Loneliness, sense of coherence and perception of teachers as a secure base among children with reading difficulties

European Journal of Special Needs EducationVol. 21, No. 1, February 2006, pp. 21–37

ISSN 0885–6257 (print)/ISSN 1469–591X (online)/06/010021–17© 2006 Taylor & FrancisDOI: 10.1080/08856250500268619

Loneliness, sense of coherence and perception of teachers as a secure base among children with reading difficultiesMichal Al-Yagona* and Malka MargalitbaBar-Ilan University, Israel; bTel Aviv University, IsraelTaylor and Francis LtdREJS126844.sgm10.1080/08856250500268619European Journal of Special Needs Education0885-6257 (print)/1469-591X (online)Original Article2005Taylor & Francis Ltd204000000November 2005MichalAl-YagonSchool of EducationBar-Ilan UniversityRamat Gan [email protected]

The study examined children’s self-reported socio-emotional characteristics (loneliness and senseof coherence) and perception of their homeroom teacher as a secure base among 3rd-graders withreading difficulties prior to formal diagnostic assessment, thus controlling for the possible impact ofdiagnosis. In line with resilience theory, this study also aimed to further explore the resilience ofthese children with reading difficulties. The sample comprised 266 3rd-graders from regularclassrooms: 118 children with reading difficulties and 148 without reading difficulties. Significantinter-group differences emerged on all three measures. Moreover, children’s perception of theirteacher as a secure base correlated significantly with feelings of loneliness and with sense ofcoherence. In addition, a subgroup of resilient children emerged among the children with readingdifficulties. The distribution of resilient and non-resilient children revealed significant inter-groupdifferences. Discussion focused on the theoretical model of developmental adaptation, providing acomprehensive and dynamic view of developing children, and pinpointing research future directionsand counselling implications.

Keywords: Attachment; Loneliness; Sense of coherence; Socio-emotional adjustment; Student–teacher relationships

Failure in reading acquisition may affect both subsequent achievement in school andsocio-emotional aspects of students’ lives (Chapman & Tunmer, 2003). Childrenwho become adults with low levels of literacy face an increasing disadvantage in West-ern society (Torgesen, 2002). Several longitudinal studies have shown that studentswho fall behind in the development of critical reading skills during the first years ofelementary school do not reach average-level reading skills by the end of elementaryschool (Francis et al., 1996; Torgesen et al., 2001).

*School of Education, Bar-Ilan University, Ramat Gan 52900, Israel. Email:[email protected]

22 M. Al-Yagon

Beyond their future academic risks, children with reading difficulties demonstratea variety of maladjustment problems such as behaviour disorders, emotional difficul-ties, lower self-esteem, and unsatisfactory peer relationships (Hinshaw, 1992;Maughan et al., 2003). These studies underscore the importance of exploring therelations of children’s reading difficulties to their socio-emotional perceptions andfunctioning. However, relatively few studies examined these characteristics amongchildren with reading difficulties prior to formal diagnostic assessment to control forthe possible impact of labelling. Therefore, unique value may be attributed to thecurrent study’s examination of how children’s experiences of reading difficulties maycontribute to their self-perceived socio-emotional characteristics as manifested intheir feelings of loneliness and sense of coherence, as well as in their perceptions ofclose relationships with teachers.

In light of the scarcity of studies on socio-emotional perceptions among childrenwith reading difficulties prior to their formal diagnostic assessment, we will nextpresent relevant data gleaned from previous research on children who were diagnosedwith learning disabilities, including those in the domain of reading.

Socio-emotional characteristics of children with learning disabilities

A comprehensive review of the factors identifying the underlying cognitive andemotional processes that facilitate successful social adjustment and peer relationshipsat all ages suggests that no simple pathway exists from problematic peer relations todisorder manifestation (Hay et al., 2004). Rather, a reciprocal relationship seems toexist between children’s difficulties with peers and significant adults and their psycho-logical adjustment from infancy to adolescence. In the case of children with learningdisabilities, these internal problems may include impulsivity, performance andproduction deficits, emotional reactivity, low self-esteem and denial of the disabilities(see for a review Culbertson, 1998).

As mentioned above, research has indicated that reading difficulties show overlapswith various maladjustment problems of both an externalizing nature like aggressionand antisocial behaviour and an internalizing nature like loneliness, anxiety anddepression (Hinshaw, 1992; Maughan et al., 2003). Studies have demonstrated thatchildren with learning disabilities are more likely to experience such maladjustmentproblems than are their typically developing peers, including a higher level of peerrejection and loneliness, lower levels of their sense of coherence, and poorer rela-tionships with peer and teachers (e.g. Margalit & Levin-Al-Yagon, 1994; Morrison& Cosden, 1997; Culbertson, 1998; Murray & Greenberg, 2000, 2001; Al-Yagon &Mikulincer, 2004b).

However, studies focusing on the strengths of children with learning disabilitieshave emphasized the importance of searching both for potential risk and protectivefactors among this population, and of identifying resilient individuals who manifestadjustment without ignoring their deficits (Margalit, 2003; Wong, 2003). Resiliencerefers to the manifestation of positive adjustment outcomes upon exposure toadversity (Luthar & Cicchetti, 2000). In line with the resilience model that highlights

Socio-emotional characteristics and perception of teachers 23

the importance of shifting from a problem-oriented approach to a strength- andnurturing-oriented one (e.g. Werner, 1993; Luthar & Cicchetti, 2000; Richardson,2002), the current study will investigate resilience among children with readingdifficulties before formal diagnosis.

Research on children with formally diagnosed learning disabilities has identifiedvulnerability and protective factors at the individual, family and community levels.Individual vulnerability factors included alterations in central nervous systemfunctioning (i.e. information-processing disorder), impulsivity, performance andproduction deficits, emotional reactivity, low self-esteem and denial of the disorder(see for review Culbertson, 1998). Individual factors that protected children withlearning disabilities from maladjustment outcomes comprised high verbal skills,high self-esteem and a delineated understanding of the nature and course of thedisorder (Morrison & Cosden, 1997). Family-level vulnerability and protectivefactors included the mother’s sense of coherence, parental disappointment, familyrigidity or disorganization, and family cohesion (e.g. Margalit et al., 1996; Al-Yagon, 2003). Community-level factors among school-age children comprised thecontribution of school policy and social support from adults in the school,especially supportive relationships with teachers (e.g. Morrison & Cosden, 1997;Wenz-Gross & Siperstein, 1998; Luthar & Cicchetti, 2000; Al-Yagon &Mikulincer, 2004b).

Focused on the individual and community levels, the current study utilized self-report measures in line with its focus on self-perceptions and with the higher reliabil-ity found for children’s self-reports on internalizing characteristics, rather than onexternalizing characteristics (Ronen, 1997). We selected three self-report measures:(a) children’s feelings of loneliness as an index of their experience of peer relationsand social alienation, (b) children’s perception of their own coping resources asmanifested in children’s sense of coherence and (c) children’s perception of theirteacher as a secure base, as an index of children’s close relationships with a significantadult figure. In line with previous studies on teacher–child relationships among high-risk children (e.g. Lynch & Cicchetti, 1992; Pianta, 1992; Al-Yagon & Mikulincer,2004b), the current study explored these relationships from an attachment theoryperspective.

Loneliness

Loneliness reflects a global indicator of dissatisfaction from the quality and/or thequantity of individuals’ social interrelations (Asher et al., 1990). Childhood lonelinessmay be considered an enduring condition of emotional distress that is often over-looked or misunderstood.

Most research on childhood loneliness has pinpointed the contributions of chil-dren’s peer relations to their feelings of well-being at school, such as children’s socialacceptance, rejection, victimization, quality of friendships and social network (Asher& Paquette, 2003; Ladd & Troop-Gordon, 2003). Importantly, loneliness rather thanpeer rejection per se appears to coincide with emotional distress, as demonstrated by

24 M. Al-Yagon

Qualter and Munn’s (2002) findings of three different groups of children: rejectedchildren who did not feel lonely; children who felt lonely but were not rejected; andrejected children who reported loneliness.

Studies have also accentuated that the construct of loneliness consists of twodimensions, emotional loneliness and social loneliness, which emerge from differentsources (Weiss, 1973; Buchholz & Catton, 1999). Emotional loneliness refers to adeficiency in intimate close relations and interpersonal bonding and has shown linkswith insecure attachment relations with significant caregivers. Social lonelinessreflects frustration related to the disruption of the social network and to the experi-ence of peer rejection.

Feelings of loneliness appear in the self-reports of children with learning disordersand even among pre-schoolers considered at risk for developing learning disabilities(see for review Margalit & Al-Yagon, 2002). Research on these children suggestedthat their personal characteristics (e.g. information-processing disorder, impulsivityand emotional reactivity) may not only predispose them to the loneliness experience,but also render it more difficult for these children to cope with these aversive socialfeelings (Margalit, 1994; Sabornie, 1994; Pavri & Monda-Amaya, 2000; Margalit &Al-Yagon, 2002). Moreover, several studies reported a negative association betweenchildren’s feelings of loneliness and their perceived coping resources as manifested bytheir sense of coherence (e.g. Margalit, 1994; Margalit et al., 1999; Al-Yagon &Mikulincer, 2004a, 2004b).

Sense of coherence

Sense of coherence comprises the core variable within Antonovsky’s (1979, 1987)health model, entitled salutogenesis in contrast to pathogenesis. This approachderived from the assumption that the human environment is stress producing. Stres-sors may stem from diverse sources such as genetic, microbiological, personal,economic, social, cultural or geopolitical. Antonovsky’s model suggested a comple-mentary approach to the pathological model of human functioning and proposed anexamination of individuals’ strategies and coping resources in dealing with thesestressors (Antonovsky & Sagy, 1986).

Sense of coherence constitutes an inner coping resource that reflects a global orien-tation or enduring tendency to see the world as more or less comprehensible, manage-able and meaningful (Antonovsky, 1987). Research has indicated that individualswith a high level of sense of coherence are less likely to perceive stressful situations asthreatening and will be more likely to appraise such situations as manageable. Personswith a high level of coherence are also less likely to perceive stressful situations asprovoking anxiety and anger.

Several studies have revealed that children with learning disabilities reported lowersense of coherence levels compared to their non-disabled peers (Margalit, 1994; Al-Yagon & Mikulincer, 2004a, 2004b). Moreover, these studies also indicated thatchildren with a high sense of coherence tended to report lower feelings of lonelinessthan did children with a low sense of coherence.

Socio-emotional characteristics and perception of teachers 25

Student–teacher relationship: an attachment perspective

The socio-emotional functioning of school-age children extends beyond the homeand the family to their school environment, with particular ramifications for the class-room in the case of children with reading difficulties. Thus, the current study exam-ined children’s specific relationship with the homeroom teacher. Indeed, Murray andGreenberg (2001) reported that children with various disabilities such as emotionaldisorders, learning disabilities and mild mental retardation demonstrated poorer affil-iations with their classes, and greater dissatisfaction with their teachers. Moreover, arecent study revealed that school-age children with learning disabilities tended toperceive their teachers as more rejecting and as less available and accepting than didtheir non-disabled peers (Al-Yagon & Mikulincer, 2004b). However, all these studiesinvestigated children with diagnosed learning disabilities.

Several studies have used attachment theory (Bowlby, 1973, 1982) as a basis forunderstanding patterns of children’s close relationships with teachers (e.g. Piantaet al., 1995; Birch & Ladd, 1997; Al-Yagon & Mikulincer, 2004b). The construct ofattachment (Bowlby, 1973, 1988) highlights the role that interactions with significantothers play in personality development. According to this theory, over the course ofthe first year of life, infants develop a specific and enduring relationship with primarycaretakers (usually parents), termed attachment (Ainsworth & Wittig, 1969; Bowlby,1982). Proximity to an available, supportive and responsive caregiver—an attach-ment figure—provides the infant with an enduring sense of ‘secure base’ (Bowlby,1973, 1982) or ‘felt security’ (Sroufe & Waters, 1977). The sense of ‘secure base’consists of a set of expectations about others’ availability and responsiveness in timesof stress (Waters et al., 1998).

Attachment theory also proposed that infants internalize their interactions withattachment figures into ‘working models of attachment’, which constitute mentalrepresentations of the significant others and the self. This internal ‘working model’forms unique attachment styles, that is stable patterns of cognitions and behavioursthat are manifested in other close relationships and social interactions. According toattachment theory, once children’s security needs are met and the parental figuresbecome the source of their secure base, the children can then direct attention andenergy towards exploring the environment and engaging in social interactions withother partners such as siblings and other relatives, peers, non-familial caregivers andteachers (Owens et al., 1995; Weiss, 1998). These new and different attachment-likerelationships with extra-familial others may serve as a context for important attach-ment experiences. Furthermore, the new partners can also act as attachment figures,providing the child with care and support when necessary and themselves becominga source of secure base for the child’s novel exploration and skill acquisition (Owenset al., 1995; Weiss, 1998).

Most measures of attachment style have utilized Ainsworth’s observations of infantsin the ‘strange situation’ (Ainsworth et al., 1978) and her classification of attachmentstyle into one secure style and two insecure styles (avoidant and anxious). However,recent studies have employed a variety of attachment-based instruments for children

26 M. Al-Yagon

at different ages, including self-report measures (e.g. Kerns et al., 1996). Numerousstudies have examined the association between children’s attachment style and variousadjustment measures (for review see Thompson, 1999). Fewer studies have focusedon the attachment-like relationships that children form with their teachers.

Existing research on teacher–child relationships among young children has indi-cated that the formation of attachment relationships with childcare settings appearsto resemble that of infant–mother attachment formation (Howes & Hamilton, 1993;Howes, 1999). Relying on teacher perceptions, these studies reported a link betweenteacher–child relationships and various aspects of pre-schoolers’ and kindergartners’adjustment (Howes & Hamilton, 1993; Pianta et al., 1995; Birch & Ladd, 1997;Howes et al., 2000). For example, Birch and Ladd (1997) reported that teacher–childrelationships with a high level of dependency correlated with kindergartners’ adjust-ment difficulties including poorer academic performance. Conversely, teacher–childcloseness linked positively with kindergartners’ academic performance.

Few studies have examined the role of teacher–child relationships among school-age children with disabilities. These studies also highlighted the important role ofthe quality of this relationship in explaining students’ socio-emotional adjustment(Pianta et al., 1995; Wenz-Gross & Siperstein, 1998; Murray & Greenberg, 2001;Al-Yagon & Mikulincer, 2004b). However, these studies examined children withdiagnosed disabilities. Taken together, unique value may be attributed to thecurrent examination of these teacher–child relations as perceived by children withreading difficulties prior to formal diagnosis.

The current study

The current study aimed to examine children’s self-reported socio-emotional charac-teristics (loneliness and sense of coherence), as well as children’s perception of theirhomeroom teacher as a secure base, among school-age children with reading difficul-ties prior to formal diagnostic assessment, thus controlling for the possible impact ofdiagnosis. In addition, in line with resilience theory (e.g. Werner, 1993; Luthar &Cicchetti, 2000; Richardson, 2002), this study aimed to further explore the resilienceof these children.

In response to researchers’ recommendations that emphasized the importance ofchildren’s self-reported perceptions of close relationships (Lynch & Cicchetti, 1997),as well as the low reliability of parents’ and teachers’ reports on children’s emotionaland over-controlling problems such as loneliness, anxiety and depression (Achen-bach, 1993; Ronen, 1997), the present study examined 3rd-graders’ perceptions oftheir own socio-emotional adjustment and their relationships with teachers. Wecompared children with and without reading difficulties prior to formal diagnosticassessment, to control for the possible impact of labelling. We predicted that:

1. Children with reading difficulties would exhibit more problems in socio-emotionaladjustment (higher loneliness and lower sense of coherence) than would childrenwithout reading difficulties.

Socio-emotional characteristics and perception of teachers 27

2. Children with reading difficulties would report a less secure relationship with theirteacher (i.e. lower perception of the teacher as a secure base) as compared to theirpeers without reading difficulties.

3. Children’s patterns of closer relationships with teachers would show a beneficialeffect associated with better socio-emotional adjustment. In other words, children’shigher perception of the teacher as a secure base would correlate with better socio-emotional adjustment (lower loneliness and higher sense of coherence).

4. In line with resilience theory, a subgroup of children with reading difficultieswould emerge who would not evidence lower socio-emotional functioning (theywould reveal low levels of loneliness and high sense of coherence).

5. Children’s affiliation to resilient and non-resilient subgroups would be associatedwith children’s perception of the teacher as a secure base.

Method

Participants

The current study comprised 266 3rd-graders: 118 children with reading difficul-ties (68 boys and 50 girls) and 148 children without reading difficulties (76 boysand 72 girls) in 16 regular schools in central Israel. The two groups showed nosignificant differences in gender, age or other demographic measures. As theybegan 3rd grade, the children with reading difficulties were identified by theirhomeroom teachers as demonstrating consistent reading difficulties and reading ata 1st-grade level. Following this identification, they became eligible to receive aschool-based intensive remedial reading programme, but we collected the currentdata prior to children’s participation in the remedial programme. Children were notreferred at this stage to formal diagnostic evaluation of learning disabilities, in linewith national educational policy to provide intensive reading instruction beforereferral to formal external assessment.

Instruments

● Loneliness and social dissatisfaction questionnaire (Asher et al., 1990): the Hebrew adap-tation (Margalit, 1991) of this children’s self-report scale consisted of 16 primaryitems tapping a child’s feelings of loneliness (e.g. ‘I have nobody to talk to in myclass’, ‘I am lonely’) and eight filler items (e.g. ‘I like school’) that covered variousactivity areas on a five-point scale ranging from Never (1) to Always (5). Higher scoresreflected more frequent feelings of loneliness (Cronbach alpha = 0.86).

● Children’s sense of coherence scale (Margalit & Efrati, 1995): this self-report scaleconsisted of 16 items tapping three dimensions of children’s sense of confidence inthe world: (a) sense of comprehensibility—feelings that one understands one’senvironment (e.g. ‘I feel that I don’t understand what to do in class’); (b) sense ofmanageability—feelings of control and confidence that positive rewards areavailable (e.g. ‘When I want something, I’m sure I’ll get it’); and (c) sense of

28 M. Al-Yagon

meaningfulness—motivation and interest in investing effort in different tasks (e.g.‘I’m interested in lots of things’), on a four-point scale ranging from Never (1) toAlways (4). Higher scores reflected a higher sense of coherence. Antonovsky(1987) recommended the computation of a single total score tapping the globalsense of coherence. In the current sample, the Cronbach alpha for the total scale(α = 0.75) allowed us to compute a total coherence score by summing up the 16items. Higher scores reflected a higher sense of coherence.

● Children’s appraisal of teacher as a secure base scale (CATSB: Al-Yagon &Mikulincer, 2004b): this children’s self-report scale assessed school-age children’sperceptions of the teacher focusing on major features of the attachment figure. The25-item scale asked children to rate the extent to which the item described theirhomeroom teacher on a seven-point scale ranging from Not at all appropriate (1) toVery appropriate (7). Items were organized around two subscales of children’sperception of the teacher as a secure base.

● Availability subscale (17 items; Cronbach alpha = 0.87): this assessed the teacher asavailable in times of need (e.g. ‘When I need the teacher’s help, she is alwaysthere’) and as a caring figure who is accepting of the child’s needs, feelings andbehaviours (e.g. ‘My teacher expresses her appreciation of me even when I try butfail’, ‘When I am worried or sad, my teacher helps me feel better’). Higher scoresreflected higher perceptions of the homeroom teacher as available and accepting attimes of need.

● Rejection subscale (8 items; Cronbach alpha = 0.76) tapped the extent to which theteacher was perceived as rejecting the child, such as ‘My teacher tends tocomplain about me to other adults (for instance, parents, teachers, principal)’.Higher scores reflected higher perceptions of the homeroom teacher as rejecting attimes of need.

Procedure

At the beginning of the 3rd grade, after obtaining parental consent, teachers assessedall students in their classrooms and identified those children who demonstrated read-ing difficulties, defined as reading at a 1st-grade level. They were trained to analysetheir students’ reading level and afterwards to provide intensive remedial teaching.After receiving approval from the Israeli Ministry of Education, we collected thecurrent data prior to this intensive remedial teaching.

Children completed the set of three self-report questionnaires: the Loneliness andsocial dissatisfaction, Sense of coherence and Children’s appraisal of teacher as a securebase scale (CATSB), as a group in their classrooms during school hours, in thepresence of either qualified teachers who were colleagues of the 3rd-grade class-room teachers or members of our research team comprising graduate students inspecial education programmes. The examiner read sample items aloud for eachquestionnaire to ensure children’s understanding, and provided additional help tostudents if necessary. Completion of the questionnaire set took from 40 to 60minutes.

Socio-emotional characteristics and perception of teachers 29

Results

Children’s reading difficulties and socio-emotional adjustment

To examine differences between children with and without reading difficulties ontheir self-reported socio-emotional characteristics, we conducted a two-way (group ×gender) multivariate analysis of variance (MANOVA), with sense of coherence andloneliness as the dependent variables. The MANOVA yielded a significant maineffect for study group, F (2, 250) = 5.60, p < 0.01. Neither the main effect for gendernor the interaction between study group and gender were statistically significant.Univariate ANOVAs revealed significant main effects for study group for both senseof coherence and loneliness (see means, standard deviations and F scores in Table 1).In line with the first hypothesis, an examination of group means indicated thatchildren with reading difficulties scored higher in loneliness and lower in sense ofcoherence than did children without reading difficulties.

Children’s patterns of close relationships with teachers. To examine differences betweenchildren with and without reading difficulties on their perception of close relationshipswith teachers, we conducted a two-way MANOVA (group × gender) with the twoCATSB scores (i.e. availability and rejection) as the dependent variables. The analysisyielded a significant main effect for study group, F (2, 250) = 6.93, p < 0.01. Neitherthe main effect for gender nor interaction between study group and gender were statis-tically significant. Univariate ANOVAs revealed significant main effects for studygroup on CATSB rejection scores but not on availability scores (see means, standarddeviations and F scores in Table 1). Thus, in line with the second hypothesis, anexamination of group means indicated that children with reading difficulties perceivedtheir teachers as more rejecting than did students without reading difficulties.Conversely, children with reading difficulties did not perceive their teachers assignificantly less available than did children without reading difficulties.

Table 1. Means and standard deviations of children’s loneliness, sense of coherence, and perception of teacher as a secure base (CATSB), according to study group

Children with reading difficulties (n = 118)

Children without reading difficulties (n = 148) F

M SD M SD (1, 251)

Socio-emotional measuresLoneliness 31.81 10.75 28.50 11.03 5.5*Sense of coherence 48.12 5.16 50.70 6.98 10.39**Teacher perceptionCATSB availability 86.46 15.00 86.80 16.06 0.01CATSB rejection 16.30 10.20 12.41 6.81 12.60***

*p < 0.05; **p < 0.01; ***p < 0.001.

30 M. Al-Yagon

Close relationships with teachers and socio-emotional adjustment

Before investigating the contribution of children’s perception of teachers as a securebase to their socio-emotional characteristics, we examined the intercorrelationsamong the various measures. In line with the present objective of scrutinizing theunique contribution of children’s difficulties to their perceptions, we conducted theseanalyses separately for children with and without reading difficulties.

With regard to children without reading difficulties, significant Pearson correlationsemerged between the two CATSB measures of availability and rejection. Childrenwho perceived their teacher as more available were significantly less likely to perceivetheir teacher as rejecting and vice versa (r = −0.45, p < 0.001). Teacher availabilityand teacher rejection on the CATSB also each revealed significant Pearson correla-tions with both measures of socio-emotional adjustment. Children’s perception oftheir teachers as more rejecting correlated significantly with a higher level of loneliness(r = 0.25, p < 0.01) and with a lower sense of coherence level (r = −0.43, p < 0.001).Similarly, children’s perception of their teachers as more available correlated signifi-cantly with a lower level of loneliness (r = −0.33, p < 0.001) and with a higher senseof coherence level (r = 0.43, p < 0.001). Significant Pearson correlations also revealeda significant association between these children’s feelings of loneliness and their senseof coherence. Children with a high sense of coherence tended to report lower feelingsof loneliness than did children with a low sense of coherence (r = −0.46, p < 0.001).

Among children with reading difficulties, the results indicated fewer significantassociations between the study measures. Conversely to children without reading diffi-culties, among this group of children, no significant correlations emerged between thetwo CATSB measures of availability and rejection. In fact, only two significant Pear-son correlations emerged among the children with reading difficulties. Similarly totheir peers without reading difficulties, children’s perception of their teachers as morerejecting correlated significantly with a higher level of loneliness (r = 0.21, p < 0.05).In addition, children with a high sense of coherence tended to report lower feelingsof loneliness than did children with a low sense of coherence (r = −0.54, p < 0.001).

The third hypothesis concerned the role of children’s close relationships with theirteachers as a protective factor; that is, when children evaluated teachers’ highavailability and low rejection. To investigate this hypothesis, we divided the totalsample into three groups (high, average and low) according to their scores on each ofthe two CATSB measures.

First, we conducted two-way MANOVA for gender and for affiliation to high/aver-age/low perceptions of the teacher as rejecting, with sense of coherence and lonelinessas the dependent variables. The MANOVA yielded a significant main effect foraffiliation to subgroup of rejection perceptions, F (4, 494) = 9.26, p < 0.001. Neitherthe main effect for gender nor the interaction between gender and rejection percep-tions subgroup showed significant effects on either of the two socio-emotional adjust-ment measures. Univariate ANOVAs revealed significant main effects for rejectionperception subgroup for both loneliness and sense of coherence (see means, standarddeviations and F scores in Table 2). As expected, an examination of group means

Socio-emotional characteristics and perception of teachers 31

indicated that children who perceived their teachers as highly rejecting reportedhigher loneliness and a lower sense of coherence than did children who perceivedtheir teachers as slightly rejecting.

Second, we conducted a two-way MANOVA for gender and for affiliation to high/average/low perceptions of the teacher as available, with sense of coherence and lone-liness as the dependent variables. The MANOVA yielded a significant main effect foraffiliation to the subgroup of availability perceptions, F (4, 494) = 5.32, p < 0.001.Neither the main effect for gender nor the interaction between gender and availabilityperception subgroup showed significant effects on either of the two socio-emotionaladjustment measures. Univariate ANOVAs revealed significant main effects for theavailability perception subgroup for both sense of coherence and loneliness (seemeans, standard deviations and F scores in Table 2). As expected, an examination ofgroup means indicated that children who perceived their teachers as highly availablereported lower loneliness and a higher sense of coherence than did children whoperceived their teachers as slightly available.

Resilient children with reading difficulties and close relationships with teachers

To explore the resilience of children with reading difficulties in line with the fourthhypothesis, we divided the total sample into three groups (high, average and low)according to their loneliness scores and into three groups (high, average and low)according to their sense of coherence scores. We defined resilience as reporting both

Table 2. Means and standard deviations of socio-emotional measures according to affiliation to low/average/high scores on the two CATSB measures: rejection and availability

Rejection perception subgroups

Low (N = 83) Average (N = 94) High (N = 89)

M SD M SD M SD F (1,247)

Loneliness 26.55 9.25 29.24 11.44 33.90 11.05 9.57***Sense of coherence

52.40 6.32 49.75 6.20 46.65 5.42 17.90***

Availability perception subgroups

Low (N = 88) Average (N = 89) High (N = 89)

M SD M SD M SD F (1,247)

Loneliness 33.29 11.5 29.12 10.10 27.01 10.50 8.10***Sense of coherence

47.54 6.12 49.71 5.5 51.51 7.11 8.00***

*p < 0.05; **p < 0.01; ***p < 0.001.

32 M. Al-Yagon

low scores on the loneliness measure (M = 19.20 for resilient children vs M = 32.24for non-resilient children who scored either average or high on loneliness) andhigh scores on the sense of coherence measure (M = 57.80 for resilient children vsM = 47.74 for non-resilient children who scored either low or average on coherence).

The analysis identified a group of 46 resilient children: 11 (9.3%) of the childrenwith reading difficulties, and 35 (23.6%) of the children without reading difficulties.A χ2 test revealed a significant difference between the two study groups on the distri-bution of resilient and non-resilient children, χ2 (1, 264) = 9.07, p < 0.01. The 11children with reading difficulties demonstrated high socio-emotional functioning,suggesting the isolated impact of their reading problems.

An important remaining question comprised whether children’s classification intoresilient and non-resilient subgroups would be associated with children’s perceptionof teachers as a secure base. We examined this association using a two-way MANOVA(group × gender) with the two measures of the CATSB, availability and rejection scores,as the dependent variables. The analysis yielded a significant main effect for studygroup, F (2, 261) = 11.40, p < 0.001. Neither the main effect for gender nor the inter-action between study group and gender were statistically significant. UnivariateANOVAs revealed significant main effects for study group on both CATSB measures.As expected, an examination of group means indicated that, compared to non-resilientchildren, the subgroup of resilient children perceived their teachers as more available(M = 94.60 for resilient children vs M = 85.22 for non-resilient children) and as lessrejecting (M = 9.91 for resilient children vs M = 15.03 for non-resilient children).

Discussion

The current findings partially supported our hypotheses concerning the contributionof children’s reading difficulties to children’s perceptions of their socio-emotionalcharacteristics and perceptions of their homeroom teacher as a secure base. Twomain theoretical issues emerged from the current findings: (a) the relevance andvalidity of attachment theory for explaining child–teacher relations and (b) expandingknowledge on the contribution of reading problems to school-age children’s percep-tion of their own socio-emotional characteristics, as well as their perception of thehomeroom teacher as a secure base.

First, the current findings highlighted attachment theory as a relevant, validexplanation for school-age children’s close relations with teachers. In line with ourpredictions, the results demonstrated that children with reading difficulties seemed toperceive their teachers as more rejecting than did children who did not exhibit readingdifficulties. In contrast with our expectations, the results indicated that the twogroups of children did not differ in their perceptions concerning their teachers’ avail-ability. Previous studies have shown that children who had already received formaldiagnoses of learning disorders reported greater dissatisfaction with their teachers(Murray & Greenberg, 2001; Al-Yagon & Mikulincer, 2004a, 2004b). However, thecurrent findings emphasized that although children with reading difficulties perceivedtheir teachers as providing adequate instrumental care (i.e. supplying extra help and

Socio-emotional characteristics and perception of teachers 33

instruction), they concurrently viewed their teachers as more rejecting than did theirpeers who exhibited no reading problems.

These outcomes highlight the role of children’s reading difficulties as a vulnerabilityfactor that may impair the quality of the child–teacher relationship. As a group, thesechildren do not perceive their teacher as providing a secure base—the basic functionsof an attachment figure—and thus may form an insecure attachment relationship withthe teacher. Considering the previous research that examined teachers’ characteristicssuch as expectations, teaching efficacy and tolerance towards maladaptive behaviours(Gersten et al., 1988; Landrum & Kauffman, 1992; Hocutt, 1996), the current resultscall for additional exploration of such characteristics’ contribution to the process ofsupporting children with reading difficulties.

Similarly, the current results revealed that child’s reading difficulties also affectedboth measures of children’s socio-emotional adjustment. The current findingsclearly indicated that as a group these children with reading difficulties reportedgreater feelings of loneliness and a lower level of coping resources as manifested intheir sense of coherence. These results supported earlier reports of children who hadalready received a formal diagnosis of learning disorders (e.g. Asher et al., 1990;Margalit & Levin-Al-Yagon, 1994; Pavri & Monda-Amaya, 2000). Overall, thepresent outcomes highlighted the possible role of reading difficulties as a vulnerabil-ity factor for the development of socio-emotional maladjustment. These results maysupport researchers’ assumptions concerning the contribution of children’s individ-ual characteristics to their socio-emotional functioning, with respect to such charac-teristics as information-processing deficits, impulsivity and performance andproduction deficits (e.g. Culbertson, 1998).

Findings from the present study may also expand our knowledge about the relationof children’s perceived relationships with extra-familial figures (i.e. teachers) and thechildren’s socio-emotional characteristics. Children’s perceptions of the teacher as asource of secure base—as a more available and less rejecting figure—were signifi-cantly related to their higher sense of coherence and lower feelings of loneliness.Overall, more secure attachment patterns in close relationships with the teacher wererelated to children’s feelings of loneliness, as well as their sense of coherence. Theseresults corroborate prior research emphasizing that extra-familial figures such aspeers, non-familial caregivers and teachers can serve as attachment-like figures,providing care and support in times of need and becoming a source of secure base forfurther exploration and learning of new skills (Owens et al., 1995; Weiss, 1998).

In accordance with the resilience approach (Antonovsky, 1979, 1987; Werner,1993; Wong, 2003), this study also identified a subgroup of children with readingdifficulties (9.5%) who reported a high sense of coherence and low feelings of lone-liness. This subgroup of children demonstrated isolated academic problems in thearea of reading, alongside well-adjusted functioning. However, the results empha-sized the significant difference between the two study groups on the distribution ofresilient and non-resilient children. These findings call for additional exploration ofthe characteristics of this subgroup of resilient children who, despite their difficul-ties, ‘beat the odds’ and manifested socio-emotional adjustment.

34 M. Al-Yagon

Interestingly, in examining whether children’s classification into resilient and non-resilient subgroups correlated with children’s perception of teachers as a secure base,our results indicated that the subgroup of resilient children tended to perceive theirteachers as more available and as less rejecting in comparison to their non-resilientpeers. These results support recent studies’ call for investigation of potential risk andprotective factors among this population, and for identification of resilient individualswho manifest adjustment without ignoring their deficits (Margalit, 2003; Wong,2003). The current findings suggest the importance of additional exploration into thecontribution of vulnerability and protective factors that may enhance the resilience ofchildren with reading difficulties.

The practical implications of this study concern two main components. First, thepresent outcomes suggest the need to develop effective interventions for children withreading problems even prior to formal diagnosis, by implementing bi-dimensionalattachment theory as well as recent studies on peer relations. These interventions couldincorporate: (a) interventions with teachers of children who have reading difficultiesto facilitate perceptions of the teacher as a secure base regarding both aspects of rejec-tion and availability, and (b) interventions at the individual and class level to promotethe abilities of children with reading difficulties to initiate, cooperate and maintain effi-cient peer relations as well as enhance their coping resources. Research must examinethe efficacy of such intervention programmes in buffering the socio-emotional diffi-culties of children with reading difficulties, and to sensitize teachers to the knowledgethat providing more time and increasing their availability to students who need themmay not reduce the children’s experience of being rejected by their teachers.

Second, professionals may utilize the current study’s social and emotional factorsto identify at-risk children with reading difficulties who may manifest socio-emotionalmaladjustment alongside their reading difficulties.

Several limitations of this study call for further research. First, the current focus onchildren’s specific attachment with teachers should be expanded to include the inter-relations between intra-familial and extra-familial attachment-based factors. Second,in line with recent studies emphasizing that the conceptualization of loneliness andattachment requires the consideration of a complexity of interacting factors over time,including significant hereditability and non-shared environmental influences(McGuire & Clifford, 2000; Larose et al., 2002), the present findings call for addi-tional exploration focusing on the role of individual factors (e.g. children’s disabili-ties, personality factors), family factors (e.g. parents’ characteristics, parental style)and community factors (e.g. the school’s readiness to accommodate itself to thechild’s special needs and social support from adults in the school) in explainingchildren’s loneliness as well as adaptive socio-emotional functioning. Moreover, asthe current study focused on children’s self-report measures, future research shouldinvestigate other information sources such as parents and teachers along severalmeasures of socio-emotional adjustment. Finally, the current correlational nature ofthis study calls for future longitudinal and experimental designs that follow up onthese children with reading difficulties to investigate the longevity of their perceptionsover time, particularly after formal diagnostic assessment.

Socio-emotional characteristics and perception of teachers 35

Acknowledgement

The authors would like to thank Dee B. Ankonina for her editorial contribution.

References

Achenbach, T. M. (1993) Implications of multi-axial empirically based assessment for behaviortherapy with children, Behavior Therapy, 24, 91–116.

Ainsworth, M. D. & Wittig, B. A. (1969) Attachment and exploratory behavior of one-year-olds ina strange situation, in: B. M. Foss (Ed.) Determinants of infant behavior, Vol. 4 (London,Methuen), 113–136.

Ainsworth, M. D., Blehar, M. C., Waters, E. & Wall, S. (1978) Patterns of attachment: a psycholog-ical study of the strange situation (Hillside, NJ, Erlbaum).

Al-Yagon, M. (2003) Children at-risk for developing learning disorders: multiple perspectives,Journal of Learning Disabilities, 36(4), 318–335.

Al-Yagon, M. & Mikulincer, M. (2004a) Patterns of close relationships and socioemotional andacademic adjustment among school-age children with learning disabilities, Learning DisabilitiesResearch and Practice, 19(1), 12–19.

Al-Yagon, M. & Mikulincer, M. (2004b) Socioemotional and academic adjustment amongchildren with learning disorders: the mediational role of attachment-based factors, Journal ofSpecial Education, 38(2), 111–123.

Antonovsky, A. (1979) Health, stress and coping (San Francisco, CA, Jossey-Bass).Antonovsky, A. (1987) Unraveling the mystery of health (San Francisco, CA, Jossey-Bass).Antonovsky, H. & Sagy, S. (1986) The development of a sense of coherence and its impact on

responses to stress situations, Journal of Social Psychology, 126, 213–225.Asher, S. R. & Paquette, J. A. (2003) Loneliness and peer relations in childhood, Current Directions

in Psychological Sciences, 12(3), 75–78.Asher, S. R., Parkhurst, J. T., Hymel, S. & Williams, G. A. (1990) Peer rejection and loneliness in

childhood, in: S. R. Asher & J. D. Coie (Eds) Peer rejection in childhood (Cambridge,Cambridge University Press), 253–273.

Birch, S. H. & Ladd, G. W. (1997) The teacher–child relationship and children’s early schooladjustment, Journal of School Psychology, 35, 61–9.

Bowlby, J. (1973) Attachment and loss: anxiety, anger, and separation (New York, Basic Books).Bowlby, J. (1982) Attachment and loss: attachment (New York, Basic Books) (original work

published 1969).Bowlby, J. (1988) A secure base: clinical applications of attachment theory (London, Routledge).Buchholz, E. S. & Catton, R. (1999) Adolescents’ perceptions of aloneness and loneliness,

Adolescence, 34(133), 203–213.Chapman, J. W. & Tunmer, W. (2003) Reading difficulties, reading-related self-perceptions, and

strategies for overcoming negative self-beliefs, Reading and Writing Quarterly, 19, 5–24.Culbertson, J. L. (1998) Learning disabilities, in: T. H. Ollendick & M. Hersen (Eds) Handbook of

child psychopathology (New York, Plenum Press), 117–156.Francis, D. J., Shaywitz, S. E., Stuebing, K. K., Shaywitz, B. A. & Fletcher, J. M. (1996) Develop-

mental lag versus deficit models of reading disability: a longitudinal, individual growth curvesanalysis, Journal of Educational Psychology, 88, 3–17.

Gersten, R., Walker, H. & Darch, D. (1988) Relationships between teachers’ effectiveness andtheir tolerance of handicapped students, Exceptional Children, 54, 433–438.

Hay, D. F., Payne, A. & Chadwick, A. (2004) Peer relations in childhood, Journal of Child Psychologyand Psychiatry and Allied Disciplines, 45(1), 84–108.

Hinshaw, S. P. (1992) Academic underachievement, attention deficits, and aggression: comorbidityand implications for intervention, Journal of Consulting and Clinical Psychology, 60, 693–703.

36 M. Al-Yagon

Hocutt, A. M. (1996) Effectiveness of special education: is placement the critical factor? Future ofChildren, 6, 77–102.

Howes, C. (1999) Attachment relationships in the context of multiple caregivers, in: J. Cassidy &P. R. Shaver (Eds) Handbook of attachment: theory, research, and clinical applications (NewYork, Guilford), 671–687.

Howes, C. & Hamilton, C. E. (1993) The changing experience of child care: changes in teachersand in teacher–child relationships and children’s social competence with peers, Early Child-hood Research Quarterly, 8, 15–32.

Howes, C., Phillipsen, L. C. & Peisner-Feinberg, E. (2000) The consistency of perceived teacher–child relationships between preschool and kindergarten, Journal of School Psychology, 38(2),113–132.

Kerns, K. A., Klepac, L. & Cole, A. (1996) Peer relationships and preadolescents’ perceptions ofsecurity in the child–mother relationship, Developmental Psychology, 32, 457–466.

Ladd, G. W. & Troop-Gordon, W. (2003) The role of chronic peer difficulties in the developmentof children’s psychological adjustment problems, Child Development, 74(5), 1344–1367.

Landrum, T. J. & Kauffman, J. M. (1992) Characteristics of general education teachers perceivedas effective by their peers: implications for inclusion of children with learning and behavioraldisorders, Exceptionality, 3, 147–163.

Larose, S., Guay, F. & Boivin, M. (2002) Attachment, social support, and loneliness in youngadulthood: a test of two models, Personality and Social Psychology Bulletin, 28(5), 684–693.

Luthar, S. S. & Cicchetti, D. (2000) The construct of resilience: implications for interventions andsocial policies, Development and Psychopathology, 12, 857–885.

Lynch, M. & Cicchetti, D. (1992) Maltreated children’s reports of relatedness to their teachers, in:R. C. Pianta (Ed.) New directions in child development: Vol. 57, Beyond the parent: the role of otheradults in children’s lives (San Francisco, CA, Jossey-Bass), 81–108.

Lynch, M. & Cicchetti, D. (1997) Children’s relationships with adult and peers: an examination ofelementary and junior high school students, Journal of School Psychology, 35(1), 81–99.

Margalit, M. (1991) Understanding loneliness among students with learning disabilities, BehaviorChange, 8, 167–173.

Margalit, M. (1994) Loneliness among children with special needs: theory, research, coping and interven-tion (New York, Springer-Verlag).

Margalit, M. (2003) Resilience model among individuals with learning disabilities (LD): proximaland distal influences, Learning Disabilities Research and Practice, 18(2), 82–86.

Margalit, M. & Al-Yagon, M. (2002) Loneliness among children with learning disabilities, in: B.Wong & R. Pearl (Eds) Learning disabilities: social-emotional aspects (Mahwah, NJ, LawrenceErlbaum), 53–75.

Margalit, M., Al-Yagon, M. & Neuberger, S. (1996) Mothers’ coherence, family climate, andpreschool children with learning disabilities: risk and resilience, paper presented at the AnnualInternational Academy for Research in Learning Disabilities Conference, Detroit, MI, October.

Margalit, M. & Efrati, M. (1995) Sense of coherence, companionship, and loneliness amongchildren with learning disorders, paper presented at the Annual Conference of the Society forResearch in Child Development, Indianapolis, IN, March.

Margalit, M. & Levin-Al-Yagon, M. (1994) Learning disability subtyping, loneliness and classroomadjustment, Learning Disability Quarterly, 17(4), 297–310.

Margalit, M., Tur-Kaspa, H. & Most, T. (1999) Reciprocal nominations, reciprocal rejections andloneliness among students with learning disorders, Educational Psychology, 19(1), 79–90.

Maughan, B., Rowe, R., Loeber, R. & Stouthamer-Loeber, M. (2003) Reading problems anddepressed mood, Journal of Abnormal Child Psychology, 31(2), 219–230.

McGuire, S. & Clifford, J. (2000) Genetic and environmental contributions to loneliness inchildren, Psychological Science, 11(6), 487–491.

Morrison, G. M. & Cosden, M. A. (1997) Risk, resilience, and adjustment of individuals withlearning disabilities, Learning Disabilities Quarterly, 20, 43–60.

Socio-emotional characteristics and perception of teachers 37

Murray, C. & Greenberg, M. T. (2000) Children’s relationships with teachers and bonds withschool: an investigation of patterns of correlates in middle childhood, Journal of School Psychol-ogy, 38(5), 423–445.

Murray, C. & Greenberg, M. T. (2001) Relationships with teachers and bonds with school: socialemotional adjustment correlates for children with and without disabilities, Psychology in theSchools, 38(1), 25–41.

Owens, G., Crowell, J. A., Pan, H., Treboux, D., O’Connor, E. & Waters, E. (1995) The proto-type hypothesis and the origins of attachment working models: adult relationships withparents and romantic partners, in: E. Waters, B. E. Vaughn, G. Posada & K. Kondo-Ikemura(Eds) Caregiving, cultural, and cognitive perspectives on secure-base behaviors and working models:Monographs of the Society for Research in Child Development, 60 (Illinois, IL, SRCD), 216–134.

Pavri, S. & Monda-Amaya, L. (2000) Loneliness and students with learning disabilities in inclusiveclassrooms: self-perceptions, coping strategies, and preferred interventions, Learning DisabilitiesResearch and Practice, 15(1), 22–33.

Pianta, R. C. (1992) New directions in child development, Vol. 57: Beyond the parent: the role of otheradults in children’s lives (San Francisco, CA, Jossey-Bass).

Pianta, R. C., Steinberg, M. & Rollins, K. (1995) The first two years of school: teacher–child rela-tionships and deflections in children’s classroom adjustment, Development and Psychopathology,7, 297–312.

Qualter, P. & Munn, P. (2002) The separateness of social and emotional loneliness in childhood,Journal of Child Psychology and Psychiatry and Allied Disciplines, 43(2), 233–244.

Richardson, G. E. (2002) The metatheory of resilience and resiliency, Journal of Clinical Psychol-ogy, 58(3), 307–321.

Ronen, T. (1997) Cognitive developmental therapy with children (Chichester, Wiley).Sabornie, E. J. (1994) Social-affective characteristics in early adolescents identified as learning

disabled and nondisabled, Learning Disability Quarterly, 17(4), 268–279.Sroufe, L. A. & Waters, E. (1977) Attachment as an organizational construct, Child Development,

48, 1184–1199.Thompson, R. A. (1999) Early attachment and later development, in: J. Cassidy & P. R. Shaver

(Eds) Handbook of attachment: theory, research, and clinical applications (New York, GuilfordPress), 265–286.

Torgesen, J. K. (2002) The prevention of reading difficulties, Journal of School Psychology, 40(1),7–26.

Torgesen, J. K., Rashotte, C. A. & Alexander, A. (2001) Principles of fluency instruction in read-ing: relationships with established empirical outcomes, in: M. Wolf (Ed.) Dyslexia, fluency andthe brain (Parkton, MD, York Press), 333–355.

Waters, H. S., Rodrigues, L. M. & Ridgeway, D. (1998) Cognitive underpinnings of narrativeattachment assessment, Journal of Experimental Child Psychology, 71, 211–234.

Weiss, R. S. (1973) Loneliness: the experience of emotional and social isolation (Cambridge, MA, MITPress).

Weiss, R. S. (1998) A taxonomy of relationships, Journal of Social and Personal Relationships, 15,671–683.

Wenz-Gross, M. & Siperstein, G. N. (1998) Students with learning problems at risk in middleschool: stress, social support and adjustment, Exceptional Children, 65, 91–100.

Werner, E. E. (1993) Risk, resilience, and recovery: perspectives from the Kauai longitudinalstudy, Development and Psychopathology, 5, 503–515.

Wong, B. (2003) General and specific issues for researchers’ consideration in applying the risk andresilience framework to the social domain of learning disabilities, Learning Disabilities Research& Practice, 18(2), 68–76.