Popular media and some scholars have shared concerns about the potential harm to children and families due to smartphones. Given the ubiquity of smartphones, scientific dialogue is sorely needed to produce the best evidence for families (Odgers, 2018). Seeking to provide a wider perspective, in Modecki et al 2020, we made use of self-report data from more than 3,000 parents nationwide, to ask – where are the links between parental phone use and parenting? Are these robust versus frail, positive versus negative? After exploring 84 analytic choices in our original study, the pattern of findings pointed to fragility – we uncovered few direct associations between phone use and parenting. Rather, this link was more complicated; more phone time was often associated with more positive parenting, mainly when technoference was low. These exploratory findings pointed to relatively few risks and suggested that risk-driven discourse regarding smartphones and families may be overstated and oversimplified.

The objective of a multiverse approach is to examine a range of potential analytic paths. We leveraged this method, asking whether a more balanced perspective might emerge regarding the potential risks and benefits of smartphones for families (Odgers, 2018; Orben, 2020). Unfortunately, concerns about potential impacts of phones, as well as the critique by McCaleb, Champion, and Schluter (in press), have oversimplified the interplay between phones as a communication and information device, versus phones as a potential distraction from parenting. As a result, we refocus here away from speculating about why data from our multiverse analysis do not align with deficit views of technology, towards understanding the patterns unveiled. We organize our response by (a) noting three logical issues that distract us from a shared agenda of moving knowledge forward, then (b) provide suggestions for future and replicable science in relation to measures used, (c) make use of the opportunity to extend our original analyses, and here re-focus on parents of very young children, and (d), finally, place this discussion in a larger context to highlight needed research moving forward.

Logical missteps Ignoring context

A common logical heuristic is termed the ‘base rate fallacy’, where people ignore conditions (e.g. moderation) when interpreting facts or theories (Bar-Hillel, 1980). Being aware of this context is critical, given the emotive issues that can emerge when studying technology use in relation to parenting (Etchells, Davidson, Kaye, Ellis, & Lieberoth, 2021). We sought to highlight this complexity in Modecki et al. 2020 and in follow-up analyses, below, by including potential moderators that might condition associations between phone use and parenting: displacement of parent–child time due to smartphone use and family conflict over parents’ phone use. In Modecki et al, we found that this moderation (technoference) was indeed the most prevalent pattern of significant effects. These types of distinctions are crucial for providing a realistic picture of complex phenomena, including families and phone use. The base rate fallacy occurs in the critique by McCaleb et al. and can be found more widely in narratives raising universal fears regarding smartphone harms to children, but with key conditioning factors such as offline vulnerabilities, overlooked (Odgers, 2018).

Misinterpreting effects

Scholars have cautioned that misinterpretation of published effects is widespread in psychology (Funder & Ozer, 2019). For this reason, we have grounded our multiverse approach within a scoping review of the literature on parental device use and parent–child relationship outcomes. In this case, we located several null results and general inconsistency in findings. Both the critique by McCaleb et al. and similar narrative reviews, claiming a corpus of research that shows broad negative effects of smartphones on youth and families, are arguably misreading where effects lie. In Table S1, we provide illustrative examples related to misinterpreted effects. This logical misstep is a challenge, broadly, for sense-making around smartphone use (Lanette, Chua, Hayes, & Mazmanian, 2018).

Conflating independent and moderating variables

We have adopted a multiverse approach in part motivated by concerns that the media and some scholars were conflating phone use with the role of phones in interfering with family life. By inaccurately merging the two, scholarly dialogue becomes muddy and parents themselves are left facing what they believe to be negative, but largely undefined, consequences from engaging with their phones (Lanette et al., 2018). As a result, it is incumbent on us as scholars to critically examine study methodologies and findings, to arrive at more precise conclusions. In Modecki et al., across 84 different study choices, our multiverse analysis showed that technoference was often a moderator of associations between phone time and parenting. In other words, phone use is distinct from, and operates differently than, technoference. McDaniel and Coyne (2016) have usefully highlighted technoference and related constructs, which measure potentially negative repercussions associated with phone use. Future work could equally benefit from parallel constructs seeking to capture the potentially positive implications of phone use, for instance the enhanced sense of support/connection from phones (Modecki, Duvenage, Uink, Barber, & Donovan, 2021).

Future directions for replicable science

Through a lens of replicable science, we reconsider several aspects of our multiverse study design, as a basis for future studies.

Purpose of a multiverse

In Modecki et al. and below, we considered many modelling options, rather than conducting a single analysis. The benefit of this exploratory approach is that a multiverse can expose how findings (previous or future) might rely on methodological choices, and thus help discern those findings in the literature that are robust versus fragile (Steegen, Tuerlinckx, Gelman, & Vanpaemel, 2016). Thus, this approach refers to methods and measures that already exist in the literature. Here, we agree with the critique presented by McCaleb et al. and others who have described the need for scholarship, which moves beyond cross-sectional associations (Orben, Dienlin, & Przybylski, 2019).

Measurement

Also important for future research is the need for better measures and clearly delineated constructs (e.g. Orben & Przybylski, 2019). We have employed 12 different measures of phone time. In Modecki et al 2020, a pattern did suggest potential differences related to immersive forms of phone use (e.g. intensive phone checking and SNS use). Immersive types of use sometimes showed negative associations with parenting, but only when parents simultaneously reported high levels of technoference. This pattern did not emerge for other operationalizations of phone use (e.g. calls or texting). Parental immersion in the phone may mean missing out on signals for attention from children and family (Hiniker et al., 2015), and we highlight this as an area for future consideration. We also concur with the critique by McCaleb et al. and articulated by Orben and Przybylski (2019) that, in future studies, behavioural measures (like passive data collection) would be an excellent complement to current self-report phone use measures.

In Modecki et al and our analyses below, we partnered with a national broadcaster to collect a breadth of data from a large sample. We pulled from measures commonly found in the family technology use literature. We wholeheartedly concur that future research should rely on multi-item parenting constructs, and better measures, overall. Generally, the literature corpus has been saturated in its early stages with ‘home grown’ measures of parenting in relation to technology (Modecki, Goldberg, Orben, & Wisniewski, 2020).

What about parents of young children?

The commentary authors did raise one point that has occupied our mind, as well – whether effects from our overall multiverse of parents would hold for parents of younger children? Although we originally accounted for child age as a continuous variable to adjust for potential demographic differences, the authors suggested a discretized alternative, focusing on very young children. Fortunately, we have the data to address this question. To progress the conversation, we reran the multiverse analyses for parents of children aged two years and under. For this sample of 818 parents, we replicated the analysis underlying Figure 1 in our original study, which characterized effect sizes for smartphone use on parenting, with moderating effects of technoference.

Replicating the multiverse on young children: Characterizing smartphone use predicting parenting, moderated by technoference. Note: Effect size values for each smartphone use variable predicting parenting. Separate panels for each independent variable. Y-axis represents effect size estimates (log-odds scaled for ease of interpretation), with horizontal line denoting zero and blue indicators denoting significant effect. X-axis shows values with no modifier in the model (N), values for smartphone use X family conflict interaction (C), and values for smartphone use X family displacement interaction (D) at low (L), medium (M) and high (H) levels. Effect sizes for parental attachment are shown in top rows, and effect sizes for warmth are shown in bottom rows

As shown below, in Figure 1, there are two main takeaways. First, the variability in how technoference moderates phone use on parenting is higher for this younger age group. Thus, any apparently significant association within the multiverse should be interpreted with caution. Second, when we focus on the only meaningful effect sizes (i.e. two significant interaction terms, blue in Figure 1), it is readily apparent that time is not the issue, but rather technoference. Plots of moderation findings indicate that for young children, at high levels of parental displacement, parental warmth tends to be lower – regardless of number of texts (or SNS use). Figures available from www.modernlifestudy.com.

Needed research moving forward

Parents stand to benefit from empirically based advice related to technology and family well-being. In particular, parents struggling with inadequate social supports or otherwise facing myriad stressors deserve our best science. Our concern is that by prematurely objecting to parents’ phone use, we may gloss over the many documented, real-world conditions known to impact these and other parents in pernicious ways – family stress, maternal mental health, financial hardship, to name a few (Clark, 2013). These and other challenges that parents face daily can make it extremely difficult for parents to engage in the positive ‘quotidian moments’ that make up the best parenting (Tronick & Beeghly, 2011). It is also entirely possible that we can leverage technologies such as smartphones to better reach these parents to support them in their roles, and apply best evidence efforts to enhance parental and child well-being (e.g. Shorey et al., 2019).

We are researchers striving to advance more rigorous research in this area. In this case, we are not pro-technology per se, nor pro-technology use while parenting. Rather, we are pro-kids and pro-evidence, believing that they deserve our best scientific efforts and most productive forms of dialogue in understanding these questions. Tensions are building between narratives of widespread risk versus accounts that suggest risk from smartphones may be much more circumscribed or may even hold potential benefits. As a result, it is worth affirming a shared goal – to produce our strongest science in support of families. The way forward is rigorous scientific dialogue to help shed light on the best ways forward for families in this postdigital age (Jandrić et al., 2018).

Acknowledgements

The authors acknowledge useful insights given by Andrew Przybylski, Paul Scuffham, and Gillian Hayes, as well as helpful feedback on the manuscript given by Candice Odgers. We thank the Australian Broadcasting Corporation and Kylie Andrews for their efforts associated with data collection and Helen Correia for her contributions to the ModernLife project. The authors have declared that they have no competing or potential conflicts of interest.

Filename Description jcpp13433-sup-0001-TableS1.docxWord document, 20.2 KB Table S1. Studies cited in Commentary and associated details.

Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

References

Bar-Hillel, M. (1980). The base-rate fallacy in probability judgments. Acta Psychologica, 44, 211– 233. Clark, L.S. (2013). The parent app: Understanding families in the digital age. Oxford, UK: Oxford University Press. Etchells, P., Davidson, B.I., Kaye, L.K., Ellis, D., & Lieberoth, A. (2021). Researchers should avoid causally attributing suicide to video game play as a single factor. Perspectives in Psychiatric Care. PsyArXiv Preprints. https://psyarxiv.com/a4nzh/ Funder, D.C., & Ozer, D.J. (2019). Evaluating effect size in psychological research: Sense and nonsense. Advances in Methods and Practices in Psychological Science, 2, 156– 168. Hiniker, A., Sobel, K., Suh, H., Sung, Y.C., Lee, C.P., & Kientz, J.A. (2015, April). Texting while parenting: How adults use mobile phones while caring for children at the playground. In Proceedings of the 33rd annual ACM conference on human factors in computing systems (pp. 727– 736). Jandrić, P., Knox, J., Besley, T., Ryberg, T., Suoranta, J., & Hayes, S. (2018). Postdigital science and education. Educational Philosophy and Theory, 50, 893– 899. Lanette, S., Chua, P.K., Hayes, G., & Mazmanian, M. (2018). How much is' too much'? The role of a smartphone addiction narrative in individuals' experience of use. Proceedings of the ACM on Human-Computer Interaction, 2, 1– 22. McCaleb, M., Champion, P.M., & Schluter, P.J. (in press). Commentary: The real (?) effect of smartphone use on parenting- a commentary on Modecki et al (2020). Journal of Child Psychology and Psychiatry. McDaniel, B.T., & Coyne, S.M. (2016). “Technoference”: The interference of technology in couple relationships and implications for women’s personal and relational well-being. Psychology of Popular Media Culture, 5, 85. Modecki, K.L., Duvenage, M., Uink, B., Barber, B.L., & Donovan, C.L. (2021). Adolescents’ online coping: when less is more, but none is worse. Clinical Psychological Science. In press. Modecki, K.L., Goldberg, R., Orben, A., & Wisniewski, P. (2020). Robustly quantifying parental monitoring and support in the digital age. Presentation for Developmental Science in the Digital Age, University of California Irvine. Modecki K. L., Low-Choy S., Uink B. N., Vernon L., Correia H., & Andrews K. (2020). Tuning into the real effect of smartphone use on parenting: a multiverse analysis. Journal of Child Psychology and Psychiatry, 61, 855– 865. http://dx.doi.org/10.1111/jcpp.13282 Odgers, C. (2018). Smartphones are bad for some teens, not all. Nature, 555, 432– 434. Orben, A. (2020). The Sisyphean cycle of technology panics. Perspectives on Psychological Science, 15, 1143– 1157. Orben, A., Dienlin, T., & Przybylski, A.K. (2019). Social media’s enduring effect on adolescent life satisfaction. Proceedings of the National Academy of Sciences of the United States of America, 116, 10226– 10228. Orben, A., & Przybylski, A.K. (2019). Screens, teens, and psychological well-being: Evidence from three time-use-diary studies. Psychological Science, 30, 682– 696. Shorey, S., Ng, Y.P.M., Ng, E.D., Siew, A.L., Morelius, E., Yoong, J., & Gandhi, M. (2019). Effectiveness of a technology-based supportive educational parenting program on parental outcomes (part 1): randomized controlled trial. Journal of Medical Internet Research, 21, e10816. Steegen, S., Tuerlinckx, F., Gelman, A., & Vanpaemel, W. (2016). Increasing transparency through a multiverse analysis. Perspectives on Psychological Science, 11, 702– 712. Tronick, E., & Beeghly, M. (2011). Infants' meaning-making and the development of mental health problems. American Psychologist, 66, 107– 119.