The Joint Media Engagement Scale (JMES): An instrument for measuring shared media use with children aged 1 to 5 years old

Abstract

The digital media landscape is rapidly shifting, and some children begin using digital media in infancy. As with book reading, young children need adult guidance to learn from digital media. Joint media engagement (JME) occurs when preschool children and their parents actively use digital media together. JME during early childhood is associated with increased learning from media. However, there are no validated scales of JME. In the present study, parents of 353 1‐to‐5‐year olds answered a series of questions about their media use and behaviours. Exploratory factor analysis yielded two scales, one for viewing video content together and one for playing digital games together. Convergent validity was established by comparison to the widely used Television Mediation Scale. Together, these findings establish the JME instrument as a valid and reliable tool for measuring joint media engagement during early childhood.

Statement of Contribution.

What is already known on this subject?

• Rapid technological advances have resulted in frequent use of digital media in different locations and multiple formats even by young children.

• Although young children can use media alone, adult guidance is needed to support young children’s learning from digital media.

• Joint media engagement (JME) refers to the activities young children and their caregivers participate in when they actively share digital media experiences.

What does the present study add?

• A validated joint media engagement scale (JMES) for young children was previously unavailable and was developed in the present study.

• JME is examined for watching digital media and playing digital games separately.

• JMES for watching video emerged with one factor: Cognitive and Affective Scaffolding.

• JMES for playing digital games includes three factors; Cognitive Scaffolding, Affective Scaffolding, and Technical and Physical Scaffolding.

INTRODUCTION

A tremendous technological change has occurred that allows even young children to engage with digital media in various forms and in different places. Television is no longer a stationary box in the living room viewed together by the family, but rather every family member can use their own screen and watch videos of their choice. In addition, video content is aimed at children from infancy. In light of this rapidly changing media landscape, it is important to investigate how families share digital media in the home.

Although young children can use media alone, adult guidance is needed to support young children's learning from digital media, just as adult guidance is needed to support learning from picture books. The term joint media engagement (JME) refers to the activities young children and their parents participate in when they actively share digital media experiences (e.g., playing digital games together, talking about TV content). JME can extend children's learning beyond the immediate media experience when parents connect what children have seen on digital media to their own lives. For example, a parent might say, “Do you remember, Big Bird also wanted ice cream?” when getting ice cream for their child.

JME has been examined in observational studies during television viewing and computer use (Fidler et al., 2010; Lauricella et al., 2009, 2015; Lavigne et al., 2015; Padilla‐Walker et al., 2020; Zack & Barr, 2016) and is associated with enhanced adult–child conversations with new opportunities for early word learning (Lavigne et al., 2015). Lavigne and colleagues found that parents introduced new words to toddlers during co‐viewing, and this carried over after the television programme ended, resulting in richer parental language during the subsequent free‐play episode.

Such language enrichment may explain why JME has been associated with better language and literacy outcomes for children. For example, Dore et al. (2020) examined the relation between 6‐year‐olds' language and literacy development, the amount of time watching video or playing digital games and the amount of JME measured using an original 7‐item scale. They found that JME moderated the association between video viewing and children's preliteracy scores such that video viewing negatively predicted preliteracy scores for children with lower levels of JME but not for those with higher levels of JME. The authors concluded that JME might be important for buffering negative effects of digital media use on language development. Using the same JME Scale, Medawar et al. (2023) replicated and extended the findings to 1‐ to 3‐year olds in an Argentinean sample. They found that higher JME scores were associated with higher language density. The findings are promising; however, the JME Scale used in these studies has not been validated.

Research also suggests there are different types of JME. The instrument that has been most frequently used to examine JME is the Television Mediation Scale (Valkenburg et al., 1999). The validated scale includes three subscales: instructive mediation, restrictive mediation and social coviewing. Instructive mediation and social coviewing subscales include elements of joint media engagement (JME). A meta‐analysis of predictors of parental mediation practices (Wang et al., 2023) showed that both positive and negative attitudes to media predict mediation. Parents with both more positive and more negative attitudes, as well as higher levels of parental involvement, reported more frequent use of instructive mediation and social coviewing (Wang et al., 2023). However, the scale has some limitations. It focuses only on television viewing and, therefore, does not fully capture the current digital media landscape. Also, it was developed for older children, and the questions do not fully capture the types of interactions common among younger children and their caregivers.

Other efforts have attempted to characterize JME across a wider age range and a broader set of media activities. Ewin et al. (2021) conducted a systematic review of 27 studies that examined JME in parent–child dyads with children aged 0–10 years and found that different types of JME emerged in the existing literature. One common type of JME was labelled ‘cognitive’. Valkenburg et al. (1999) capture aspects of cognitive scaffolding in the instructive and the coviewing subscales. The cognitive aspect of JME denotes scaffolding conversations aimed at increasing children's understanding of the media content, which can occur both during and after digital media use (Yelland & Masters, 2007). Another type of JME is ‘affective’, which denotes parents' attempts to scaffold their children's emotion regulation (e.g., “Soon they will find Mr. Rabbit again”). Yet another type of JME denotes technical aspects and, more specifically, relates to digital media devices themselves. That is, parents try to help their children with using a device or an app (Ewin et al., 2021).

Ewin et al. (2021) found that rates of JME varied widely between studies and also between dyads. One reason for the differences between studies might be that digital media use was undifferentiated. Digital media activities are not limited to watching video content. Children also engage in digital games on smartphones, tablets and laptops. Digital games offer different challenges and opportunities for young children's learning, and as such, they present unique JME opportunities. For example, while watching a video, parents might talk about the characters in the video and explain their motives. Conversely, while playing digital games parents might focus more on explaining the technical aspects or rules of digital games. Moreover, parents may differ in their amount of JME when children watch videos versus play digital games and apps. Therefore, it may be necessary to measure JME separately for activities that occur while watching video versus playing digital games. It is also important to capture JME that occurs after media use has ended, such as when parents refer back to aspects of a show and connect the media content to children's everyday experiences (e.g., “Let's bounce just like Big Bird did”).

The current research meets the need for a reliable and valid measure of JME with young children that reflects the complexity of the digital media landscape. The validation of a new Joint Media Engagement Scale (JMES) will aid in our understanding of how preschool children learn from new opportunities presented by digital media and how they engage in activities with adult facilitation. When parents and preschool children use digital media together, parents can scaffold their children's learning and tailor the content to their children's needs and interests. Hence, in the current study, we present a new instrument with subscales that assess JME between parents and their preschool children when viewing video content and playing digital games, separately.

METHOD

Participants

In total, 493 parents of children between 18 and 60 months were recruited through social media and flyers distributed at libraries and open childcare centres in various cities in Sweden to participate in an Internet‐based survey. Out of the initial 493 participants who agreed to take part, 330 parents completed the survey and accurately responded to the quality control questions, making them eligible for inclusion in the final sample. Children's age was evenly distributed across the age range, with a median age of 36.95 months and an interquartile range from 27.4 to 47.52 months. The respondents were 86% mothers and 14% fathers. Just over half (54%) of children were boys and the rest (46%) were girls. Parents' mean age was 35.1 years (SD = 4.9, range 24–51). The parents were well‐educated, with 75% stating graduation from university (bachelor or higher university degree; missing data: 2%). In the current sample, 66% of the families had an economic standard above the median for Sweden, and 22% had an economic standard below the median (missing: 12%). Most families spoke solely Swedish at home (72%) or were bilingual with Swedish as the main (19%) or one of the main languages at home (5%). Some families did not speak Swedish at home (3%) or spoke Swedish less than other languages at home (1%; missing data 1%). Most parents' birth country was Sweden (87%), whereas 8% of parents were born in a different country (missing 5%).

Ethics

The present study was conducted according to guidelines laid down in the Declaration of Helsinki. The caregivers were informed of the study and asked to proceed to the online study if they consented to participate. No identifying information was obtained. The survey was approved by the Swedish Ethical Review Authority (Dnr 2021–01313).

Survey data collection

Parents used an anonymous link to access the Media Assessment Questionnaire (MAQ, version 2.1, Barr et al., 2020) electronically. As an anonymous link was used, parents were not reminded to complete the survey. However, parents could continue answering the survey once they started if they used the same device. The MAQ was administered through Qualtrics™ software (Qualtrics, 2024).

Materials

The MAQ (Barr et al., 2020) was developed by the Comprehensive Assessment of Family Media Exposure Consortium to assess digital media use as well as other media use (such as picture book reading) in the child's home and to measure both child media use and parent media use, attitudes and practices. The MAQ has been standardized for use in multiple countries and languages (including Swedish). The parents are asked to provide demographic data about the parent, child and household, including childcare and languages spoken in the home. In the current study, the MAQ version 2.1 was used, which focused assessment on four digital activities: watching video, playing digital games, video chatting and reading E‐books. Parents answered detailed questions for each activity that their child engaged in during the last 2 weeks. Following sections about each digital activity, other non‐digital activities were assessed, as well as parents' own use and attitudes towards digital media. Finally, parents answered the Television Mediation Scale (Valkenburg et al., 1999).

Parents reported that 98% of children watched television, 48% played digital games, 83% video chatted, and 14% used digital books. We focused here on developing new questions to assess JME during video viewing (JMES Video) and playing digital games (JMES Games).

Joint Media Engagement Scale

Items were compiled to capture parental JME with their children while watching video content as well as playing digital games. Items were designed to capture three types of JME (following Ewin et al., 2021): cognitive, affective and physical/technical scaffolding. For each item, one or two specific examples were included for clarity. Each JME item was answered by choosing one of four ordinal response options that suited the circumstances of the respondent best. Response options were Never/very rarely, rarely, occasionally and often/always.

Parents completed 15 items regarding watching video content. See Appendix S1 for all initial Swedish items and their English translations. Eight items focused on how frequently parents engaged in cognitive scaffolding with their children, asking how frequently parents engaged in conversations regarding the content at hand with their child, either during or after watching. Three items explored how frequently parents engaged in affective scaffolding with references to the emotional states of the child, including expressions of encouragement or concerns, as well as references to the emotional states of characters in the viewed material. Four items focused on parental behaviours concerning help (physical or technical) with devices or apps.

In the context of playing digital games, 17 JME items were assessed. See Appendix S2 for all initial Swedish items and their English translations. Eight items focused on how frequently parents engaged in cognitive scaffolding with their children, either during or after playing, specifically referring to the content matter of the game. Four items explored how frequently parents engaged in affective scaffolding, referring to the emotional states of their children during digital gameplay, including expressions of encouragement or concerns. Five items addressed parental behaviours related to physical and technical assistance with devices, apps or games.

Data reduction and statistical analyses

Data were analysed using R version 4.4.1 (R Core Team, 2024). Exploratory factor analysis was used for identifying relevant items and factors (R package psych version 2.4.6, Revelle, 2024). As the response options are ordinal, polychoric calculations were used for examining correlations and factor structures. Polychoric calculation assumes that each item instantiates a latent continuous variable. Data for JMES Video and JMES Games were first examined for how many factors would be appropriate for further exploratory factor analyses, using methods MAP and VSS complexity 2. Factor loadings were then estimated using weighted least square and compared with estimations based on maximum likelihood. Factor structures were rotated using the oblique transformation technique oblimin. In order to examine the robustness of these findings, these findings were compared to results based on rotation methods geominQ and cluster techniques estimations. We tested how children's age interacted with the obtained factors for each scale, using a median split by age for comparing the younger half to the older half of the sample. Further, we obtained individual scores for each subscale by calculating means for each participant for all items that composed a factor. Then, we examined the Pearson's correlations between children's age and individual scores. We examined convergent and divergent validity for the JMES Video with Pearson correlation to subscales of Television Mediation Scale (Valkenburg et al., 1999). Preliminary analyses indicated that children's gender was not associated with scores on any of the final subscales (ps > .05) and gender was not considered further in this study. The data and analysis scripts are available on Open Science Framework: osf.io/p57ys/.

RESULTS

For video, 27% of parents Always or almost always watched with their children, 17% did so More than half the time, 19% About half the time, 26% Less than half the time and 9% Never or almost never (missing: 2%). On average, children watched video content for about 9 hours per week (Table 1).

TABLE 1.

Mean and SD for children's time spent with watching video and digital games per week.

	N	Mean	SD
Time spent watching video	322	9 h and 11 min	5 h and 59 min
Time spent playing digital games	159	2 h and 45 min	3 h and 17 min

For games, 8% of parents reported that they Always or almost always played digital games with their children, 6% did so More than half the time, 8% About half the time, 15% Less than half the time and 10% Never or almost never did. The remaining 53% of children did not play digital games. On average, children who played digital games did so for about 2.75 h a week (Table 1).

Joint media engagement while watching video

All 15 JMES Video items were answered by 309 participants. A further 13 cases had one or two missing items. The sample's median for an item was used to impute these missing values. After imputation, complete answers for JMES Video were available from 322 respondents. Initial items used to assess JMES Video can be found in Appendix S1. Initially, three categories were used to create items for JMES while watching video:

1. JMES Video–Cognitive Scaffolding.

2. JMES Video–Affective Scaffolding.

3. JMES Video–Technical and Physical Scaffolding.

A 3‐factor solution did not yield stable results (Appendix S1). While MAP suggested a 2‐factor solution, VSS complexity 2 suggested a 3‐factor solution. In order to identify a theoretically and statistically sound factor structure, we iteratively removed items based on uniqueness values above .75 and complexity loadings above 1.7 (none of these items was theoretically supposed to load on several factors) while exploring 3‐ and 2‐factor solutions. We found that a 2‐factor solution with 8 items fit the data best. However, the factor based on two items was a mathematical artefact, meaning that it was created due to overall lower means. Means of both items in this factor were significantly lower than all other items using pairwise comparisons (all ps < .001). Therefore, these items were dropped. A one‐factor model consisting of Cognitive and Affective Scaffolding for JMES Video is the most appropriate and retained solution for JMES Video Table 2. The factor accounts for 55% of the variance found.

TABLE 2.

Final solution of JMES Video, with 6 items loading on one factor, consisting of items tapping cognitive and affective scaffolding.

Item	Cognitive and affective scaffolding factor	Communality	Uniqueness
1. I relate what we are watching to what the child has experienced, for example, “Remember, we also saw ducks at the park” or “we also went swimming!”	0.83	0.68	0.31
2. I point out things on the screen that the child might not know or see, for example, “Look, there is a hammer, you can hit nails with that” or “Look there's the boat.”	0.76	0.57	0.43
3. I ask the child to tell me what they see on the screen, for example, “What is that over there?” or “What colour is this?”	0.70	0.50	0.51
4. I ask my child about what we are watching, for example, “Was Big Bird in the house earlier?” or “Do you think they will be friends now?”	0.82	0.67	0.32
5. I talk about what we have watched together when we do something else, for example, “Let's jump just like Peppa Pig did!”	0.60	0.36	0.64
6. I clarify feelings and thoughts in the videoclips, for example, “Skye did not think Rocky was there” or “The Grinch became very angry.”	0.65	0.43	0.58

Convergent and divergent validity

To provide a measure of convergent and divergent validity with the JMES Video, parents also completed the 15‐item Television Mediation Scale (Valkenburg et al., 1999). We computed scores for each of the three subscales from the Television Mediation Scale: instructive mediation, restrictive mediation and social coviewing. The JMES Video was related to the instructive mediation and the social coviewing subscales, indicating convergent validity for that measure as the JMES Video should capture instructive as well as social aspects of JME (Table 3). Demonstrating divergent validity, the JMES Video was not related to the restrictive mediation subscale. While the instructive mediation subscale was related to the age of the children, JMES Video was not, showing that parents' engagement was captured equally well across the age range in the JMES Video, meeting the goal of developing a more age‐appropriate scale.

TABLE 3.

Bivariate correlations for JMES Video and Television Mediation subscales (Valkenburg et al., 1999) and children's age.

	Television Mediation Scale			JME Video
	Social coviewing	Restrictive mediation	Instructive mediation
Restrictive mediation	0.18***	–
Instructive mediation	0.47***	0.34***	–
JME Video	0.42***	0.1	0.57***	–
Child's age	−0.03	0.09	0.11*	−0.04

***p < .001.

*p < .05.

Joint media engagement while playing digital games

All 17 JMES Games items were answered by 114 parents. A further six cases had one missing item. The sample's median for an item was used to impute these missing values. Five cases that had more than 2 items missing were excluded. After imputation, complete answers for JMES Games are available from 120 respondents. For the JMES Games, three categories were used to create items for JMES while playing digital games:

1. JMES Games–Cognitive Scaffolding.

2. JMES Games–Affective Scaffolding.

3. JMES Games–Technical and Physical Scaffolding.

We used the same general strategy as for JMES Video to examine factor structure for JMES Games. Initially, we were not able to use weighted least squares estimations, as model estimations did not converge. Therefore, we based our analyses on maximum likelihood estimations (Appendix S2). Several methods suggested a 3‐factor solution suited these data (MAP, VSS complexity 2). Once again, to identify a theoretically and statistically sound factor structure, we iteratively removed items based on uniqueness values above .75 and complexity loadings above 1.7 (none of these items was theoretically supposed to load on several factors). During this process, 5 items were removed, either due to uniqueness over .75 (2 items) or due to theoretically irrelevant complexity over 1.7 (3 items). A 3‐factor solution was stable for the remaining 12 items, even when using different methods for rotation (Oblimin, GeoMinQ, Cluster; see Table 4).

1. JMES Games–Cognitive Scaffolding (7 items).

2. JMES Games–Affective Scaffolding (2 items).

3. JMES Games–Technical and Physical Scaffolding (3 items).

TABLE 4.

Final solution of JMES Games, with 12 items loading on 3 factors: JMES Games–Cognitive Scaffolding, JMES Games–Affective Scaffolding and JMES Games–Technical and Physical Scaffolding.

Item	Scaffolding factors			Communality	Unique‐ness
	Cognitive	Affective	Technical
1. I relate the game we are playing to what the child has experienced, for example, ““Remember, we also saw ducks at the park“!” or “we also went swimming!”	0.83			0.68	0.32
2. I point out things on the screen that the child might not know or see, for example, “Look, there is a hammer, you can hit nails with that” or “Look there's the boat.”	0.84			0.72	0.29
3. I comment on events from earlier in the game, for example, “Do you remember, you have to get the scissors first?”	0.72			0.59	0.41
4. I ask the child to tell me what it sees, for example, “What is that over there?”, “What colour is this?”, or “Which one is the rabbit?”	0.79			0.64	0.36
5. I ask questions regarding what we are doing, for example, “Where does the car go?” or “Where do you have to put the vegetables?”	0.79			0.61	0.39
6. I ask about why things are happening in the game, for example, “Why to do you drive the car there?” or “Why do all the blocks need to be there?”	0.75			0.67	0.33
7. I talk about events from games when we do something else, for example, “Should we put all the fruits there, just like in Toca Boca?”	0.67			0.41	0.59
8. I encourage the child if I notice that the child is afraid/worried, for example, “It is not dangerous.”, “You can do it!”, or “Watch out, the ghost is there.”		0.73		0.50	0.50
9. I try to calm the child if they get upset about the game, for example, “It is okay, you can try again.”		0.96		0.99	0.01
10. I help the child by holding the tablet/mobile phone so the child can play more easily.			0.49	0.27	0.72
11. I show the child what they can do in different games/apps to complete tasks/win. For example, “Once you catch all the green ones you'll go to the next level.” Or “You have to click here and choose first.”			0.54	0.35	0.64
12. I help the child to click on the correct button to move forward in the game.			0.91	0.86	0.14

Examining relations to age for factors in JMES Games

To assess whether factor structure was stable across age, the data were split in half, based on a median split for children's age (median = 43.7 months). The median age was higher than the median used as a cut‐off for JMES Video since playing digital games was less common in younger children. Comparing the two models for the younger and older children, one item, “I help the child by holding the tablet/mobile phone so the child can play more easily,” changed its factor loading, suggesting that JME dynamics around games changed in this age range. In the older sample, the item loaded with JMES Games–Affective Scaffolding, while in the younger sample, it loaded with JMES Games–Technical and Physical Scaffolding. As such, the factor JMES Games–Technical and Physical Scaffolding changed its meaning somewhat with age. In the younger sample, the item “I help the child to click on the correct button to move forward in the game,” had a communality of above .95, which indicated that this item explained a lot of variance in the factor. In the older sample, the item “I show the child what they can do in different games/apps to complete tasks/win. e.g., ‘Once you catch all the green ones you'll go to the next level.’ or ‘You have to click here and choose first.’” had a communality of above .95, explaining more variance for this factor in the older sample. While all three items were related to technical aspects of using the device, in younger children, there was more focus on using the physical device, whereas in older children, there was more focus on how to use apps installed on the device.

Removing the item “I help the child by holding the tablet/mobile phone so the child can play more easily” from the final solution did not improve the consistency across the age range. Actually, the opposite happened: The consistency across the age range suffered from removing the item. This might be due to the low sample size. Therefore, this item was retained in the final solution.

Exploring relations between JMES measures and with age

Correlations between the JMES and child age are shown in Table 5. JMES Games–Technical and Physical Scaffolding use was the only factor related to age. Older children received less help than younger children with devices when playing games. JMES Video and JMES Games–Cognitive Scaffolding were strongly correlated with each other. Parents who frequently engaged in cognitive scaffolding with their children when watching video also did so when their children were playing digital games. As hypothesized, all scales for JMES were correlated with each other. However, there are some differences. Affective items were grouped in a separate factor for games, whereas affective items were grouped with cognitive items for the video scale. This suggests emotions might be discussed differently for video and for games.

TABLE 5.

Bivariate Pearson's correlations between JMES Video and Games and children's age.

	JMES Video	JMES Games–Cognitive	JMES Games–Affective	JMES Games–Technical
JMES Games–Cognitive	0.58***	–
JMES Games–Affective	0.18*	0.20*	–
JMES Games Technical	0.17	0.17	0.22*	–
Children's age	−0.05	−0.04	0.05	−0.2*

***p < .001.

*p < .05.

DISCUSSION

In the present study, we developed a novel Joint Media Engagement Scale (JMES) specifically for children aged five and under. We designed the scale to capture collaborative interactions regarding video viewing and playing digital games, respectively. Overall, this new scale demonstrated a reliable factor structure for children aged 1–5 years.

Video content and JME

The initial scale included items that were intended to tap different aspects of JME (cognitive scaffolding, affective scaffolding, technical/physical scaffolding) as described in a metanalysis by Ewin et al. (2021). However, the final JMES for video was comprised of one subscale consisting of six items, capturing cognitive and affective scaffolding around content. This JMES Video demonstrated age invariance from 1 to 5 years in terms of factor loading and did not show any differences due to gender. During JME, parents discussed important aspects needed to understand the story or characters' actions and emotions. The retained items for JMES Video highlighted that parents focused on cognitive and affective scaffolding. As such, the current scale addressed several aspects of JME that have been shown to be of importance to children's learning (Ewin et al., 2021). Additionally, we followed Ewin et al.'s suggestion to assess whether scaffolding interactions occur after media has ended to help children connect the digital content with their lived reality. For instance, parents reported that they talked about what they had watched after the show when they were doing something else. Thus, the current scale captured discussion of digital media beyond the digital media experience, extending previous research conducted using other mediation scales (Dore et al., 2020; Medawar et al., 2023).

Convergent and divergent validity were determined by examination of correlation with the Television Mediation Scale (Valkenburg et al., 1999). We found that the JMES Video correlated positively with Valkenburg et al.'s instructive mediation and social coviewing subscales but not with the restrictive mediation subscale, as expected. The restrictive mediation subscale asked parents about how they grant children access to television, a general media practice that is not part of JME. Hence, this demonstrated the specificity of the new scale designed to measure JME, rather than general media practices.

The hypothesized structure of cognitive scaffolding, affective scaffolding and technical and physical scaffolding was not supported for JMES Video. Items intended to capture affective aspects of JME assessed the affect of characters appearing on the screen, rather than children's own affective states. Therefore, cognitive and affective scaffolding merged into one factor. Furthermore, technical and physical scaffolding were rarely reported when parents were asked about watching video and therefore these items were not retained in the final factor structure.

Digital games and JME

Three subscales, Cognitive Scaffolding (7 items), Affective Scaffolding (3 items) and Technical and Physical Scaffolding (2 items), did emerge as predicted for the JMES Games. The JMES Games also captured discussions of content beyond the digital experience per se, which could play a vital role in making the content more understandable and memorable. JMES Games factor structure separated cognitive and affective scaffolding. Affective scaffolding captures how the parent tried to help the child regulate their emotions when playing. For instance, parents reported that they offered emotional support if the child got upset when they lost or if they could not do something in the game. This differed from JME when watching video.

Another difference between JME when watching video and when playing digital games was how parents helped children with technical issues (Ewin et al., 2021). Contrary to watching video, parents tended to help children during digital game play both with using devices and with tasks that were part of the games. However, the kind of help parents offered varied with age. For younger children using the device was more in focus, whereas for older children using apps and solving problems in the games was more in focus for the technical/physical scaffolding. Moreover, the negative correlation between JMES Games–Technical and Physical Scaffolding and children's age suggested that technical and physical scaffolding decreased with age, perhaps as children gain experience and skill with digital media.

Limitations and future research directions

There are some limitations to this scale development. It was tested on a relatively homogenous, well‐educated sample in Sweden. Additional samples with more diverse cultural, educational and economic backgrounds are needed to replicate the findings and establish generalizability across different populations and countries. Video viewing is still the most common media activity for children under 5 years of age. The JMES Video was the most robust. The JMES Games subscales were robust for content and affect but less stable for technical/physical device use, which changed as a function of age. Given the limited sample size, the interpretation of the factor structure of JMES Games should be viewed with caution. In particular, future research should test the robustness of factors that included 2 or 3 items. Additional samples are needed to assess this pattern further.

In light of the rapidly changing media landscape, researchers should move past solely examining the amount of time children view digital media and begin to analyse the context of children's digital media use, too (Barr et al., 2020). The current JMES is intended to elucidate the context of digital media use by capturing how families use digital media together. Further research needs to investigate JME in different cultures and in relation to parental attitudes towards digital media. If we understand how digital media is used, we will better understand how it might be used to support children's development. Hence, examining JME in relation to, for example, language or cognitive development, will further our understanding of the consequences of digital media use in children. This scale will help to assess change and outcomes of intervention studies that attempt to impact how parents use digital media together with their child. Additionally, since digital media is integrated in everyday life, JME might be a useful indicator for studies examining parent–child relationships in general, even if the intervention is not directly aimed at JME.

Conclusion

Overall, increasing the validity and age‐appropriateness of digital media use assessments in children under 5 years of age is essential given the rapid change in the digital media landscape and the high prevalence of media use in this age range (Barr et al., 2024). For video watching, one JMES Video emerged which combined Cognitive and Affective Scaffolding. When playing digital games, three different and separable JME scaffolding strategies emerged: Cognitive Scaffolding, Affective Scaffolding and Technical and Physical Scaffolding. The current work will be extended by incorporating the JMES Video and JMES Games into an updated version of the MAQ (Barr et al., 2020) that is currently available in 6 languages (Swedish, German, Italian, Spanish, French, Portuguese, Hebrew, English) and in use in 12 countries. Data for this scale will be collected and collated across multiple projects to retest the reliability and validity of the new JMES in different contexts. This future work will also assess how the JMES is related to family demographic factors, family media use practices and child outcomes including language, social competence, executive functioning and emotion regulation.

AUTHOR CONTRIBUTIONS

Felix‐Sebastian Koch: Conceptualization; methodology; data curation; validation; formal analysis; visualization; writing – original draft; writing – review and editing; funding acquisition; project administration; resources. Rachel Barr: Conceptualization; investigation; validation; writing – original draft; writing – review and editing; resources; project administration. Annette Sundqvist: Conceptualization; methodology; investigation; validation; formal analysis; writing – review and editing; funding acquisition.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

Supporting information

Appendix S1.

Appendix S2.

Koch, F.‐S. , Barr, R. , & Sundqvist, A. (2025). The Joint Media Engagement Scale (JMES): An instrument for measuring shared media use with children aged 1 to 5 years old. British Journal of Developmental Psychology, 43, 98–110. 10.1111/bjdp.12526

DATA AVAILABILITY STATEMENT

Data and analysis script are available on Open Science Framework: https://osf.io/p57ys/?view_only=7de0a587f9644c64a15cb80d60450090