Overview
Phone calls, despite facing competition from digital communication methods, remain a prevalent means of interaction in both personal and professional settings. However, many individuals, particularly younger generations, have expressed aversion to phone calls, especially from unknown numbers.
Some even claim that such calls trigger anxiety, as evidenced by the growing number of memes in pop culture that caricature the supposed issue. But as interesting and important as it is, research to-date has been largely silent on the matter.
To investigate this intriguing hypothesis and understand its implications for different age groups, we conducted a simple experiment testing whether phone calls from known and unknown numbers can actually trigger anxiety.
The Experiment
For our experiment, we recruited 450 participants from Amazon MTurk to take part in an image/writing activity about their phones. Unbeknownst to participants, the activity was actually a randomized controlled trial.
Each participant was randomly assigned to one of the three writing tasks, as follows: (a) describe your phone, (b) describe what you would do if you received a call from someone you know, or (c) describe what you would do if you received a call from an unknown number. Each writing activity was accompanied by a corresponding image, a screenshot of an iPhone, as depicted below.
On the next page, participants were asked a set of survey questions to measure their anxiety on a 1-4 scale. To test whether certain groups are more affected by phone anxiety than others, participants also completed a series of demographic survey questions asking about their age, gender, income, extroversion vs. introversion, and perceived time constraints.
Results
Our findings revealed that calls from unknown numbers did, in fact, cause anxiety, particularly among younger people. Calls from known sources had a limited impact on anxiety overall, but Millennials and Gen Z still displayed increased anxiety in response to such calls. No significant interactions were observed for introverts, low-income individuals, or those with time constraints. However, specific negative relationships with anxiety were identified for income, age, and extroversion.
Known Callers
On average, calls from known numbers led to a slight 7% increase in anxiety relative to the control condition (1.76 vs. 1.65), but this difference was not statistically significant (p = 0.135). However, among Millennials and Gen Z, anxiety significantly increased by 15% (1.81 vs. 1.57, p=0.013). This equated to a medium standardized effect size, an increase of approximately 0.40 standard deviations. Other demographic groups did not exhibit such differences in anxiety.
Unknown Callers
Calls from unknown numbers increased anxiety even more, by 14.5% compared to the control condition (1.65 vs. 1.89, p=0.004). In particular, Millennials and Gen Z experienced a substantial 25% increase in anxiety in the unknown call condition (1.96 vs. 1.57, p<0.001), over half a standard deviation (d = 0.57). Although introverts and time-constrained individuals showed a 10% and 15% increase, respectively, these effects were not statistically significant (p’s = 0.11).
Non-Causal Relationships
Several variables were found to be correlated with anxiety, such as income, age, and extroversion, but causality could not be established. Introversion, low-income, and time constraints were also positively correlated with anxiety. Notably, no significant relationship between anxiety and Millennials was observed.
Conclusion
Our experiment provides evidence that phone call anxiety is a real phenomenon, particularly affecting younger generations and in response to unknown callers. So, if you're considering calling a Millennial or Gen Z, consider giving them a heads up. Or perhaps just send them an email or text.
Methods Note
Statistical analyses included two-sample t-tests to identify significant differences in anxiety between each condition. Effect sizes were measured using Cohen’s d. OLS regression analyses with interaction terms were employed to explore differences between demographic groups. P-values less than 0.05 were considered statistically significant. For more comprehensive details, including pre-registration, survey materials, and data, please submit your request here.