This study examined factors contributing to shared and unique perspectives among fathers’, mothers’, and teachers’ ratings of school-aged children’s psychosocial adjustment among military families. Utilizing baseline data from three randomized controlled trials of a preventive parenting program (N = 870, 51.7% girls; Mage = 8.13; 12.7% fathers and 12.8% mothers identified as people of color), we first described the pattern of informant (dis)agreement on children’s psychosocial adjustment rated using the Behavior Assessment System for Children (BASC). Using trifactor models, this study explored factors associated with informants’ shared and unique perspectives. Moderate-to-strong correlations between similar informants and small-to-moderate correlations between distinct informants were observed. Parental efficacy, parental distress, and couple relationships were related to parental shared and unique perspectives of children’s internalizing problems, externalizing problems, and adaptive functioning. These results emphasize the complexity of accessing child psychosocial adjustment, and highlights the need for multi-informant assessment in future research and practice.
In both research and clinical settings, optimal assessment of children’s psychosocial adjustment utilizes multiple data sources (e.g., rating scales from multiple informants, interviews, behavioral observations) that are tailored to fit children’s developmental stage and the resources available. Clinically validated and normed rating scales, such as the Behavior Assessment System for Children (BASC; [33, 34], are typically simple to administer, time-efficient, can be administered across settings, can reflect the level of impairment, and are useful in monitoring children’s response to intervention [1, 28]; Kamphaus et al. 2004). Generally, assessment in early childhood largely depends on feedback from parents and teachers, with self-report measures becoming increasingly important as children progress to middle childhood and adolescence [35]. This study aimed to examine patterns of agreement and discrepancy among father, mother, and teacher ratings of school-aged children’s psychosocial adjustment, and to identify factors that associate with shared and unique perspectives across informants.
From an ecological perspective [7], both family and school represent critical microsystems in children’s development, and incorporating information from both contexts is essential for understanding children’s behavior. However, it is widely acknowledged that informant discrepancies often exist across informants [13]. Achenbach and colleagues [2] first introduced the idea of situational specificity after synthesizing robust evidence from 119 studies, describing how children’s behaviors vary in different contexts (e.g., home vs. school), leading to different informants observing and reporting different aspects of children’s behaviors. Since then, numerous studies and reviews have consistently demonstrated that similar informants (e.g., fathers vs. mothers) shared moderate correspondence (e.g., De Los Reyes et al. [14] while distinct informants (e.g., parents vs. teachers) shared low-to-moderate correspondence (e.g., Rescorla et al. [32]. Moreover, evidence suggests that informant discrepancy also varies across different dimensions (e.g., De Los Reyes et al. [14, 16], with higher correspondence observed on more overtly observable behaviors (e.g., externalizing problems) and lower correspondence observed on less salient emotions and behaviors (e.g., internalizing problems), yet few studies have explored informant discrepancy in ratings of adaptive functioning.
Historically, discrepancies in children’s psychosocial adjustment from multiple informants were often dismissed as measurement “error” or “noise” while only shared perspectives (e.g., shared variance) were considered to be “true” and “valid” indicators of children’s psychosocial adjustment [18]. However, a significant paradigm shift over the past two decades has resulted in researchers becoming more interested in the meaningful situational information that can be garnered from informant discrepancies, potentially providing insights into the multifaceted nature of mental health (e.g., De Los Reyes et al. [13, 15]. The trifactor model (TFM; Bauer et al. [5] is a psychometric model developed specifically to examine sources of differences among raters. The model separates the variance in each item into three sources: variance shared across informants’ perspectives (consensus factors or C), variance unique to each informant’s perspective (perspective factors or P), and variance that is specific to the item (specific factors or Si). Therefore, if one had reports from a five-item instrument administered to a parent and a teacher, there would be one consensus factor, two perspective factors, and five specific factors. The consensus factor represents the shared perspective or agreement among informants, whereas the perspective factors highlight the unique perspective each informant has. Such a model is particularly useful in enhancing our understanding of how children’s behavior varies across different settings and how these behaviors are interpreted differently depending on the context.
Previous research indicates that parent-teacher discrepancies on school-aged children’s psychosocial adjustment are influenced by individual and family-level variables. For example, parents with more ineffective parenting behaviors, higher parenting stress, and/or more negative parental attitudes tended to perceive their children as having more mental health problems compared to other caregivers [10, 27, 38], teachers [30], and independent observers [29]. Parents with higher distress levels (e.g., higher depression, anxiety, or posttraumatic stress) were more likely to perceive their children as experiencing more internalizing problems compared to other caregivers (e.g., Chesmore et al. [10, 12]) and teachers (e.g., Youngstrom et al. 2000). Stressful family contexts have been shown to influence parents’ perceptions of children’s psychosocial adjustment. For example, mothers with lower socioeconomic status (SES) were more likely to report greater mental health problems in their children compared to teachers [36]. Findings regarding the impact of race and ethnicity on informant discrepancy are not entirely consistent. For example, Takeda and colleagues [37] found that parents of color tend to perceive their children as having more severe mental health problems compared to teachers, while Kuhfeld and colleagues [25] found that teachers were more likely to perceive Black students as having lower self-control. However, it is noteworthy that existing literature rarely incorporates teacher-level or school-based variables, potentially due to lack of data.
Military families provide a special opportunity to examine informant discrepancy within a stressful family context. Following the attacks of 9/11/2001, more than two million US service members were deployed to the wars in Iraq and Afghanistan, half of whom were National Guard or Reserves (Reserve Component) service members [23, 31]. An unprecedented number of National Guard personnel were deployed for lengthy periods of time (more than a year), almost half of whom were parents to dependent children (IOM, 2010). A significant body of research has documented the impact of this stressful context of parental deployment to war, long separations, and parents’ reintegration following deployment, on children and families (see, e.g., NASEM, 2019; Gewirtz & Simenec, 2022). However, just one paper has examined informant discrepancies in this context: Chesmore and colleagues [10] found that military parents with higher posttraumatic stress symptoms were more likely to perceive their children as having more internalizing problems compared to their partners.
The Current Study
While existing research acknowledges that informant discrepancy reflects meaningful situational information rather than purely measurement bias, less work has been done to identify factors that influence shared and unique perspectives in school-age children’s psychosocial adjustment, including both problems and adaptive functioning. Most studies only focus on mother-father or mother-teacher discrepancy, leaving a gap in understanding what might be associated with the unique perspectives of fathers, mothers, and teachers. This is particularly important for school-aged children, as parents and teachers represent the two most important environments they interact with: home and school. Additionally, most studies only focus on children’s internalizing and externalizing problems, overlooking children’s strengths and adaptive functioning, such as social skills, adaptability, communication skills, etc. Early studies also focused on demographic variables (e.g., age, gender, race/ethnicity, SES) or parental depression, without considering a broader range of potential influencing factors. This highlights a critical need for a more comprehensive evaluation of factors that influence the unique perspectives of fathers’, mothers’, and teachers’ ratings of mental health problems and positive functioning among school-aged children.
Utilizing baseline data from three randomized controlled trials of a behavioral parenting program for military families, this study first described correlations between fathers’, mothers’, and teachers’ ratings on school-aged children’s psychosocial adjustment, including internalizing and externalizing problems as well as adaptive functioning (Aim 1). We hypothesized higher agreement between similar informants (e.g., mother-father) than between distinct informants (e.g., parent-teacher). This study then identified factors associated with fathers’, mothers’, and teachers’ shared and unique perspectives of school-aged children’s psychosocial adjustment (Aim 2). As discussed in Bauer (2013) and demonstrated in Kuhfeld et al. [25], the TFM was selected as it allows for external factors to be included as predictor variables, enhancing our understanding of how those factors affect the consensus and unique perspectives provided by each informant. In this study, those external factors included parental efficacy, parental posttraumatic stress symptoms, couple functioning, parental race, marital status, household income, and length of parental deployment. As previous evidence was not always consistent and mostly focused on two informants instead of three, our analysis was primarily exploratory, aiming to identify which variables may differentially predict shared and unique perspectives. In this model (see Fig. 1 as an example for internalizing problems), discrepancies among informants are primarily captured by the perspective factors, and a significant predictor of a perspective factor can be interpreted as a predictor of that informant’s disagreement relative to the other informants.
Fig. 1
Path Model of the Trifactor Model. Demonstrating the Analysis of Internalizing Problems in Children as Reported by Mothers, Fathers, and Teachers. M = Mother Report; F = Father Report; T = Teacher Report; Anx = Anxiety; Dep = Depression; Som = Somatic Symptoms; With = Withdrawal; S = Specific Factors
This study utilized baseline data from three randomized control trials of a behavioral parenting program for military families in the United States. Eligible families were those with at least one child aged 4–12 years and at least one parent who had been deployed in support of post-9/11 military operations. Detailed information for the three trials, conducted between 2011 and 2022, can be found elsewhere (Gewirtz et al. 2018; Gewirtz et al. 2024; DeGarmo & Gewirtz, 2019). A total of 870 families, encompassing 689 fathers, 813 mothers, and 870 children (51.7% girls; Mage = 8.13 years, range 4–12), were included in this study. Most parents were White/European American (87.2% mothers and 87.3% fathers), followed by Black/African American (4.9% mothers and 6.7% fathers), and others (7.9% mothers and 6.1% fathers). Additionally, 7.6% of mothers and 6.9% of fathers identified as Latinx. Detailed demographic information for the overall sample and the specific breakdown of each dataset can be found in Table 1.
Table 1
Summary of participant demographic information for the total sample, and from each of the three randomized controlled trials
Full Sample
Trial 1
Trial 2
Trial 3
F value/Chi square
Family N
870
336
244
290
Child Age M (SD)
8.13 (2.35)
8.39 (2.52)
7.67 (2.29)
8.22 (2.15)
6.95***
Girl n (%)
450 (51.7%)
180 (53.6%)
120 (49.2%)
150 (51.7%)
1.09
Mother’s Ethnicity (Latinx)
61 (7.6%)
10 (3.2%)
10 (4.6%)
41 (15.1%)
33.13***
Mother’s Race (White)
694 (87.2%)
286 (94.4%)
193 (88.9%)
215 (77.9%)
35.98***
Mother’s Race (Black)
39 (4.9%)
5 (1.7%)
10 (4.6%)
24 (8.7%)
15.44***
Mother’s Race (Other)
63 (7.9%)
12 (4.0%)
14 (6.5%)
37 (13.4%)
18.56***
Father’s Ethnicity (Latinx)
46 (6.9%)
10 (3.4%)
9 (4.9%)
27 (14.2%)
22.76***
Father’s Race (White)
577 (87.3%)
254 (90.1%)
165 (88.2%)
158 (82.3%)
6.44*
Father’s Race (Black)
44 (6.7%)
13 (4.6%)
14 (7.5%)
17 (8.9%)
3.60
Father’s Race (Other)
40 (6.1%)
15 (5.3%)
8 (4.3%)
17 (8.9%)
3.95
Mother Age
35.63 (5.64)
35.69 (5.89)
35.92 (5.67)
35.34 (5.34)
0.68
Father Age
37.12 (6.07)
37.76 (6.54)
37.57 (5.81)
35.8 (5.39)
7.041***
Married or Cohabiting n (%)
751 (86.6%)
281 (83.9%)
200 (82.6%)
270 (93.1%)
15.99***
Mother Ever Been Deployed
165 (20.3%)
56 (17.9%)
60 (27.4%)
49 (17.5%)
9.29**
Father Every Been Deployed
665 (96.7%)
282 (95.9%)
180 (94.2%)
203 (100%)
10.97 **
Mother Deployment Length
2.86 (7.23)
2.37 (6.27)
3.89 (7.68)
2.59 (7.78)
3.115*
Father Deployment Length
19.03 (13.17)
19.08 (11.46)
18.66 (12.54)
19.31 (15.84)
0.123
Household Income
$51,000–80,000
$51,000–80,000
$51,000–80,000
$51,000–80,000
*p <.05. **p <.01. ***p <.001
Procedure
Participants were recruited through various methods, including post-deployment reintegration events, targeted mailings at a Veterans Affairs Medical Center, social and traditional media marketing, and via word of mouth. After expressing interest, participants underwent an online screening to determine eligibility. A total of 870 families gave informed consent, with the target child giving assent in the presence of the parent(s) and subsequently completed baseline online and in-home assessments across the three trials. Parents also gave permission for research staff to contact the target child’s teacher, who was then contacted via email and asked to complete questionnaires online. Each parent who completed the online survey received a $25 gift card, and each family who completed the in-home assessment received a $50 gift card. Children who completed their self-report questionnaires received a small gift worth $5-$10. Teachers who completed the questionnaire received a $10 gift card.
Measures
Child Psychosocial Adjustment
Children’s psychosocial adjustment was measured using the Behavioral Assessment Scale for Children, Parent and Teacher Rating Scales (BASC-2 and BASC-3; [33, 34]. The BASC-2 was utilized in the first two trials and the BASC-3 was utilized in the third trial. T-scores across BASC-2 and BASC-3 were intentionally designed to be comparable to ensure the consistency of the test across time [34]. Age and gender-normed T-scores for each subscale were used in the current study. Three dimensions of children’s psychosocial adjustment were included in the current study: internalizing problems, externalizing problems, and adaptive functioning. Internalizing problems included four subscales: anxiety (e.g., “Worries about things that cannot be changed.”), depression (e.g., “Cries easily.”), somatization (e.g., “Expresses fear of getting sick.”), and withdrawal (e.g., “Avoids other children.”). Externalizing problems included hyperactivity (e.g., “Acts without thinking.”), aggression (e.g., “Threatens to hurt others”), and attention problems (e.g., “Has a short attention span.”). Adaptive functioning included adaptability (e.g., “Recovers quickly after a setback.”), social skills (e.g., “Makes others feel welcome.”), leadership (e.g., “Is a self-starter.”), activities of daily living (e.g., “Organizes chores or other tasks well.”), and functional communication (e.g., “Responds appropriately when asked a question.”). Those subscales were selected based on content, consistency across informants, and psychometric analyses. Fathers, mothers, and teachers rated children’s behaviors on a four-point Likert scale ranging from 1 (never) to 4 (almost always). Higher scores represent more severe symptoms in internalizing and externalizing subscales and higher functioning in the adaptive subscales. These assessment scales are widely used and well-established with both theoretical and psychometric support [8, 33, 34]. Cronbach’s alpha showed acceptable to excellent internal consistency for all subscales in our sample, and the average alpha by subscale for parents and teacher reports are presented in Table 2.
Table 2
Mean scores, standard deviations, mean differences, and inter-informant correlations on child internalizing problems, externalizing problems, and adaptive functioning as reported by fathers, mothers, and teachers
Mother
Mean (SD)
Father
Mean (SD)
Teacher
Mean (SD)
Parent alpha
Teacher alpha
M vs. F
t test
M vs. T
t test
F vs. T
t test
rM, F
rM, T
rF, T
Internalizing
Problems
Anxiety
53.71 (11.74)
52.59 (10.98)
50.29 (10.66)
0.86
0.77
2.62**
6.15***
3.27***
0.40***
0.20***
0.20***
Depression
53.05 (11.46)
51.41 (10)
50.75 (11.97)
0.84
0.89
3.66***
4.57***
0.11
0.60***
0.35***
0.37***
Withdrawal
48.1 (11.1)
47.92 (10.78)
49.7 (10.81)
0.80
0.84
−0.49
0.28
−0.01
0.52***
0.28***
0.25***
Somatization
50.55 (10.78)
50.81 (9.77)
50.6 (10.42)
0.83
0.81
0.83
−3.18**
−3.86***
0.56***
0.33***
0.28***
Externalizing
Problems
Aggression
51.41 (10.12)
51.72 (10.44)
49.57 (9.84)
0.79
0.89
−1.12
4.27***
3.27***
0.58***
0.33***
0.25***
Attention
53.88 (9.29)
54.81 (8.9)
50.48 (9.91)
0.85
0.95
−3.51***
8.22***
9.57***
0.57***
0.42***
0.42***
Hyperactivity
53.77 (11.16)
54.42 (10.69)
51.84 (11.11)
0.82
0.93
−2.40*
4.15***
4.39***
0.60***
0.36***
0.34***
Adaptive
Function
Adaptability
46.69 (9.12)
45.74 (8.53)
51.15 (9.61)
0.79
0.86
3.63***
−10.54***
−10.88***
0.42***
0.26***
0.17***
Communication
48.64 (10.11)
46.5 (10.15)
50.25 (9.25)
0.81
0.87
6.19***
−4.22***
−7.97***
0.57***
0.41***
0.38***
Leadership
50.27 (9.16)
49.52 (8.59)
52.79 (9.13)
0.82
0.87
1.78
−5.91***
−5.53***
0.39***
0.35***
0.25***
Social skills
49.55 (9.3)
47.17 (9.03)
52.86 (10.64)
0.87
0.92
6.20***
−7.36***
−10.46***
0.34***
0.22***
0.14**
ADL/SS
46.02 (9.71)
44.97 (9.76)
50.79 (9.14)
0.71
0.91
0.67
−10.41***
−11.05***
0.55***
0.36***
0.35***
SD = Standard deviation; M = Mother, F = Father, T = Teacher; r = Correlation
ADL: Activities of Daily Living; SS: Study Skills; *p <.05. **p <.01. ***p <.001
Parental Efficacy
In all three trials, both maternal and paternal efficacy was assessed using the Parental Locus of Control–Short Form Revised (PLOC-SFR; Hassall et al. [19]. The PLOC-SFR consists of 24 self-reported items across four domains: parental efficacy (e.g., “I am often able to predict my child’s behavior in situations”), parental responsibility (e.g., “When my child is well-behaved, it is because he/she is responding to my efforts”), child control of parents’ life (e.g., “I feel like what happens in my life is mostly determined by my child”), and parental control of child’s behavior (e.g., “I always feel in control when it comes to my child”). Parents rated items on a five-point Likert-style scale, ranging from one (strongly agree) to five (strongly disagree). The mean score of PLOC-SFR for each parent was calculated, with higher scores indicating higher parental efficacy. The measure had adequate reliability across all trials at baseline (maternal α = 0.78, 0.80, 0.76 and paternal α = 0.78, 0.76, 0.78, for the three trials respectively).
Posttraumatic Stress Symptoms
The Post-Traumatic Stress Disorder Checklist (PCL; Weathers et al. [39] is a standardized, validated 17-item self-report assessment tool. In all three trials, parents who were previously deployed completed the PCL-Military while nondeployed parents used the PCL-Civilian. Both PCL-Military and PCL-Civilian evaluated PTSD symptoms (intrusion, avoidance, and hyperarousal) according to DSM-IV criteria (American Psychiatric Association, 2000) and showed solid psychometric properties (Wilkins et al. 2011). Parents rated the degree to which they were distressed by each symptom over the past month on a five-point Likert-style scale, ranging from one (not at all) to five (extremely), with higher average scores indicating a greater presence of PTSD symptoms. Both fathers (α = 0.95, 0.97, 0.94) and mothers (α = 0.91, 0.92, 0.95) exhibited high internal consistency in their responses across all trials.
Couple Relationship Quality
Both fathers and mothers in all three study trials completed the seven-item version of the Dyadic Adjustment Scale (DAS-7; Hunsley et al. [22]. The DAS-7 assesses agreement (e.g., the extent of agreement about the amount of time spent together), dyadic cohesion (e.g., how often you calmly discuss something together), and overall relationship satisfaction using a six-point Likert-style scale, ranging from zero (always disagree/never/extremely unhappy) to six (always agree/more often/perfect). The total score for the scale was calculated, with higher scores indicating a higher level of positive relationship adjustment. The DAS-7 has been well-validated [22]. Within this study, responses of both fathers (α = 0.86, 0.84, 0.81) and mothers (α = 0.87, 0.85, 0.84) exhibited high internal consistency across all trials.
Covariates
As child race was not collected for all children, parental race and ethnicity were coded into three binary variables for each parent, Black (0 = non-Black, 1 = Black), Latinx (0 = not Latinx, 1 = Latinx), and Other. In the ‘Other’ category, we combined Asian, Pacific Islander, Native American, and Multiracial groups due to each group’s very small sample size. Family annual income was coded from one to seven (1 = Less than $25,000; 2 =$26,000 to $50,000; 3 = $51,000 to $80,000; 4 = $81,000 to $100,000, 5 = $101,000 to $120,000, 6 = $121,000 to $150,000, 7 = $151,000 or more). Marital status was coded as binary for each family (0 = divorced, separated, widowed, or never married, 1 = married or domestic partnership). Total deployment length was reported as the total numbers of months in two of the trials. In the third trial, deployment months were categorized into six-month intervals (0 = not been deployed, 1 = 6 months or less, 2 = 7–12 months, … 7 = 37 or more months), and the median number of months from within each range was used for analyses.
Analytic Plan
Data analyses were conducted in the following stages using Mplus 8.8 (Muthen & Muthen, 1998–2021).
Step 1: Unconditional Trifactor Model/TFM
In the initial step, an unconditional TFM was fit to T-scores from the BASC Parent and Teacher Rating Scales for each of the three constructs. This initial model only had constraints that were necessary for model estimation and to freely estimate the mean and variance for the teacher’s perspective factor. These constraints included fixing the first loading on the consensus factor to 1.0, constraining the perspective factor loadings for one item between the teacher and one other reporter, and constraining one item’s intercept for the teacher and one other reporter. The means and variances for the consensus factor, parent perspective factors, and specific factors were fixed to 0 and 1.0, respectively. These constraints allowed the teacher’s perspective factor mean and variance to be freely estimated. The number of specific factors varied based on the number of subscale scores for each construct, with the adaptive functioning construct having the largest number of specific factors. The specific factors, though not of primary interest, are important to include to avoid inflating consensus factor loadings (Marsh, 1993). Modifications were implemented based on both theoretical considerations (e.g., more overlapping contexts between mother-father compared to parent-teacher) and empirical indicators (e.g., modification indices or estimation issues) to improve model fit or achieve proper solutions. All modifications and their justifications are described in the Supplementary Materials.
Step 2: Measurement Invariance Assessment
After ensuring that it was possible to fit an unconstrained TFM to each construct, measurement invariance was examined. Following the recommendation of Bauer et al. [5], a series of models were executed to assess measurement invariance across reporters, ensuring that items and T-scores were comparable. It is noteworthy that although Bauer et al. [5] recommended that reporters of similar types, such as parents, should have invariant loadings on the common and perspective factors while reporters that were different, such as teachers, did not necessarily need to have invariant loadings, we chose to be conservative and empirically tested for invariance across all reporters. These models progressed from the least constrained to the most constrained, which included the Configural Invariance Model, allowing differences in loadings and intercepts across reporters beyond identification constraints, the Metric Invariance Model, equating loadings for shared items across reporters for both the consensus factor and perspective factors, and the Scalar Invariance Model, setting item intercepts as equal for shared items across reporters. Invariance was assessed through likelihood ratio tests, employing the model Chi-square statistic, which compared the fit of more constrained models (e.g., metric or scalar invariant models) to less constrained models. A non-significant hypothesis test indicates that the more constrained model fits the data as well as the less constrained model. Change in CFI > 0.01, change in RMSEA > 0.015, and change in SRMR > 0.030 for metric invariance and change in SRMR > 0.015 for strict invariance were also considered as being indictive of measurement noninvariance [9, 11].
If a model did not meet the criteria for a certain level of invariance, partial invariance models were fit by making model modifications in which corresponding parameters (e.g., loadings or intercepts) were freed until either the likelihood ratio test became non-significant, or until changes in the other fit indices (CFI, RMSEA, SRMR) indicated that the models fit equally well. As recommended in the SEM literature (e.g., Kline [24], Brown, 2015), modifications were both data-driven (i.e., using modification indices) and informed by theory so that modifications were only made if there was a theoretical argument that could explain the source of noninvariance.
Step 3: Conditional Trifactor Model
Following the establishment of measurement invariance, predictors were introduced for both the common and perspective factors within each of the three constructs to form a conditional TFM model.
Results
Preliminary Analyses
The means, standard deviations, and bivariate correlations among mothers’, fathers’, and teachers’ ratings of psychosocial adjustment can be found in Table 2. Missing data analysis revealed 18% missing values. Families with complete data were more likely to be married, have female children, older children, longer father deployment length, higher household income, higher maternal and paternal perceived couple relationship satisfaction, lower paternal distress, and lower child’s functional communication (a subscale of the BASC adaptive skills scale) reported by father, mother, and teacher. Missing data did not vary based on children’s race, maternal deployment lengths, parental efficacy, maternal distress, and children’s mental health subscales. Full Information Maximum Likelihood (FIML) was used to handle missing data.
Descriptive statistics, inter-informant correlations, and paired t-tests among all BASC subscales are presented in Table 2. The results showed the sample was generally representative of the children’s age and gender-matched peers as the mean scores of all problem scales and adaptive functioning scales were within one standard deviation of the mean (T = 50, SD = 10). Paired t-tests showed that compared to fathers, mothers perceived their children as having higher depression symptoms, higher adaptability, functional communication, social skills, and activities of daily living, while fathers perceived their children to have more attention challenges. Compared to teachers, parents tended to perceive their children to have more symptoms and less adaptability on almost all subscales other than withdrawal and somatization. Teachers tended to perceive children as experiencing fewer somatization symptoms compared to parents. On average, the effect sizes of the correlations between fathers’ and mothers’ ratings were medium-to-high for internalizing problems (r =.52), externalizing problems (r =.58), adaptive functioning (r =.45), and all mental health domains (r =.51). The effect sizes of the correlations between parents’ and teachers’ ratings were small-to-moderate on internalizing problems (father r =.28; mother r =.32), externalizing problems (father r =.28; mother r =.29), adaptive functioning (father r =.34; mother r =.37), and all mental health domains (father r =.26; mother r =.32).
Trifactor Models
The results from trifactor models are organized by construct, beginning with the results from the measurement invariance tests and followed by analyses with the predictors. In the trifactor model, the residual variance remaining for an item after partitioning the variance into three factors is often quite small and indistinguishable from zero, which can cause negative variance estimates. To avoid these estimation problems, those residual variances were fixed to small values and the models re-estimated.
Measurement Invariance
Measurement invariance was assessed for the three reporter groups within the trifactor models. In sum, the measurement invariance models suggested that mother and father reports were invariant for loadings, with non-invariant loadings being associated with the teacher-reported subscales for externalizing and adaptive functioning. The pattern of non-invariance in the intercepts was more variable as to which reporter had a different parameter value. Overall, the final models had adequate fit.
Predicting Factors in TFM
Models were fit in which the consensus and perspective factors from the TFM were regressed onto relevant predictors. All standardized estimates are reported in Table 3.
Table 3
Results of the trifactor models assessing factors associated with shared and unique perspectives
Internalizing Problems
Externalizing Problems
Adaptive Functioning
β
SE
t
p
β
SE
t
p
β
SE
t
p
Shared Perspective
Mother’s Ethnicity (Latinx)
−0.07
0.07
−1.07
0.285
−0.08
0.06
−1.38
0.168
0.03
0.08
0.40
0.690
Mother’s Race (Black)
−0.01
0.07
−0.13
0.896
0.03
0.07
0.47
0.640
0.10
0.08
1.17
0.243
Mother’s Race (Other)
−0.02
0.07
−0.23
0.820
0.01
0.06
0.23
0.815
0.14
0.07
1.84
0.065
Father’s Ethnicity (Latinx)
−0.02
0.07
−0.37
0.713
0.09
0.06
1.57
0.116
0.14
0.07
1.89
0.059
Father’s Race (Black)
0.17
0.08
2.24
0.025*
0.09
0.07
1.34
0.181
−0.07
0.08
−0.87
0.386
Father’s Race (Other)
−0.06
0.07
−0.93
0.351
−0.09
0.06
−1.60
0.109
−0.08
0.07
−1.13
0.259
Marital Status
0.03
0.11
0.26
0.796
−0.15
0.13
−1.17
0.243
0.19
0.08
2.29
0.022*
Family Income
−0.08
0.11
−0.78
0.437
−0.18
0.14
−1.31
0.191
0.16
0.07
2.22
0.026*
Father Deployment
0.06
0.06
0.85
0.393
0.11
0.06
1.90
0.058
−0.10
0.07
−1.40
0.162
Mother Deployment
−0.03
0.06
−0.54
0.593
−0.04
0.06
−0.69
0.492
0.04
0.07
0.52
0.603
Father DAS
−0.02
0.07
−0.32
0.751
−0.06
0.07
−0.91
0.365
0.00
0.08
−0.04
0.967
Mother DAS
−0.16
0.07
−2.30
0.022*
−0.05
0.07
−0.78
0.437
−0.02
0.08
−0.24
0.810
Father PLOC
−0.15
0.07
−2.18
0.029*
−0.13
0.06
−2.19
0.028*
0.08
0.07
1.15
0.252
Mother PLOC
−0.27
0.06
−4.37
< 0.001***
−0.32
0.06
−4.97
< 0.001***
0.20
0.08
2.44
0.015*
Father PCL
−0.13
0.07
−1.94
0.053
0.06
0.06
1.02
0.309
−0.13
0.08
−1.76
0.079
Mother PCL
0.07
0.07
1.12
0.262
0.06
0.06
0.99
0.323
−0.23
0.07
−320
0.001**
Father Perspective
Father’s Ethnicity (Latinx)
0.07
0.07
1.02
0.310
−0.02
0.07
−0.30
0.768
−0.07
0.05
−1.45
0.148
Father’s Race (Black)
−0.18
0.07
−2.38
0.017*
−0.22
0.08
−2.72
0.007**
0.05
0.05
1.18
0.238
Father’s Race (Other)
−0.08
0.07
−1.15
0.250
−0.03
0.07
−0.34
0.736
0.04
0.04
0.86
0.388
Marital Status
−0.03
0.11
−0.26
0.793
0.19
0.16
1.18
0.239
−0.09
0.05
−1.92
0.055
Family Income
0.17
0.11
1.48
0.139
0.14
0.18
0.76
0.448
0.08
0.05
1.63
0.103
Father Deployment
−0.11
0.07
−1.72
0.085
−0.06
0.07
−0.93
0.353
−0.02
0.04
−0.38
0.703
Father DAS
0.03
0.07
0.45
0.651
−0.01
0.08
−0.11
0.915
0.15
0.05
3.20
0.001**
Father PLOC
−0.12
0.07
−1.71
0.088
−0.29
0.07
−4.01
< 0.001***
0.42
0.04
10.25
< 0.001***
Father PCL
0.33
0.07
4.94
< 0.001***
0.09
0.08
1.23
0.218
0.07
0.05
1.48
0.140
Mother Perspective
Mother’s Ethnicity (Latinx)
0.01
0.07
0.18
0.857
0.07
0.08
0.87
0.387
−0.01
0.05
−0.21
0.836
Mother’s Race (Black)
−0.11
0.08
−1.45
0.148
−0.02
0.08
−0.22
0.824
−0.03
0.05
−0.74
0.461
Mother’s Race (Other)
−0.07
0.07
−0.90
0.368
−0.05
0.08
−0.69
0.489
0.09
0.05
1.87
0.061
Marital Status
−0.02
0.12
−0.16
0.872
0.14
0.17
0.85
0.394
−0.05
0.05
−1.03
0.305
Family Income
0.06
0.12
0.48
0.632
0.12
0.19
0.64
0.522
−0.03
0.05
−0.61
0.542
Mother Deployment
0.06
0.07
0.87
0.383
0.03
0.08
0.34
0.733
−0.02
0.05
−0.43
0.668
Mother DAS
−0.02
0.08
−0.31
0.759
−0.08
0.08
−0.96
0.336
0.23
0.05
5.12
< 0.001***
Mother PLOC
−0.07
0.07
−0.93
0.354
−0.27
0.08
−3.41
0.001**
0.40
0.04
9.27
< 0.001***
Mother PCL
0.20
0.07
2.78
0.005**
0.17
0.08
2.22
0.026*
0.02
0.05
0.30
0.763
Teacher Perspective
Mother’s Ethnicity (Latinx)
−0.08
0.07
−1.10
0.271
−0.04
0.07
−0.61
0.544
0.02
0.06
0.28
0.778
Mother’s Race (Black)
0.03
0.09
0.32
0.748
0.03
0.07
0.35
0.724
−0.05
0.07
−0.76
0.447
Mother’s Race (Other)
−0.06
0.07
−0.79
0.431
−0.06
0.06
−0.92
0.358
−0.01
0.06
−0.09
0.929
Father’s Ethnicity (Latinx)
0.09
0.07
1.28
0.201
0.04
0.06
0.70
0.486
−0.18
0.06
−3.08
0.002**
Father’s Race (Black)
−0.02
0.09
−0.27
0.784
−0.02
0.07
−0.24
0.807
−0.11
0.07
−1.54
0.123
Father’s Race (Other)
0.10
0.07
1.40
0.161
0.06
0.06
1.02
0.309
0.09
0.06
1.50
0.135
Marital Status
−0.06
0.10
−0.53
0.596
0.06
0.11
0.59
0.554
−0.07
0.06
−1.05
0.293
Family Income
0.06
0.10
0.57
0.571
0.13
0.11
1.16
0.246
0.01
0.06
0.16
0.871
Father Deployment
−0.02
0.07
−0.21
0.831
−0.16
0.06
−2.88
0.004**
0.05
0.05
1.00
0.318
Mother Deployment
0.01
0.07
0.13
0.896
0.01
0.06
0.12
0.907
−0.04
0.06
−0.67
0.505
Father DAS
−0.09
0.08
−1.06
0.287
−0.07
0.07
−0.97
0.333
0.13
0.07
2.00
0.046*
Mother DAS
0.14
0.08
1.80
0.071
0.09
0.07
1.37
0.170
0.08
0.06
1.21
0.228
Father PLOC
0.01
0.07
0.07
0.941
−0.01
0.06
−0.09
0.925
0.06
0.06
0.99
0.323
Mother PLOC
0.06
0.07
0.91
0.364
−0.01
0.07
−0.20
0.844
−0.03
0.06
−0.51
0.610
Father PCL
−0.01
0.08
−0.08
0.934
−0.02
0.07
−0.31
0.756
0.09
0.06
1.48
0.140
Mother PCL
0.09
0.07
1.22
0.223
0.00
0.06
−0.06
0.951
0.07
0.06
1.08
0.281
DAS: Dyadic Adjustment Scale; PLOC: Parental Locus of Control; PCL: Post-Traumatic Stress Checklist; Deployment: coded in months, and participants who were never deployed were coded as 0; *p <.05. **p <.01. ***p <.001
Internalizing Problems. The model with predictors fit well (χ2(191) = 226.98, p =.04; RMSEA = 0.020, 90% CI (0.005-0.030); CFI = 0.984, TLI = 0.978; SRMR = 0.044). Lower mother’s perceived relationship satisfaction (β = − 0.16, p =.022), and lower parental efficacy (father: β = − 0.15, p =.029; mother: β = − 0.27, p <.001) were significantly related to higher overall (consensus factor) internalizing problems. There were no statistically significant correlates of teacher reports of internalizing symptoms. Parents with higher distress (father: β =. 33, p <.001; mother: β = 0.20, p =.005) tended to perceive their children as having greater internalizing problems. Compared to White fathers, Black fathers tended to perceive their children to have fewer internalizing problems, holding all else constant (β = − 0.18, p =.017).
Externalizing Problems. The model with predictors fit well (χ2(117) = 213.76, p <.001; RMSEA = 0.042, 90% CI (0.033-0.051); CFI = 0.96, TLI = 0.939; SRMR = 0.043). Greater parental efficacy for either parent was associated with lower levels of reported overall externalizing problems (father: β = − 0.13, p =.028; mother: β = − 0.32, p <.001). Teachers tended to perceive children as having fewer externalizing problems as fathers were deployed for longer periods (β = − 0.16, p =.004). Parents with greater parental efficacy perceived their children’s externalizing problems to be lower (father: β = − 0.29, p <.001; mother: β = − 0.27, p <.001). Mothers who had higher distress scores perceived their children’s externalizing problems to be higher (β = 0.17, p =.026). Compared to White fathers, Black fathers tended to perceive their children to have fewer externalizing problems (β = − 0.22, p =.007).
Adaptive Functioning. The model with predictors had acceptable fit in terms of RMSEA and SRMR (χ2(272) = 417.55, p <.001; RMSEA = 0.034, 90% CI (0.028-0.041); CFI = 0.965, TLI = 0.956; SRMR = 0.075). The residual variance for one of the mother’s items was constrained to zero to avoid a negative variance estimate. Overall adaptive functioning was higher for married families (β = 0.19, p =.022), higher income families (β = 0.16, p =.026), and when mothers reported lower distress (β = − 0.23, p =.001) and higher maternal efficacy (β = 0.20, p =.015). Teachers perceived target children as having higher adaptive functioning when fathers’ relationship satisfaction was higher (β = 0.13, p =.046), and perceived children with Latinx fathers as having lower adaptive functioning than children with White fathers (β =−0.18, p =.002). Parents tended to perceive their children as having higher adaptive functioning when they perceived higher relationship satisfaction (father: β = 0.15, p =.001; mother: β = 0.23, p <.001) and higher parental efficacy (father: β = 0.42, p <.001; mother: β = 0.40, p <.001).
Discussion
While informant discrepancy has been widely acknowledged in children’s mental health assessment, there is still a limited understanding of what factors contribute to the shared and unique perspectives of fathers, mothers, and teachers on school-aged children’s psychosocial adjustment. Utilizing baseline data from three randomized controlled trials, this study described informant discrepancies and then explored potential sources of shared and unique perspectives of fathers’, mothers’, and teachers’ ratings on internalizing problems, externalizing problems, and adaptive functioning. Incorporating both clinical and adaptive scales allowed us to have a more comprehensive and holistic view of school-age children’s psychosocial adjustment, incorporating children’s strengths rather than solely assessing problem behaviors. By seeking to understand the sources of discrepancies (i.e., the unique perspectives), this study is in line with the contemporary view that discrepancies among reporters can provide important insight about the target child (e.g., De Los Reyes et al. 2023). This study provided an opportunity to examine contextual factors that contribute to shared and unique informant perspectives on children’s psychosocial adjustment.
Consistent with previous meta-analyses [2]; De Los Reyes et al. 2019; Duhig et al. 2006), our study supports the domain-specific hypothesis [2] that children’s behaviors vary in different contexts. The results revealed moderate-to-strong correspondence between similar informants within the same context (e.g., fathers vs. mothers) and small-to-moderate correspondence among distinct informants across contexts (e.g., parents vs. teachers). In line with previous research (e.g., Berg-Nielsen et al. [6, 28], mothers perceived their children as exhibiting more challenges and demonstrating lower adaptability in comparison to fathers and teachers.
Further, we examined whether such informant discrepancies reflect meaningful information rather than informant biases. Measurement invariance analyses of the three constructs resulted in well-fitting models. The loadings were invariant for father and mother reports, while there were some differences for teacher-reported loadings for externalizing and adaptive functioning. However, as discussed in Bauer et al. [5], this is likely acceptable since teachers are a different type of reporter than parents. There were some differences in intercepts meaning that some raters tended to have higher average subscale scores. Across the predictor models, findings demonstrated the important role of parental distress, parental efficacy, couple satisfaction, race, and household income on parents’ unique perspectives on their children’s mental health.
For shared perspective or consensus factors, our results indicated a robust association between parental efficacy and stronger shared perspective on child psychosocial adjustment, underscoring the important protective role of parental efficacy [3]. Interestingly, mother-reported couple relationship satisfaction was related to the shared perspective on internalizing problems, while mother’s distress was related to the shared perspective on adaptive functioning. On the other hand, father-reported couple relationship or distress did not seem to have a consistent relationship with shared perspective on child psychosocial functioning. Additionally, higher family income and married parents were associated with stronger shared perspective on adaptive functioning, potentially indicating socioeconomic stability as beneficial to children’s adjustment.
Perhaps the most intriguing finding of this study is that results related to parents’ unique perspectives revealed consistent patterns that align with the domain-specific hypothesis [2], which suggests that the impact of specific parental psychosocial factors on their perceptions of children’s mental health outcomes depends on the resources available to parents and the specific domain of functioning being examined. These findings indicate that parents with higher levels of distress tend to perceive their children as having more internalizing problems. Parental distress has been widely documented as a significant factor influencing how parents perceive their children’s mental health (e.g., Bajeux et al. [4, 17]. This distress likely affects the child’s home environment and behavior, altering how the child feels at home, which in turn may change how parents perceive their children’s emotions. Consistent with previous research [4, 38], parents who are more confident in their role and in their ability to guide their children’s behaviors effectively likely perceive their children as exhibiting fewer problematic observable behaviors and as more adaptable. It is possible that more efficacious parents may elicit better behavior from their child, particularly in situations where they are interacting one-on-one. Moreover, results highlighted that a strong couple relationship is related to unique perspectives on children’s psychosocial adjustment, specifically in adaptive functioning. While few studies have examined the impact of couple relationships on parents’ unique perspectives on their children [20], results suggest that a supportive marital environment may help parents focus on the positive qualities of their children.
Overall, these results suggest that discrepancies in parent-reported internalizing problems may arise as a function of parental distress; discrepancies in parent-reported externalizing problems may be influenced by parental efficacy; and discrepancies in parent-reported adaptive functioning appear to be affected by both the quality of the couple’s relationship and parental efficacy. As pointed out by Kuhfeld and colleagues [25], potential sources for discrepancies among informants included children behaving differently in different contexts, expectations per reporter, response styles, and potential bias. Future studies should further differentiate the various sources of these differences to better understand situational specificity.
Findings indicate that while family income and marital status are associated with the shared perspective on adaptive functioning, these sociodemographic factors do not appear to influence parents’ unique perspectives. Additionally, parental deployment length did not seem to impact parental unique perspectives on children’s mental health. Furthermore, while having a Black father was linked to higher shared perspective on internalizing problems, Black fathers themselves tended to perceive their children as having lower levels of internalizing and externalizing problems compared to White fathers. However, these results should be interpreted with caution due to the small sample size and limited demographic categories. Future studies should further investigate how broader and more diverse demographic factors may be related to shared and unique perspectives on children’s mental health.
It is not surprising that findings for teachers’ unique perspectives were somewhat limited, as data related to the teacher or school environment were scarce in this dataset. However, results indicated that teachers generally perceived children whose fathers had longer deployments as exhibiting fewer externalizing problems, suggesting that teachers might show greater understanding and empathy towards behavioral challenges in children whose parents have been deployed for prolonged periods. Additionally, the findings that teachers perceive lower adaptive functioning in children with a Latinx father might be consistent with those reported by Takeda et al. [37], with greater parent–teacher discrepancies found for ethnic minority families compared to White/European American families. Furthermore, teachers tended to perceive higher adaptive functioning in children when fathers reported better relationship satisfaction, which may be reflective of the complex interaction between family and school contexts. However, these results should be interpreted with caution, as limited data were available from school contexts. Future studies should further explore how various family and school contextual factors may influence parents’ and teachers’ unique perspectives on children’s mental health (e.g., Heyman et al. [21].
Strengths and Limitations
The current study has several unique strengths. First, this study provided an opportunity to explore informant perspectives on mental health problems and adaptive functioning in a relatively large sample of military families, a population affected by high levels of stress and trauma exposure. From a risk and resilience perspective, the study was able to include both behavioral and emotional problems and adaptive functioning, which is critical as mental health research has often taken a deficit-based rather than strengths-based perspective. The inclusion of father, mother, and teacher ratings provides a more comprehensive view of children’s mental health than is typical in a field in which fathers’ perspectives are rarely explored [26]. Finally, the trifactor model enabled the analysis of each reporter’s unique perspective in addition to the agreement among reporters.
Nevertheless, the study is not without limitations. First, the sample of predominantly White, middle-class, military families is relatively homogeneous. While this allowed us to explore unique perspectives within a highly specific and stressful population, the findings may not generalize to more diverse or non-military sample. Although model modifications were guided by both theoretical rationale and empirical indicators, they may still reflect a data-driven approach that could limit generalizability. Second, the lack of additional information about teachers and schools (e.g., metrics on parent-teacher relationships) might also limit the ability to understand factors that influence teachers’ unique perspectives. Third, children’s perspectives were not included in the model, given the fact that half of the participants were not old enough to provide self-report data, and the BASC utilized different subscales for children compared to those used for parents and teachers. Fourth, only cross-sectional data were utilized, which limits the ability to infer causal relationships over time. Although we used age-normed T-scores to account for developmental differences, the broad age range (4–12 years) may have introduced variability in informant perspectives that was not fully captured in our analyses. Additionally, the study only included parental posttraumatic stress symptoms, and the lack of a consistent measure of depression and anxiety across the three randomized controlled trials limits the ability to disentangle the impact of depression, anxiety, and post-traumatic stress symptoms. Future studies should include more differentiated measures of parental distress to understand further associations of depression and anxiety with parents’ unique perspectives of children’s mental health. Finally, it is not possible to differentiate reporter bias from each reporter’s “true” experiences with the target child. For instance, if the father’s perspective factor had a lower mean than the mother’s perspective factor for a certain behavior, it would not be possible to know whether that was due to the child behaving differently while they are around each parent on average, fathers underreporting that behavior on average, or mothers overreporting that behavior on average. Because many of our predictors were based on self-report (e.g., parental efficacy, parental distress), it is possible that similar response biases affected both their self-reports and their BASC (child) ratings. A model that can better account for context might help to distinguish these sources of variability in future research. For instance, Kraemer’s (2003) Satellite Model is useful for disentangling sources of variance in multiple informant data using principal components analysis with composite scores. However, it was not possible to apply the Satellite model in our case as it would have required examining children’s self-report as well.
Implications
The study highlights the role of parental efficacy, parental distress, and family dynamics in parents’ and teachers’ assessment of children’s internalizing and externalizing problems and adaptive functioning. Results suggest that both in research and clinical practice, adopting a family systems perspective would enable exploration of the broader context of children’s mental health and how it is influenced by various family-related factors. The findings also underscore the importance of comprehensive assessments that incorporate multiple perspectives both within and outside the family to understand children’s mental health as well as considering both strengths and challenges. Recognizing that each informant brings unique insights into a child’s psychosocial adjustment can lead to more tailored and effective interventions; for example, context-specific intervention strategies. Understanding the value of different informant perspectives can also encourage clinicians to consider the wider family environment and its role in intervention and maintaining intervention-related improvements in children’s behavior and emotional well-being. Ultimately, by embracing a more holistic approach that incorporates multiple informants and considers both family and other key adult roles in child adjustment, clinicians and researchers can work toward better understanding, assessing, and addressing children’s mental health needs.
Declarations
Competing Interests
The authors declare no competing interests.
Informed Consent
Informed consent was obtained from all participants included in the study.
Ethical Approvals
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee at the University of Minnesota, Twin Cities (IRB number: 1005S82692; 1407S52001) and Arizona State University (STUDY00015375) and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Achenbach TM (2011) Commentary: definitely more than measurement error: but how should we understand and deal with informant discrepancies? J Clin Child Adolesc Psychol 40(1):80–86. https://doi.org/10.1080/15374416.2011.533416CrossRefPubMed
2.
Achenbach TM, McConaughy SH, Howell CT (1987) Child/adolescent behavioral and emotional problems: implications of cross-informant correlations for situational specificity. Psychol Bull 101(2):213–232. https://doi.org/10.1037/0033-2909.101.2.213CrossRefPubMed
3.
Albanese AM, Russo GR, Geller PA (2019) The role of parental self-efficacy in parent and child well‐being: a systematic review of associated outcomes. Child Care Health Dev 45(3):333–363. https://doi.org/10.1111/cch.12661CrossRefPubMed
4.
Bajeux E, Klemanski DH, Husky M, Leray E, Chan Chee C, Shojaei T, Fermanian C, Kovess-Masfety V (2018) Factors associated with parent–child discrepancies in reports of mental health disorders in young children. Child Psychiatry Hum Dev 49(6):1003–1010. https://doi.org/10.1007/s10578-018-0815-7CrossRefPubMed
5.
Bauer DJ, Howard AL, Baldasaro RE, Curran PJ, Hussong AM, Chassin L, Zucker RA (2013) A trifactor model for integrating ratings across multiple informants. Psychol Methods 18(4):475–493. https://doi.org/10.1037/a0032475CrossRefPubMedPubMedCentral
6.
Berg-Nielsen TS, Solheim E, Belsky J, Wichstrom L (2012) Preschoolers’ psychosocial problems: in the eyes of the beholder? adding teacher characteristics as determinants of discrepant parent–teacher reports. Child Psychiatry Hum Dev 43(3):393–413. https://doi.org/10.1007/s10578-011-0271-0CrossRefPubMed
7.
Bronfenbrenner U (1996) The ecology of human development: experiments by nature and design. Harvard University Press
8.
Canivez GL, Embse VD, N. P., McGill RJ (2021) Construct validity of the BASC-3 teacher rating scales: independent hierarchical exploratory factor analyses with the normative sample. School Psychol 36(4):235–254. https://doi.org/10.1037/spq0000444CrossRef
Datta Gupta N, Lausten M, Pozzoli D (2016) Does mother know best? Parental discrepancies in assessing child behavioral and educational outcomes. Rev Econ Househ. https://doi.org/10.1007/s11150-016-9341-1CrossRef
13.
De Los Reyes A, Epkins CC (2023) Introduction to the special issue. A dozen years of demonstrating that informant discrepancies are more than measurement error: toward guidelines for integrating data from multi-informant assessments of youth mental health. J Clin Child Adolesc Psychol 52(1):1–18. https://doi.org/10.1080/15374416.2022.2158843CrossRef
14.
De Los Reyes A, Augenstein TM, Wang M, Thomas SA, Drabick DAG, Burgers DE, Rabinowitz J (2015) The validity of the multi-informant approach to assessing child and adolescent mental health. Psychol Bull 141(4):858–900. https://doi.org/10.1037/a0038498CrossRefPubMedCentral
Duhig AM, Renk K, Epstein MK, Phares V (2000) Interparental agreement on internalizing, externalizing, and total behavior problems: a meta-analysis. Clin Psychol Sci Pract 7(4):435–453. https://doi.org/10.1093/clipsy.7.4.435CrossRef
17.
Fergusson DM, Lynskey MT, Horwood LJ (1993) The effect of maternal depression on maternal ratings of child behavior. J Abnorm Child Psychol 21(3):245–269. https://doi.org/10.1007/BF00917534CrossRefPubMed
Hassall R, Rose J, McDonald J (2005) Parenting stress in mothers of children with an intellectual disability: the effects of parental cognitions in relation to child characteristics and family support. J Intellect Disabil Res 49(6):405–418. https://doi.org/10.1111/j.1365-2788.2005.00673.xCrossRefPubMed
20.
Hay DF, Pawlby S, Sharp D, Schmücker G, Mills A, Allen H, Kumar R (1999) Parents’ judgements about young children’s problems: why mothers and fathers might disagree yet still predict later outcomes. J Child Psychol Psychiatry Allied Discip 40(8):1249–1258. https://doi.org/10.1111/1469-7610.00541CrossRef
21.
Heyman M, Poulakos A, Upshur C, Wenz-Gross M (2018) Discrepancies in parent and teacher ratings of low-income preschooler’s social skills. Early Child Dev Care 188(6):759–773. https://doi.org/10.1080/03004430.2016.1236257CrossRef
22.
Hunsley J, Best M, Lefebvre M, Vito D (2001) The seven-item short form of the dyadic adjustment scale: further evidence for construct validity. Am J Fam Ther 29(4):325–335. https://doi.org/10.1080/01926180126501CrossRef
23.
Institute of Medicine (2010) Committee on the initial assessment of readjustment needs of military personnel, veterans, and their families. National Academies Press US, Washington DC
24.
Kline RB (2023) Principles and practice of structural equation modeling (Fifth edition.). Guilford
25.
Kuhfeld M, Soland J, Lewis K (2023) Investigating differences in how parents and teachers rate students’ self-control. Psychol Assess 35(1):23–31. https://doi.org/10.1037/pas0001187CrossRefPubMed
26.
Lechowicz ME, Jiang Y, Tully LA, Burn MT, Collins DAJ, Hawes DJ, Lenroot RK, Anderson V, Doyle FL, Piotrowska PJ, Frick PJ, Moul C, Kimonis ER, Dadds MR (2019) Enhancing father engagement in parenting programs: translating research into practice recommendations. Aust Psychol 54(2):83–89. https://doi.org/10.1111/ap.12361CrossRef
27.
Mascendaro PM, Herman KC, Webster-Stratton C (2012) Parent discrepancies in ratings of young children’s co-occurring internalizing symptoms. Sch Psychol Q 27(3):134–143. https://doi.org/10.1037/a0029320CrossRefPubMed
28.
McDonald CA, Lopata C, Donnelly JP, Thomeer ML, Rodgers JD, Jordan AK (2016) Informant discrepancies in externalizing and internalizing symptoms and adaptive skills of high-functioning children with autism spectrum disorder. School Psychol Q 31(4):467–477. https://doi.org/10.1037/spq0000150CrossRef
29.
Moens MA, Weeland J, Van der Giessen D, Chhangur RR, Overbeek G (2018) In the eye of the beholder?? Parent-observer discrepancies in parenting and child disruptive behavior assessments. J Abnorm Child Psychol 46(6):1147–1159. https://doi.org/10.1007/s10802-017-0381-7CrossRefPubMedPubMedCentral
30.
Park JH, Rha J-H, Kim YK, Lee D (2023) Maternal attribution, perceptions of parenting, and behavior management strategies as predictors of mother–teacher discrepancies in rating behavior problems of Korean preschoolers. Psychol Rep 126(4):1847–1870. https://doi.org/10.1177/00332941221076761CrossRefPubMed
Rescorla LA, Bochicchio L, Achenbach TM, Ivanova MY, Almqvist F, Begovac I, Bilenberg N, Bird H, Dobrean A, Erol N, Fombonne E, Fonseca A, Frigerio A, Fung DSS, Lambert MC, Leung PWL, Liu X, Marković I, Markovic J, Verhulst FC (2014) Parent–teacher agreement on children’s problems in 21 societies. J Clin Child Adolesc Psychol 43(4):627–642. https://doi.org/10.1080/15374416.2014.900719CrossRefPubMed
Takeda T, Nissley-Tsiopinis J, Nanda S, Eiraldi R (2020) Factors associated with discrepancy in parent–teacher reporting of symptoms of ADHD in a large clinic-referred sample of children. J Atten Disord 24(11):1605–1615. https://doi.org/10.1177/1087054716652476CrossRefPubMed
38.
Vallotton C, Harewood T, Froyen L, Brophy-Herb H, Ayoub C (2016) Child behavior problems: mothers’ and fathers’ mental health matters today and tomorrow. Early Child Res Q 37:81–93. https://doi.org/10.1016/j.ecresq.2016.02.006CrossRef
39.
Weathers FW, Litz BT, Herman DS, Huska JA, Keane TM (1993) The PTSD checklist (PCL): reliability, validity, and diagnostic utility. the Annual Convention of the International Society for Traumatic Stress Studies, San Antonio, TX
