Teacher–Student Relationships in Special Education: The Value of the Teacher Relationship Interview

Abstract

Research on teacher–student relationships is mainly based on questionnaires investigating teachers’ explicit cognitions. However, it is also important to investigate implicit processes in social interactions, such as internalized feelings. The Teacher Relationship Interview (TRI) is a narrative method aimed at assessing teachers’ (implicit) mental representations of dyadic teacher–student relationships. The TRI may provide—for researchers as well as practitioners—a deeper understanding of teachers’ implicit cognitive and emotional processes elicited in teacher–student relationships, which can be useful for psychodiagnostic assessment and relationship-focused consultation with teachers. This study investigated the psychometric properties of the TRI in a special education sample of students with symptoms of attachment disorders (N = 80). Expected interrelations of the TRI scales were found with a widely used questionnaire of teacher–student relationships, independent observations of teacher–student interactions, and teacher perceptions of student behavior.

Keywords

teacher–student interactions special education narrative interview mental representations

Teacher–student relationships are meaningfully associated with students’ academic and behavioral outcomes. Especially, the development of students with special needs, including students with attachment and/or behavior problems, have been found to be affected by the teacher–student relationship quality (McGrath & Van Bergen, 2015; Verschueren & Koomen, 2012). The attachment perspective applied to teacher–student interactions indicates that a sensitive teacher may act as an (ad hoc) attachment figure with a safe haven and secure base function at school (Verschueren & Koomen, 2012). Accordingly, sensitive teachers can help students to feel safe to explore the environment and to cope with the demands of school, thereby impacting students’ learning behaviors and development (Pianta, 1999).

The attachment perspective assumes that teachers develop mental representations of dyadic (one-on-one) relationships with students through multiple daily interactions. Teachers’ mental representations of relationships with individual students contain internalized representations of the self as a teacher in various teaching roles (e.g., caregiver, instructor), of the needs of the student, and of the self as a teacher in relation to the specific student and his or her development (Pianta, 1999; Pianta, Hamre, & Stuhlman, 2003). These mental representations are assumed to automatically influence teachers’ social information processing. Mental representations may trigger specific cognitions and affective states that guide teachers’ emotional and behavioral responses toward students (Bowlby, 1969/1982; Pianta, 1999). If the content of teachers’ mental representations is negative due to a conflictual teacher–student relationship, this can cause biases in teachers’ social information processing including reasoning errors (e.g., perceiving the student being disruptive “on purpose”; Crick & Dodge, 1994; Thijs & Koomen, 2009) and congruent negative affect. As a result, teachers may become less sensitive to needs of students (e.g., Stuhlman & Pianta, 2002) which, in turn, will reinforce student problem behavior. This often results in a vicious circle of negative processes in which students’ problem behaviors reinforce the maladaptive content of teachers’ mental representations and consequently teachers’ insensitive behaviors, which in turn will intensify students’ problem behaviors (cf. Doumen et al., 2008).

Up to now, teacher–student relationship quality at the dyadic level has mostly been assessed using teacher-report questionnaires. The Student–Teacher Relationship Scale (STRS; Koomen, Verschueren, van Schooten, Jak, & Pianta, 2012; Pianta, 2001) is the most widely used questionnaire for assessing teachers’ perceptions of the quality of the interpersonal relationship with a specific student, the interactive behavior of this student toward the teacher, and teachers’ feelings and thoughts about the student. The STRS is often used as screening tool to get an efficient overall impression of the quality of the relationship (Koomen, Verschueren, & Thijs, 2006). Questionnaires primarily assess teachers’ explicit perceptions of the teacher–student relationship relying on those aspects teachers can consciously reflect upon (Bosmans & Kerns, 2015). However, teachers may not be (fully) aware of their internalized, implicit representations of teacher–student relationships. Attachment research has shown that narrative interview techniques are suitable to investigate implicit mental representations of relationships (e.g., Adult Attachment Interview; Main, Kaplan, & Cassidy, 1985). Narrative techniques are, for example, recently used to measure caregivers’ implicit processes and its associations with their behavior (Borelli, West, Decoste, & Suchman, 2012; Stuhlman & Pianta, 2002). Narrative interviews are expected to have a complementary value to questionnaires which can be of interest for both researchers as practitioners (Bosmans & Kerns, 2015). However, empirical evidence and (predictive) validity of the use of narrative techniques are still scarce.

The Teacher Relationship Interview (TRI; Pianta, 1999) is a semistructured interview to tap teachers’ mental representations of their relationships with individual students (Spilt & Koomen, 2009). The TRI is an adapted version of the Parent Development Interview (Button, Pianta, & Marvin, 2001; Pianta, 1999) that was developed to investigate specific dimensions of parental representations of child–parent relationships. The dimensions were based on the attachment literature on mental models of caregiving (Bowlby,1969/1982). The oldest studies focused on the content and process of thoughts (George & Solomon, 1996). The content dimension reflects the extent to which a particular role of the caregiver is represented (e.g., Secure Base). The process dimension reflects how represented information is admitted or denied indicated by reluctance to talk about certain themes or negative feelings (e.g., Neutralizing of Negative Affect and Coherence of the Narrative). Later, the affective dimension (e.g., Helplessness, Positive Affect, and Anger) was included acknowledging the internalized feelings and affective states of caregivers (Bowlby, 1969/1982; Button et al., 2001). Consequently, the TRI captures three theoretic dimensions of mental representations of teacher–student relationships (Button et al., 2001; Spilt & Koomen, 2009; see Supplemental Appendix A). Offering a differentiated picture of the dyadic teacher–student relationship, the use of the TRI can be useful for (trained) teacher consultants as a starting point for psychodiagnostic assessment and relationship-focused consultation with teachers (Koomen et al., 2006).

Only two studies in regular primary schools have examined the TRI as a measure of teachers’ mental representations of dyadic relationships. First, Stuhlman and Pianta (2002) found that more narrated negative emotions of the teacher were strongly associated with more interactions between teacher and student (r = .42) and more negative behavior of teachers toward that student (r =.33, e.g., angry tone of voice). Second, Spilt and Koomen (2009) found associations between the TRI and the STRS. As expected, they found moderate convergence between teacher-reports with the STRS and teacher-narratives elicited with the TRI (r = .28-.56): Conflict related negatively to expressed anger, whereas closeness in teacher–student interactions were positively associated with narrated positive affect and low levels of helplessness. Teachers’ narrated sensitive practices were associated with closeness, conflict, and dependency in teacher–student interactions. In addition, narratives about relationships with externalizing in comparison with nonexternalizing students revealed more anger (d = 0.72) and helplessness (d = 0.42) of the teacher.

This study builds upon the previous studies to support the validity of the TRI in a special education sample. Until now, there are no validity studies of the TRI in special education or in clinical samples. However, the TRI can also be useful in facilitating the consultation process with teachers in relation to students with special needs (Koomen et al., 2006). Students with attachment problems may have more difficulties in forming close relationships with their teacher because they are less likely to trust their caregivers due to a history of insufficient caregiving and more likely to interact with others in socially inappropriate ways (Schwartz & Davis, 2006). Nevertheless, positive teacher–student relationships for these students are of great importance because they have been found to buffer negative developmental outcomes for these students (McGrath & Van Bergen, 2015). This study is the first to investigate the psychometric properties of the TRI in a special education sample including students with mild to severe attachment problems.

The Current Study

The current study investigated the reliability and the validity of the TRI in a special education sample of students with mild to severe attachment problems. First, this study investigated the interrater reliability of the scales of the TRI (cf. Cicchetti et al., 2006). Second, this study investigated the convergent and the concurrent validity of the TRI. To investigate the convergent validity, this study examined associations of the TRI with a teacher-report questionnaire (STRS) and with independent observations of teacher–student interactions (teacher sensitivity and emotional security of the student). We expected that positive affect of the TRI was positively associated with closeness of the STRS and that anger of the TRI was negatively associated with conflict of the STRS (cf. Spilt & Koomen, 2009). Concerning observed teacher–student interactions, we expected, for example, that sensitive practices would be positively and anger would be negatively associated with observed teacher sensitivity and emotional security of the student (cf. Stuhlman & Pianta, 2002). To investigate the concurrent validity, this study also examined associations of the TRI with measures of teacher-perceived student behavior (prosocial behavior, school adjustment, externalizing problems, and emotional problems). We expected that positive scales of the TRI (e.g., positive affect) would be positively related to school adjustment and prosocial behavior of the student and negatively to students’ externalizing and emotional problems, and vice versa for negative scales of the TRI (e.g., anger).

Method

Sample

The sample consisted of 85 students and 70 teachers (15 teachers reported on two students of their class) in 19 elementary schools providing special education for students with emotional and behavioral disorders in Flanders (Dutch-speaking region of Belgium). Students were selected based on the 75% highest scores on a scale of attachment problems with a maximum of two students per teacher in order not to overload teachers (for more information about the selection procedure, see Vervoort, Verschueren, & Bosmans, 2013). Two teachers dropped out of the study, and of three interviews the sound quality was too low for encoding. Accordingly, the final sample consisted of 80 teacher–student dyads including 67 teachers (13 teachers reported on two students) in 19 schools. Teachers (8.8% men) were on average 33.76 years old (SD = 7.74; range = 22.23-56.51) and had on average 8.56 years of experience in education (SD = 7.46; range = 0-33). Students (83.8% boys) were on average 8.29 years old (SD = 0.99; range = 6.22-10.39). The majority of the students had at least one psychiatric diagnosis (86%), including autism spectrum disorder (39%), reactive attachment disorder (31%), attention deficit hyperactivity disorder (30%), conduct disorder (29%), mental retardation (14%), or mood or anxiety disorder (3%); and 48% of the students received more than one diagnosis.

Procedure

All information was collected in the first semester of the school year (fall), and all measurements were completed by a team of trained researchers and graduate students. Three different teams coded the variables of the TRI, the Teacher Sensitivity or Emotional Security of the teacher–student interactions (to prevent inflated associations). Each team was trained in several sessions until excellent interrater reliability was obtained on practice data (intraclass correlation coefficient [ICC] >.75, Cicchetti et al., 2006).

Measurements

TRI

The TRI (Pianta, 1999; Dutch version, Koomen & Lont, 2004) is a semistructured interview that takes approximately half an hour. The TRI contains 12 questions referring to teachers’ interpersonal experiences with a target student (e.g., “Describe a time in the last week when you and your student really clicked”). The teachers are asked to give real-life examples and to be as specific as possible. The TRI comprises nine constructs (see overview in Supplemental Appendix A): Sensitivity of Discipline (sensitive and proactive management style), Secure Base (understanding that emotional support is linked to the child’s social, emotional, and cognitive skills), Perspective Taking (being aware of the student’s internal states, and providing possible reasons for these states), Neutralizing of Negative Affect (avoiding discussing negative emotions), Intentionality (seeking opportunities to promote the student’s growth in the social-emotional and academic domain), Helplessness (expressing feelings of hopelessness and ineffectiveness, and refraining from trying new strategies), Anger (expressing feelings of anger, hostility or disapproval toward the child), Positive Affect (expressing positive feelings, including happiness, closeness, joy, etc.), and Coherence (presenting experiences in a reasonable and understandable manner; Pianta, 1999; Spilt & Koomen, 2009, p. 100). Following the current administration guidelines, ratings were given on a 7-point scale ranging from 1 (little or no evidence of the construct) to 7 (construct is articulated in a clear way and fresh examples are given that support the presence of the construct). Only Coherence was rated on a 5-point scale (for an example of the ratings, see Supplemental Appendix B). Each interview was double coded by two coders. When large rating differences were observed on a construct (i.e., ⩾3), a third coder (the first author) independently rated the interview again and replaced the most extreme score by the third coder’s score (2% of the ratings). Sensitivity of Discipline, Secure Base, Perspective Taking, and Intentionality were averaged to compute the scale Sensitive Practices (based on factor patterns explored by Spilt & Koomen, 2009; α_{coder 1} = .65; α_{coder 2} = .61).

STRS

The STRS is a questionnaire to measure teacher-perceived teacher–student relationship quality (Koomen, Verschueren, & Pianta, 2007; Pianta, 2001). Teachers reported their relationship quality with individual students on three dimensions: Closeness (11 items, e.g., “I share an affectionate, warm relationship with this child”; α = .89), Conflict (11 items, e.g., “Dealing with this child drains my energy”; α = .91), and Dependency (6 items, e.g., “This child reacts strongly to separation from me”; α = .80) on a 5-point Likert-type scale ranging from not at all applicable (1) to highly applicable (5). Many studies (e.g., in Dutch sample: Koomen et al., 2012) supported the construct and predictive validity of the STRS, for example, indicating associations with behavioral engagement or current and future academic skills (e.g., Doumen, Koomen, Buyse, Wouters, & Verschueren, 2012; Hamre & Pianta, 2001).

Observations

Teacher–student interactions were observed in a dyadic setting outside the classroom in three activities of 10 min each: a free choice activity, a cognitive task, and an emotion event task. In the free choice task (T1), the student could choose among many toys (e.g., Lego, puzzles, Playmobil toys, crayons, . . .). The student could play alone with the chosen toys or ask the teacher to play along, or the teacher could take initiative to play together. In the cognitive task (T2), the student had to solve as many difficult puzzles (smartgame) as possible. The level of difficulty of the puzzles was high and required teacher support. In the emotion events task (T3), the teacher presented the student four cards with a pictorial representation of a facial expression of joy, anger, fear, or sadness. The teacher asked the student to recall an emotional experience for each of the four cards and talk about what he or she felt, thought, and did in that situation. For the second and third task, the teachers were instructed to guide and motivate the student, if needed, in performing these tasks. The teachers received a sheet with the solutions of the smartgame (T2) and questions to discuss with the student (T3, e.g., “What did the student feel in that situation?”). Teacher–student interactions were videotaped and coded afterward. The scale Teacher Sensitivity was inspired by the Classroom Assessment Scoring System (CLASS; La Paro, Pianta, & Stuhlman, 2002). The CLASS was adapted by Verschueren, Van de Water, Buyse, and Doumen (2006) to measure the quality of dyadic teacher–student interactions (instead of classroom-level interactions) in a reliable and valid way (Doumen et al., 2012). The scale Teacher Sensitivity referred to a teacher’s responsiveness to the student and provision of comfort, reassurance, and encouragement with respect to the students’ academic and emotional functioning (Verschueren et al., 2006). The scale Emotional Insecurity referred to a student’s insecure reactions to the teacher, as indicated by responding minimally to questions, displaying nervous gestures, and making an overall very passive, stressed, or timid impression (Koomen, van Leeuwen, & van der Leij, 2004). This scale showed good reliability and validity as evidenced by associations with measures of task-related student behavior and teacher support (e.g., Koomen et al., 2004; Thijs & Koomen, 2008). Teacher Sensitivity and Emotional Insecurity were double coded on a scale ranging from 1 to 7. The scale Emotional Insecurity was reversed to represent a positive outcome measure, that is, Emotional Security. The ICCs for the three tasks were for Teacher Sensitivity, respectively, .53, .63, and .72, and for Emotional Security, respectively, .78, .76, and .80, indicating moderate to substantial interrater reliability (Cicchetti et al., 2006).

Student behavior

Two questionnaires were completed by the teacher to assess student behavior. First, the scale Cooperative Classroom Participation (TSSRA; Ladd, 1992) measures cooperative and responsible behavior of the student as well as the extent to which the student accepts authority of the teacher (Cornelissen & Verschueren, 2001). Eight items (e.g., “This student listens carefully to the teacher’s instructions and directions”; α = .90) were presented to the teachers, who scored them on a 5-point Likert-type scale ranging from certainly not true (1) to certainly true (5). Construct validity have been provided by studies showing associations between classroom participation and student behavior, academic achievement, and relationship quality (e.g., Doumen et al., 2012; Ladd & Burgess, 2001). In addition, factorial validity and internal consistency was found in a Dutch sample (Cornelissen & Verschueren, 2001).

Second, teachers completed four scales of the Strengths and Difficulties Questionnaire (SDQ; Goodman, 1997) with five items each. The 10 items of externalizing problems (Conduct problem and Hyperactivity) were aggregated into the scale Externalizing Problems (e.g., “This student is restless, overactive and cannot stay still for long”; α = .81). In addition, teachers reported on Emotional Problems (e.g., “This student has many fears, is easily scared”; α = .72) and Prosocial Behavior (e.g., “This student is helpful if someone is hurt, upset or feeling ill”; α = .71). Each item has three possible answers (“not true,” “somewhat true,” “certainly true”) scored 0, 1, or 2. The psychometric properties of the SDQ were found to be satisfactory (Goodman, 2001), also in a Dutch sample (van Leeuwen, Meerschaert, Bosmans, De Medts, & Braet, 2006).

Results

Reliability

Table 1 indicates the ICCs and descriptive statistics of the six scales of the TRI. Except for Coherence, the ICCs of all the scales were higher than .60, which indicates good interrater agreement (Cicchetti et al., 2006). Coherence was coded less reliably (<.40) and therefore excluded from further analyses. Table 2 reports the descriptive statistics of the criterion variables.

Table 1.

Descriptive Statistics and Correlations of the TRI Subscales.

TRI subscales	ICC	M (SD)	Range	1	2	3	4
1. Sensitive Practices	.73	4.44 (0.55)	2.88-5.63	—
2. Neutralizing of Negative Affect	.66	3.18 (0.93)	2.00-5.50	.04	—
3. Helplessness	.78	3.06 (1.06)	1.50-6.00	−.45**	−.37**	—
4. Anger	.77	2.78 (1.06)	1.00-5.00	−.34**	−.27*	.33**	—
5. Positive Affect	.70	3.90 (0.90)	2.00-5.50	.62**	−.13	−.34**	−.29**
6. Coherence	.34	3.86 (0.53)	2.50-5.00

Note. TRI = Teacher Relationship Interview; ICC = intraclass correlation coefficient.

p < .05. **p < .01.

Table 2.

Descriptive Statistics and Correlations of All Variables.

Variables	M (SD)	1	2	3	4	5	6	7	8	9	10	11	12
1. Closeness	3.62 (0.70)	—
2. Conflict	2.37 (0.88)	−.37**	—
3. Dependency	2.67 (0.84)	.18	.36**	—
4. Teacher Sensitivity T1	5.03 (0.66)	.05	−.10	−.02	—
5. Emotional Security T1	5.47 (1.09)	.17	.08	−.07	.23*	—
6. Teacher Sensitivity T2	5.09 (0.71)	−.07	.06	−.09	.43**	.20^†	—
7. Emotional Security T2	5.08 (0.98)	.13	−.02	−.05	.15	.63**	.31**	—
8. Teacher Sensitivity T3	4.95 (0.79)	−.06	−.05	−.15	.34**	−.07	.41**	.13	—
9. Emotional Security T3	4.54 (1.00)	.16	−.02	−.04	.09	.43**	.10	.46**	.18	—
10. School Adjustment	3.54 (0.75)	.40**	−.79**	−.32**	−.02	−.04	−.14	.04	.06	−.01	—
11. Externalizing Problems	0.90 (0.45)	−.12	.56**	.36**	−.07	.09	−.06	.06	−.15	−.04	−.57**	—
12. Emotional Problems	0.82 (0.51)	−.17	.28*	.12	.17	.21^†	.06	.01	.11	.07	−.18	.32**	—
13. Prosocial Behavior	1.12 (0.50)	.30*	−.32**	−.13	−.03	.08	−.12	.15	.12	.15	.43**	−.34**	−.16

Note. T1 = Task 1(free choice task); T2 = Task 2(cognition task); T3 = Task 3(emotion event task).

†

p < .10. *p < .05. **p < .01.

Convergent Validity

Table 3 reports the correlations of TRI with the STRS and observational data of teacher–student interactions. The magnitude of the effects is indicated between parentheses (based on Cohen, 1988). Closeness of the STRS was positively associated with Sensitive Practices (medium effect) and Positive Affect (large effect) and negatively associated with Helplessness (medium effect) and Anger (small effect). Conflict of the STRS was positively associated with Helplessness (medium effect) and Anger (small effect) and negatively associated with Positive Affect (small effect). Concerning the teacher–student observations in the free choice task (T1), Teacher Sensitivity and Emotional Security of the student were positively associated with Sensitive Practices (small and medium effect) and Positive Affect (small effects), and negatively associated with Helplessness (small effects). Concerning the observations in the two other tasks (T2-T3), only the associations between Emotional Security of the student and Helplessness in both tasks (medium and small effect) and Positive Affect in T3 (medium effect) were significant.

Table 3.

Correlations of the TRI Subscales With the STRS, Observational Data of Teacher–Student Interactions, and Teacher-Reported Student Behavior.

Variables	TRI subscales
Variables	Sensitive practices	Neutralizing of negative affect	Helplessness	Anger	Positive affect
Closeness	.44**	−.04	−.39**	−.19^†	.52**
Conflict	−.08	−.12	.31**	.26**	−.27*
Dependency	.11	−.02	−.00	.08	.00
Teacher Sensitivity T1	.25*	.05	−.19^†	.04	.25*
Emotional Security T1	.34**	.01	−.20^†	−.11	.23*
Teacher Sensitivity T2	−.04	−.08	−.04	.16	.05
Emotional Security T2	.17	−.04	−.23*	−.01	.17
Teacher Sensitivity T3	−.16	−.09	.02	.05	.00
Emotional Security T3	.04	.10	−.30**	−.06	.20^†
School Adjustment	.18	.05	−.32**	−.36**	.35**
Externalizing Problems	.08	−.21^†	.22^†	.26*	−.08
Emotional Problems	.05	−.17	.15	.11	−.16
Prosocial Behavior	.03	−.01	−.09	−.06	.28*

Note. TRI = Teacher Relationship Interview; STRS = Student–Teacher Relationship Scale; T1 = Task 1(free choice task); T2 = Task 2(cognition task); T3 = Task 3(emotion event task).

†

p < .10. *p < .05. **p < .01.

Concurrent Validity

Table 3 also reports the correlations of TRI with teacher-reported student behavior. Students’ Prosocial Behavior (small effect) and School Adjustment (medium effect) were positively associated with Positive Affect. In addition, School Adjustment was negatively associated with Helplessness and Anger (medium effects). Externalizing Problems was positively associated with Helplessness and Anger and negatively with Neutralizing of Negative Affect (small effects). No associations between Emotional Problems of the student and the TRI scales were found.

Discussion

Research on teacher–student relationships is mainly based on questionnaires merely investigating teachers’ explicit cognitions. However, it is also important to investigate implicit processes in social interactions, such as internalized feelings and beliefs. The TRI is a narrative interview that assesses teachers’ implicit mental representations of dyadic teacher–student relationships. This multimethod study in a sample of students with attachment problems aimed to extend the support for the psychometric properties of the TRI.

Reliability and Validity of the TRI

The findings of our study supported the reliability and validity of the TRI. First, the interrater reliability of five of the six constructs was good. Only the construct coherence was not reliable (Cicchetti et al., 2006). Spilt and Koomen (2009) already suggested that the 5-point scale for coherence may not discriminate well between the narratives. A review of the coherence scale is required.

Second, the expected associations between the TRI and two other measurements of the quality of teacher–student relationships were found, thus confirming the convergent validity of the TRI. The affective scales (anger and positive affect) of the TRI were moderately related to questionnaire reports of closeness and conflict of the STRS in the expected directions (cf. Spilt & Koomen, 2009). In addition, the expected positive association between sensitive practices in the teachers’ narratives and closeness as reported by teachers in a questionnaire was found.

Concerning the observations of teacher–student interactions, teachers’ narratives did, as expected, correlate with observed teacher sensitivity and student emotional security. Sensitive practices and positive affect were positively associated with observed teacher sensitivity and student emotional security. Yet the associations with teacher sensitivity were only found in the free choice task. Possibly, the more structured nature of the second and third task in comparison to the free choice task accounts for these differences in findings. We did not find the expected association between anger in the teachers’ narratives and lower levels of observed teacher sensitivity. Perhaps, negative emotions and challenges of this special education sample may have less influence on special education teachers’ practices because they are more prepared to handle those challenges (cf. Gavish, 2017).

Third, the expected associations between the different scales of the TRI and student behavior were found, confirming the concurrent validity of the TRI. Narrated positive affect positively correlated with students’ classroom participation and prosocial behavior. In contrast, teachers expressed more anger and helplessness toward students who were not cooperative and displayed behavioral problems. This is probably due to the negative teacher–student interactions elicited by negative student behaviors causing tensions and feelings of failure in teachers (Chang, 2009; Stuhlman & Pianta, 2002).

The Three Dimensions of the TRI: Affect, Content, and Process

Similar to previous research, the affective dimension of the TRI (helplessness, anger, and positive affect) was more strongly associated with observed teacher–student interactions (teacher sensitivity and student’s emotional security) and student (mal)adjustment in comparison with the content and process dimensions of the TRI. However, within the affective dimension, feelings of helplessness appeared to play a more prominent role than anger in this special education sample (in contrast to Stuhlman & Pianta, 2002). Perhaps, teachers’ helplessness is a more prominent feeling among teachers of students with emotional and behavioral disorders. Teachers in special education may attribute student problem behavior to stable and unintentional student behavior (e.g., social skill deficits; cf. Schwartz & Davis, 2006), whereas teachers in regular education may be more inclined to attribute student problem behavior to students’ negative intent (Thijs & Koomen, 2009). Consequently, teachers in special education may experience more personal failure and helplessness rather than anger in negative or nonclose teacher–student interactions (cf. Spilt & Koomen, 2009).

As expected, the content dimension of the TRI (sensitive practices) significantly correlated with teachers’ self-reported closeness, observed teacher sensitivity, and student’s emotional security in the free choice task. However, sensitive practices were not related to reports of conflict, teacher–student interactions in the more structured tasks, or student (mal)adjustment. Therefore, the content of teachers’ narratives may be less important than the affective quality of what is narrated (cf. Button et al., 2001; Spilt & Koomen, 2009).

The process dimension of the TRI (coherence and neutralizing negative affect) was more difficult to interpret. First, the construct coherence was not measured reliably. As mentioned above, adjustments to this scale are needed to increase the reliability to validate its value for research and practice. Second, we found limited support for the validity of the construct neutralizing of negative affect. We expected a positive association of neutralizing of negative affect with relational conflict and with student maladjustment on the teacher-report questionnaires because teachers may find it difficult to acknowledge that they are having negative feelings for a student, and negative feelings are more common in conflictual teacher–student relationships or in relationships with disruptive students (Spilt & Koomen, 2009). Unexpectedly, however, a negative correlation with externalizing student problems was found. Perhaps, teachers in special education sample are more aware of the problems of the students (e.g., social skill deficits) in their classroom and are more willing to talk about the conflictual teacher–student interactions and their negative emotions that come along with these (behavioral) problems. In addition, recurrent interpersonal conflicts, caused by a student’s misbehavior, may make it easier for the teachers to acknowledge and express (e.g., recall examples) negative feelings in their narratives (cf. Spilt & Koomen, 2009).

Limitations and Further Research

Several limitations of this study have to be considered. First, the focus on the special education sample of students with mild to severe attachment problems has to be accounted for when interpreting and generalizing the results. Second, the structured observations allowed for a more standardized and accurate assessment of teacher–student interactions. However, the ecological validity of these observations in a one-to-one setting outside the classroom is limited. In addition, the impact of teachers’ negative affect on teacher sensitivity might be masked by social desirability because teachers were aware that they were observed. Unstructured classroom observations (cf. Stuhlman & Pianta, 2002) may decrease social desirability bias and increase ecological validity.

Further research on the validity of the TRI could investigate the predictive validity of the TRI on teacher and student outcomes using a longitudinal study. In addition, it would be interesting to use more objective measurements (e.g., independent observations) of student behavior instead of teacher-perceived reports.

Practical Implications

The TRI can be useful for needs-based assessment in a practice-oriented psychodiagnostic approach or so-called “assessment for intervention.” The TRI can give an impression of difficulties and needs experienced by the teacher in relation to a specific student which facilitates the consultation process with a mentor or school psychologist (Koomen et al., 2006). Narratives are seen as means to facilitate reflection and professional growth offering the possibility of going from implicit beliefs to explicit thoughts, from unawareness and custom to self-knowledge and reflection (Clemente & Ramírez, 2008; Pianta, 1999). Using the TRI, the consultant can help teachers to become aware of their mental representations by helping to put words to implicit feelings and beliefs about their relationship with a specific student and to connect these with their behavior toward the student. Accordingly, the TRI may provide insight in the transactional processes between teacher and student, which may help to prevent vicious circles of negative transactional processes (Pianta, 1999).

The TRI has been used in the Relationship-Focused Reflection Program (RFRP) to target teachers’ mental representations of dyadic relationships using the scores of each scale in a unique relational profile as a starting point for more in-depth reflection (Spilt, Koomen, Thijs, & van der Leij, 2012). Spilt et al. (2012) observed improvements in teachers’ sensitive behavior toward the students for all teachers in the RFRP. More research about the use of relationship-focused narrative interviews in teacher consultation and its effect on teacher and student outcomes is warranted.

Supplemental Material

Appendix_A – Supplemental material for Teacher–Student Relationships in Special Education: The Value of the Teacher Relationship Interview

Supplemental material, Appendix_A for Teacher–Student Relationships in Special Education: The Value of the Teacher Relationship Interview by Anne-Katrien Koenen, Eleonora Vervoort, Karine Verschueren and Jantine L. Spilt in Journal of Psychoeducational Assessment

Supplemental Material

Appendix_B – Supplemental material for Teacher–Student Relationships in Special Education: The Value of the Teacher Relationship Interview

Supplemental material, Appendix_B for Teacher–Student Relationships in Special Education: The Value of the Teacher Relationship Interview by Anne-Katrien Koenen, Eleonora Vervoort, Karine Verschueren and Jantine L. Spilt in Journal of Psychoeducational Assessment

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 study was supported by a grant of the Fund for Scientific Research—Flanders (G.0555.09N) to Karine Verschueren.

ORCID iDs

Anne-Katrien Koenen

Jantine L. Spilt

Supplemental Material

Supplemental material for this article is available online.

References

Borelli

J. L.

West

J. L.

Decoste

Suchman

N. E.

(2012). Emotion language in the parenting interviews of substance-dependent mothers: Associations with reflective functioning, recent substance use, and parenting behavior. Infant Mental Health Journal, 33, 506-519. doi:10.1002/imhj.21340

Bosmans

Kerns

(2015). Attachment in middle childhood: Progress and prospects. New Directions for Child and Adolescent Development, 148, 1-14. doi:10.1002/cad.20100

Bowlby

(1982). Attachment and loss: Vol. 1. New York, NY: Basic Books. (Original work published 1969)

Button

Pianta

R. C.

Marvin

R. S.

(2001). Mothers’ representations of relationships with their children: Relations with parenting behavior, mother characteristics, and child disability status. Social Development, 10, 455-472. doi:10.1111/1467-9507.00175

Chang

(2009). An appraisal perspective of teacher burnout: Examining the emotional work of teachers. Educational Psychology Review, 21, 193-218. doi:10.1007/s10648-009-9106-y

Cicchetti

Bronen

Spencer

Haut

Berg

Oliver

Tyrer

(2006). Rating scales, scales of measurement, issues of reliability: Resolving some critical issues for clinicians and researchers. The Journal of Nervous and Mental Disease, 194, 557-564. doi:10.1097/01.nmd.0000230392.83607.c5

Clemente

Ramírez

(2008). How teachers express their knowledge through narrative. Teaching and Teacher Education, 24, 1244-1258.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

Cornelissen

Verschueren

(2001). Nederlandse vertaling van de Teacher-Rating-Scale-of-School-Adjustment (TSSRA) [Dutch translation of the Teacher-Rating-Scale-of-School-Adjustment]. Leuven: Centrum voor schoolpsychologie, KU Leuven.

10.

Crick

N. R.

Dodge

K. A.

(1994). A review and reformulation of social information-processing mechanisms in children’s social adjustment. Psychological Bulletin, 115, 74-101. doi:10.1037/0033-2909.115.1.74

11.

Doumen

Koomen

Buyse

Wouters

Verschueren

(2012). Teacher and observer views on student–teacher relationships: Convergence across kindergarten and relations with student engagement. Journal of School Psychology, 50, 61-76. doi:10.1016/j.jsp.2011.08.004

12.

Doumen

Verschueren

Buyse

Germeijs

Luyckx

Soenens

(2008). Reciprocal relations between teacher–child conflict and aggressive behavior in kindergarten: A three-wave longitudinal study. Journal of Clinical Child & Adolescent Psychology, 37, 588-599. doi:10.1080/15374410802148079

13.

Gavish

(2017). Special education trainee teachers’ perceptions of their professional world: Motives, roles, and expectations from teacher training. Teachers and Teaching, 23, 153-170. doi:10.1080/13540602.2016.1204285

14.

George

Solomon

(1996). Representational models of relationships: Links between caregiving and attachment. Infant Mental Health Journal, 17, 198-216. doi:10.1002/(SICI)1097-0355(199623)17:3<198:: AID-IMHJ2>3.0.CO;2-L

15.

Goodman

(1997). The Strengths and Difficulties Questionnaire: A research note. Journal of Child Psychology and Psychiatry, 38, 581-586. doi:10.1111/j.1469-7610.1997.tb01545.x

16.

Goodman

(2001). Psychometric properties of the Strengths and Difficulties Questionnaire. Journal of the American Academy of Child and Adolescent Psychiatry, 40, 1337-1345. doi:10.1097/00004583-200111000-00015

17.

Hamre

B. K.

Pianta

R. C.

(2001). Early teacher-child relationships and the trajectory of children’s school outcomes through eighth grade. Child Development, 72, 625-638.

18.

Koomen

H. M. Y.

Lont

T. A. E.

(2004). Teacher relationship interview: Qualitative coding manual (Ongepubliceerde Nederlandse vertaling). Amsterdam, The Netherlands: Universiteit Amsterdam.

19.

Koomen

H. M. Y.

van Leeuwen

M. G. P.

van der Leij

(2004). Does well-being contribute to performance? Emotional security, teacher support, and learning behaviour in kindergarten. Infant and Child Development, 13, 253-275.

20.

Koomen

H. M. Y.

Verschueren

Pianta

R. C.

(2007). Leerling Leerkracht Relatie Vragenlijst-handleiding [Manual of Student Teacher Relationship Scale]. Houten, The Netherlands: Bohn Stafleu van Loghum.

21.

Koomen

H. M. Y.

Verschueren

Thijs

J. T.

(2006). Assessing aspects of the teacher-child relationship: A critical ingredient of a practice-oriented psycho-diagnostic approach. Educational and Child Psychology, 23, 50-60.

22.

Koomen

H. M. Y.

Verschueren

van Schooten

Jak

Pianta

R. C.

(2012). Validating the Student-Teacher Relationship Scale: Testing factor structure and measurement invariance across child gender and age in a Dutch sample. Journal of School Psychology, 50, 215-234. doi:10.1016/j.jsp.2011.09.001

23.

La Paro

K. M.

Pianta

R. C.

Stuhlman

M. W

. (2002). Classroom Assessment Scoring System (CLASS). Charlottesville: University of Virginia.

24.

Ladd

G. W.

(1992). The Teacher Rating Scale of School Adjustment. Champaign: University of Illinois at Urbana–Champaign.

25.

Ladd

G. W.

Burgess

K. B.

(2001). Do relational risks and protective factors moderate the linkages between childhood aggression and early psychological and school adjustment? Child Development, 72, 1579-1601. doi:10.1111/1467-8624.00366

26.

Main

Kaplan

Cassidy

(1985). Security in infancy, childhood, and adulthood: A move to the level of representation. Monographs of the Society for Research in Child Development, 50, 66-104. doi:10.2307/3333827

27.

McGrath

K. F.

Van Bergen

(2015). Who, when, why and to what end? Students at risk of negative student–teacher relationships and their outcomes. Educational Research Review, 14, 1-17. doi:10.1016/j.edurev.2014.12.001

28.

Pianta

R. C.

(1999). Enhancing relationships between children and teachers. Washington, DC: American Psychological Association.

29.

Pianta

R. C.

(2001). Student-teacher relationship scale: Professional manual. Lutz, FL: Psychological Assessment Resources.

30.

Pianta

R. C.

Hamre

Stuhlman

(2003). Relationships between teachers and children. In Reynolds

W. M.

Miller

G. E.

(Eds.), Handbook of psychology (pp. 199-234). New York, NY: John Wiley. doi:10.1002/0471264385.wei0710.

31.

Schwartz

Davis

A. S.

(2006). Reactive attachment disorder: Implications for school readiness and school functioning. Psychology in the Schools, 43, 471-479. doi:10.1002/pits.20161

32.

Spilt

J. L.

Koomen

H. M. Y.

(2009). Widening the view on teacher–child relationships: Teachers’ narratives concerning disruptive versus non-disruptive children. School Psychology Review, 38, 86-101.

33.

Spilt

J. L.

Koomen

H. M. Y.

Thijs

van der Leij

(2012). Supporting teachers’ relationships with disruptive children: The potential of relationship-focused reflection. Attachment & Human Development, 14, 305-318. doi:10.1080/14616734.2012.672286

34.

Stuhlman

M. W.

Pianta

R. C.

(2002). Teachers’ narratives about their relationships with children: Associations with behavior in classrooms. School Psychology Review, 31, 148-163.

35.

Thijs

J. T.

Koomen

H. M. Y.

(2008). Task-related interactions between kindergarten children and their teachers: The role of emotional security. Infant and Child Development, 17, 181-197. doi:10.1002/icd.552

36.

Thijs

J. T.

Koomen

H. M. Y.

(2009). Toward a further understanding of teachers’ reports of early teacher–child relationships: Examining the roles of behavior appraisals and attributions. Early Childhood Research Quarterly, 24, 186-197. doi:10.1016/j.ecresq.2009.03.001

37.

Van Leeuwen

Meerschaert

Bosmans

De Medts

Braet

. (2006). The Strengths and Difficulties Questionnaire in a community sample of young children in Flanders. European Journal of Psychological Assessment, 22, 189-197. doi:10.1027/1015-5759.22.3.189

38.

Verschueren

Koomen

H. M. Y.

(2012). Teacher–child relationships from an attachment perspective. Attachment & Human Development, 14, 205-211. doi:10.1080/14616734.2012.672260

39.

Verschueren

Van de Water

Buyse

Doumen

(2006). Adaptation of the classroom assessment scoring system-kindergarten to measure the individual teacher-child relationship [in Dutch]. Leuven: Centrum voor Schoolpsychologie, KU Leuven.

40.

Vervoort

Verschueren

Bosmans

. (2013). Teacher-child relationships and indiscriminate fr iendliness in early elementary school children: Measurement, challenges, and protective effects [Unpublished Doctoral Thesis]. KU Leuven, Faculteit Psychologie en Pedagogische Wetenschappen, School- en Ontwikkelingspsychologie van Kind en Adolescent, Leuven.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.11 MB