Who Do Students Ask for Help With Classwork? Sources of Help and Changes in Help-Seeking From Peers During Early Adolescence

Abstract

This study used peer nominations to identify who students ask for help in class. We describe the level of friendship affiliation that students' share with their peer helpers, and their classroom behavior reputations. Participants were 1037 fifth (49%) and sixth (51%) graders with varied gender and ethnic-racial backgrounds (51% girls; 43% White, 29% Black, 6% Hispanic, 6% Asian, 17% multiracial and other). We used multilevel modeling to examine how peer helpers and teachers’ help were related to changes in peer help-seeking. Teacher help was related to higher adaptive and lower expedient help-seeking. Peer academic reputation was related to changes in expedient help-seeking. When students asked higher achieving non-friends for help this was positively related to expedient help seeking, whereas this association was weaker among friends and best friends. Teachers need to explain how and when to seek help from peers. High achieving peers can reinforce classmates’ requests for expedient help.

Keywords

peer relationships friendship classroom behavior/environment motivation

Peers provide support across multiple developmental domains and thus play a central role in adolescents’ functioning and adjustment at school (Wentzel, 2017). Peers fulfill adolescents' social and emotional needs for acceptance, relatedness, and companionship, without which, they are much less likely to participate and do well in school (Ladd et al., 2008). The academic benefits afforded by the emotional quality of peer relationships has been well-documented (Wentzel, 2017). However, peers can also provide direct instrumental support for learning in the classroom. When students encounter academic difficulties, such as a challenging problem, or the need for conceptual clarification, peers are an abundant, and likely immediate, resource for seeking direct assistance with schoolwork. To date, much less is known about peer relationships as they pertain to instrumental, compared to emotional, support in the classroom. This is surprising given the extensive theoretical work on the importance of peers for social learning (Vygotsky, 1978). Current knowledge about adolescents’ tendencies to seek help is largely centered on an individual’s motivation toward certain kinds of help (Roussel et al., 2011; Ryan & Pintrich, 1997; Ryan & Shin, 2011). Yet, the process of help seeking involves a longer chain of decisions, which also includes selecting from whom to seek help. As such, some peers may be a better resource than others.

To provide a better understanding of instrumental support that occurs through adolescents’ help seeking among classroom peers, our study draws on sociometric peer relationships. Specifically, we examine two components of adolescents’ peer helper context. First, we consider the closeness of the relationship students share with a specific peer helper. In the classroom, students can choose to ask for help from a peer with whom they have no close personal relationship (simply, a classmate), or they may turn to a friend or best friend. Second, students have different reputations in the peer system that may be important for understanding changes in help-seeking behaviors. Some students are highly known for their academic skills and effort, others are noticed for prosocial behaviors, like being nice or cooperative, and still there are others who get into trouble or don’t follow rules (Bukowski et al., 2017). Incorporating relational ties and characteristics that describe both the adolescent and their helpers are key to understanding how sources of help may regulate the success and reinforce the nature of adolescents’ help-seeking exchanges. Finally, we affirm that classroom help seeking is embedded into a complex ecological system in which peer interactions further depend on their teacher’s practices and expectations (Farmer et al., 2019; Vollet et al., 2017). Therefore, we examine adolescents’ help seeking with peers in conjunction with their perceptions of the teacher’s effort to provide individual academic help (Ryan & Shim, 2012).

The Help Seeking Process

Help seeking is especially important in adolescence because students have increased cognitive capacity to regulate their own learning, including both awareness about academic difficulties as well as problem-solving strategies. Additionally, adolescents are developing more advanced social skills to know who can help and how to ask for help (Newman, 2000). Both teacher-reports and classroom observations show that students engage in a variety of help-seeking strategies in late elementary and middle school (Calarco, 2011; Ryan, 2000). Naturalistic observations of help seeking in the classroom have found that by 5^th grade, students increasingly turn to their peers for help and the overall rates of help seeking tend to be higher when students have opportunities to work on tasks independently or in groups (Nelson-Le Gall & Glor-Scheib, 1985). Classroom observations of teacher-student interactions in math and science indicate that one strategy teachers use to promote content understanding includes opportunities for students to practice procedures and skills (McKellar et al., 2019). Therefore, math and science classrooms in early adolescence are an important and relevant context for investigating help seeking.

The overall help-seeking process includes several steps, in which learners: (1) become aware of a problem and make a decision to seek help (or not), (2) determine the kind of help they need, (3) consider the people or resources available to meet their need, (4) elicit help, and (5) evaluate the success of the help exchange (Makara & Karabenick, 2013; Nelson-Le Gall, 1981). Students’ engagement in this process is likely to be dynamic and not necessarily sequential. For example, eliciting a certain type of help may be contingent on who is available. Researchers use this framework to examine different aspects of the help-seeking process. In this study, we focus on the first three phases using adolescents’ peer nominations of who they ask for help and self-report ratings of what kind of help they ask for.

Students’ Help-seeking Behaviors

Our first research question investigates adolescents’ individual characteristics related to their decision to seek help, which can be observed both by the presence or absence of a peer helper, as well as their level of help avoidance (Ryan et al., 2001). Students who avoid asking for help are stunted at the very beginning of the help seeking process. They may be aware that there is a problem but choose not to request help and thus do not engage further. Students who avoid asking for help tend to have lower confidence and experience greater academic difficulties (Ryan et al., 1998). These avoidance patterns are explained in the context of how adolescents weigh the costs and benefits of seeking help (Roussel et al., 2011; Ryan & Pintrich, 1997). Deterrents include the fear of looking dumb, that their need is burdensome and unwelcome, or that ultimately asking for help will not actuate gains (Peeters et al., 2020). Consistent with this research using students’ self-reported help-seeking behavior, we expect that students who do not nominate a peer helper will be characterized by a less positive academic profile, including reputations for low achievement, help-avoidance, and poor social behavior (Hypothesis 1).

Once a student recognizes the need and elects to seek help, another relevant decision in the help-seeking process to consider is: what kind of help? Adaptive help is defined as appropriate and needed requests to improve one’s comprehension or mastery (Nelson-Le Gall, 1981). Students who have higher tendencies toward adaptive help seeking also demonstrate more adaptive motivation, including higher mastery goals, academic confidence, and higher perceived support and social efficacy with teachers (Ryan et al., 2005). In contrast, expedient help seeking is viewed as a maladaptive strategy because the underlying goal is to accelerate a practicable and convenient solution to a problem without engaging in the learning process (Nelson-Le Gall, 1981). As such, expedient help requests undermine one’s learning and are associated with declines in achievement (Ryan & Shim, 2012). When students request expedient help, they are asking for the answer or to solicit someone else to do the work (Karabenick, 2011; Nelson-Le Gall, 1981). Expedient help is usually higher when adolescents are focused on competitive performance, popularity, or striving for dominance among peers (Kiefer & Shim, 2016; Schenke et al., 2015). There is a paucity of research examining the interplay between the kinds of help students request along with their choice or preferences for peer helpers. As students evaluate how their needs are met by certain peers, these sources of help may play a role in shaping changes in adolescents’ help seeking behavior over time (Makara & Karabenick, 2013).

Who are Peer Helpers?

Next, our second and third research questions investigate individual characteristics related to who adolescents’ select as helpers. Most of what we know about who students turn to for help comes from interviews with children and adolescents, while there is limited research beginning to incorporate peer nominations. This sociometric perspective on help seeking with peers is important because it is a window into the peer context in which help exchanges take place. Additionally, it is a relational approach that has underlying directionality; when adolescents recognize the need for help, their next step is to consider: who is a good resource to meet their needs? During help exchanges, there is potential for peer influence to occur. We know from studies on adolescents’ friendships that peers influence one another on a variety of beliefs and behaviors, like effort, disruptive behavior, and intrinsic value (Berndt & Keefe, 1995; Shin & Ryan, 2014); but they also tend to select friends who are similar to them on these characteristics (Duriez et al., 2013; Wang et al., 2018). Together, these processes lead to homophily (or similarity) in friendship networks. Concerning adolescents’ help seeking with peers, it is possible that their choices follow similar tendencies toward homophily. Friends are an attractive choice for help requests because, by definition, they are expected to provide help, mutual support, and intimacy (Newman, 2000). Moreover, asking a friend for help may mitigate the social costs associated with increased help-avoidance (Ryan & Pintrich, 1997). As children get older, they tend to more often to explain their choice of helper in terms of social roles; like choosing a work partner because they are a friend (Ladd et al., 2014). However, requests for help may also follow complementary patterns, where there is a preference for dissimilarity because their helper choice is goal directed – toward a peer who has a reputation for high achievement or being more prosocial and thus more likely to provide help. To better understand the interplay between adolescents’ peer helper friendship affiliations and characteristics one aim of this study is to describe adolescents and their choice of peer helpers as they naturally occur in a classroom setting. We describe the classroom behavior reputations (prosocial, academic, and disruptive) of focal students along with their peer helpers across three levels of friendship: classmates, friends, and best friends. There is not enough work that investigates peer help seeking using sociometric data, so we do not posit specific hypotheses about who adolescents ask for help (i.e., hypotheses 2 and 3 are exploratory).

Changes in Classroom Help Seeking

Next, we examine changes in help seeking across the school year to gain insight into how social partners reinforce different help-seeking behaviors over time. The longitudinal link is important because behavior patterns are learned over time. Consistent with social learning theory (Bandura, 1986), behavior is determined through reciprocal transactions with students’ environment and personal characteristics. Examining help-seeking behavior in the spring, while controlling for their own behavior in the fall, is necessary to isolate how students’ social learning environment (as distinct from their personal characteristics), is related to their future learning behavior. Longitudinal research among adolescents suggests that adaptive help seeking declines over time whereas expedient help increases (Ryan & Shim, 2012). However, these changes are not uniform. The classroom social environment, inclusive of peer and teacher characteristics, like support, perceptions of belonging, and mastery goals, matter for adolescents’ help-seeking behaviors (Ryan & Shim, 2012; Schenke et al., 2015). In the following sections, we elaborate on three aspects of the classroom environment that we believe to be important for help seeking with peers: the level of closeness of peer helpers via friend affiliations, as well as the classroom behavioral reputations of peers, and teacher’s help provisions. These factors associated with changes in help seeking encompass our next set of research questions.

Friend Affiliations

Friendships can be conceptualized along a continuum that represents how much someone knows and likes another person. Strangers are on the lowest end of the spectrum. As knowledge and liking of another person increases, then someone might be described as a close or best friend (Berndt & McCandless, 2008). To capture the full range of affiliations in the classroom, we consider level of closeness with peer helpers along three points on this continuum. Classmates are on the lowest end because, like acquaintances, these are students who adolescents know but do not necessarily like or dislike. Next are friends who students identify as someone they know and like, and a best friend is one who students distinguish as having the greatest knowledge and affection toward among all their friends. As one moves along the friendship continuum, students’ friendships become more selective. These choices signify that not all friendships are equal. Moreover, there is evidence from prior research that relationship quality with best friends is related to higher involvement in school, whereas when positive friendship provisions were averaged across multiple friendships the association with school involvement dissipated (Berndt & Keefe, 1995). Thus, important information about help seeking in the context of friendships might be lost without using the full range of closeness on the friendship continuum.

In the help seeking context, provisions of academic help are one way that support might be demonstrated between friends. During help exchanges, friendship skills like perspective-taking, resolving conflict, and cooperation may be an important resource for facilitating adaptive help. Compared to interactions that occur between non-friends, friends tend to engage in more discussion and are better able to work through differences in understanding (Newcomb & Bagwell, 1995). More specifically, there is evidence that demonstrates the benefits of friendship for facilitating help seeking. One social network study that examined the overlap between adolescents’ friends and sources of help found that mutual friendships contributed to positive changes in help seeking, supporting the idea that reciprocity and intimacy lay a foundation for the willingness to ask for help (van Rijsewijk et al., 2020). This is consistent with self-report data as well — students’ perceptions of a safe and supportive peer climate overall (when students are respectful and nice to each other) is associated with declines in help-avoidance (Shim et al., 2013). Adolescents who endorse social goals that strive toward stronger and more intimate friendships also report higher levels of adaptive help seeking (Roussel et al., 2011). This prior work suggests that closeness with a peer helper could be an important attribute that is positively associated with adaptive help-exchanges (Hypothesis 4a).

At the same time, peers can also undermine learning by reinforcing expedient help-exchanges. Much less attention has been directed to understand the role of friends for expedient help. Research shows that high quality peer support has been associated with lower tendencies for expedient help seeking (Shim et al., 2013). The nature of peer interaction between non-friends has also been characterized by higher levels of competition and control (Newcomb & Bagwell, 1995). Such competitive peer contexts, and the need to demonstrate superior performance or popularity, have been positively associated expedient help seeking (Kiefer & Shim, 2016; Ryan & Shim, 2012). Students who seek help from a peer who is not a friend may be more likely to engage in expedient help seeking because it is occurring outside the context of a supportive relationship that facilitates mastery-oriented discussion. Accordingly, we expect that close friendships with one’s peer helper will be negatively related expedient help seeking, whereas exchanges between classmates who are not friends will be related to higher expedient help (Hypothesis 4b).

Peer Classroom Behavior Reputations

It is also likely that the characteristics of the peer helper need to be taken into account along with the closeness of the relationship (i.e., friendship) to better understand exactly how peers are related to changes in help-seeking behaviors. Social learning theories suggest that peer collaboration on schoolwork is important to advance learning and development (Vygotsky, 1978). When peers work together through help exchanges, each one is an opportunity to reinforce or deter future help-seeking behaviors. For one, peer modeling occurs during help-exchanges and adolescents gain a closer perspective on their peers’ academic values, learning strategies, and performance (Ryan, 2000). Additionally, adolescents turn to their peers as a source of help for information exchange. Peers who understand a concept in class can often explain it in a way that is easy for another student to follow. For students who are seeking expedient help, peers are both approachable and accessible (Peeters et al., 2020). Thus, considering the characteristics of peers can provide important information about the nature of help-exchanges that occur in the classroom. Students may be more inclined to invest time in an adaptive help exchange when they believe the peer is capable of fulfilling their need; or their expedient requests may be reinforced if they are often successful at obtaining quick answers.

To that end, this study investigates the reputations of adolescents and their peer helpers. In the classroom, adolescents develop reputations through their day-to-day interactions and observations of one another’s behavior. These reputations may be important for who students choose as helpers, especially when asking a peer who is not a close friend. One reputation that is relevant to the help-seeking context is peer academic reputation. These students are observed by their peers as putting in effort, getting good grades, receiving praise for their work, and being asked for help more often (Altermatt, 2014). In turn, they develop a positive academic self-concept (Gest et al., 2005). Additionally, students’ reputations for prosocial behavior may be relevant to the quality of help students request from their peers. Helping is one form of engaging in prosocial behavior (K. R. Wentzel et al., 2007). Highly visible prosocial students are seen by their peers as being supportive, nice, cooperative, and likely to reciprocate a request for help. Moreover, these students can be leaders for shaping the peer environment such that it is conducive to learning (Ladd et al., 2014). When students do ask for help from their socially and academically well-adjusted peers, we expect this will facilitate adaptive help exchanges and decrease the tendency toward expedient help (Hypothesis 4c).

On the other hand, students’ interactions with peers may be driven not only by their academic goals to receive help, but also their social goals to achieve status or dominance in the classroom (Kiefer & Shim, 2016). In early adolescence, behavioral norms begin to favor academic disengagement and antisocial behavior (Galván et al., 2011). When students give or receive answers for the purpose of avoiding work, this demonstrates low value for learning and therefore may also be a strategy to gain attention and establish dominance among peers. While disruptive students are likely an unattractive source for adaptive help because this behavior is negatively correlated with engagement (Shim & Finch, 2014), expedient help may be common from disruptive peers who favor social over academic gains. Indeed, students who endorse popularity goals, or the desire to achieve social status, are also more likely to engage in expedient help seeking (Kiefer & Shim, 2016). Across adolescence, there are coinciding increases in the importance of social status (LaFontana & Cillessen, 2010) and expedient help seeking with peers (Ryan & Shim, 2012). For adolescents looking for expedient help, their peers who have higher reputations for disruptive behavior might be more receptive to these requests, thus reinforcing this type of help seeking over time. Indeed, students’ own level of disruptive behavior is positively related to expedient help and negatively related to adaptive help (Shim & Finch, 2014). Therefore, we expect peers’ levels of disruptive behavior may be associated with positive changes expedient help seeking (Hypothesis 4d).

Taken together, both peer helper affiliation and classroom behaviors are important dimensions of the peer context because they provide information about the closeness of affiliation with the peer helper, as well as the behavioral content of peer interactions (Berndt & Murphy, 2003). Although these specific peer characteristics have not been examined in the context of help seeking, research has found both aspects of peer relationships are related to youth’s learning and adjustment. For example, in dyadic partnerships where youth report higher affinity (or liking) for their partner, this was related to greater knowledge gains (Hartl et al., 2015). Additionally, peers’ expectations for behavior are related to their own academic and prosocial behavior (Wentzel et al., 2016). Moreover, social learning theories suggest that students are more likely to learn and adopt behaviors from peers with whom they perceive to be good models (Bandura, 1986; Schunk & DiBenedetto, 2020). As such, friendships are distinctive peer role models (from classmates or other grade-level peers) because students have more opportunities to interact with friends, students value and desire to be more like their friends, and they have greater affection toward friends than other peers (Hartup, 1996; Barry & Wentzel, 2006). Therefore, our fifth research question investigates the interactive associations between peer helper’s affiliation and classroom behaviors. It follows that peers’ classroom behaviors may have a stronger association with help seeking when such peers are friends or best friends (Barry & Wentzel, 2006; Berndt & Murphy, 2003; Hypothesis 5).

Teachers’ Help Provisions

Lastly, classrooms contain overlapping developmental microsystems wherein both peers and teachers have integral roles in shaping the learning environment (Bronfenbrenner, 1994). Pertaining to help seeking, both the teacher and peers are sources of academic help. The intersection of peer and teacher relationships in the classroom can affect adolescents’ help-seeking behavior because both can factor into the perceived costs and benefits of seeking help (Newman, 2000). Given the interactions between teachers and peers in the classroom social context, our sixth research question examines students’ help seeking with peers along with their perceptions of the teacher’s help.

Drawing on Bronfenbrenner’s ecological model (1994), there are two ways that students’ perceptions of their teacher (i.e., the teacher microsystem) could be related to their help-seeking behavior with peers (i.e., the peer microsystem). First, there could be a direct association (or spillover linkage; see (Skinner et al., 2022), such that when students have a supportive relationship with their teacher who they perceive as helpful, this is related to more adaptive academic behaviors, like help seeking with peers (Marchand & Skinner, 2007; Ryan & Shim, 2012). Students who perceive teachers value learning and care about their well-being also engage in more adaptive help seeking and are less likely to ask for expedient help seeking (Kiefer & Shim, 2016; Ryan & Shim, 2012). One aspect of teacher support that has not been examined is students’ perceptions of the effectiveness of teachers’ help when explaining concepts or adjusting instruction. Teachers who provide help effectively are also modeling adaptive help exchanges for students to be able to provide help to their peers independently. Positive help exchanges with the teacher may also alleviate fears about help seeking because it shows that provisions of help are valued (Newman, 2000). Thus, our first hypothesis is that teachers’ help sets the expectation for adaptive help seeking with peers by encouraging and modeling these exchanges thereby directly relating to students’ help seeking with peers (hypothesis 6a). Most research supports this hypothesis (Kiefer & Shim, 2016; Ryan et al., 1998; Ryan & Shim, 2012).

Second, help seeking with peers could depend on their perception of the teacher, such that help seeking with peers increases only in the absence of teacher support. This compensatory pattern is plausible given that teacher help in our study is operationalized as each individual student’s perceptions. Qualitative research has found that even if one student perceives the teacher is helpful, another student in the same classroom may not agree. These differences are especially true when the teacher’s explanations assume prior knowledge that not everyone shares (Peeters et al., 2020). When a student perceives their teacher as not helpful, it may heighten the costs associated with help seeking from the teacher specifically – thus putting greater weight on help-seeking exchanges that occur among peers (Peeters et al., 2020). Indeed, research has found that in low teacher emotional support classrooms, peer connections matter more for shaping students’ engagement (Kim & Cappella, 2016). Thus, we also explore this alternative hypothesis where students’ level of help seeking with peers is dependent on their perceptions of the teacher, such that we observe greater help seeking from peers (either expedient or adaptive) when the teacher is perceived as a less adequate source (Hypothesis 6b).

Overview of the Current Study

Peers play a developmentally important role in the help-seeking process as sources of academic support, information exchange, and behavioral models in the classroom. The primary aim of this study is to learn more about adolescents’ explicit peer helpers and their associations with changes in help-seeking behaviors within the classroom context and in the presence of teachers’ help. First, we examine individual characteristics related to adolescents’ decision to seek help. Are there differences in students’ own peer reputations and help-avoidance between those who identify a peer helper and those who do not (RQ1)? Furthermore, are there differences in students’ classroom behavior reputations depending on whether they select a classmate, friend, or best friend for help (RQ2)? Next, we investigate from whom adolescents seek help. What are the behavioral reputations of adolescents’ peer helpers and do these vary at increasing levels of friendship affiliation (i.e., classmate, friend, and best friend; RQ3)? Finally, we investigate how the sources of help (from teachers and peers) independently and jointly relate to changes in adolescents’ help-seeking behavior across the school year. Specifically, this question includes three components. First, focusing on the main effects, how do peer helper’s level of friendship affiliation, classroom behavior, and perceptions of teachers’ help relate to changes in help seeking (RQ4)? Do associations between friendship affiliation and help seeking depend on peer helpers’ classroom behaviors (RQ5)? Lastly, are changes in help seeking with peers dependent on teachers’ help (RQ6)? We examine these questions in the context of early adolescents’ math and science classrooms.

Methods

Participants

Data were collected as part of a larger longitudinal project designed to learn about adolescents’ academic and social adjustment. Project participants were recruited from three school districts in the Midwestern region of the United States. All three districts were comparable in achievement levels (62–74% of students met state standards) and served a moderate percentage of low-income families (between 50% and 71% of families were eligible for free or reduced-price lunch). Within each school district, all middle schools agreed to participate (N = 6), and we then recruited two local elementary schools that fed into each middle school. This resulted in a sample of 12 (from 24 total) elementary schools across these districts. Our study is a cross-sectional sample of fifth graders in elementary school (49%) and sixth graders in middle school (51%). Elementary schools served students in kindergarten through fifth grade. Fifth graders were assigned one primary teacher. Middle school students in sixth grade rotated among several teachers within their grade-level teams. To coordinate 6^th graders participation, we worked with math and science teachers, who each selected one of their class periods to complete the survey. In total, students were nested within 57 classrooms. The number of participating students in each class ranged from 11 to 26 students with a mean of 18 students per class for fifth graders and 19 students per class for sixth graders.

There were 1130 students who participated in the survey in either the fall, spring or both times during the school year. Within this participating sample, 79% of students participated in both the fall and spring, and 14% (N = 159) participated only in the fall. The remaining 6% of the sample (N = 75) who joined in the spring were excluded from analysis because they were missing data on all predictor variables. We further excluded one classroom whose students (N = 14) did not participate in peer nomination measures, as well as four students who were missing a peer nomination about who they would ask for help. Therefore, our analytic sample was 1037 students. Missing data on each variable in the fall was minimal, ranging from 1-5%. The outcome variables were students’ spring help-seeking behaviors, and these had a larger portion of missing data (15% - 17%). We looked at several predictors of missingness in the spring, including parent education, students’ educational attainment expectations, grade-level, gender, and race. Only grade-level was associated with missingness, indicating that fifth graders were missing data more than would be expected by chance compared to 6th graders. We included grade-level in our imputation model to increase confidence that data are missing at random (Graham, 2009).

We used two different strategies to account for missing data, depending on the research question. For research questions examining mean differences in adolescents’ reputations and help-seeking behaviors at wave 1, we used estimation maximization (EM) to generate a single imputed data set. This method is efficient and preferred to listwise deletion; it is unlikely to bias estimates given the small portion of data missing (<5%) at random (Dong & Peng, 2013). In contrast, our questions examining changes in adolescents’ self-reported help-seeking behavior at wave 2 required more complex multi-level modeling procedures and had a higher percentage of missing data. Therefore, we used a model-based imputation procedure that was carried out in Blimp software (Keller & Enders, 2017) and analyzed in RStudio. Bayesian model-based imputation has been found to perform well in simulation studies when estimating multi-level models with random coefficients and interaction effects (Enders et al., 2020). Since the imputed data is dependent on the model, we generated a unique set of five imputed data sets for each outcome variable. We excluded four additional students who were missing data on Race because including this nominal variable in the imputation model led to non-convergence.

Procedures

Data were collected in the fall and spring the of school year, approximately 6 months apart in October-November and again in April-May. In all participating schools and grade-levels, parents received letters to invite their children’s participation and 84% responded by returning a slip with consent for their child to participate. The percentage of students with consent ranged from 58 – 100% across classrooms with all but two classrooms exceeding a two-thirds participation rate. Those students having parent consent also assented to participation on the day the survey was administered in class. Consistent with institutional review board procedures, to obtain students’ assent, the research team explained that all participation was voluntary, students could withdraw at any time or refrain from answering any questions on the survey and were reassured that their answers would be kept private. All students (both 5^th and 6^th graders) were informed that the purpose of the survey was to learn more about their classroom experiences in math and science. Trained researchers administered the survey by reading the instructions and items aloud while students read along on their own and responded. All students were provided a blank sheet of paper to use as a coversheet that kept their responses confidential. These survey procedures were repeated when students participated in the survey in both the fall and spring. To obtain information about students’ class-specific experiences, the survey items were written to reflect a specific subject-area class, either math or science. For sixth grade-students, the subject-area matched the class in which the survey was administered. Although fifth graders had one primary teacher who taught multiple subjects, students were asked to think about one subject – either math or science. To obtain a comparable sample of fifth and sixth graders, the subject-area was designated for fifth graders by matching the number of participating math and science classrooms in middle schools. Of the 1037 students in the sample, about 60% reflected on their experience in math class while 40% responded about their science class.

Measures

Help Seeking with Peers

We assessed multiple aspects of adolescents’ perceived help-seeking behavior with peers. The standard approach to assessing help-seeking strategies is to frame items conditionally, to control for differences in students’ help-seeking needs and opportunities to seek help (Karabenick & Knapp, 1991; Newman, 1990; Ryan & Pintrich, 1997; Schenke et al., 2015). First, we asked about students’ general tendency to avoid help when needed using four items that have been used in prior research with young adolescents (Ryan et al., 1998; Ryan & Pintrich, 1997). All these items asked students about what they would do if they needed help, for example: “When I don’t understand my (math/science) work, I often guess at the answers instead of asking for help.” A composite scale for avoidant help seeking was created by taking the mean of all four items (α = .79).

Next, we asked adolescents to identify who they would ask for help and what kind of help they would seek from their peer. Students were initially prompted: “We’d like you to think of a time when you are doing some math/science work at your desk. You realize that you do not understand how to do some of it, and you need help. Which classmate would you ask for help with your math/science work?” Below this prompt, there was a blank line for students to write-in the name of a classmate who they would ask for help. This item uses a peer nomination technique to identify which classmates get asked for help from peers. Almost all students identified a single peer helper and for the six students who identified two classmates, their first choice was used for analysis. Immediately after writing in their peer helper, the survey continued and then asked: “What type of help would you ask this classmate for?” which was followed by eight items – four describing expedient help and four related to adaptive help. These scales have also been used in prior research on peer help seeking and there is evidence supporting its factor structure and measurement invariance over time (Ryan & Shim, 2012). Expedient requests for help focus on getting the correct answer. A sample item is: “I would ask someone to tell me if it was right or wrong and not explain it to me,” (T1 α = .69; T2 α = .78). In contrast, adaptive requests for help are oriented to improve learning. For example, “I would ask someone to explain it to me, not just give me the answer;” (T1 α = .73; T2 α = .74). For both scales, a composite measure was created by taking the mean of the four items on each.

Teacher Helps

Four items captured adolescents’ perceptions about how much the teacher helps them by explaining concepts and adjusting instruction based on their needs. These items are from Skinner and Belmont’s (1993) work measuring students’ perceptions of the teaching context. Specifically, provisions of help are one way in which the teacher can cultivate a structured learning environment. Example items include, “If I can’t solve a problem, my teacher shows me different ways to try,” and “My teacher checks to see if I’m ready before he/she starts a new topic.” These items were averaged to create a single observed score and higher values indicate that students perceive their teacher is more helpful (α = .80).

Peer Nominations

Peer nominations were used to assess adolescents’ friendships and classroom behaviors. Peers have a unique vantage point to report about one another’s behaviors and abilities that may not be observable by teachers or other adults. Moreover, these measures are based on a large number of peer informants, such as all members of the class or grade-level, thereby representing more than a single viewpoint (Cillessen & Marks, 2017). In this study, we used peer nominations where the reference group was within classrooms for both 5^th and 6^th graders. All participating students in the class could nominate any number of their classmates for each item and these nominations were not restricted to study participants.

Friendships

Students were asked, “Who are your friends in this class? Who do you talk do and hang around with the most?” A class roster was embedded into the survey and students could select as many or as few friends as they wanted. After selecting all of their friends in the class, students were then prompted to identify their one best friend. These nominations were used to create a nominal indicator for the level of affiliation of adolescents’ peer helpers. We examined students’ friendship and best friend nominations to see if their peer helper occurred on these lists. Peer helpers who were classmates but did not appear on either friend list served as the reference group for peer helpers who were either friends or best friends.

Classroom Behavior Reputations

Peer nominations were used to assess three kinds of classroom behaviors: prosocial behavior, peer academic reputation, and disruptive behavior. Students were prompted with: “All students act differently in school. Which students in your class…” and were able to select as many students as they wanted from the class list for each item. The number of nominations that students received for each behavior was first summed and then standardized within class to control for differences in class sizes, which is a common method for quantifying peer nomination measures (Cillessen & Marks, 2017). Additionally, because all students in the class were included on the roster to receive peer nominations, participating students’ classroom behavior reputations are standardized relative to all peers who received nominations, regardless of whether that classmate participated in the study. Two items for prosocial behavior asked which students “are really cooperative and willing to help others,” and “are really nice.” Peer academic reputation included one positive (i.e., “Who gets good grades”) and one negative item (“Who does not get good grades”). To create an aggregate across valences, we subtracted the number of nominations received for the negative item from the positive, such that high positive values indicate a reputation for good grades and negative values represent a reputation for poor grades. Similarly, there was one positive (“who follows school rules”) and one negative item (“who does not follow school rules – gets in trouble”) for disruptive behavior. Disruptive behavior is represented by the negative item, so positive values represent more nominations for getting into trouble whereas negative values represent reputations for following rules. For all three behavior reputations, items were first summed or subtracted to aggregate before standardizing by class.

Analytic Plan

We began with a concurrent analysis to examine the characteristics of adolescents and their peer helpers in the fall, including the demographics for individuals and dyads, as well as correlations between peer reputations and help-seeking behaviors. Next, we used analysis of variance to compare the characteristics of adolescents who had a peer helper with those who did not. For students who named a peer helper, we also examined whether their own levels of prosocial, academic, and disruptive behavior were related to their choice of helper (i.e., classmate, friend, and best friend). We then examined mean differences in peer helpers’ levels of prosocial, academic, and disruptive behavior for helpers who were classmates, friends, or best friends. After examining the descriptive information about adolescents and their peer helpers, we proceeded to multi-level modeling (MLM) to investigate whether peer helper characteristics were related to changes in help-seeking behavior across the school year.

Results

Preliminary Analysis

Demographic information about adolescents and their peer helpers are presented in Tables 1 and 2. 80% of adolescent’ peer helper dyads were same gender, while 8% were girls who asked a boy for help and 12% were boys who asked a girl for help. There was a much larger percentage of cross-race dyads, where 62% of students asked a different race peer for help. For dyads that were between same race peers, this included 24% White, 8% Black, 1% Hispanic, 2% other, and 3% multiracial. Prosocial behavior and peer academic reputation were strongly correlated (r = .78), and these were both negatively correlated with disruptive behavior (r = -.74 and -.81, respectively). Reputations for classroom behavior were not significantly correlated with adolescents’ self-reported adaptive help seeking but were related to avoidant and expedient help. Adaptive help was negatively correlated with avoidant and expedient help seeking. Peer helpers, on average, were around one standard deviation above the mean for prosocial behavior and academic reputations.

Table 1.

Demographic Characteristics for Individuals, Peer Helpers, and Peer-Helper Dyads.

	N Individuals		N Dyads
	Student	Helper	Non-friends	Friends	Best friends	Total
Gender
Girls	525 (51%)	456 (56%)	51	141	165	357 (44%)
Boys	499 (49%)	363 (44%)	47	132	115	294 (36%)
Girls ask boys for help			31	26	7	64 (8%)
Boys ask girls for help			35	45	14	94 (12%)
Race
White	440 (43%)	407 (49%)	30	79	91	200 (24%)
Black	302 (29%)	153 (18%)	5	28	37	70 (8%)
Hispanic	62 (6%)	49 (6%)	1	0	5	6 (1%)
Asian	57 (6%)	93 (11%)	3	6	7	16 (2%)
Multiracial	153 (15%)	113 (14%)	5	4	12	21 (3%)
Other	19 (2%)	19 (2%)	0	0	0	0
Cross-race dyad			122	238	156	516 (62%)

Table 2.

Means, Standard Deviations, and Correlations for Reputations and help-Seeking Behavior, along with Perceptions of Teacher Help.

Variable	1	2	3	4	5	6	7	8	9
Fall
1. Prosocial reputation	1	.78**	-.74**	-.20**	.04	-.18**	.01	-.01	-.13**
2. Peer academic reputation		1	-.81**	-.26**	.10	-.22**	.01	.00	-.15
3. Disruptive reputation			1	.24**	-.05	.19**	-.02	-.05	.17**
4. Help avoidance				1	-.13**	.42**	-.14**	-.12**	.37**
5. Adaptive help					1	-.27**	.46**	.46**	-.29**
6. Expedient help						1	-.17**	-.22**	.47**
7. Teacher helps							1	.30**	-.18**
Spring
8. Adaptive help								1	-.37**
9. Expedient help									1
Mean (SD)	.12 (1.02)	.11 (1.01)	-.10 (1.0)	1.99 (.96)	3.76 (.97)	1.98 (.90)	3.77 (1.04)	3.64 (.97)	2.13 (1.01)

**indicates p < .01.

The Decision to Ask for Help

First, we used multivariate analysis of variance to compare the characteristics of adolescents who identified a peer helper (N=927) with those who do not (N = 110). Approximately 11% of the sample did not identify a classmate who they would ask for help with their work. Box’s M test suggested the homogeneity of covariances assumption was met (χ² (10) = 12.68, p = .24)). Table 3 summarizes differences in peer reputations and avoidant help-seeking behavior between these groups of students (F (4, 1032) = 4.00, p < .003; Wilk’s Λ = 0.98). Adolescents who did not identify a peer helper had a below average academic reputation (F (1, 1035) = 8.40, p = .004; M = -.16, SD = 1.09) and were more help avoidant (F (1, 1035) = 11.01, p < .001; M = 2.28, SD = .96) compared to their classmates who did identify a peer for help (academic reputation M = .14, SD = .99; help avoidance M = 1.96, SD = .95). These differences were significant using Bonferonni’s adjusted p-value (p < .013). Students who did not identify a peer helper also had lower reputations for prosocial behavior (F (1,1035) = 5.81, p = .02) and were more known for disruptive behavior (F (1,1035) = 4.40, p = .04).

Table 3.

Means and (standard Deviations) of Classroom Behavior and Help-Avoidance by Students Who Do and Do Not have a Peer Helper.

	No Peer Helper n = 110	Yes Peer Helper n = 917	Mean Difference (No - Yes)	F
Prosocial reputation	-.10 (1.01)	.14 (1.02)	-.24	5.81*
Peer academic reputation	-.16 (1.09)	.14 (.99)	-.30	8.40**
Disruptive reputation	.09 (1.06)	-.12 (.99)	.21	4.40*
Help avoidance	2.28 (.96)	1.96 (.95)	.32	11.01***

Note. *p < .05, **p < .01, ***p < .001.

Among students who did nominate a peer helper, about 20% identified a classmate who they would ask for help, while 43% picked a friend, and 36% selected a best friend. Again, we used multivariate analysis of variance to examine differences in adolescents’ peer reputations to explain their choice of helper (classmate, friend, and best friend). Homogeneity of covariances was assumed (Box’s M: χ² (12) = 8.05, p > .05). The omnibus MANOVA test was significant F (2, 924) = 3.91, p < .001; Wilk’s Λ = 0.98). Follow-up tests revealed students who asked their best friend for help were more well-adjusted, with higher levels of prosocial behavior (F (2, 924) = 5.73, p = .003; M = .26, SD = 1.03) and lower levels of disruptive behavior (F (2, 294) = 7.51, p < .001; M = -.29, SD = 1.00) than students who asked a classmate or friend for help (see Table 4). They also had higher peer academic reputations (F (2, 924) = 3.56, p =.03; M = .25, SD = 1.03) than students who ask a classmate for help (M = .03, SD = .99).

Table 4.

Means and (Standard Deviations) for Individuals’ and Helpers’ Classroom Behavior by Level of Friendship.

	Individual Students’ classroom behaviors				Peer helpers’ classroom behaviors
	Asked a classmate	Asked a friend	Asked best friend	F	Classmates	Friends	Best friends	F
Prosocial reputation	-.05 (.99)_a	.14 (1.01)_a	.26 (1.03)_b	5.73**	.92 (1.04)	1.05 (1.06)	.86 (1.11)	3.01
Academic reputation	.03 (.99)_a	.10 (.95)_ab	.25 (1.03)_b	3.56*	1.07_a (.78)	1.01_a (.85)	.79_b (.89)	9.24***
Disruptive reputation	.00 (1.0)_a	-.04 (.96)_a	-29 (1.00)_b	7.51***	-.78 (.82)	-.77 (.88)	-.64 (.94)	2.32

Note. Classroom behavior reputations were standardized and thus these averages reflect standard deviation units from the mean. Means in the same row that do not share the same subscript are significantly different from each other. Pairwise follow-up comparisons were corrected with Bonferonni’s criteria.

*p < .05, **p < .01, ***p < .001.

Adolescents’ Peer Helpers

Next, we examined the characteristics of adolescents’ peer helpers. Moreover, we examined whether their peer helpers’ characteristics varied depending on whether they were a classmate, friend, or best friend (see Table 4). For this analysis, homogeneity of covariances was not assumed (Box’s M: χ² (12) = 30.92, p = .002) and we therefore conducted a robust MANOVA. The omnibus test was significant χ² (6) = 27.39, p < .001; Wilk’s Λ = 0.97). Follow-up analysis of variance and pairwise comparison’s applying Bonferonni’s correction (p < .017) revealed only one difference between groups. Peer helpers who were a best friend had slightly lower academic reputations (F (2, 924) = 9.14, p < .001; M = .79, SD = .89) than their counterparts who were friends (M = 1.01, SD = .85) or classmates (M = 1.07, SD = .78). In general, we see that peer helpers on average are well-adjusted and are sought out for their positive characteristics (which range from .79 – 1.07 standard-deviations above the sample mean), even when that person is not a friend.

Changes in Help Seeking from Peers

Finally, we examined changes in adolescents’ adaptive and expedient help seeking with peers across the school year. We built multi-level models predicting each type of help-seeking behavior in the spring, while controlling for their fall help seeking. These models addressed our research questions concerning the independent and joint contributions of the teacher’s help provisions along with peer helpers’ affiliations and characteristics for adolescents’ own help-seeking behavior.

Adaptive Help Seeking

First, we included several covariates in the model explaining adolescents’ adaptive help seeking in the spring. Gender was not significantly related to help-seeking behavior, but sixth grade students reported lower levels of adaptive help than fifth graders (b = -.24, p = .007). To include students’ race in the analytic model, we combined Hispanic, Asian, multiracial, and other students into a single “Other” category due to low sample sizes within each. As the numerical majority, White students (43%) served as the reference group and we included a dummy coded variable for Black students (29%) and other minoritized students (29%). Black students reported higher levels of adaptive help than White students (b = .17, p = .03). To compare this model with the null model, we pooled maximum likelihood-ratio tests across the imputed data sets, which indicated improved model fit (F (4, 411.03) = 3.46, p = .008). Including these covariates reduced the unexplained variance at level-2 by 26% and by 1% at level-1. In model 2, we added students’ help seeking with peers in the fall, as well as their perceptions of the teacher’s help, and interaction term for teacher help and peer help seeking. Both adaptive help seeking with peers (b = .39, p < .001) and perceptions of the teacher’s help (b = 13, p = .002) were positively associated with changes in adaptive help seeking. However, the interaction term was not significant (b = .01, p > .05). Students’ help seeking with teachers and peers also improved model fit (F (3, 67.503) = 47.417, p < .001) and reduced the unexplained variance from the previous model by 19% and 48% at levels 1 and 2, respectively. Since the interaction term with teacher’s help and peer help seeking were non-significant, this was removed from subsequent models. Next, adding adolescents’ friend helper affiliations (friend and best friend, compared to classmate) were non-significant and did not improve model fit (F (1, 38.07) = .731, p = .40).

The next set of models examined each peer characteristic (prosocial, academic, and disruptive behavior) individually, while controlling for students’ own level of these behaviors. Across all three of these models (see Table 5, Models 4a, 4b, and 4c), none of these factors were significant. Adolescents’ peer helper characteristics were unrelated to changes in adaptive help seeking from peers and this was consistent across all levels of peer helper affiliation. We were not able to explain any additional variance in adaptive help seeking with peers and model fit was not improved.

Table 5.

Multilevel Model Predicting Spring Adaptive Help-Seeking Behavior From Individual And Peer Helper characteristics.

	Model 1: Covariates	Model 2: Help	Model 3: Affiliation	Model 4a: Prosocial	Model 4b: Peer Academic Reputation	Model 4c: Disruptive
	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)
Fixed effects
Intercept	3.78*** (.09)	3.71*** (.08)	3.65 *** (.10)	3.66*** (.10)	3.67*** (.10)	3.63*** (.10)
Individual
Boys (compared to girls)	-.13 (.07)	-.03 (.06)	-.03 (.06)	-.06 (.07)	-.05 (.07)	-.11 (.07)
Grade 6 (compared to 5)	-.24** (.09)	-.10 (.07)	-.10 (.07)	-.10 (.07)	-.11 (.07)	-.11 (.07)
Black (compared to White)	.17* (.08)	.06 (.07)	.06 (.07)	.04 (.08)	.05 (.08)	.09 (.08)
Other (compared to White)	.04 (.09)	.01 (.09)	.01 (.09)	.01 (.09)	.01 (.09)	.02 (.09)
Fall adaptive help		.39*** (.04)	.39*** (.04)	.39*** (.04)	.39*** (.04)	.39*** (.04)
Fall teacher helps		.13** (.04)	.13** (.04)	.13** (.13)	.13** (.04)	.12** (.04)
Adaptive help x teacher help		.01 (.03)
Classroom behavior				-.4 (.04)	-.03 (.04)	-.02 (.04)
Helper affiliation
Fall friend			.06 (.10)	.07 (.10)	.07 (.10)	.05 (.10)
Fall best friend (BF)			.08 (.10)	.09 (.10)	.10 (.10)	.09 (.10)
Helper characteristics
Classroom behavior				.02 (.07)	-.03 (.10)	.06 (.09)
Behavior x friend				-.07 (.08)	-.06 (.11)	-.09 (.10)
Behavior x BF				.02 (.09)	.11 (.12)	-.13 (.11)
Random effects
Student-level (residual)	.89	.71	.71	.71	.71	.71
Class-level (intercept)	.04	.02	.02	.02	.02	.02

*p < .05, **p < .01, ***p < .001.

Expedient Help Seeking

We followed identical procedures to examine changes in adolescents’ expedient help seeking with peers. Model 1 indicated that boys (b = .21, p = .003), sixth graders (b = .24, p = .006), and minoritized students (Black: b = .23, p = .005; Other: b = .18, p = .049) reported higher levels of expedient help in the spring. Including these covariates in the model improved model fit (F (4, 326.78) = 5.92, p < .001) while reducing the unexplained variance at level-2 by 28% and 2% at level-1. Like results for adaptive help, Model 2 demonstrated main effects for fall help seeking with peers (b = .49, p = .002) and teacher’s help (b = -14, p = .002), but the interaction term was non-significant. Students who reported higher expedient help seeking with peers in the fall also reported higher expedient help in the spring, whereas when students perceived their teacher’s as being more helpful, this was related to lower expedient help seeking with peers across the school year. This set of variables improved model fit (F (3, 158.32) = 65.18, p < .001), but given the non-significant interaction term, this was removed from subsequent models. Again, the helper’s friend and best friend affiliations was unrelated to changes in help seeking (see Table 6) and did not improve the model (F (1, 43.71) = 1.35, p = .25).

Table 6.

Multilevel Model Predicting Spring Expedient Help-Seeking Behavior From Individual and Peer Helper Characteristics.

	Model 1: Covariates	Model 2: Help	Model 3: Affiliation	Model 4a: Prosocial	Model 4b: Peer Academic Reputation	Model 4c: Disruptive
	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)
Fixed effects
Intercept	1.77*** (.08)	1.85*** (.07)	1.94*** (.10)	1.95*** (.10)	1.94*** (.10)	1.98*** (.11)
Individual
Boys (compared to girls)	.21** (.07)	.14* (.06)	.14* (.06)	.13* (.07)	.11 (.06)	.09 (.07)
Grade 6 (compared to 5)	.24** (.09)	.13 (.07)	.13 (.07)	.14 (.07)	.14 (.07)	.14 (.07)
Black (compared to White)	.23** (.08)	.24** (.08)	.23** (.08)	.21** (.08)	.18* (.08)	.17* (.09)
Other (compared to White)	.18* (.09)	.16 (.08)	.16 (.08)	.14 (.08)	.14 (.08)	.13 (.08)
Fall expedient help		.49*** (.03)	.49*** (.03)	.49*** (.04)	.48*** (.04)	.48*** (.04)
Fall teacher helps		-.14*** (.03)	-.14*** (.03)	-.14*** (.03)	-.14*** (.03)	-.14 ***(.03)
Expedient help x teacher help		-.02 (.03)
Classroom behavior				-.02 (.04)	-.05 (.04)	.07* (.04)
Helper affiliation
Fall friend			-.09 (.09)	-.09 (.09)	-.06 (.09)	-.08 (.09)
Fall best friend (BF)			-.13 (.09)	-.13 (.09)	-.10 (.09)	-.11 (.09)
Helper characteristics
Classroom behavior				.11 (.08)	.22* (.09)	-.11 (.08)
Behavior x friend				-.04 (.09)	-.21* (.10)	.05 (.50)
Behavior x BF				-.14 (.10)	-.28* (.11)	.14 (.11)
Random effects
Student-level	.98	.76	.76	.75	.75	.75
Class-level	.026	.013	.013	.014	.015	.015

*p < .05, **p < .01, ***p < .001.

Last, we examined peer helper characteristics in conjunction with peer helper affiliations to see if these explained changes in expedient help seeking with peers. Models 4a, 4b, and 4c, each examined prosocial, academic, and disruptive behavior. We found that the academic reputation of adolescents’ peer helper was positively related to expedient help seeking (b = .22, p = .006). Moreover, this pattern was contingent on adolescents’ peer helper affiliation. Asking for help from a friend (b = -.21, p = .03) or best friend (b = -.28, p = .01) with a higher peer academic reputation attenuated the positive association with expedient help seeking (see Figure 1). Adding these variables to our model did not significantly improve the variance explained (F (4, 262.25) = 2.29, p = .06). No other peer characteristics were significantly associated with changes in expedient help from peers and this was consistent among classmates, friends, and best friends. However, individual’s own level of disruptive behavior was related to higher expedient help seeking (b = .07, p = .04).

Figure 1.

Changes in expedient help by level of friendship.

Discussion

In the classroom, peers have the potential to provide critical informational support that contributes to adolescents’ academic adjustment in school. Specifically, help seeking with peers creates opportunities for provisions of support through information exchange and modeling (Ryan et al., 2001). The primary aims of this study were twofold: to better understand who adolescents turn to for help in the classroom by examining peer nominations for helpers, friends, and behavioral reputations. Furthermore, how does the classroom social context, including peer helpers’ characteristics and teachers’ provision of help relate to changes in help-seeking behaviors across the schoolyear? In general, our findings suggest that peer helpers are not limited exclusively to adolescents’ friendships. Moreover, peers contribute to changes in expedient help seeking, whereas teachers’ provisions of help are important for explaining higher levels of adaptive help seeking over time.

Similar to findings from self-report surveys, adolescents’ peer nominations revealed that around one in 10 students did not identify a peer helper (Ryan et al., 2005; Ryan & Pintrich, 1997). These students self-reported higher levels of avoidance and were also noticed by peers as low achievers. Not having access to help when needed is one way that students can become more disengaged in learning (Ryan et al., 2001). Teachers need to be particularly aware of who these students are to explicitly communicate the benefits and expectations for seeking help. Among students who did identify helpers, they selected peers who were academically well-adjusted, including high reputations for academic success, cooperation, and kindness along with lower levels of disruptive behavior. Consistent with emerging research, there was a subset of students who asked a classmate with these characteristics for help, even when they were not friends (Hoffman et al., 2020; van Rijsewijk et al., 2020).

Prior research suggests that adolescents avoid seeking help due to social costs around what their peers or teachers may think (Ryan & Pintrich, 1997). By asking for help from non-friends who have higher academic and social skills, adolescents in our sample demonstrated a willingness to seek out opportunities to learn from peers in spite of these costs. Students who asked classmates for help had average reputations, whereas students who asked a best friend tended to be academically and socially more well-adjusted. This is not surprising, as it is well known that best friends tend to share similar characteristics (Shin & Ryan, 2014). When it comes to help seeking, we found that peer helper relationships were both complementary and homophilic in nature when comparing individual and peer helper characteristics on average. Future work observing how peer helper dyads interact can better inform if certain compositional qualities yield productive exchanges of information or modeling.

We hypothesized that adolescents’ inclination toward adaptive help seeking would be higher when it occurs with peers who are high achieving, nice, and cooperative friends. This hypothesis was not supported. Instead, only teachers’ provisions of help were positively associated with adaptive help seeking in the spring, whereas peer characteristics were not. There are times when it might be more appropriate to ask the teacher rather than peers for help with explanations, examples, or hints to continue working. In fact, research on parents’ socialization of academic help seeking indicates that middle class parents often coach their children to ask teachers rather than peers for help because teachers are the ultimate authority on academic matters required for school success (Calarco, 2018). Although peers were unrelated to self-reported adaptive help seeking, we did find evidence that adolescents sought help outside of their friend group – suggesting a more purposeful selection of peers with higher levels of achievement and prosocial behavior. This effortful selection provides the opportunity for a productive learning exchange to occur. When seeking help from peers, adolescents may not make it clear whether their request is expedient or adaptive in nature – leaving the decision on the kind of help obtained to their peer. For example, a student may turn to a classmate and simply say, “I don’t get #2,” and the peer could respond either by giving the answer or by giving an explanation (Rodriguez-Mojica, 2019). Adaptive help with peers may be more complex and difficult to define than adaptive help seeking with teachers, as the success of such exchanges are less certain.

In contrast, this study suggests that peers play a more salient role in shaping expedient help-seeking behaviors. Teachers need to monitor what kind of help students obtain from peers – especially when help is solicited from peers with high academic reputations. Our results show that when adolescents asked peers who get good grades for help, this was associated with higher expedient help seeking. In other words, when students’ peer helpers were known for high achievement, this was associated with an increasing tendency to get answers or find out if their work was right or wrong. Yet, this pattern was dependent on adolescents’ friendship with their helper. Adolescents who asked a friend or best friend with high academic reputations for help also reported lower expedient help seeking (like requesting an answer without explanation). This difference may be related to the close personal relationships that exist among friends and best friends, compared to classmates. Friends mutual caring for one another’s academic success and wellbeing may suppress this maladaptive orientation toward help seeking. The results of this study also suggest the need for future research to observe the process of expedient help seeking among friends and classmates while doing classwork, as we can only speculate about the social interactions that took place during help-exchanges.

Importantly, our results show that adaptive and expedient help seeking with peers are not antithetical to one another. Although friendship contexts attenuated the positive association between expedient help seeking and high-achieving peers, we do not see that close friendships are related to higher adaptive help with peers. If expedient help seeking is less present, but adaptive help is not more likely, what could be happening instead? From ethnographic observations of children’s classroom help seeking (Calarco, 2018), one possible inference is that friendship mitigates expedient help seeking by tapping into the emotional rather than informational support that friends provide. Asking a friend for help may reveal that both the student and their peer helper need assistance. Although they might ask for an answer, this request may reveal dissonance that requires clarification, and they may support each other in seeking adaptive help from the teacher or identifying another equally appropriate source. Similarly, we do not find that peer helpers’ prosocial behaviors were related to either adaptive or expedient help seeking. This could also be a reflection of the multifaceted nature of peer help. Prosocial peers, who are kind and cooperative, may be a more attractive source for emotional rather than academic support because they might offer reassurance or listen to their classmates needs even if they cannot explicitly answer the academic question at hand.

Additionally, the contrasting findings that peers are related to differences in expedient, but not adaptive help, reveal an opportunity for further conceptualization around help seeking with peers. As mentioned above, it is possible to imagine instances when expedient help seeking (i.e., asking for the answer) may not be inherently maladaptive, but rather a step in the learning process that is not captured in our measure. Conducting cognitive interviews (Karabenick et al., 2007) with adolescents to unpack their own perceptions of help seeking with peers could better delineate expedient help along a continuum from adaptive efforts that facilitate learning to maligned efforts that covert cheating.

Finally, we observed that help seeking with peers is supported by teachers’ accessibility as a resource for help, rather than as a substitute if teacher help is inadequate. When teachers proactively provided help and checked in with students about their need for additional explanation, this was related positively to adaptive help seeking and negatively to the overall need for expedient help from peers. This aligns with prior research showing that teacher support and emphasis on learning and mastery supports the development of adaptive help (Ryan & Shim, 2012). Similarly, the actual quality of help peers provide could be contingent on the classroom context. Future research could examine both help requested, and help received, to better understand peer helping dynamics. In classrooms where teachers take the time to explain and model expectations for help, then asking a prosocial or high achieving peer may be more effective and thus more strongly relate to changes in help-seeking behavior over time. Observing the overall classroom quality would better differentiate not only which peers are sources of help, but also when peers are more or less effective helpers.

Implications for Teachers

Although friendship alone does not guarantee the nature of help-exchanges being non-expedient, it does suggest the need for teachers to intentionally monitor help exchanges that do not occur between friends. Explicitly modeling expectations for both help giving and help receiving could make exchanges more productive overall. Teachers’ expectations for help seeking are often implicit and highly variable from day-to-day and for different tasks (Calarco, 2018). Sometimes students are expected to work out problems on their own and other times help seeking with teachers and peers is encouraged. Additionally, students are often bad at giving help when asked. Observations of peer help exchanges in a 4^th grade classroom documented that students’ requests for help from peers was ignored one in four times and only five percent of exchanges were elaborated responses (Rodriguez-Mojica, 2019). Our results add to the current literature, showing that students need support from teachers to facilitate adaptive help with peers (Calarco, 2018; Rodriguez-Mojica, 2019) and this is likely important for adaptive help with peers to improve achievement (Ryan & Shim, 2012).

Facilitating opportunities for the formation of friendships among students with diverse academic skills may be foundational for quality help exchanges. However, in classrooms, students interact with a variety of peers with whom they have different relationships. While it may not always be possible for students to seek help from friends, teachers might instead aim to promote feelings of responsiveness, reciprocity, and positive affect so that all students care for one another in the class, even if they are not best friends. This positive interaction style, typical but not exclusive to friendships, has been referred to as dyadic mutuality (Piehler & Dishion, 2007). Corroborating these ideas, recent research examining students’ peer collaborator preferences in math found that when students feel safe and supported by all classroom peers, this increases the attractiveness of peer collaborators based on engagement and a willingness to work together in math, whereas when students had greater fears of being embarrassed in front of peers, they were more likely to seek high achieving peer collaborators (Hoffman et al., 2020). Thus, attending to these underlying relational qualities that guide how classmates interact are important for attenuating the likelihood that students will solely rely on high-achieving peers for expedient help in adolescence (Ryan & Shim, 2012).

Strengths, Limitations and Future Directions

This study has both strengths and limitations. We used multiple reporting methods to examine how adolescents’ peer contexts contribute to changes in help seeking. Peer nominations for social behavior (e.g., prosocial) are not susceptible to self-report bias and are ecologically valid because peers often interact in contexts that are hidden from teachers and adults (Cillessen & Marks, 2017). Although we examine the role of friendship and reputations of peer helpers as they naturally occur in the classroom setting, it is not a randomized design that allows us to directly compare the kind or quality of help exchanged in friend compared to non-friend dyads. Additionally, we controlled for adolescents’ initial levels of help seeking in the fall to examine how classroom peers explained changes across the school year. Yet we also only asked students to nominate their friends and peer helpers at one point in time. Therefore, we do not know if their choice of peer helpers would be dynamic to different needs, if it is stable, or constrained by the physical context of the classroom – like seating arrangements. Furthermore, a limitation of the study sample is that 5^th graders were in elementary school whereas 6^th graders were in middle school. These different school contexts may mean that 5^th graders have greater familiarity with peers who they share the same class with all day. Lastly, this study is limited to the perspective of focal students who identified their need for help. As such, friendships were operationalized as outgoing, rather than reciprocal nominations. Since we were interested in who adolescents chose as helpers, we believe their own perception of whether or not a peer is a friend is an accurate representation of the meaning they ascribed to this choice.

More research on adaptive help seeking with peers is needed given that the quality of help received from peers is likely to vary. Peers may not always be able to provide a clear explanation or help solve a problem, but perhaps an expedient request to compare answers may lead to an adaptive help request from the teacher to resolve uncertainty. Additionally, students’ classroom peer groups contain more than just their friend groups. Future research could examine whether peer helpers or collaborators are a distinct source of influence for adolescents’ adjustment in school. Some work on peer socialization finds that higher status peers (i.e., popular or well-accepted) are more likely to influence lower status peers (DeLay et al., 2016). As such, does being a helper (or someone who is frequently sought out for help) confer higher social status based on their academic and prosocial behavior that relates to changes in their adjustment over time? Finally, our study examined peer characteristics on average and general tendencies toward adaptive or expedient help with peers. Future research that uses an unlimited nomination procedure to identify all peers who adolescents ask for help, and what kind of help they get from each peer, would be insightful for describing within-person differences in help-seeking behavior and the implications for academic adjustment.

Conclusion

Peers are an abundant and increasingly attractive source for academic assistance in adolescence (Peeters et al., 2020). On average, fifth and sixth graders in our study turned to more academically and socially well-adjusted peers for help when they needed it. It is important for teachers to attend to peer dynamics around help seeking – especially given the associations between friendships, peer academic reputation, and expedient help that can undermine learning, or at times border on cheating (i.e., getting an answer from a peer). Working with peers who have a diverse set of academic skills is inherent to learning and development in school. Findings suggests that facilitating friendships or friendship skills, such as dyadic mutuality, among academically diverse students could be helpful to attenuate positive associations between peer helper’s achievement and expedient help seeking in early adolescence. At the same time, when teachers meet adolescents’ need for help by explaining information or providing tips for problem-solving, this may serve as a model for desired help-seeking exchanges that is positively related to adaptive help from peers across the school year.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported, in part, by a grant from the Spencer Foundation awarded to Allison M. Ryan. The funding sponsor had no role in the study design, analysis, interpretation of data, writing of the report, nor the decision to submit the article for publication.

ORCID iD

Jessica E. Kilday

Author Biographies

Jessica E. Kilday is a doctoral candidate in the combined program of education and psychology at the University of Michigan, Ann Arbor. Her research focuses on school social contexts, particularly how peer and teacher relationships are related to youth’s social and academic adjustment during early adolescence and the transition to middle school.

Allison M. Ryan is a professor in the combined program in education and psychology at the University of Michigan, Ann Arbor. Her research interests include achievement beliefs and behaviors, peer relationships, classroom climate, and development during early adolescence, especially around the transition to middle school.

References

Altermatt

E. R.

(2014). Children’s achievement-related discourse with peers: Uncovering the processes of peer influence. In Ryan

A. M.

Ladd

G. W.

(Eds.), Peer relationships and adjustment at school (pp. 109–134). Information Age Publishing, Incorporated.

Bandura

(1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall.

Barry

C. M.

Wentzel

K. R.

(2006). Friend influence on prosocial behavior: The role of motivational factors and friendship characteristics. Developmental Psychology, 42(1), 153–163. https://doi.org/10.1037/0012-1649.42.1.153

Berndt

T. J.

Keefe

(1995). Friends’ influence on adolescents’ adjustment to school. Child Development, 66(5), 1312–1329. https://doi.org/10.2307/1131649

Berndt

T. J.

McCandless

M. A.

(2008). Methods for investigating children’s relationships with friends. In Rubin

K. H.

Bukowski

W. M.

Laursen

(Eds.), Handbook of peer interactions, relationships, and groups (pp. 63–81). Guilford Publications.

Berndt

T. J.

Murphy

L. M.

(2003). Influences of friends and friendships: Myths, truths, and research recommendations. In Kail

R. V.

(Ed.), Advances in Child Development and Behavior (Vol. 30, pp. 275–310). JAI. https://doi.org/10.1016/S0065-2407(02)80044-1

Bronfenbrenner

(1994). Ecological models of human development. In Husén

Neville

Postlethwaite (Eds.), The International encyclopedia of education (pp. 1643–1647). Pergamon ; Elsevier Science.

Bukowski

W. M.

Castellanos

Persram

R. J.

(2017). The current status of peer assessment techniques and sociometric methods. New Directions for Child and Adolescent Development, 2017(157), 75–82. https://doi.org/10.1002/cad.20209

Calarco

J. M.

(2011). “I need help!” Social class and children’s help-seeking in elementary school. American Sociological Review, 76(6), 862–882. https://doi.org/10.1177/0003122411427177

10.

Calarco

J. M.

(2018). Negotiating opportunities: How the middle class secures advantages in school. Oxford University Press.

11.

Cillessen

A. H. N.

Marks

P. E. L.

(2017). Methodological choices in peer nomination research. New Directions for Child and Adolescent Development, 2017(157), 21–44. https://doi.org/10.1002/cad.20206

12.

DeLay

Laursen

Kiuru

Poikkeus

A.-M.

Aunola

Nurmi

J.-E.

(2016). Friend influence and susceptibility to influence: Changes in mathematical reasoning as a function of relative peer acceptance and interest in mathematics. Merrill-Palmer Quarterly, 62(3), 306–333, https://doi.org/10.13110/merrpalmquar1982.62.3.0306

13.

Dong

Peng

(2013). Principled missing data methods for researchers. SpringerPlus, 2(1), 1–17. DOI:10.1186/2193-1801-2-222

14.

Duriez

Giletta

Kuppens

Vansteenkiste

(2013). Extrinsic relative to intrinsic goal pursuits and peer dynamics: Selection and influence processes among adolescents. Journal of Adolescence, 36(5), 925–933. https://doi.org/10.1016/j.adolescence.2013.07.009

15.

Enders

C. K.

Keller

B. T.

(2020). A model-based imputation procedure for multilevel regression models with random coefficients, interaction effects, and nonlinear terms. Psychological Methods, 25(1), 88–112. https://doi.org/10.1037/met0000228

16.

Farmer

T. W.

Hamm

J. V.

Dawes

Barko-Alva

Cross

J. R.

(2019). Promoting inclusive communities in diverse classrooms: Teacher attunement and social dynamics management. Educational Psychologist, 0(0), 1–20. https://doi.org/10.1080/00461520.2019.1635020

17.

Galván

Spatzier

Juvonen

(2011). Perceived norms and social values to capture school culture in elementary and middle school. Journal of Applied Developmental Psychology, 32(6), 346–353. https://doi.org/10.1016/j.appdev.2011.08.005

18.

Gest

S. D.

Domitrovich

C. E.

Welsh

J. A.

(2005). Peer academic reputation in elementary school: Associations with changes in self-concept and academic skills. Journal of Educational Psychology, 97(3), 337–346. https://doi.org/10.1037/0022-0663.97.3.337

19.

Graham

J. W.

(2009). Missing data analysis: Making it work in the real world. Annual Review of Psychology, 60(1), 549–576. https://doi.org/10.1146/annurev.psych.58.110405.085530

20.

Hartl

A. C.

DeLay

Laursen

Denner

Werner

Campe

Ortiz

(2015). Dyadic instruction for middle school students: Liking promotes learning. Learning and Individual Differences, 44, 33–39. https://doi.org/10.1016/j.lindif.2015.11.002

21.

Hartup

W. W.

(1996). The company they keep: Friendships and their developmental significance. Child Development, 67(1), 1–13. https://doi.org/10.2307/1131681

22.

Hoffman

A. S.

Hamm

J. V.

Lambert

Meece

(2020). Peer-collaborator preferences of African American and White sixth-graders in mathematics classrooms. The Journal of Experimental Education, 88(4), 559–577. https://doi.org/10.1080/00220973.2019.1615857

23.

Karabenick

S. A.

(2011). Methodological and assessment issues in research on help seeking. In Zimmerman

Schunk

D. H.

(Eds.), Handbook of self-regulation of learning and performance (pp. 267–281). Routledge Handbooks Online. https://doi.org/10.4324/9780203839010.ch17

24.

Karabenick

S. A.

Knapp

J. R.

(1991). Relationship of academic help seeking to the use of learning strategies and other instrumental achievement behavior in college students. Journal of Educational Psychology, 83(2), 221–230. https://doi.org/10.1037/0022-0663.83.2.221

25.

Karabenick

S. A.

Woolley

M. E.

Friedel

J. M.

Ammon

B. V.

Blazevski

Bonney

C. R.

Groot

E. D.

Gilbert

M. C.

Musu

Kempler

T. M.

Kelly

K. L.

(2007). Cognitive processing of self-report items in educational research: Do they think what we mean? Educational Psychologist, 42(3), 139–151. https://doi.org/10.1080/00461520701416231

26.

Keller

B. T.

Enders

C. K.

(2017). Blimp user’s manual (version 2.2).

27.

Kiefer

S. M.

Shim

S. S.

(2016). Academic help seeking from peers during adolescence: The role of social goals. Journal of Applied Developmental Psychology, 42, 80–88. https://doi.org/10.1016/j.appdev.2015.12.002

28.

Kim

H. Y.

Cappella

(2016). Mapping the social world of classrooms: A multi-level, multi-reporter approach to social processes and behavioral engagement. American Journal of Community Psychology; Macon, 57(1/2), 20–35. http://dx.doi.org.proxy.lib.umich.edu/10.1002/ajcp.12022

29.

Ladd

G. W.

Herald-Brown

S. L.

Reiser

(2008). Does chronic classroom peer rejection predict the development of children’s classroom participation during the grade school years? Child Development, 79(4), 1001–1015, https://doi.org/10.1111/j.1467-8624.2008.01172.x

30.

Ladd

G. W.

Kochenderfer-Ladd

Visconti

K. J.

Ettekal

(2014). Classroom peer relations and children’s social and scholastic development. In Ryan

A. M.

Ladd

G. W.

(Eds.), Peer relationships and adjustment at school (pp. 11–50). Information Age Publishing, Incorporated.

31.

LaFontana

K. M.

Cillessen

A. H. N.

(2010). Developmental changes in the priority of perceived status in childhood and adolescence. Social Development, 19(1), 130–147. https://doi.org/10.1111/j.1467-9507.2008.00522.x

32.

Makara

K. A.

Karabenick

S. A.

(2013). Characterizing sources of academic help in the age of expanding educational technology. In Karabenick

Stuart A.

(Ed.), Advances in help-seeking research and applications: The role of emerging technologies (pp. 37–63). Information Age Publishing.

33.

Marchand

Skinner

(2007). Motivational dynamics of children’s academic help-seeking and concealment. Journal of Educational Psychology, 99(1), 65–82. DOI:10.1037/0022-0663.99.1.65

34.

McKellar

S. E.

Cortina

K. S.

Ryan

A. M.

(2019). Teaching practices and student engagement in early adolescence: A longitudinal study using the classroom Assessment scoring system. Teaching and Teacher Education, 89, 102936. https://doi.org/10.1016/j.tate.2019.102936

35.

Nelson-Le Gall

(1981). Help-seeking: An understudied problem-solving skill in children. Developmental Review, 1(3), 224–246. https://doi.org/10.1016/0273-2297(81)90019-8

36.

Nelson-Le Gall

Glor-Scheib

(1985). Help seeking in elementary classrooms: An observational study. Contemporary Educational Psychology, 10(1), 58–71. https://doi.org/10.1016/0361-476X(85)90006-2

37.

Newcomb

A. F.

Bagwell

C. L.

(1995). Children’s friendship relations: A meta-analytic review. Psychological Bulletin, 117(2), 306–347. https://doi.org/10.1037/0033-2909.117.2.306

38.

Newman

R. S.

(1990). Children’s help-seeking in the classroom: The role of motivational factors and attitudes. Journal of Educational Psychology, 82(1), 71–80. https://doi.org/10.1037/0022-0663.82.1.71

39.

Newman

R. S.

(2000). Social influences on the development of children’s adaptive help seeking: The role of parents, teachers, and peers. Developmental Review, 20(3), 350–404. https://doi.org/10.1006/drev.1999.0502

40.

Peeters

Robinson

Rubie-Davies

(2020). Theories in use that explain adolescent help seeking and avoidance in mathematics. Journal of Educational Psychology, 112(3), 533–550. https://doi.org/10.1037/edu0000423

41.

Piehler

T. F.

Dishion

T. J.

(2007). Interpersonal dynamics within adolescent friendships: Dyadic mutuality, Deviant talk, and patterns of antisocial behavior. Child Development, 78(5), 1611–1624, https://doi.org/10.1111/j.1467-8624.2007.01086.x

42.

Rodriguez-Mojica

(2019). “I don’t get it. I need help”: Emergent bilinguals seeking and receiving help from peers in language arts. Literacy Research and Instruction, 58(2), 105–121. https://doi.org/10.1080/19388071.2019.1585492

43.

Roussel

Elliot

A. J.

Feltman

(2011). The influence of achievement goals and social goals on help-seeking from peers in an academic context. Learning and Instruction, 21(3), 394–402. https://doi.org/10.1016/j.learninstruc.2010.05.003

44.

Ryan

A. M.

(2000). Peer groups as a context for the socialization of adolescents’ motivation, engagement, and achievement in school. Educational Psychologist, 35(2), 101–111. https://doi.org/10.1207/S15326985EP3502_4

45.

Ryan

A. M.

Gheen

M. H.

Midgley

(1998). Why do some students avoid asking for help? An examination of the interplay among students’ academic efficacy, teachers’ social–emotional role, and the classroom goal structure. Journal of Educational Psychology, 90(3), 528–535. https://doi.org/10.1037/0022-0663.90.3.528

46.

Ryan

A. M.

Patrick

Shim

S.-O.

(2005). Differential profiles of students identified by their teacher as having avoidant, appropriate, or dependent help-seeking tendencies in the classroom. Journal of Educational Psychology, 97(2), 275–285. https://doi.org/10.1037/0022-0663.97.2.275

47.

Ryan

A. M.

Pintrich

P. R.

(1997). “Should I ask for help?” the role of motivation and attitudes in adolescents’ help seeking in math class. Journal of Educational Psychology, 89(2), 329–341. https://doi.org/10.1037/0022-0663.89.2.329

48.

Ryan

A. M.

Pintrich

P. R.

Midgley

(2001). Avoiding seeking help in the classroom: Who and why? Educational Psychology Review, 13(2), 93–114. https://doi.org/10.1023/A:1009013420053

49.

Ryan

A. M.

Shim

S. S.

(2012). Changes in help seeking from peers during early adolescence: Associations with changes in achievement and perceptions of teachers. Journal of Educational Psychology, 104(4), 1122–1134. https://doi.org/10.1037/a0027696

50.

Ryan

A. M.

Shin

(2011). Help-seeking tendencies during early adolescence: An examination of motivational correlates and consequences for achievement. Learning and Instruction, 21(2), 247–256. https://doi.org/10.1016/j.learninstruc.2010.07.003

51.

Schenke

Lam

A. C.

Conley

A. M.

Karabenick

S. A.

(2015). Adolescents’ help seeking in mathematics classrooms: Relations between achievement and perceived classroom environmental influences over one school year. Contemporary Educational Psychology, 41, 133–146. https://doi.org/10.1016/j.cedpsych.2015.01.003

52.

Schunk

D. H.

DiBenedetto

M. K.

(2020). Motivation and social cognitive theory. Contemporary Educational Psychology, 60, 101832. https://doi.org/10.1016/j.cedpsych.2019.101832

53.

Shim

S. S.

Finch

W. H.

(2014). Academic and social achievement goals and early adolescents’ adjustment: A latent class approach. Learning and Individual Differences, 30, 98–105. https://doi.org/10.1016/j.lindif.2013.10.015

54.

Shim

S. S.

Kiefer

S. M.

Wang

(2013). Help seeking among peers: The role of goal structure and peer climate. The Journal of Educational Research, 106(4), 290–300. https://doi.org/10.1080/00220671.2012.692733

55.

Shin

Ryan

A. M.

(2014). Early adolescent friendships and academic adjustment: Examining selection and influence processes with longitudinal social network analysis. Developmental Psychology, 50(11), 2462–2472. https://doi.org/10.1037/a0037922

56.

Skinner

E. A.

Belmont

M. J.

(1993). Motivation in the classroom: Reciprocal effects of teacher behavior and student engagement across the school year. Journal of Educational Psychology, 85(4), 571–581. https://doi.org/10.1037/0022-0663.85.4.571

57.

Skinner

Rickert

N.P.

Vollet

J.W.

Kindermann

T.A.

(2022). The complex social ecology of academic development: A bioecological framework and illustration examining the collective effects of parents, teachers, and peers on student engagement. Educational Psychologist, 57(2), 87–113. DOI:10.1080/00461520.2022.2038603

58.

van Rijsewijk

L. G. M.

Snijders

T. A. B.

Dijkstra

J. K.

Steglich

Veenstra

(2020). The interplay between adolescents’ friendships and the exchange of help: A longitudinal multiplex social network study. Journal of Research on Adolescence, 30(1), 63–77. https://doi.org/10.1111/jora.12501

59.

Vollet

J. W.

Kindermann

T. A.

Skinner

E. A.

(2017). In peer matters, teachers matter: Peer group influences on students’ engagement depend on teacher involvement. Journal of Educational Psychology, 109(5), 635–652. https://doi.org/10.1037/edu0000172

60.

Vygotsky

L. S.

(1978). Mind in society: The development of higher psychological processes ( Cole

John-Steiner

Scribner

Souberman

, Eds. Revised ed.). Harvard University Press.

61.

Wang

M.-T.

Kiuru

Degol

J. L.

Salmela-Aro

(2018). Friends, academic achievement, and school engagement during adolescence: A social network approach to peer influence and selection effects. Learning and Instruction, 58, 148–160. https://doi.org/10.1016/j.learninstruc.2018.06.003

62.

Wentzel

(2017). Peer relationships, motivation, and academic performance at school. In Elliot

A. J.

Dweck

C. S.

Yeager

D. S.

(Eds.), Handbook of competence and motivation, second edition: Theory and application. Guilford Publications.

63.

Wentzel

K. R.

Filisetti

Looney

(2007). Adolescent prosocial behavior: The role of self-processes and contextual cues. Child Development, 78(3), 895–910. https://doi.org/10.1111/j.1467-8624.2007.01039.x, JSTOR.

64.

Wentzel

K. R.

Russell

Baker

(2016). Emotional support and expectations from parents, teachers, and peers predict adolescent competence at school. Journal of Educational Psychology, 108(2), 242–255. https://doi.org/10.1037/edu0000049