because of their pervasiveness. Several studies have shown that there is a negative correlation between social media and academic performance since they can lead to behaviors that hurt students' careers, e.g., addictedness. However, these studies either focus on smartphones and social media addictedness or rely on surveys, which only provide approximate estimates. We propose to bridge this gap by i) parametrizing social media usage and academic performance, and ii) combining smartphones and time diaries to keep track of users' activities and their smartphone interaction. We apply our solution on the 72 students participating in the SmartUnitn project, which investigates students' time management and their academic performance. By analyzing the logs of social media apps on students' smartphones and by comparing them to students' credits and grades, we can provide a quantitative and qualitative estimate of negative and positive correlations. Our results show the negative impact of social media usage, distinguishing different influence patterns of social media on academic activities and also underline the need to control the smartphone usage in academic settings.
1 Introduction
Social media and smartphone use are associated with poorer academic performance, but prior work relies largely on addiction measures or approximate survey reports. The paper proposes combining usage metrics, smartphone logs, and time diaries to quantify these relationships among students.
- Prior studies report a negative association between social media or smartphone use and academic performance.
- Survey-based studies can misrepresent smartphone behavior because users inaccurately estimate usage time and checking frequency.Reported usage can be underestimated or overreported depending on the study.
- The proposed approach defines social-media usage metrics and combines smartphone tracking with time diaries.This coupling is intended to connect reported activities with actual app logs.
- Using SmartUnitn data, the study extracts social-media use during studying and classes and compares it with GPA and credits.
- The paper reports a negative correlation between social-media use and academic performance, with patterns varying by activity.
2 Literature Review and Hypothesis
Earlier research links social-media and smartphone use with academic outcomes, but its evidence is divided between surveys and smartphone studies focused on addictedness. This literature motivates correlating logged usage patterns with academic performance.
- Smartphone studies of students commonly classify potential addicts and non-addicts using the Smartphone Addiction Scale.The scale contains ten six-point Likert-type items.
- The Copenhagen Networks Study combines smartphone data with face-to-face interactions and Facebook usage but does not relate those data to academic performance.
- The extended-version note identifies this paper as an expanded version of work presented at SocInfo 2017.
- SmartGPA did not consider social-media usage when analyzing students’ academic careers, although such information was collected.
- Survey-based correlations may approximate actual usage because respondents must recall activities and summarize them with aggregate estimates.
- Smartphone-use research either studies addictedness alone or does not correlate usage patterns with academic performance.
3 Social media usage and academic performance
The paper operationalizes social-media use through app categories and interaction metrics, and academic performance through GPA and CFU. Smartphones and time diaries jointly capture activity, context, and usage behavior while reducing reliance on averaged self-reports.
- The proposed solution defines smartphone social-media usage metrics and combines time diaries with smartphone data to correlate usage with academic performance.
- Social-media applications are grouped into social network sites, instant messaging applications, and browsers.Examples include Facebook, WhatsApp, and Chrome.
- The three application categories distinguish usage patterns and allow different relationships with student performance to be represented.Browsers may support either academic or non-academic activity.
- Social-media interaction is represented by session occurrences, average session duration, and average inactivity time between app interactions.
- GPA measures the qualitative dimension of performance, whereas CFU measures quantitative academic progress.
- Time diaries record activities, locations, and social relations, while smartphones administer entries and collect application and sensor data.The combined records can match reported activity, location, and social relation with smartphone status.
4 Methods
The study validates its smartphone-and-time-diary approach using SmartUnitn data from University of Trento students, combining behavioral, demographic, and academic records. The analyzed sample is smaller than the initial cohort because some application logs and one credit record were unusable.
- SmartUnitn investigates how students’ time allocations affect academic performance using smartphone-based behavioral data.
- The i-Log application collects smartphone sensor and application data and administers time-diary questions about activities, locations, and social relations.Time-diary questions were administered every 30 minutes.
- The project selected 72 University of Trento students who met survey, attendance, and Android-device criteria.
- Participants consented to data handling, received privacy information, and were covered by university ethical approval.
- The project lasted two weeks, with time-diary reporting during the first week and passive data collection during both weeks.
- The dataset merges smartphone behavioral data, socio-demographic characteristics, and academic performance records.The dataset contains 110 Gb of data.
- The overall analysis used 67 students rather than 72 because of incompatible smartphone logs and one unusable CFU outlier.
5 Quantifying social media usage
The paper quantifies social-media usage through temporal distributions and average parameter values across studying and lesson attendance. These analyses reveal broadly distributed use, with distinct duration patterns across the day.
- Two dimensions capture usage behavior: temporal distribution identifies patterns, while average values summarize mean usage.The parameters analyzed are S̄, D̄, and Ī.
- 5.1 Distribution: The temporal distributions use one-hour slots arranged by hour of day and day of week, with darker shades indicating more students using apps.White slots indicate no students were studying or attending classes during that interval.
- 5.1 Distribution: During studying, social-media usage is mostly uniform across days and hours, while duration rises around midnight–3AM and 7–9AM and falls during weekday 9AM–6PM.Usage therefore occurs while studying across different times, but duration varies across daily periods.
- 5.1 Distribution: During lessons, social-media use also largely ignores time intervals, with Friday showing more students checking apps and longer connection early in the morning.Duration later returns to similar levels as the day progresses, especially by Friday afternoon.
- 5.1 Distribution: Overall, the parameters are broadly distributed throughout the week during both studying and lessons, consistent with multitasking behavior.The paper interprets this temporal pattern as multitasking permeating students’ everyday lives.
- 5.2 Average mean: Average-usage tables organize rows by activity, app category, and parameter, and columns by mean, standard deviation, and student count.The activities include general use, studying, and attending lessons; app categories include all apps, social-media apps, SNS, IM, and Web.
6 Social media usage vs GPA and CFU
The analysis relates social-media usage parameters to CFU and GPA using Pearson correlations, revealing activity-, app-, and demographic-specific patterns. More frequent or longer usage generally aligns with lower performance, whereas longer inactive periods tend to align with higher performance.
- Correlation framework: Pearson correlations relate average usage, duration, and inactivity parameters to CFU, GPA, and sociodemographic variables.Rows cover application types and activities; columns cover gender, faculty, their combination, GPA, and CFU.
- Significant correlations: During general activities, average usage has 9 significant correlations and duration has 6, while inactivity has 2; SNS usage alone reaches 7 average-usage values, including 4 with p < .01.These counts are reported in Table 4a.
- Significant correlations: 28 significant-value occurrences appear for studying versus 21 for attending lessons, with similar distributions across significance levels.The counts summarize Tables 4b and 4c.
- Activity-specific patterns: For studying and lessons combined, inactivity has 25 significant values, followed by duration with 18 and average usage with 6.Within social media, IM is most associated with studying correlations, whereas SNS is most associated with lesson correlations, especially duration.
- Activity-specific patterns: While studying, longer IM usage is identified as most harmful, whereas longer IM inactivity aligns with higher performance; during lessons, SNS duration and checking align negatively with performance.The study therefore identifies different influence patterns across applications and academic activities.
- CFU and GPA: 33 significant correlations are reported for both CFU and GPA, with stronger average influence for scientific than humanities students, 7 versus 4.The authors also identify male scientific students and female humanities students as the groups most at risk of decreased performance.
7 Conclusions and Limitations
The study couples smartphone logs with time diaries to identify behavioral patterns associated with better or worse academic performance. It finds that activity-specific social-media use is informative, while the study is constrained by its short observation window and sample size.
- Conclusions: Coupling smartphone logs with time diaries identified behavioral patterns that could hurt or improve academic performance.Examples included messaging while studying or staying on social-networking services in class, versus limiting instant-messaging use.
- Conclusions: Controlling social-media use during studying or class attendance was more informative than considering general use without distinguishing activities or faculties.
- Limitations: The study’s observation window was limited to two weeks.The authors note that this is shorter than windows used in several comparison studies, though longer than typical sociology time-diary recording.
- Limitations: The sample size was smaller than those of several sociology studies, although larger than some computational social-science studies.The authors identify both the sample and observation window as limitations to be addressed in a later SmartUnitn iteration.