Abstract
The people who regularly interact with an adolescent form that youth's social network, which may impact participation. We investigated the relationship of social networks to participation using personal network analysis and individual interviews. The sample included 36 youth, age 11 – 16 years. Nineteen had diagnoses of learning disability, attention disorder, or high-functioning autism and 17 were typically developing. Network analysis yielded 10 network variables, of which 8 measured network composition and 2 measured network structure, with significant links to at least one measure of participation using the Children's Assessment of Participation and Enjoyment (CAPE). Interviews from youth in the clinical group yielded description of strategies used to negotiate social interactions, as well as processes and reasoning used to remain engaged within social networks. Findings contribute to understanding the ways social networks are linked to youth participation and suggest the potential of social network factors for predicting rehabilitation outcomes.
Keywords: personal network analysis, adolescence, community participation, environment, qualitative research
Engagement in occupations within clients’ roles and life situations is a primary outcome of interest for occupational therapists. The World Health Organization (WHO) describes involvement in life's situations as participation, where, in interplay with other personal and environmental factors, participation contributes to outcomes of functioning, disability and health (WHO, 2007). For children and youth, participation refers to engagement in everyday activities that entail naturally occurring interactions, such as during mealtime, playing or engaging in the classroom. Participation is a health-related construct whereby involvement (i.e. participation) in the real world is critical to healthy development for children with and without disabilities (Forsyth & Jarvis, 2002).
Children and youth growing up with disabilities are at risk for lower participation in everyday life activities (King et al, 2004). Within childhood and adolescent development, activity engagement moves along a continuum from family-based activities toward more peer and community-based behaviors and undertakings. Increases in competence and independence accompany developmental maturation, and coincide with gradual environmental shifts of daily life from home and school to more diversified environments involving community and society (WHO, 2007). However, for many youth with disabilities, participation doesn't readily move toward more diverse participatory activities (Imms, Reilly, & Dodd, 2008). Rather, socially partnered participation in activities remains more family-based in contrast to typically developing youth who move toward engaging in more socially partnered activities with peers in the community (Engel-Yeger, Jarus, & Law, 2009; Jarus, Anaby, Bart, Engel-Yeger, & Law, 2010).
As the nature and complexity of life situations and participatory expectations shift throughout childhood and adolescence, physical, social and attitudinal environments remain important contextual factors for youth growing up with chronic conditions (WHO, 2007). Social networks, and the relationships of those within the networks, are important aspects of the youth's social environment contributing to development (WHO, 2007). Social networks refer to the people within the social structures surrounding an individual, such as a family or a classroom, and are a part of the developing youth's social environment that function as dynamic contexts of adolescent participation in occupations. Social networks impact the health and development of those within the network (Smith & Christakis, 2008; Holland, Reynolds, & Weller, 2007), and the perceptions, beliefs and behaviors of those within the group (Knoke & Yang, 2008).
In practice, occupational therapists regularly involve individuals from their client's social network (e.g. family members, teachers, classmates) as a way of facilitating their client's engagement in natural contexts. For youth growing up with disabilities, occupational therapy (OT) contributes a unique disciplinary understanding of the interlinked nature of the developing youth, environment and engagement in occupations and activities within the home and community. Integral to the OT process is the identification and understanding of environmental factors that are salient to the client (American Occupational Therapy Association, 2014). This exploratory study aimed to contribute to the understanding of the interrelated nature of the social environment and participation. We sought to identify and describe social network (SN) factors relevant to participation and to identify social network variables for further investigation. We asked, ‘How are social network metrics related to participation?’ and ‘How do youth from the clinical group participate within their social networks?’ We hypothesized that two types of social network variables, compositional (i.e. what types of people make up the network) and structural (e.g. groupings of people, connectedness of people in the network), would have strong associations with measures of participation. However, because the meaning of some social network metrics (e.g. level of connectedness of close friends) are not easily understood within the everyday lives of youth with disabilities, qualitative interviews from youth in the clinical group were used to expand understanding of participatory experiences within their social networks.
Methods
Design and Study Procedures
The research design was a mixed method embedded design – correlation model (Creswell & Plano Clark, 2007). Cross-sectional quantitative and qualitative data were collected simultaneously but analyzed separately whereby the qualitative data from the subset of youth in the clinical group was used to more fully elaborate the results of the correlation analyses and enhance interpretation (Creswell & Plano Clark, 2007).
Data were collected in a home-like community-based research lab or in the participant's home, whichever the parent preferred. All participants, with assistance of a parent as needed, underwent collection of data used to describe the sample and measurement of the youth's SN and participation. Immediately after collection of SN data, the youth's SN was graphically mapped and shared with the youth. Maps were checked for accuracy, relationships elaborated upon and the map validated as representative of each youth's SN. On each map, network members were labeled and represented by a circle whose size indicated the network member's relative connectivity to others in the network (i.e. degree centrality), such that larger circles indicated more connections. Lines on the map indicated the presence of an existing relationship between the network members (Figure 1). Network visualizations (i.e. maps) and metrics were used to individualize interview of youth from the clinical group aimed at elucidation of participation experiences within their SN. With the youth's network map available to reference during the interview, the youth were asked about their experiences, interactions and activities with network members – the network map was used to ask about specific individuals by name and relationships with each other.
Figure 1. Example of personal network visualizations.
Note: network on left: 15/16-year-old youth from clinical group; network on right: age and gender matched youth from typically developing group; shape = age category (square/circle=adult, circle=child/youth); color = relationship (grey=kin by blood or marriage, white=non-kin); size = number of connections to others in the network (i.e. degree centrality; larger shape = more connections to others in the network); names have been changed to protect privacy
Participants & Recruitment
A convenience sample of thirty-six youth participated in this study. Following approval from the sponsoring university's Institutional Review Board, 19 youth ages 11-16 years (mean age 13.9 ± 1.3; 84.2% male) with a diagnosis of learning disability (LD), attention deficit hyperactivity disorder (ADHD) or autism spectrum (ASD), and 17 typically developing youth (mean age 13.9 ± 1.2; 82.4% male) matched for age and gender distribution were recruited from regional schools and clinics via word of mouth, recruitment flyers, and professional referral. Study recruitment was conducted when the youth were most available to participate, which was during the last two months of the youths’ school year and the first two months of the summer break. All parents who contacted study researchers agreed to participate in the study. Only two parents who agreed to participate were unable due to scheduling conflicts. Additionally, one youth, whose parent agreed to participate, declined invitation to participate. Written informed consents from parents, and verbal and written assent from the youth, were obtained after discussion of study risks, benefits and the study's voluntary nature. All participating youth were offered a $10.00 gift certificate as a token of thanks. Of the youth in the clinical group, 14 had diagnosis of LD, ADHD or ASD and the remaining 5 reported a combination of diagnoses to include ASD plus LD, ADHD plus ASD, or ADHD plus LD. Parents served as secondary study participants and provided diagnostic, demographic, performance information, and when necessary, assisted in reporting on the youth's SN and participation.
Quantitative Methods
Measures of the Social Network
SN data were collected and analyzed using conventions of personal network analysis and network visualization established in the social sciences (Borgotti, Everett, & Johnson, 2013). Twenty network variables were measured consisting of two types of SN variables, compositional and structural (Table 1). Network composition variables provided quantitative information regarding who and what types of people made up each youth's network (e.g. who are kin, proportion providing social support). Structural network variables provided quantified information regarding the social structure (e.g. social groupings, degrees of connectedness between network members). An interviewer-assisted paper and pencil survey was developed for this study, which was used to establish the names and characteristics of people within each youth's network; each youth identified 25 people within their SN (15 with whom they do things, and 10 acquaintances) and reported characteristics of network members (i.e. age, gender, kinship, and types of social support provided). Afterwards, Egonet (version 2012-05-18; McCarty, 2012), an open-source Java-based network software program, was used to collect information about network members’ relationships to each other (i.e. who had existing relationships), calculate network structure (e.g. measures of connectedness within the network), and generate each youth's network map.
Table 1.
Social network variables
| Network variable | Description |
|---|---|
| Entire network - Compositional network variables | |
| Number same gender * | Number of network members same gender as youth. |
| Number kin * | Number of network members related to youth by blood or marriage. |
| Number adults | Number of network members youth considers a grown-up. |
| Number weak ties | Number of network members youth reports to know a little or just know who he/she is. |
| Entire network - Structural network variables | |
| Density | Gross measure level of integration of entire network. Number of existing network ties proportional to number of total possible ties (sum total of raw degree centrality divided by 300). |
| Average number ties | Gross measure of level of network integration. Mean number of direct connections each network member has to others in network (mean degree centrality). |
| Number ties most central * | Refers to social power that can be derived from being directly connected to others. Number of connections to others in network by most connected member of network. |
| Sibling network (subgroup) variables | |
| Average number ties - all siblings | Gross measure of integration of sibling subgroup within network. For all participants with siblings (n = 33), average of each sibling's number of connections to others in entire network (mean degree centrality of all siblings). |
| Support network (subgroup) variables | |
| Size - support network | Number of network members providing social support. |
| Average number ties - support network | Gross measure of integration of support subgroup within network. For all network members providing at least one kind of social support (i.e., can share feelings with, is helpful, gives information, sticks up for youth), average of the number of connections to others in entire network (mean degree centrality of all supportive network members). |
| Peer network (subgroup) variables | |
| Size - peer network * | Number of network members neither kin nor grown-up. |
| Same gender - peers * | Proportion of peers same gender as youth. |
| Weak tie peer network (subgroup) variables | |
| Size - weak tie peer network * | Number of peer acquaintances or network peers youth reports to know a little or just know who he/she is. |
| Same gender - weak tie peer network * | Proportion of peer acquaintance network same gender as youth. |
| Average number ties - weak tie peer network | Gross measure of integration of acquaintance peer subgroup within network (mean degree centrality of peer acquaintance subgroup). |
| Social support - weak tie peer network * | Proportion of peer acquaintance network members providing social support. |
| Strong tie peer network (subgroup) variables | |
| Size - strong tie peer network | Number of close peers or network peers youth reports to know really well or is close to. |
| Same gender - strong tie peer network | Proportion of close peers same gender as youth. |
| Average number ties - strong tie peer network* | Gross measure of integration of close peer subgroup within network (mean degree centrality of strong tie peer subgroup). |
| Social support - strong tie peer network * | Proportion of all strong tie peer network members providing social support. |
Note
Significant correlation (p < 0.01) to a participation variable
Measure of Participation
The Children's Assessment of Participation and Enjoyment (CAPE) was used to measure participation in activities beyond those mandated in school (King et al., 2004). The CAPE is a valid and reliable self-report instrument designed to assess the manner in which youth, with and without disability, ages 6 through 21 participate in everyday activities (King et al., 2004; King et al., 2006). Participation variables used in this study were the CAPE dimension scores for activity Diversity (number of activities engaged in), Intensity (activity frequency), With Whom (indicating with whom the activity occurred most often), and Where (indicating location activity occurred most often). Dimensions were further classified by domain (Informal, referring to more spontaneous types of activities such as playing cards) and activity type (Recreational, Physical, and Social). For example, the CAPE score for Where-Social was used as the participation variable indicating the range of locations where the youth engaged in social activities; lower Where-Social scores correspond to proportionally more social activities engaged in at home or at a relative's home rather than at a friend's home or in the community.
Correlation Analysis and Results
Correlation analyses of network and participation data were conducted using SPSS version 21. Spearman's rank-order correlation coefficients were calculated for all combinations of network and participation variables, which resulted in a 16 by 20 correlation matrix. Due to increased potential for Type 1 error of this exploratory analysis, non-parametric statistics were used with two-tailed significance level set at p < 0.01 and scatter plots inspected for all combinations of significant correlations.
Ten SN variables had strong correlations with at least one participation variable from the CAPE for a total of 17 statistically significant correlations, of which eight were compositional network variables and two were structural. Table 2 details statistically significant correlations.
Table 2.
Correlation coefficients (Spearman's Rho) and p-values for statistically significant (p < 0.01; 2-tailed assumed) associations of network and participation variables.
| Participation variable from | Compositional Network Variables |
Structural Network Variables |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| CAPE scores | Nmbr same gend | Same gend peers | Same gend weak tie peer ntwk | Size peer ntwk | Size weak tie peer ntwk | Supp weak tie peer ntwk | Supp strg tie peer ntwk | Nmbr kin | Nmbr ties most cent | Ave nmbr ties strg tie peer ntwk | |
| Diversity | Informal | -- | -- | -- | -- | -- | -- |
rs = .510 p = 0.001 |
-- | -- | -- |
| Recreational | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | |
| Physical | -- | -- | -- | -- | -- | -- | -- | -- | -- |
rs = .452 p = 0.006 |
|
| Social | -- | -- | -- | -- | -- |
rs = .432 p = 0.008 |
-- | -- | -- | -- | |
| Intensity | Informal | -- | -- | -- | -- | -- | -- | -- |
rs = −.481 p = 0.003 |
-- | -- |
| Recreational | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | |
| Physical | -- | -- | -- | -- | -- | -- | -- |
rs = −.484 p = 0.006 |
-- | -- | |
| Social |
rs = .447 p = 0.006 |
-- | -- | -- | -- | -- | -- | -- | -- | -- | |
| With Whom | Informal | -- |
rs = .504 p = 0.002 |
-- | -- | -- | -- | -- | -- | -- | -- |
| Recreational | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | |
| Physical | -- |
rs = .505 p = 0.002 |
rs = .441 p = 0.007 |
-- |
rs = .580 p = 0.000 |
-- | -- | -- | -- | -- | |
| Social | -- |
rs = .470 p = 0.004 |
-- |
rs = .515 p = 0.001 |
-- | -- | -- |
rs = −.508 p = 0.002 |
-- | -- | |
| Where | Informal | -- | -- | -- | -- |
rs = .467 p = 0.004 |
-- | -- |
rs = −.520 p = 0.001 |
-- | -- |
| Recreational | -- | -- | -- | -- | -- | -- | -- | -- |
rs = −.475 p = 0.003 |
-- | |
| Physical | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | |
| Social | -- | -- | -- | -- | -- | -- | -- |
rs = −.524 p = 0.001 |
-- | -- | |
Note: CAPE = Children's Assessment of Participation and Enjoyment; nmbr = number; gend = gender; ntwk = network; supp = support; strg = strong; cent = central
Qualitative Methods
Qualitative Analysis
Qualitative analysis of interview data from youth in the clinical group was conducted to the level of thematic survey, which focused on moving beyond the cataloging of topics brought up by participants toward a description of patterned responses discerned from participants’ discussions (Sandelowski & Barroso, 2003). Additionally, in an effort to advance qualitative analysis toward greater conceptualization, we incorporated interpretive use of conceptualizations to recast portions of the data relevant to this research question (Sandelowski & Barroso, 2003). An iterative process of data collection, analysis and interview modification was used to enable discussion and refinement of emerging conceptualizations with subsequent participants. For instance, if a participant brought up or expanded on experiences with individuals not inquired about or named on the youth's SN map, researchers encouraged the discussion. From these unprompted discussions, inquiry in subsequent interviews was expanded to include inquiry of sibling interactions with network members. Interviews were audio-recorded, transcribed verbatim, checked for accuracy, and sections of the text coded using the participant's words as initial coding labels. Codes with similar meanings were then consolidated and reduced to create more conceptual codes that were then grouped into broad descriptions of the data (Ryan & Bernard, 2003). For instance, the initial coding labels of Avoiding Trouble, Stay Quiet and Move Away were combined into the conceptual code Passive Strategies, which was later combined with other conceptual codes to form the broader descriptive category of Social Negotiation. Two researchers repeatedly read each transcript, comparing emergent categories or conceptualizations to the data from previous interviews within the group, then annotating the data and modifying the categories to ensure consistency in defining category properties and application of data to them (Ryan & Bernard, 2003).
Qualitative Results
Qualitative descriptions of the ways that youth in the clinical group fit into their social groups resulted in the emergence of two related conceptualizations within the data, social negotiation and working hard socially.
Social Negotiation
Multiple (7 of 19) youth conveyed stories of being part of a social group that used social processes of negotiation, compromise, turn taking and consensus. “We [are] kind of like a republic...we all [take part in making] a decision.” [Youth #15] However, not all youth in the clinical group described high levels of active and egalitarian engagement within their groups. Some described use of passive strategies. From one youth with diagnosis of ADHD
Well, it's not that I don't have a choice [about what the group does or talks about], I know that would just cause way too many problems...so even if I wanted to [do something different or change the topic of conversation], I wouldn't ‘cause I...don't want to cause trouble. [Youth #5]
This youth actively chose to avoid social negotiations. He was also a youth who, despite a desire for peer friendships, reported he did not have any school friends. Another youth with diagnosis of ASD described a different strategy for avoiding social negotiations: “I just stop talking or maybe even move away.” [Youth #16] Descriptions of social avoidance strategies illustrate reasons youth with disability can be perceived as being social followers or on the periphery of their peer groups.
For some youth in the clinical group, diagnostically related processing lags and/or social awkwardness impacted negotiation of interactions involving siblings when peers were present. For these youth, sibling interactions could take on tones of exclusion that were difficult for them to negotiate. In describing interactions with peers known by both the youth and the youth's sibling, this youth with ASD explained, “They are nice to me. We try and talk but sometimes we can't because...[sibling] walks up...[then] it's over, [sibling] dominates.” This youth's mother went on to clarify that the sibling puts down the youth when others are involved in the interaction [Youth #2] Some endured overt, and at times condescending, sibling interactions that worked to exclude or marginalize the youth, while others endured inadvertent exclusion. From a youth with diagnosis of ADHD:
Well, it'll be the two of them [brother and family friend] talking in the back, and me just overhearing and making comments when they say something...We'll be talking and they'll get excited with each other and they'll [laugh and], even do like weird noises just messing with each other, like [talking in] weird code names. [Youth #5]
Difficulty in comprehending jokes and interpreting social tone contributed to this youth being the outsider in sibling interactions containing a peer. Despite some siblings’ exclusionary tendencies, several parents expressed expectations that the youth from the clinical group would be included in the sibling's activities and social interactions with others – even when these expectations were met with resistance from the sibling.
Working Hard Socially
Some youth were able to remain actively engaged in their peer groups, but had to work hard to do so. Of these youth, some took on the role of social facilitators to ensure they had someone with whom to engage, even when interactions were socially forced or awkward. A youth with ADHD explained, “They didn't keep a conversation going for long. I was the main one who kept the conversation going.” [Youth #14] He reported that he had to keep interactions going, otherwise he would be faced with hanging out by himself. This youth felt as if he had to keep his loosely associated social group from disbanding in order to remain socially engaged.
A different youth described his strategy for keeping a peer engaged, “I try and make him happy...I try and talk about something that will make him happy.” [Youth #16] This strategy did not always work; the youth was referring to a peer with whom he had an on-again off-again friendship. Another youth with ADHD described walking up to peer groups and attempting to start conversations; he reported, “I'll just start trying to make conversation...I'll try and get him to talk...I guess it's one of my fatal flaws...it makes me seem annoying.” [Youth #11] In conveying stories of their social interactions, multiple youth gave no indication of attentiveness to nuances of social interactions such as body language. Lack of attention to social nuances contributed to instances of missed social cues such as messages of social censorship or rejection conveyed through extended group silence. For several participants, interjecting themselves into social groups contributed to marginalizing rather than integrating them into the group.
Consistent with the diagnoses represented in the study sample, some youth described difficulties related to language processing. A respondent with ASD put it this way, “Well, most of the time they are usually talking about something...stuff that doesn't really make a lot of sense to me.” [Youth #20] From a different youth with diagnosis of ADHD, “When it came to something that we would ... talk about it and I would just listen in and see if I, if anything made sense to me that I would like to talk about.” [Youth #14] Another youth described how he was able to use in-the-moment awareness of comprehension difficulties to assist with maintaining social appropriateness during peer interactions. “Basically I would just kind of twiddle my thumbs and listen to what their conversation and see if it, it peaked my interests and then I would get into the conversation if it peaked my interest.” [Youth #14] This youth's meta-cognitive strategy helped him remain integrated within his peer groups. Participants from each diagnostic group relayed stories of how their processing lags impacted their ability to read in-the-moment social interactions. For others participants with LD or ASD, the sheer language load was, at times, enough to take them out of social collaborations. Youth who had developed some insight into the shortcomings of their social and or communication skills were able to employ socially acceptable strategies that helped keep them engaged within their social networks.
Discussion
We hypothesized that SN methods could be used to identify and quantify compositional and structural network factors having significant correlations to childhood participation, and used qualitative data to expand understanding of quantitative findings. Eight compositional network variables (e.g. gender, kin) and two structural network variables (e.g. number of ties of most connected) had significant correlations to at least one measure of participation.
Acquaintances
For study participants, the number of peer acquaintances (i.e. size of weak tie peer network) had significant positive associations with whom and where activities occurred. The greater the number of peer acquaintances, the greater the proportion of activities partnered by non-family members and engaged in beyond home or a relative's home. This finding suggests a potential strength in having peer acquaintances, or weak ties, during adolescence. Within the adult social network literature, weak ties have been shown to provide important opportunities for bridging social groups and opportunities available beyond those offered by close or immediate network members (Granovetter, 1973). Our study findings suggest that weak ties in adolescence can potentially play a role in facilitating developmentally appropriate shifts in participation away from family and home-based activities toward activities partnered by members of the youth's community and engaged in within the community.
Peers
Strong positive relationships were observed for activity diversity and the level of connectedness of participant's good friends. We also found strong positive relationships between activity diversity and the amount of social support received from the good friends. Notably, same-gender variables had strong positive correlations with participation - youth engaged in activities with more peers and others not related to them, as well as engaged in more activities outside of the home, when the youth's network contained more peers and proportionally more same-gender network members.
For youth from the clinical group who were able to use social negotiation during their peer interactions, qualitative data elaborated on the ways that they engaged with peers from larger and primarily same-gender peer groups. These youth described engaging in a broader range of activities with their peers - activities that often included physical and spontaneous hanging-out types of activities such as horsing around. Findings regarding peers points to the potential importance of developing interventions that target social skills important for developing and maintaining same-gender interactions and social navigations. Greater understanding of the relationship between same gender network members and participation for youth with disabilities can be useful in guiding development of interventions that focuses on development and maintenance of same-gender relationships.
Connectedness of Network Members
One structural network measure, the number of ties held by the most connected person in the network, had a significant negative relationship to where recreation activities occurred. This finding contributes to general understanding of the kinds of influence on behaviors and actions potentially held by central, or highly connected, people within networks. For study participants, having both highly connected central individuals in the network and higher numbers of kin in the network, each had strong negative relationships with CAPE participation scores. These negative relationships suggest mechanisms of potential constraints to the development of more distal patterns of participation when family and central individuals are highly involved within the youth's network. While supportive networks are generally understood to be beneficial, having overly integrated or protective networks can potentially constrain the type of social risk-taking (e.g. moving in social circles beyond one's comfort zone) that can expand the bounds of a youth's SN (Holland et al., 2007). Additional questions remain concerning highly central (i.e. connected) individuals, such as ‘What types of highly connected individuals are better suited to facilitate participation?’ and ‘What is too connected?’ Future investigations should include a focus on the effect on participation of highly central individuals possessing various characteristics (e.g. parent, teacher/coach, sibling, cousin, peer, personality types, social roles).
Despite observed strong correlations between participation and two structural network variables, only a modest number of youth in the clinical group indicated any awareness of the social structures surrounding them - let alone awareness of potentials that may be brought about by the network. For instance, only a few youth in the clinical group explicitly understood that individuals on the edges of their network were potential links to other networks and the people they contained. During the qualitative interviews, most youth in the clinical group discussed a focus on staying engaged with the various sectors of their social network. These youth had only modestly, if ever, contemplated their own current or potential social navigations within their network - social navigations that may have the potential to create additional or expanded opportunities for participation. Greater understanding is needed of the ways that youth with LD, ADHD and ASD perceive and conceptualize their SN, as well as navigate within their network. Such understanding is important for development of interventions that move beyond development of social skills toward development of social networks and social negotiations within networks that are supportive of desired participatory outcomes.
Understanding Enhanced by Qualitative Data
Qualitative data was used to extend understanding beyond the quantification of relationships between network variables and participation. While multiple network variables had strong relationships to participation, qualitative findings clearly explicated just how persistent many youth from the clinical group needed to be in order to participate with members of their network. When asked specifically about activities engaged in with members of their network, several youth simply catalogued their activities. Despite repeated prompting, the majority of youth did not elaborate on either the activities engaged in or the social network context of their activity engagement. These youth described a narrower range of engagement, which primarily consisted of social interactions, activities involving families or family gatherings, and engagement in more formal or structured activities such as summer camp or scouts. These youth focused their discussion on interpersonal experiences – the level of challenge for many in the clinical group. Their discussions primarily described one-to-one social interactions with specific network members, which are reflected in the qualitative analysis.
The handful of study youth who did elaborate on their activities and the social network context of their activity engagement did describe engaging in a wider range of activities. Their activities included activities that occurred when they were hanging out with their peers (such as food fights, joking around, video games), as well as engagement in informal or more spontaneous group activities such as bicycling together and playing with neighborhood youth. For these youth, activities were described as occurring within larger, primarily same-gender peer groups, where the youth was an active participant in the social negotiations of the group. These descriptions are consistent with the strong positive relationship observed between the number of same gender peers and the intensity of participation in social activities.
Qualitative findings point to ways that diagnostically related social-cognitive and communication impairments may impact a youth's social network and participation. Additional qualitative inquiry should focus on deepening understanding and advancing conceptual understanding of participatory experiences within social network contexts – deeper understanding that may one day contribute to greater understanding of models of participation for youth with cognitive impairments. Future quantitative social network investigations should also incorporate ways of distinguishing the effect of individual level differences.
Future Directions and Study Limitations
Social network analysis provided a strategy for quantifying multiple nuanced aspects of the social network. However, network inquiry was limited to individuals the youth interacted with in person; it excluded inquiry as to individuals only interacted with through remote modalities such as online gaming or social media. Future SN studies should consider inclusion of online interaction partners, especially those whose focus is on groups with mobility or social communication constraints, or where respondents are old enough to have some degree of autonomy with their online interactions. This study expands current descriptive understanding of the social networks that youth with disability interact with in person. Moreover, observed SN associations to participation contribute evidence that suggests potential functional meanings of SN differences for youth with disabilities. However, observed quantitative associations were for the entire sample; the small sample size did not allow for meaningful analysis of correlations specific to youth in the clinical group. Future studies should investigate SN relationships to participation specific to youth growing up with disabilities and compare them to relationships for typically developing youth. Better understanding of the ways social network factors are connected to rehabilitation outcomes of interest can potentially be useful to researchers, clinicians and families in guiding decision making. For instance, greater understanding of social networks’ linkages to participation can be used to inform development of occupational therapy interventions that focus on social skills development for the purpose of helping the youth understand and navigate one's individual network.
In future studies, if predictive network factors are identified, holistic OT interventions considering SN factors can be developed and tested. For example, if features of a sibling's connectedness are shown to impact engagement in activities, OT interventions can more precisely recruit sibling involvement in reinforcing desired behaviors within the child's natural social contexts. Elucidating relationships between social environmental factors and participation in occupations are important early steps in the development of testable holistic OT interventions that harness the influencing power of the social environment on participation in activities and occupations. Testing interventions that holistically address the person, environment and task in affecting performance of occupations within life's situations can provide evidence for the fundamental tenants of OT while demonstrating effectiveness of OT interventions. More research is needed to provide the evidence base for potential clinical use of network analysis with adolescents.
This exploratory investigation used a small sample with a clinical group diagnosed with multiple types of cognitive impairments related to development. While diagnosis of LD, ADHD or ASD is qualitatively determined by the presence of a minimum number of heterogeneous symptoms, symptoms overlap between these diagnoses. Continued investigation with a larger and more focused sample is essential before generalizable conclusions can be made. Additionally, while several network variables were explored in this study, not all variables investigated can be potential targets of OT intervention. For instance, the number of kin in a youth's network may not be modifiable by rehabilitation efforts. However, four types of network variables should be further investigated, which include compositional variables concerning same gender, peers, weak ties, and structural variables investigating highly connected network members. Research investigating the influence of sibling and peer networks on participation outcomes of youth with disabilities is also warranted.
Continued investigation of network variables will be important for identifying and quantifying network factors having predictive value in studying rehabilitation outcomes of interest and determining strength of relationships to participation for clinical groups. Such analyses will require larger and more focused studies. Additionally, because research conventions are established for typically functioning adults, future studies utilizing methods of personal network analysis would benefit from refinement. Refinement of network data collection tools and/or strategies tailored for pediatric populations and/or those with clinical diagnoses impacting cognitive processing and attention is needed. Many questions remain, including those involving potential strengths of weak ties (i.e. peer acquaintances) for youth with and without disability. Additionally, research investigating the influence of sibling and peer networks on participation outcomes of youth with disabilities is warranted.
Important for advancing the science of rehabilitation, this study provides the beginning of an evidence base of the ways in which social networks are linked to youth participation. The development or restoration of individual capacities is achievable through rehabilitation efforts aimed at change to both the person and the surrounding environment (Institute of Medicine, 1997). Evidence is steadily increasing demonstrating the effectiveness of pediatric OT interventions that reflect the interplay between the youth, environment and participation in the occupations of childhood and adolescence (Kreider, Bendixen, Huang, & Lim, 2014). This study identifies SN factors that may have potential to predict youth participation and contributes to explication of the interplay between the person, environment and participation in adolescent activities.
Acknowledgements and Declaration of Interests
The authors acknowledge Christen Fechtel Stevens for assistance in data collection and analysis. CK's work on this project was supported in part by Award Number K12 HD055929 from the National Center Medical Rehabilitation Research (NICHD) nad the National Institure Neurological Disorders and Stroke RB's work on this project was supported in part by Award Number K01HD064778 from the Eunice Kennedy Shriver National Institute of Child Health &. Human Development. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Eunice Kennedy Shriver National Institute of Child Health & Human Development or the National institutes of Health.
Contributor Information
Consuelo M. Kreider, Department of Occupational Therapy, University of Florida, PO Box 100164, Gainesville, FL.
Roxanna M. Bendixen, Occupational Therapy, University of Pittsburg, 5025 Forbes Tower, Pittsburg, PA, 15260, 412-383-6603, bendixen@pitt.edu.
William C. Mann, Department of Occupational Therapy, University of Florida, PO Box 100164, Gainesville, FL 32610-0164, 352-273-6817, wmann@phhp.ufl.edu.
Mary Ellen Young, Department of Behavioral Science and Community Health, University of Florida, P.O. Box 100175 Gainesville, Florida 32610, 352-273-6745, meyoung@phhp.ufl.edu.
Christopher McCarty, Health Services Research Management & Policy, University of Florida, Suite 156, Ayers Building, 720 SW 2nd Ave. University of Florida Gainesville, Florida 32611, 352-392-2908, ufchris@ufl.edu.
References
- American Occupational Therapy Association Occupational therapy practice framework: Domain and process (3rd ed.). American Journal of Occupational Therapy. 2014;68(Suppl. 1):S1–S48. http://dx.doi.org/10.5014/ajot.2014.682006. [PubMed] [Google Scholar]
- Borgotti SP, Everett MG, Johnson JC. Analyzing social networks. Sage; Thousand Oaks, CA: 2013. [Google Scholar]
- Creswell JW, Plano Clark VC. Designing and conducting mixed methods research. Sage; Thousand Oaks, CA: 2007. [Google Scholar]
- Engel-Yeger B, Jarus T, Law M. Differences in patterns of participation between youths with cerebral palsy and typically developing peers. American Journal of Occupational Therapy. 2009;63(1):96–104. doi: 10.5014/ajot.63.1.96. [DOI] [PubMed] [Google Scholar]
- Forsyth R, Jarvis S. Participation in childhood. Child: Care, Health & Development. 2002;28:277–279. doi: 10.1046/j.1365-2214.2002.00272.x. [DOI] [PubMed] [Google Scholar]
- Granovetter M. The strength of weak ties. American Journal of Sociology. 1973;78:1360–1380. [Google Scholar]
- Holland J, Reynolds T, Weller S. Transitions, networks and communities: the significance of social capital in the lives of children and young people. Journal of Youth Studies. 2007;10:97–116. [Google Scholar]
- Imms C, Reilly S, Dodd K. Diversity of participation in children with cerebral palsy. Developmental Medicine & Child Neurology. 2008;50:363–369. doi: 10.1111/j.1469-8749.2008.02051.x. [DOI] [PubMed] [Google Scholar]
- Institute of Medicine . Enabling America: Assessing the role of rehabilitation science and engineering. The National Academies Press; Washington, DC: 1997. [PubMed] [Google Scholar]
- Jarus T, Anaby D, Bart O, Engel-Yeger B, Law M. Childhood participation in after-school activities: what is to be expected? British Journal of Occupational Therapy. 2010;73:344–350. [Google Scholar]
- King G, Law M, King S, Hurley P, Rosenbaum P, Hanna S, Kertoy M, Young N. Children's assessment of participation and enjoyment & preferences for activities of children manual. Harcourt Assessment, Inc.; San Antonio, TX: 2004. [Google Scholar]
- King GA, Law M, King S, Hurley P, Hanna S, Kertoy M, Rosenbaum P. Measuring children's participation in recreation and leisure activities: construct validation of the CAPE and PAC. Child: Care, Health and Development. 2006;33:28–39. doi: 10.1111/j.1365-2214.2006.00613.x. [DOI] [PubMed] [Google Scholar]
- Knoke D, Yang S. Social network analysis. 2nd ed. Sage; Thousand Oaks, CA: 2008. [Google Scholar]
- Kreider CM, Bendixen RM, Huang YY, Lim Y. Centennial Vision - Review of occupational therapy intervention research in the practice area of children and youth 2009–2013. American Journal of Occupational Therapy. 2014;68:e61–e73. doi: 10.5014/ajot.2014.011114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- McCarty C. EgoNet [Computer Software] 2012 Retrieved from http://sourceforge.net/projects/egonet/
- Parten MB. Social participation among pre-school children. Journal of Abnormal and Social Psychology. 1932;27:243–269. [Google Scholar]
- Ryan GW, Bernard HR. Techniques to identify themes. Field Methods. 2003;15:85–109. [Google Scholar]
- Sandelowski M, Barroso J. Classifying the findings in qualitative studies. Qualitative Health Research. 2003;13:905–923. doi: 10.1177/1049732303253488. [DOI] [PubMed] [Google Scholar]
- Smith KP, Christakis NA. Social networks and health. Annual Review of Sociology. 2008;34:405–429. [Google Scholar]
- World Health Organization . International classification of functioning, disability and health: Children & youth version. WHO Press; Geneva, Switzerland: 2007. [Google Scholar]