How Technology Use Differs in Families from Different Racial and SocioEconomic Backgrounds
This blog summarizes a paper about practices of technology use in White and Asian Indian families in the U.S. This paper will be presented at the 22nd ACM Conference on Computer-Supported Cooperative Work and Social Computing, a top venue for social computing scholarship. It will also be published in the journal Proceedings of the ACM (PACM).
Have you ever thought about how one’s identity, as defined by race and social class, influence technology use in a family? How parents use or avoid its use around children, share technology with children, and mediate the technology use of their children?
We see our work as a crucial preliminary step towards the broad agenda of investigating technological practices of people from different backgrounds in the U.S. by investigating those of White families (racial group currently in majority in the U.S.) and Asian Indian families (fastest growing racial minority in the U.S.) who are from different socioeconomic backgrounds. To this end, we conducted a diary study for eight weeks wherein 40 participants logged in-situ instances of technology use by, with, and around children in combination with participating in semi-structured interviews after every four weeks (two in total).
All of the Asian Indian participants in the study were first-generation immigrants to U.S, although, at the time of study, the total duration for which these participants had been in U.S. varied from six to 25 years, and therefore four out of nine of the working-class and six out of 13 of the middle-class Asian Indian participants were first-generation U.S. citizens as well. Furthermore, we specifically explored families’ practices in respect to two devices — smart phones and smart speakers with voice assistant. These devices were chosen because, first, both types are widely adopted in the U.S. — 77% of the U.S. population own a smart phone, and as of first quarter of 2019, i.e., within four years of its U.S. wide release, 31% of U.S. households already own a smart speaker. Second, smart phones are designed to be a personal devices, with touch as the primary mode of interaction, whereas smart speakers are inherently sharable, with voice being the main mode of interaction; these differences might affect the technology practices of families. All of the selected participants had to own atleast one device of each type.
While we acknowledge that social mobility exists and that technology practices cannot be detached from the participants’ backgrounds or the broader milieus in which they live, we also want to draw attention to the structural, racial, and class differences that emerged in our analysis in regards to the preferences for technology use in families. Therefore, our work argues for the inclusion of heterogeneous voices to develop and analyze technology, as they can act (a) as an analytical lens to better reveal how adults and children from different backgrounds are affected by technology materially and emotionally, and (b) as a means to avoid the insinuation of stereotyping and racial biases into designs. Here we list few of main findings and corresponding design recommendations.
(a) The Influence of Conversational Smart Speakers on Children
All of the parents reported that children, specifically those under the age of 10, were heavily attracted to the use of smart speakers, and several children as young as four years old could successfully give commands and respond to them. We posit that this attraction could be partly due to the human-like attributes that they afford or which are attributed to them by users (e.g., the ability to converse and to respond to their name [e.g., “Hey, Google,” “Hey, Alexa”] and the possession of a personality [e.g., the ability to generate humorous responses] and intelligence). However, in our findings, the most prominent reasons for use, and parents’ opinions and concerns about children’s interacting with and using conversational devices, were different.
While, the Asian Indian parents appreciated their children’s use of the speakers to learn or to ask trivia or questions on a broad range of topics, including math, english, science and technology, and history, the Asian Indians who are from the middle classes were concerned about their children’s use of these functionalities to complete their homework. One reason for this could lie in the fact that Asian Indian families consider education and associated homework to be of the highest value to their children, and the smart speakers were proving to hinder their children from acquiring knew skills through practice (i.e., homework). The White parents were more likely to use the devices for fun activities during family time or for asking trivia or factual questions that their children might be interested in. Based on in-home interviews with seven families. In our study, while none of the Asian Indian parents expressed such concerns, White parents were apprehensive about the impact that voice-based interactions could have on their children’s behavior and conversational style in daily life. Motivated by these challenges and concerns, one of our design recommendation is:
Conversational In-home Tutor: To leverage parents’ and children’s inclination to use smart speakers for learning, we propose that they provide a “tutor-mode” that can help children to understand concepts and practice using them, instead of providing direct answers. It could be activated by a parent’s or a child’s voice, and a few functionalities that this mode could support might include (a) short tutorial sessions that could explain the concepts of specific subjects through features that have proven to be beneficial, such as storytelling or displaying illustrations on a connected device (e.g., a smart phone or tablet) or on the display screen of the smart speaker itself (e.g., Google Hub) and (b) interactive simulation through personalized, adaptive, open-ended questions (e.g., sequences of questions and answers that provide incremental information and test a child’s level of understanding); these would be used to construct learning experiences related to a child’s interest and to leverage a child’s inclination to use the device to ask trivia questions or acquire knowledge. This in turn would facilitate self-directed learning, which has been found to generate better information retention in children.
(b) Restricting the Technology Use By and Around Children
The working- and middle-class participants were subject to different structural and economic realities. For example, children in the middle-class families were actively overseen by at least one parent or a caregiver at all times (an intensive child-rearing style that sociologist Annette Lareau calls “concerned cultivation” ), but working-class parents had longer work hours, and their children were expected to be on their own (a style Lareau calls “natural growth”), or their children were with members of the extended family or a close circle of friends who were lenient or did not monitor technology at all.
As is characteristic of “concerned cultivation,” middle-class parents had strong involvement in their children’s lives, which led them to persistently monitor and restrict their children’s interactions with devices through either the direct enactment of pre-defined rules (as seen in Indian middle-class families) or the enactment of pre-defined rules based on the context (as seen in White middle- class families). In the context of our study, this distinction between families from two different racial backgrounds could be due to the greater emphasis Asian Indian families placed on respect for authority; obligations to, and dependency on, parents and family; and strict control, as opposed to White families, who primarily sought to cultivate self-regulation in children. Within working-class households, active mediation was the most common form of parental mediation, where parents relied on talking to their children about their concerns and goals regarding technology use and thereafter hoped the children would use the technology responsibly when they were on their own. This reflects the sentiment that collaborative rule-setting fosters feelings of trust and fairness among family members and leads to compliance with the rules, which is especially needed in working-class households, as the parents are often not around their children.
Our study revealed that parents across demographics try to reduce or modify their own use of smart phones and speakers around children in various contexts and in relation to several activities. Our study suggests that there are overlaps between the technology restrictions that parents impose on children and on themselves. This is because, for both, the driving factors are the values important to the family and the impact technology use has on children and the family social system. Specifically, for our Asian Indian families, the desire to adhere to self-imposed restrictions was driven by the need to either act as role models for children (as seen in working-class families) or to gain leverage for times when children tried to negotiate strict parental technology restrictions (as seen in middle-class families). In contrast, White American households primarily employed context- specific non-use, because their children’s reactions characterized the parents’ use of technology as alienating and emotionally dissatisfying. In addition, a few of the White middle-class parents felt guilty about using technology around their children, because they thought it inhibited them from being sufficiently responsive to their children or from spending quality time with them, and therefore they felt that they needed to further restrict their use.
Targeted Non-Use: Parents are the ones most aware of their values, circumstances (e.g., a child has behaved properly recently and has earned more technology time), and plans (e.g., the parent wants to further restrict her own technology use during dinner time) and how these factors impact their own and their children’s technology use or non-use. While parents can be asked to input these details into the application, such engagement must be constrained, as parents might either not be willing or have no time to do so. Therefore, we propose that one way designers can account for the different and dynamic values and preferences of families, in inclusive designs that can be seamlessly integrated into daily lives, is by embracing a balance of user-centric and agent-centric design principles. Such application designs could prompt parents in different contexts (e.g., time of the day, location of use) to report how they feel at a given moment about their technology use around children (user-centered), while tracking actual usage patterns in the background (agent-centered). The application could then send subtle nudges or enforce restrictions on users (e.g., lock-down of applications for a certain time period) when they repeat a use pattern that normally leads to dissatisfaction. Overall, these nudges and restrictions would serve as interventions that could lead people towards more thoughtful, informed, and appropriately contextual use or non-use decisions.
For more details please check out our paper: Radhika Garg, Subhasree Sengupta: When you can do it, why can’t I?’: Racial and Socioeconomic Differences in Family Technology Use and Non-Use. Proc. ACM Human Computer Interaction, Vol. 3, No. CSCW, Article 63. Publication date: November 2019. For questions and comments about the work, please drop an email to Radhika Garg at rgarg01[at] syr [dot] edu.
