Abstract
Parents play an important role in scaffolding autonomy and independence as their children transition to adulthood. In the digital age, mobile phones allow for increased connection at this important developmental transition, but we know little about the extent to which digital connection may help (i.e., through developmentally appropriate support) or hinder (i.e., through intrusiveness or helicopter parenting) emerging adult (EA) autonomy development. We tested whether digital parent-EA interactions tapping engagement, monitoring, and responsiveness were associated with EA perceptions of parental autonomy support in a sample of 238 college students (M age = 19.85) who contributed all text messages exchanged with their parents over 2 weeks. Results indicate that many dimensions of parent-EA text message interactions are unrelated to perceived parental autonomy support, but those that did emerge point towards a potentially maladaptive role of overparenting in associations with less perceived parental autonomy support. Results underscore that, for most EAs, parental text messaging is not likely to be perceived as autonomy inhibiting, but that for a small minority of parent-EA dyads, intense levels of digital connection with parents may be associated with perceived autonomy inhibition.
Introduction
Emerging adulthood (ages 18–25; Arnett, 2000) is a unique life stage at the intersection of adolescence and adulthood. This time is characterized by multiple transitions, new experiences, and increasing independence and self-reliance (Soenens et al., 2007). Emerging adult theory suggests that gaining autonomy from parents is necessary for the health of emerging adults (henceforth “EA”s) as well as a healthy transition into adulthood (Arnett, 2000). Many EAs strive to strike a balance between increasing independence in how they think and act and maintaining an emotional connection with their parents (Van Petegam et al., 2013). Although considerable empirical evidence supports the importance of parental influences on children and adolescents, fewer studies examine whether and how parental behaviors remain influential into emerging adulthood (Nelson et al., 2010).
Monitoring in emerging adulthood
Parents who are striving to support EA autonomy must navigate and balance how (and how much) to engage in traditional tasks of parenting such as provision of responsive, warm caregiving and monitoring and guidance of their child’s behavior. This balance is best captured as the distinction between parental responsiveness and parental demandingness, which are often considered the two main dimensions of parenting (Bornstein, 2019).
Among the constructs that define parental demandingness is parental monitoring. Traditionally, parental monitoring encompassed parental behaviors intended to keep tabs on children such as attending to children’s locations, friends, and activities (Dishion & McMahon, 1998). Today, consistent with Stattin and Kerr’s (2000) reconceptualization, scholars (and this study’s operationalization of monitoring) recognize monitoring as more than a one-sided endeavor, comprising distinct parent behaviors like asking questions (solicitations) and enforcing rules (control) alongside distinct youth behaviors like disclosures of information about their whereabouts and activities. Studies with adolescents suggest that youth disclosures may contribute more to parental knowledge than parental monitoring behaviors (Kerr et al., 2010). Thus here, we conceptualize parental monitoring as a dyadic process that includes parental solicitation, parental control, and EA disclosures.
For the most part, parental monitoring tends to be associated with positive adjustment in adolescents (Dishion & McMahon, 1998; Yap et al., 2014), but less is known about monitoring in EAs. In adolescence, lower levels of parental monitoring are a strong predictor of more adolescent externalizing (Lopez-Tamayo et al., 2016) and internalizing (Pinquart, 2017b) symptoms. Among EA samples, there is some evidence to suggest that disclosures to fathers (though interestingly, not mothers) may protect against delinquency and depression (Son & Padilla-Walker, 2021). High parental control (a component of monitoring), in contrast, may be linked with more internalizing and externalizing problems in emerging adulthood (Padilla-Walker et al., 2021). Evidence suggests that more parental solicitations are associated with higher levels of deviant behaviors (Pesola et al., 2015), and that EAs may perceive online solicitations as intrusive (Hessel et al., 2017). Overall, parental control and solicitations do not seem to be beneficial for psychosocial adjustment in EAs, whereas the results for disclosure are a bit more mixed.
Such mixed findings may reflect a curvilinear effect of parental monitoring, suggesting that a good thing may be best in moderation. Despite the generally protective nature of parental monitoring in adolescence, over monitoring or harsh monitoring can become intrusive and thus hinder psychosocial development (especially as EAs strive for increased independence; Barber et al., 2005). Parents who are high in control may display intrusive behaviors, use coercion to invalidate children’s perspectives, undermine their individuality, and overall deny adolescents’ independence (Soenens et al., 2010). Harsh parental monitoring can negatively affect children’s emotional wellbeing as it is associated with externalizing and internalizing behaviors during adolescence and can predict these behaviors over time (Pinquart, 2017a, 2017b). Findings seem to generalize to emerging adulthood, as those EAs who perceive their parents as more psychologically and behaviorally controlling report greater psychological distress and less psychological well-being (García Mendoza et al., 2019). It may be that parental monitoring behaviors in emerging adulthood lead to the highest EA perceived levels of parental autonomy support when delivered at moderate levels that are neither excessively low nor excessively high.
Responsiveness in emerging adulthood
Parents must also decide how (and how much) to provide warm, nurturant responsiveness to their children during the transition to emerging adulthood. Responsiveness often encompasses warm and supportive parenting-EA interactions that are physically and emotionally affectionate, approving, loving, and caring (Openshaw et al., 1984). Warmth provides numerous benefits for well-being in childhood and adolescence (Tubman & Lerner, 1994) which seem to extend into emerging adulthood (Inguglia et al., 2015). A large body of research shows that adolescents whose parents express low levels of warmth and support are more likely to exhibit internalizing and externalizing symptoms and less likely to report life satisfaction (Parra et al., 2015).
Scholars have identified three distinct forms of parental support: emotional support, advice (or informational) support, and instrumental (or tangible) support (House, 1981). Parental emotional support is characterized by care, concern, and listening (House, 1981). Advice-giving from a parent can help a child cope with personal and environmental problems (House, 1981). Instrumental support includes provision of assistance through methods such as labor, money, or time (House, 1981). Adolescents who perceive more parental support tend to also experience more positive psychosocial outcomes (Auerbach et al., 2011). Conversely, a lack of support has been linked to greater depression and more relationship strain in emerging adulthood (Gomez & McLaren, 2006). For EAs, parental financial and residential support seem to provide necessary scaffolds as the adolescent transitions into adulthood (Mortimer, 2012). Like monitoring, social support in parent-EA relationships is best conceptualized as a dyadic process in which EAs both seek and receive support (Wills & Shinar, 2000).
The idea that all good things are best in moderation may also apply to parental responsiveness, especially in emerging adulthood. There is rising concern around “overparenting” or “helicopter parenting,” which is characterized by a developmentally inappropriate amount of responsiveness (Segrin et al., 2013) and can include behaviors such as excessive advice giving, problem solving and tangible aid (Segrin et al., 2012). Some studies suggest that overparenting in emerging adulthood may be linked with poorer psychosocial outcomes such as depression, anxiety, and perceived stress (Schiffrin et al., 2014; Segrin et al., 2013 though see Kwon et al., 2016). Overparenting is also associated with increased levels of maladaptive traits relevant to autonomy for EAs, such as entitlement and narcissism (Segrin et al., 2012). Here, we expand the usual conceptualization of the responsive side of parenting to include the dyadic process of warmth conveyance (by EAs and parents), support provision (by parents) and support seeking (by EAs) in domains of emotional, advice, and instrumental supports.
Self-determination theory, autonomy, and emerging adulthood
The Self-Determination Theory (SDT) is critical in understanding the importance of EAs perceptions of parental autonomy support in emerging adulthood. SDT posits that there are three universal psychological human needs: competence, relatedness, and autonomy (Deci & Ryan, 2000) and highlights the importance of the social context, specifically focusing on the role of caregivers, who can either facilitate or undermine a child’s progress towards autonomy development (Ryan & Deci, 2006). EAs fulfilling these needs is necessary for EAs personal and social development, and their overall well-being (Ryan & Deci, 2017). Given that increasing independence and autonomy are key developmental tasks of the EA period (Arnett, 2000), the ways in which parents can support effective autonomy development are of great interest (Kouros & Garber, 2014). Broadly, parental autonomy support is defined as the parent’s active support of the child’s capacity to be self-initiating and independent (Ryan & Deci, 2006).
Based mostly on research involving children and young adolescents, autonomy support has been operationalized to include parental behaviors such as providing rationale and explanation for behavioral requests, recognizing the feelings and perspective of the child, offering choices, encouraging initiative, and minimizing the use of controlling techniques (Ryan & Deci, 2017). Parental behaviors that could impede autonomy include fostering an environment where children feel reliant and dependent on their parents for a developmentally inappropriate amount of support (Steinberg & Silk, 2002). Thus, it is important to attend to how and how much parents are providing support to their EA child and to consider amounts of support, out of proportion to the developmental needs of the child, might be inhibiting EA’s perception of parental autonomy support.
Responsiveness, monitoring and parental autonomy support in emerging adulthood
It is not yet clear the extent to which parent-child interactions may help or hinder perceptions of parental autonomy support as adolescents become EAs. There are few studies that consider that importance of parental support autonomy support in emerging adulthood and even fewer concerned with how perceptions of parental autonomy support are related with other parenting dimensions (Padilla-Walker & Nelson, 2019). The current understanding of autonomy support and parental control are that they are not mutually exclusive (Benito-Gomez et al., 2020; Soenens et al., 2009). Parents who are low in autonomy granting are not necessarily exhibiting higher levels of behavioral control, while the absence of control does not imply autonomy granting behavior. Thus, parents can be engaging in controlling and autonomy supportive behaviors simultaneously. However, levels of parental monitoring that are too low or too high could both be detrimental to EA perceptions of autonomy support and related constructs (LeMoyne & Buchanan, 2011). In adolescent samples, higher levels of parental control were negatively associated with perceptions of parental autonomy support (Soenens et al., 2009). Thus, developmentally appropriate, moderate amounts of monitoring may help scaffold youth autonomy development (another instance of a potential “best in moderation” effect; Karabanova & Poskrebysheva, 2013).
Parental autonomy support and responsiveness often co-occur within healthy parent-child relationships (Soenens et al., 2009) and we have some evidence that parental responsiveness may also be best in moderation when it comes to autonomy development. In adolescent samples, authoritative parents (who are high in responsiveness) were viewed as more supportive of their children’s autonomy (Baumrind, 1991). When parental responsiveness is too high however, emerging-adult children report higher levels of entitlement (Segrin et al., 2012), and lower self-efficacy (Bradley-Geist & Olson-Buchanan, 2014). Further, EAs with “helicopter parents” are found to have lower levels of competence and to perceive their parents as less supportive of their autonomy (Schiffrin et al., 2014). Therefore, like monitoring, moderate amount of digital responsivness may be the most benefical for perceptions of parental autonomy support for EAs development (aligned with a potential “best in moderation” effect).
Importance of digital communication
Many EAs leave their family homes for the first time during this period, and thus the use of digital communication as a form of parent-EA contact at a distance is common. Over 90% of parents report using text messages to communicate with their adolescents (Rudi et al., 2015), though young people communicate far more with peers than parents via text (Ehrenreich et al., 2020). Adolescents use calls and text messages to engage in managerial communications with their parents as well as to connect emotionally (Fletcher et al., 2018), and many parents report that the mobile phone is a useful tool for achieving parenting goals and ultimately makes parenting easier (Lauricella et al., 2016). There is also initial evidence that parents utilize digital communication to stay connected with their EA children (Miller-Ott et al., 2014) but we still know little about what role digital devices play in the maintenance and evolution of EA-parent relationships.
Co-construction theory asserts that online and offline behaviors often mirror and impact each other (Subrahmanyam et al., 2006), which is likely the case for digital and face-to-face parenting. Although there are few studies of digital parenting, those that do exist suggest that parenting practices remain consistent, whether face-to-face or digital. For instance, those parents who engage in the most offline monitoring also report higher digital monitoring of their children (Rudi & Dworkin, 2018). We assume that the literature on traditional parenting practices reviewed above is likely still highly relevant to the online parenting environment, and thus test whether our hypotheses based on face-to-face parenting studies hold true for digitally enacted parent-EA interactions using the rich content of digital communications that offer a unique window through which naturalistic parent-EA interactions can be observed.
Studies suggest that more frequent parent-child digital interactions are associated with positive parent-child relationship quality, less parent-child conflict, and greater perceived parental support (Weisskirch, 2011). New research, including in this same sample, suggests that parent-youth phone contact can serve as a conduit for both the transmission of responsive, warm parenting-EA interactions (Jensen, George, et al., 2021; Fletcher et al., 2018; Weisskirch, 2011) as well for remote monitoring of youth (Jensen, George, et al., 2021). Although some research suggests that the use of the phone in these ways can help parents provide support and guidance in-the-moment for youth who need it (Jensen, George, et al., 2021), the omnipresence of the mobile phone also opens the door to potentially intrusive parent contact at any hour of the day or night. Indeed, some EAs point out that their parent’s use of the mobile phone to monitor and control their behaviors yields access to what was once considered personal time and space (Williams & Williams, 2005), which is associated with feelings of intrusiveness and privacy invasion (Racz et al., 2017). More recent findings in the same sample used here have suggested that EAs who text more with their mothers (though not fathers) perceive higher levels of intrusiveness and pressure around contact with parents by phone or online (Jensen, Hussong, & Haston, 2021). It is important to consider both the ways in which the mobile phone may serve as a vital lifeline to positive parent-youth interactions in emerging adulthood as well as the ways in which it may serve as a mechanism for potentially intrusive boundary crossings that are perceived as restricting youth autonomy.
Present study
The present study leverages 2 weeks of observed parent-EA text message conversations (coded for traditional features of monitoring and responsiveness) to examine how parent-EA text message interactions are associated with EA perceptions of parental autonomy support. Specifically, we tested whether parent-EA 1) texting frequency, 2) digital responsiveness, and 3) digital monitoring were associated with higher or lower EA perceived parental support for autonomy. First, we tested linear and quadratic (a “best in moderation” hypothesis) associations between parent-EA texting frequency and perceived parental autonomy support; analyses were conducted separately for mother- and father-EA dyads. We hypothesized—consistent with past studies suggesting that parent mobile phone contacts can be intrusive (Williams & Williams, 2005), but also literature which suggests that both very low and very high amounts of parental engagement (including digital) can be unsupportive of EA autonomy development (Jensen, George, et al., 2021)—that overall more frequent mother- and father-EA texting frequency would be tied with lower perceived parental autonomy support (a negative linear hypothesis), but also that those EAs who texted with their mothers and fathers at moderate frequency would perceive their parents as more supportive of their autonomy (relative to those whose parents texted them only infrequently or at very high frequency).
Second, we tested linear and quadradic associations between indicators of mother- and father-EA digital responsiveness and EA perceptions of mother and father autonomy support, over and above (potential) associations with mother- and father-EA texting frequency (separately for mother- and father-EA dyads). We hypothesized both an overall positive linear association—based on research that suggests that low parental responsiveness is detrimental to EA autonomy development (Gomez & McLaren, 2006)—and a quadratic “best in moderation” association—consistent with literature that suggests that helicopter parenting and overparenting can undermine EA autonomy development (Schiffrin et al., 2014)—such that those EAs whose mothers and fathers exhibited moderate levels of digital responsiveness would perceive the highest levels of parental autonomy support relative to those whose parents exhibited very low levels of digital responsiveness and very high levels of digital responsiveness.
Third, we tested linear and quadradic associations between indicators of mother- and father-EA digital monitoring and EA perceptions of mother and father autonomy support, controlling for potential associations with parent texting frequency (separately for mother- and father-EA dyads). We again hypothesized both an overall negative linear association—consistent with research that parent mobile phone contacts for monitoring can be intrusive (Hessel et al., 2017)—and a quadratic “best in moderation” association consistent with literature that both low monitoring and over monitoring can be harmful for autonomy development (García Mendoza et al., 2019).
Methods
Sample and procedures
Sample Characteristics of Sample.
Measures
Demographic covariates
All participants in the sample (N = 238; n = 215 for mother-EA dyads, n = 182 father-EA dyads) reported on their gender, age, and parent education computed as the highest of the mother’s or the father’s education. These covariates were chosen because decades of research have suggested that parenting-EA interactions (including responsiveness and monitoring) and parental autonomy support vary based on the age of the child (Inguglia et al., 2015), family socioeconomic status (Roubinov & Boyce, 2017), and gender (Bumpus et al., 2001).
Perceptions of parental autonomy support
One purpose of the larger study was to test data harmonization techniques, and thus participants were randomly assigned to one of two surveys (X or Y) that contained overlapping but not identical subsets of measures. Relevant here, 126 participants were randomly assigned to receive our measure of EA perceived parental autonomy support (the psychological autonomy granting scale; Steinberg et al., 1992), yielding 105 mother-child dyads and 94 father-child dyads randomized to be administered the outcome of interest in the parent text-message sample. Six items asked how often each parent (mother and father separately) encouraged or discouraged the expression of their ideas or individuality, on a 0–4 scale ranging from ‘never’ to ‘very often'. A confirmatory factor analysis of these six items in the larger study initially yielded poor fit for both mothers (χ2 (9) = 115.132, p < .001; RMSEA = .174 [.147 to .203]; SRMR = .060) and fathers (χ2 (9) = 90.405, p < .001; RMSEA = .155 [.127 to .185]; SRMR = .049). Item 6 (“If you did something she/he didn’t like, how often did your mother/father react by not being in contact with you for a while?”) appeared to contribute to misfit, and indeed the fit was acceptable once this item was removed for both mothers (χ2 (5) = 12.260, p = .031; RMSEA = .061 [.017 to .105]; SRMR = .021; standardized λs .675 to .822) and fathers (χ2 (6) = 10.850, p = .054; RMSEA = .056 [<.001 to 0.102]; SRMR = .021; standardized λs .745–.807). Thus, mother and father support for autonomy are modeled here as latent variables with five indicators.
Text-analysis measures
Texting frequency
Descriptive Statistics of Coded PCTICs Domains.
Note. Means (M), standard deviations (SD), and range reported across dyads over the 2-week study period.
Coded digital responsiveness and monitoring
All parent-EA text messages were coded using the Parent-Child Text Interaction Coding Scheme (Jensen, Hussong, & Haston, 2021). Micro-codes were assigned by an undergraduate primary coder and a postdoctoral master coder and codebook developer at the level of the text message and were neither mutually exclusive nor exhaustive. The primary coder was trained to an acceptable inter-rater reliability (κ > .80) with the master coder, with 20% of all messages double coded to monitor coder drift and reliability. Here the codes are a frequency of how many times the digital interaction occurred over the 2-week period; means, standard deviations, and ranges can be found in Table 2, with detailed code definitions found in supplemental Table S1. The present study utilizes parent-EA codes reflective of responsiveness including parent and EA warmth alongside EA seeking and parent provision of emotional support, instrumental support, and advice. Parent-EA monitoring is captured by three discrete codes: parent solicitation, parent control, and EA disclosure.
Analytic approach
We sought to answer study questions in a series of structural equation models in Mplus version 8.2 (Muthen & Muthen, 2017) with the MLR estimator, which is robust to non-normality through adjustment of standard errors. Predictors in curvilinear interaction terms were mean centered. Missing data were handled with FIML given that the substantial missingness on the primary outcome of interest is missing completely at random (MCAR).
To test our first hypothesis, we regressed EA perceptions of parental autonomy support on parent-EA texting frequency (linear and quadratic terms), controlling for demographic covariates (age, gender, parent education), with separate models for mother- and father-EA dyads. Non-significant quadratic terms were trimmed. To test our second and third hypotheses, we expanded this initial model to include each indicator of parent-EA digital responsiveness and monitoring (linear and quadratic terms, alongside parent-EA texting frequency and covariates); non-significant quadratic terms were trimmed.
Results
EA perceptions of mother and father support for autonomy were not significantly associated with covariates of age (bmothers = .086, SEmothers = .069, pmothers = .209; bfathers = .003, SEfathers = .065, pfathers = .966), gender (bmothers = −.213, SEmothers = .255, pmothers = .344; bfathers = −.278, SEfathers = .243, pfathers = .253), and parental education (bmothers = .117, SEmothers = .095, pmothers = .218; bfathers = .105, SEfathers = .100, pfathers = .293) but were nonetheless controlled for in all models.
Parent-EA texting frequency
As seen in Table 2, and inconsistent with our hypotheses, parent-EA texting frequency was not associated with EA perceived parental autonomy support (linearly or curvilinearly) for mothers or fathers. However, adopting a conservative approach, the linear term was included as a covariate in all subsequent analyses.
Parent-EA digital responsiveness
Associations between EA Perceived Parental Autonomy Support and Digital Parent-EA Interactions.

Quadratic regions of significance plots depict the slope of the coefficient for parent-EA digital interactions (dark solid line) predicting EA perceived parent autonomy support (Y-axis) across the entire observed range of each parent-EA digital interactions (X-axis). Dashed lines represent bounds of the 95% confidence interval. The light grey and dark grey shading represent the range within which most of the data fall (the 95th and 99th percentiles, respectively).
Parent-EA digital monitoring
The bottom portion of Table 3 conveys results of tests of our third hypothesis about associations between parent-EA digital monitoring and EA perceived parental autonomy support, controlling for covariates and the (non-significant) linear association with parent-EA texting frequency. The frequency of parental digital solicitations was not significantly associated with EA perceived autonomy support among mother- or father-EA dyads. Digital parental control in mother-EA dyads was not significantly associated with perceived autonomy support, however for father-EA dyads there was a negative linear association between father digital control and perceived autonomy support; those EAs whose fathers enacted more digital control tended to perceive their fathers as less supportive of their autonomy. Among mother-EA dyads, although there was not a significant linear association between EA disclosure to mothers and perceived mother support for autonomy, there was a quadratic association (depicted in Figure 1), such that the association between the frequency of EA digital disclosures to mothers and perceived mother autonomy support grew more strongly negative at higher frequencies of EA disclosures to mother, with those EAs who engage in the most frequent disclosures to mothers perceiving their mothers as the least supportive of their autonomy.
Sensitivity analysis: Parent-EA digital interactions behaviors and winsorized texting frequency
As seen in wide ranges and large standard deviations for many of our coded parent-EA digital interactions, there was considerable variability across parent-EA dyads, and the plots of quadratic associations highlight that many associations seemed to be driven by dyads at the extremes of digital responsiveness and monitoring. The potential impact of these outliers was explored by winsorizing outlying cases that fell three or more standard deviations above the mean. Winsorization lowers the influence of outliers by assigning an outlier a lower weight (Dixon & Tukey, 1968).
As seen in Table S2, in winsorized analyses parent-EA texting frequency and all indicators of parent-EA digital responsiveness were not significantly associated with EA perceived parental autonomy support for mothers or fathers. For mother-EA dyads, there were significant linear and quadratic association between parent solicitation and EA perceived parental autonomy support such that the association between the frequency of mother solicitation and perceived mother autonomy support grew more strongly negative at higher frequencies of mother solicitations (as depicted in Supplemental Figure S1). Also, there was a quadratic (but no linear) association for EAs’ disclosure to mothers, where EA perceived their mother as more supportive of their autonomy at higher frequencies of disclosures (Supplemental Figure S1). Of note, for both quadratic associations, across virtually the entire range of the digital interactions the confidence intervals span zero, and thus we cannot be sure that these associations ever differ from zero at a level that could not have occurred by chance. Thus, our discussion below balances the fact that primary results appear to be heavily influenced by the dyads who evidence the highest frequencies of parent-EA digital interactions, but also the fact that outlying values here are real data (objectively collected and thus not subject to self-reporting biases or exaggeration that sometimes drive outlying survey responses) and there may be something important about these very high digital interaction dyads that is worthy of attention and interpretation.
Discussion
In the current study, we examined the role of parent-EA digital relationships in EA perceptions of parental autonomy support. As EAs seek to balance increased independence and maintaining emotional connections with their parents, parents can engage in behaviors that inhibit or promote autonomy development. Here we found that, for the most part, neither the quantity nor the quality of parent-EA digital interactions was consistently associated with EA perceptions of parental autonomy support. Moreover, we saw little evidence consistent with our “best in moderation” hypothesis, in that results suggest that low and moderate levels of most digital interactions impart little risk for low perceived autonomy support, and that any risks seem to lie at the extremes of parent-EA digital interactions.
Overparenting
Rather, those significant associations that did emerge tended to support an overparenting/helicopter parenting hypothesis (consistent with self-determination theory’s emphasis on the important balance of relatedness and autonomy) wherein high levels of some digital parent-EA interactions could potentially be perceived as unsupportive of EA autonomy. Specifically, EAs who experienced more mother advice provision and father control via text message tended to perceive their parents as less supportive of their autonomy. Those quadratic associations which emerged largely supported this same pattern, as seen in findings wherein those EAs who sought and received more mother emotional/esteem support via text message and who engaged in more digital disclosures to their mothers tended to perceive their mothers as less supportive of their autonomy.
Overall, most findings here are inconsistent with traditional parenting literatures (largely based on adolescence) that have suggested mostly linear protections conveyed by responsiveness and monitoring behaviors for adolescent autonomy development (Auerbach et al., 2011) and inconsistent with previous findings suggesting that moderate levels of parental behaviors are most beneficial for EA autonomy development (Karabanova & Poskrebysheva, 2013). These findings are also somewhat inconsistent with the co-construction theory which suggests that the impacts of digital parent-EA interactions should mirror those seen in face-to-face interactions. Rather, we see a pattern here in which those associations that did emerge tended to point towards a potentially maladaptive role of over-responsiveness and over-monitoring in inhibiting autonomy (but seemingly only once over responsiveness and over monitoring reached very high levels), which is somewhat aligned with the overparenting and helicopter parenting literatures that suggest that high levels of parental behaviors can negatively impact the psychosocial adjustment of EAs (Padilla-Walker et al., 2012). These findings contribute to and extend our understanding of overparenting and helicopter parenting by offering support for the hypothesis that high levels of digital parent-EA interactions can be perceived as inhibiting of EA autonomy development.
The dyadic nature of overparenting
Some findings here highlight the role of the EA in co-constructing a parent-EA dynamic around autonomy. In primary analyses, two EA-driven behaviors (EA disclosure and emotional/esteem support seeking) were linked with perceived lack of parental support for autonomy; those EAs who engaged in the most frequent disclosures and emotional/esteem support seeking perceived their mothers as the least supportive of their autonomy. This intriguingly suggests that a portion of “overparenting” may potentially be driven by the EA child rather than intrusive parent-driven behaviors in isolation. Although overparenting is usually thought of as something that parents inflict upon their unreceptive EA children, these results suggest that overparenting may have a dyadic component in which the EA takes a more active role in contributing to highly engaged parent-EA relationships that may be thought of as indicative of an overparenting dynamic. Future overparenting research should take a more nuanced approach to understanding the EA’s role in these interactions.
Of note, the negative quadratic disclosure association seen in the primary analyses changed directions in the winsorized sensitivity analyses where EAs who disclosed at higher frequencies seemed to perceive their mothers as more supportive of their autonomy (though as seen in Figure S1 the confidence interval on this was very wide, and thus perhaps unreliable). This suggests that there may be something unique about those very heavy disclosing EA-mother dyads that needs exploration in future research. Indeed, it may be that, for most parent-EA dyads, more EA disclosures could be indicative of a more autonomy-supportive dynamic, but that those extremely heavy disclosers may have a unique parent-EA dynamic around autonomy.
Emotional/esteem support from mothers versus fathers
One exception to this overall overparenting pattern of findings was that those fathers who provided higher levels of emotional/esteem support via text message tended to be perceived as more supportive of EA autonomy whereas mothers who conveyed more emotional/esteem support via text message were perceived as less autonomy supportive. Our results are partially aligned with findings from a differing sample (Miller-Ott et al., 2014), which also uncovered differences between EAs’ perceptions of mother and father cell phone communications (though the direction of association differed, uncovering an association between mother’s violating cell phone rules and boundaries and poorer perceptions of the relationship (somewhat consistent with an overparenting hypothesis), though no such association emerged for fathers).
We offer several possible explanations for this finding. One, it is important to note that mothers and fathers are enacting emotional/esteem support at considerably different absolute levels; the maximum frequency of emotional/esteem support provision over the 2-week span for mothers was 58 (M = 2.09, SD = 5.93; 60 percent of mothers never texted to convey emotional/esteem support), whereas the maximum frequency for fathers was 10 (M = .38, SD = 1.41; 87 percent of fathers never texted to convey emotional/esteem support). Thus, it is possible that objectively higher levels of emotional/esteem support could lead EA children to perceive the emotional/esteem support from their mothers as a threat to their autonomy development, but to perceive these same types of emotionally supportive texts (albeit in lower quantities) as autonomy promoting when they come from fathers.
Another related explanation concerns traditional gender roles, as mothers are often expected to be more emotionally supportive, whereas fathers are rarely prescribed the role of emotional support person. As a result, EAs may perceive fathers who provide emotional support as more remarkable, and even as supporting their autonomy. It is also worth noting that, as emotional/esteem support is not a common text message behavior amongst most dyads, there could be some other unmeasured distinguishing factor (such as higher levels of enmeshment) about those dyads that do engage in emotion supportive conversations via text that is driving observed associations here. Future research is needed to help to illuminate what process may be driving EA’s differing perceptions of emotion/esteem support from their mothers versus their fathers.
Limitations, future directions and conclusions
The present study used novel methods to examine intersections between parent-EA digital interactions and EA perceptions of parental support for autonomy. Some limitations nonetheless merit consideration. First, it is important to note that some features of the dataset may limit the generalizability of the findings. The sample was comprised of college students (M = 19.85), many of whom lived on their college campus, and during a specific macro temporal moment (Navarro & Tudge, 2022) in 2014–2015; thus, it is possible that the findings here would not generalize to non-college going EAs or to college students in the post-COVID-19 world (which seems to have shifted patterns of college going; Hoover, 2020). Additionally, participants were not asked about sexuality nor disability status, which limits our ability to understand these dimensions of diversity in this sample. Second and importantly, the cross-sectional data makes it impossible to determine whether parent-EA digital interactions shape EA perceptions of parental autonomy support, if EA perceptions of parental autonomy support in some way shape parental-EA digital interactions, or if there is some third variable driving both. Nonetheless, we can learn much from this study and use it to inform future longitudinal, experimental, or quasi-experimental research (in which causation can be better inferred) in more diverse samples in the post-pandemic landscape.
Third, the study assumed a mother-father family structure in the survey prompts and text message designations, and we cannot know what other types of family structures are represented in our data. EAs that come from single parent homes, have same sex parents, or who were raised by other family members were not fully captured in this study’s design and may have different norms or cultural values that shape perceptions of their family’s digital interactions. Future studies should comprehensively assess family structure to better understand implications for parental autonomy support. Finally, it is possible that, instead of individual digital parent-EA digital interactions having the most impact on EA perceived autonomy support, it could instead be that the interplay amongst parental-EA digital behaviors shapes perceptions of autonomy support. Examining profiles of digital parenting-EA behaviors is an important direction for future research.
The present study helped to answer the question of how parents may be supporting or undermining EA autonomy development via mobile phone, with implications for future research and practice. Methodologically, this study suggests that it is valuable to move beyond self-report data to methods that offer a window into naturalistic parent-EA interactions. Ultimately, it seems that virtual parent-EA connections are increasingly normative and being integrated into the daily context of family life. Given that results suggested that most dimensions of parent-EA digital interactions were not strongly linked to EA perceptions of parental autonomy support, parents should feel encouraged that text messaging with their child at low to moderate levels is unlikely to be perceived as autonomy inhibiting. However, some results did suggest that there were some types of digital parent-EA interactions which were tied to less perceived parental support for autonomy, especially among a small minority of parent-EA dyads which saw very high levels of certain types of texts. Thus, we might encourage those parents who are very highly engaged with their EA child via text message to convey control or extreme levels of support to possibly consider tempering this type of interaction or consider alternate methods of communication. Importantly, most dyads were not enacting these behaviors at such high levels, and thus we must also consider that other special features in these dyads (e.g., pre-existing parent-EA dynamics) might be driving observed associations.
Findings also have clear implications for clinicians and higher educational professionals who seek to better understand the interplay between parent-EA digital engagement and EA perceptions of parental autonomy support. For example, parental education programs would do well to extend beyond adolescence into the college period to emphasize the key role that parents continue to play for their EA children. Programs could potentially highlight that for only a small minority of families does high digital engagement lead to EA perceptions of autonomy-limiting parenting behaviors. Results further suggest that clinicians and higher education professionals might benefit from the recognition that digital overparenting may not be very relevant to most EAs with whom they work, though a small minority might warrant additional support to help facilitate developmentally appropriate parent-EA connections during the transition from adolescence to adulthood.
Supplemental Material
Supplemental Material - Parent-Emerging Adult Text Interactions and Emerging Adult Perceived Parental Support of Autonomy
Supplemental Material for Parent-Emerging Adult Text Interactions and Emerging Adult Perceived Parental Support of Autonomy by Morgan Taylor Brown, Michaeline R. Jensen, Andrea M. Hussong in Journal of Social and Personal Relationships.
Footnotes
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research reported in this article was supported the National Institute on Drug Abuse of the National Institutes of Health under award number 1R01DA034636-01A1, and by a postdoctoral fellowship provided by the National Institute of Child Health and Human Development (T32- HD07376) through the Center for Developmental Science, University of North Carolina at Chapel Hill, to Michaeline Jensen.
Open research statement
As part of IARR’s encouragement of open research practices, the authors have provided the following information: This research was not pre-registered. The data used in the research are not available. The materials used in the research are available. The materials can be obtained by emailing:
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
