Lock of Ages

Research conducted by Lina Qiu, Alexander De Luca, Ildar Muslukhov, Konstantin Beznosov

Abstract

In this study, we explore how age influences smartphone authentication use. To do this, we performed a two-month-long field study with a diverse pool of North American participants (N = 137). Among other factors, we examined how the authentication usage of our participants correlates with their age. Our results suggest that for many usage patterns age makes a significant difference. For instance, older participants interacted with their devices significantly less frequently, and for a significantly shorter amount of time per day. As such, authentication took up a much bigger proportion of their overall device use and is thus more likely to cause annoyance.

a REPRESENTATIVE STUDY SAMPLE

Of the 137 participants whose data we included in the analysis, 123 (89.8%) were from the US, 81 (59.1%) were female and 56 (40.9%) were male. Ages ranged from 19 to 63 years, with a mean age of 40 and median age of 38 (SD = 12.5). Participants had diverse education levels, with around 66% of the participants having either a high school diploma or a Bachelor’s degree. Occupations and salaries of participants also varied, as shown in the following table.

.tg {border-collapse:collapse;border-spacing:0;border-color:#ccc;}
.tg td{font-family:Arial, sans-serif;font-size:14px;padding:0px 20px;border-style:solid;border-width:0px;overflow:hidden;word-break:normal;border-top-width:1px;border-bottom-width:1px;border-color:#ccc;color:#333;background-color:#fff;}
.tg th{font-family:Arial, sans-serif;font-size:14px;font-weight:normal;padding:0px 20px;border-style:solid;border-width:0px;overflow:hidden;word-break:normal;border-top-width:1px;border-bottom-width:1px;border-color:#ccc;color:#333;background-color:#f0f0f0;}
.tg .tg-0ord{text-align:right}
.tg .tg-l711{border-color:inherit}
.tg .tg-us36{border-color:inherit;vertical-align:top}
.tg .tg-lqy6{text-align:right;vertical-align:top}
.tg .tg-dvpl{border-color:inherit;text-align:right;vertical-align:top}
.tg .tg-yw4l{vertical-align:top}

Parameter Property # of participants
Residence US 123
Canada 14
Gender Females 81
Males 56
Age 19-24 15
25-34 42
35-44 26
45-54 31
55-63 23
Education Less than High School 1
High School 56
Professional School 23
University (Bachelor’s) 35
Master or PhD 17
Other 5
Occupation Managers 9
Professionals 27
Clerical Support Workers 16
Service and Sales Workers 18
Craft and Trades Workers 6
Machine Operators 1
Elementary Occupations 3
Students 14
Self-employed 3
Unemployed/Retired/Disabled 28
Salary (US, N=123) Less than $30,000 37
$30,000 – $49,999 17
$50,000 – $74,999 34
$75,000 – $99,999 17
$100,000+ 17
Prefer not to specify 1
Salary (Canada, N=14) Less than $30,000 5
$30,000 – $49,999 4
$50,000 – $74,999 1
$75,000 – $99,999 0
$100,000+ 0
Prefer not to specify 4

 

AGE MAKES A DIFFERENCE

Fig. 1 shows the distribution of unlocking mechanisms for participants among the different age groups. We conducted significance tests on usage statistics, including session lengths, the number of sessions, daily usage lengths, error rates, and the number of auto/manual locks among the predefined age groups and different unlocking mechanisms. We consider all of these factors relevant for authentication behavior as they are potentially influenced by it or, opposed to this, influence (choice of) authentication. For example, if a group has more sessions per day, they are exposed to authentication more often.

Fig. 1 – Unlocking mechanism distribution among age groups. Numbers below each age group represent the total number of participants from that group.

First of all, our data shows that smartphone usage patterns indeed differ based on age.
For instance, we found that participants in older age groups interacted with their devices significantly less frequently than younger groups (Fig. 2). They also interacted with the devices for a significantly smaller amount of time each day (Fig. 3). This means two things: 1) they are exposed to the authentication mechanism less frequently than younger groups and 2) authentication takes up a bigger portion of the overall interaction with their devices.

Fig. 2 – Average number of sessions per age group, N = 137. The red squares represent means.
Fig. 3 – Average daily usage length per age group (minutes), N = 136. The red squares represent means.

In addition, age also significantly correlated with the locking behaviour of our participants. We found that older groups (specifically those from the “55-63” group) relied more on autolocks than the other groups (Fig. 4). Overall, the “55-63” participants were more than three times as likely to use the autolock feature than participants from the “19-24” and “35-44” groups, and about two times more likely than those from the “25-34” and “45-54” groups. As the autolock feature presents a trade-off between security and usability, future research into the popularity of autolock among older users and their awareness of respective security implications is necessary.

Fig4. – Distribution of locking types among age groups, N = 46. Numbers below each age group represent the total number of participants from that group.