Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Dec 22.
Published in final edited form as: J Abnorm Child Psychol. 2019 Sep;47(9):1547–1558. doi: 10.1007/s10802-019-00535-4

A Daily Diary Analysis of Preschool Depressive Behaviors: Prospective Associations and Moderators Across 14 Days

Katherine A Leppert 1, Sara J Bufferd 2, Thomas M Olino 3, Lea R Dougherty 1
PMCID: PMC7755310  NIHMSID: NIHMS1652549  PMID: 30911869

Abstract

Depressive disorders can be observed in early childhood and are associated with significant concurrent and prospective impairment; however, little is known about day-to-day variations in common depressive behaviors in children. This study examined the day-to-day variability of two common depressive behaviors in preschool-aged children, sadness and irritability, and factors associated with the daily occurrence of these behaviors. Participants included 291 parents of preschool-aged children, and parents completed a 14-day daily diary. Results indicated that sleep quality did not prospectively predict next-day sadness or irritability the following day. We observed between-person stability, but within-person variability, in children’s sadness and irritability across 14 days. We observed greater between-person stability and greater within-person variability in sadness and irritability for males and for children with fewer baseline psychiatric symptoms and lower baseline impairment. Findings provide a developmental perspective on normative patterns of sadness and irritability in young children and can inform prevention and individualized intervention efforts to reduce negative sequelae in at-risk preschoolers.

Keywords: Depression, Irritability, Preschool, Daily diary, Stability, Sleep quality

Introduction

Depression is a major public health concern for children, adolescents, and adults and an increasing public health concern for young children. Studies have reported that rates of preschool depression range from 0 to 2% and documented associations with significant impairment in psychosocial functioning (Dougherty et al. 2015). Preschool depression demonstrates continuity, predicting both depression (homotypic continuity) and other psychiatric disorders (heterotypic continuity) later in childhood (Finsaas et al. 2018). Despite these gains in understanding the clinical characteristics of depression in young children, preschool psychopathology is under-detected by physicians (Tolan and Dodge 2005), which is likely attributable to difficulties recognizing when normative behaviors and mood changes in this developmental period become clinically significant (Bufferd et al. 2016). Understanding the phenomenology of early emerging depressive behaviors is critical for guiding pediatricians, psychologists, and other health providers in the accurate identification of clinical depression in young children.

Characterizing depressive behaviors in young children necessitates a fine-grained, day-to-day analysis of sadness and irritability. Investigating these behaviors is particularly important given that sadness and irritability are the most common depressive behaviors in preschoolers (Bufferd et al. 2017) and, although they occur normatively during this age, findings suggest that normative versus clinical levels can be differentiated in the preschool years (e.g., Wakschlag et al. 2012; Wiggins et al. 2018). Luby et al. (2002) proposed that to identify depression in preschoolers, depressive symptoms applicable for diagnosis in adults require developmental modifications, such as symptom presence but not necessarily persistence over a two-week period. Yet, little is known about day-to-day variability of these depressive behaviors and whether variability is associated with risk status. Understanding the phenomenology of these behaviors and their day-to-day variability can inform the identification of high-risk children and subsequent intervention.

Prior research in young children has largely examined between-person differences (i.e., how a child on average compares to his/her peers) in depressive behaviors rather than day-to-day variability of depressive behaviors within a child. Between-person analysis is used to establish group-level norms against which a child’s behavior is compared to determine relative standing along a spectrum of behaviors. In contrast, within-person analysis provides a fine-grained approach to understanding affective variability that indicate how a child’s behavior on a given day compares to that of his/her own average behavior and yield important information regarding person-specific antecedents and consequences. Within-person variability may also provide information about risk status, as research shows that clinical depression and depressive symptoms in adolescents and adults are associated with intra-individual variability in daily affect (Peeters et al. 2006; Silk et al. 2003). Although fluctuations at the within-person level contribute to between-person differences, patterns at one level may diverge from those at the other level and thus examination of both inter- and intra-individual affective dynamics is critical to gain a developmental perspective on normative patterns of behavior.

Daily Diaries of Mood and Affect

Prior work has investigated daily affective experiences and variability in older youth and adults using daily diaries. Within-person estimates of daily variation in negative affect range from 39 to 58% in adults (42–61% between-person; Sin et al. 2017; Thompson et al. 2012) and 60–62% in children (38–40% between-person; Aunola et al. 2013; Könen et al. 2015). However, the extant literature has been mixed regarding the adaptive or maladaptive nature of affective variability in adults and older adolescents. On the one hand, affective variability may be an indicator of vulnerability and reflect difficulties with coping. Better well-being has been linked to less variable and more stable emotions in adolescents and adults (Gruber et al. 2013; Houben et al. 2015), and greater intraindividual variability in negative affect has been associated with more internalizing and externalizing symptoms in adolescents (Silk et al. 2003) and with depression, neuroticism, and lower well-being in adults (Houben et al. 2015; Peeters et al. 2006). Frequent, minor emotional disturbances can have negative cumulative effects and increase risk for affective disorders (Wichers 2014) due to the need for coping on a regular basis. Thus, day-to-day variability in adolescent and adult mood may reflect “frailty” or “lack of robustness” and be suggestive of persistent difficulties with coping when variability does not decrease over time (Röcke and Brose 2013).

On the other hand, some research suggests that greater affective variability may indicate an adaptive, flexible emotional response to daily events and may be an indicator of better psychological health in adolescents and adults (Houben et al. 2015; Röcke and Brose 2013). Despite these data in older samples, no studies to our knowledge have examined affective variability in young children, and thus it is unknown whether dynamic emotion in young children is indicative of risk for depression or is adaptive. It is possible that greater affective variability is more common in younger children relative to older youth and adults given their developing emotion regulation capacities (López-Pérez et al. 2016), although relatively larger variability may be indicative of increased risk. Preschoolers likely experience variable success in applying newly-acquired emotion regulation skills, rendering affective variability normative in the early childhood period.

Gaps in the Literature

Despite these advances in the understanding of day-to-day emotion dynamics, little is known about their manifestations in young children. First, no prior work to our knowledge has examined the daily dynamics of affect or depressive behaviors in a preschool-aged sample. Depressive behaviors such as sadness and irritability are common in preschool-aged children (Bufferd et al. 2017; Stringaris 2011); thus, fine-grained examination of these behaviors during the preschool period may help differentiate typical from atypical developmental pathways. Second, no study has examined factors affecting the variability of depressive behaviors in young children. By examining factors affecting daily depressive behaviors, such as sleep quality, triggers of negative emotion dynamics can be identified and mitigated. Lastly, many prior studies have examined emotion dynamics in clinical populations of adults and adolescents. Given the lack of research on daily affect during this developmental period, an understanding of emotion dynamics in non-clinical, community populations is necessary to provide developmentally sensitive information about common depressive behaviors. Finally, this investigation is critical to developing medical guidelines for the screening of young children at-risk for depression and offers targets for intervention and prevention.

Current Study

The current study aimed to address these gaps in the literature by examining the daily dynamics of two depressive behaviors (sadness and irritability) in a community sample of 291 preschool-aged children. First, we examined the day-to-day patterns of children’s sadness and irritability and prospective associations between children’s sadness and irritability. Given previous research demonstrating within-person variation in negative affect around one’s own mean from one day to the next in adults (Sin et al. 2017) and children (Könen et al. 2015), we hypothesized that sadness and irritability would demonstrate within-person change (“variability”) from day-to-day; however, we hypothesized that mean levels of affect would be stable across participants. We further examined how child sleep quality was associated with daily sadness and irritability across 14 days. In older youth, sleep quality has been linked to same-day and next-day affect (e.g., Könen et al. 2016). As younger children have even greater sleep needs than older youth, disturbances in sleep may be particularly linked to daily affect in early childhood. Further, given that change in sleep habits is a symptom of depression and that sleep quality prospectively predicts internalizing symptoms in middle childhood (El-Sheikh et al. 2010), examining sleep-affect associations in young children may provide a critical opportunity for intervention and prevention of depression. We thus hypothesized that poorer sleep quality the previous night would predict increases in depressive behaviors the next day.

Lastly, we explored moderators of children’s day-to-day patterns of sadness and irritability across a 14-day diary period to determine which variables predicted differences in day-to-day variability of sadness and irritability. Moderators included average daily sadness and irritability, child sex, daily sleep quality, and baseline measures of functional impairment and symptoms of depression and oppositional defiant disorder (ODD), given that sadness and irritability are key characteristics of these disorders. We had no specific hypotheses regarding how these factors would moderate children’s affective variability given there are no available data on this topic.

Method

Participants

The sample consisted of 299 parents of 3–5 year-old children without medical or developmental disabilities. Participants were recruited using flyers sent to local pediatricians, pre-schools/daycares, and community institutions within a 20-mile radius of two study sites – University of Maryland in College Park, MD and California State University in San Marcos, CA. Eligible parents had a child who was three to five years of age, were English-speaking, had at least 50% legal custody, and had nightly internet access. One child per family participated, and most parents completing the diary were mothers (93.5%). Participants who completed at least one diary (n = 291; 97.3%) were included in the study. This study was approved by the Institutional Review Boards at both universities. Informed consent was obtained from parents both verbally after the completion of the phone screen and online at the time of the baseline questionnaire. See Table 1 for demographic characteristics of the sample.

Table 1.

Demographic characteristics of study sample

Demographic variable
Child sex, male [n (%)] 137 (47.1%)
Child age, mean (SD [range]), years 4.20 (0.80 [3.00–6.42])
Child race [n (%)]
 White, European-American 176 (60.5%)
 African-American 26 (8.9%)
 Asian 17 (5.8%)
 Mixed/Other 71 (24.4%)
Child ethnicity [n (%)]
 Hispanic/Latino descent 48 (16.7%)
Number of siblings, mean (SD [range]) 1.23 (1.28 [0–13])
Parent completing diaries, mother [n (%)] 273 (93.8%)
Parent employment status [n (%)]
 Mother employed at least part-time 161 (56.3%)
 Mother stay-at-home parent or unemployed 125 (43.7%)
 Father employed at least part-time 258 (94.5%)
 Father stay-at-home parent or unemployed 15 (5.5%)
Family income [n (%)]
 < $20,000 18 (6.3%)
 $20,001 to $40,000 43 (15.1%)
 $40,001 to $70,000 56 (19.7%)
 $70,001 to $100,000 70 (24.6%)
 > $100,000 97 (34.2%)
Parent age
 Mother, mean (SD [range]), years 33.60 (5.65 [21.00–49.00])
 Father, mean (SD [range], years 35.91 (6.72 [21.00–60.00])
Parent marital status, married/living together [n (%)]a 266 (91.4%)
Parental education: graduated 4-year college [n (%)]
 Mothers 196 (67.4%)
 Fathers 171 (58.7%)
Total number of diaries completed by family [n (%)]
 0 diaries 8 (2.7%)
 1–5 diaries 7 (2.3%)
 6–10 diaries 9 (3.0%)
 11–14 diaries 275 (92.0%)
Open in a new tab

N = 291.

a

Parents were permitted to endorse more than one marital status if applicable. One family (0.3%) did not report race; 3 families (1.0%) did not report the ethnicity; 7 families (2.4%) did not report parental education; 7 families (2.4%) did not report their yearly income; 5 mothers (1.7%) did not report employment status; 18 fathers (6.2%) did not report employment status; 2 mothers (0.7%) did not report their age; 11 fathers (3.8%) did not report their age

Procedure

Interested participants completed a phone screen to assess study eligibility. All eligible parents were trained by research staff on the completion of the electronic daily diary. Participants were emailed a link to an online baseline questionnaire and started their daily diaries for the next 14 days on the Monday following completion of the baseline questionnaire. Parents of children who recently (past two weeks) began a new daycare, camp, or school program started their diaries a minimum of two weeks after the start of the program to minimize over-reporting of normative behavioral and emotional symptoms associated with the transition. Participants were instructed to complete the daily diary after the child’s bedtime each evening. Diaries were emailed through Qualtrics to participants at 6:00 PM each evening and parents had until noon the following day to complete the diary for the prior day. Participants received monetary compensation for completion of diaries and an incentive for completion of all 14 diaries. The total number of diaries completed by each family is included in Table 1.

Measures

Baseline Psychiatric Symptoms

To assess behavioral and emotional difficulties at baseline, parents completed the Early Childhood Inventory-4 (ECI-4; Sprafkin and Gadow 1996), a 108-item measure which was used to assess symptoms of depression and ODD. Parents rated the frequency of their children’s symptoms on a 4-point Likert scale, ranging from 0 (never) to 3 (very often). Behaviors endorsed by parents were summed in their respective disorder categories to create a composite score reflecting the severity of child psychiatric symptoms: depressive (11 items; M = 15.79, SD = 1.73, range = 10–21, α = 0.69) and ODD (8 items; M = 14.19, SD = 3.84, range = 8–27, α = 0.85) symptoms.

Baseline Impairment

Child baseline impairment was assessed using the Impairment Rating Scale (IRS; Fabiano et al. 2006). The IRS is an 8-item parent-report measure assessing child functioning across a variety of domains, including with peers, siblings, parent, academics, self-esteem, family, and global functioning. Of note, the IRS was not specific to depressive symptoms but rather provided a global assessment of impairment due to any psychiatric symptoms. Parents were asked to rate how their child’s problems affected functioning in each of these areas using a 7-point scale ranging from 0 (No problem) to 6 (Extreme problem/definitely needs treatment or special services). Scores in each IRS domain were averaged to create a composite impairment score (8 items; M = 1.36, SD = 0.62, range = 1.00–4.38, α = 0.87).

Daily Diary

Daily Depressive Behaviors

Items assessing depressive behaviors were derived from two reliable and validated measures, the ECI-4 (Sprafkin and Gadow 1996) and the Preschool Age Psychiatric Assessment (PAPA; Egger et al. 1999). The ECI-4 demonstrates satisfactory test-retest reliability and predictive and concurrent validity (Sprafkin et al. 2002), while the PAPA demonstrates good interrater reliability, adequate test-retest reliability, and good content and construct validity (Dougherty et al. 2008). As detailed in Bufferd et al. (2017), parents were asked to report the daily frequency of twelve depressive behaviors: sadness, irritability, tantrums, tearfulness, low interest, thoughts of death, low self-esteem, fatigue, and changes in appetite or weight, sleep habits, activity level, or concentration. We previously established the frequent and normative nature of sadness, irritability, tantrums, and tearfulness (Bufferd et al. 2017), and thus these behaviors were used for the purposes of the current study. Given overlap in constructs, the daily frequencies of sadness and tearfulness were averaged to create a daily sadness score and the daily frequencies of irritability and tantrums were averaged to create a daily irritability score.

Daily Sleep Quality

Child sleep quality was assessed daily with a single item asking parents to report how well their child slept the previous night on a 5-point scale, ranging from 1 (not at all restful) to 5 (extremely restful) (M = 4.01, SD = 0.79, range = 1–5). This single-item sleep quality assessment has good criterion validity compared to the Pittsburgh Sleep Quality Index (rs = 0.40; p < 0.001) and good reliability (Cronbach’s α = 0.89) (Fung et al. 2014). We observed an intraclass correlation coefficient (ICC) of 0.11 for sleep quality across the 14 days, demonstrating greater within-person, relative to between-person, consistency in the sleep quality item across days, which is consistent with prior studies (e.g., Bouwmans et al. 2017; Kalmbach et al. 2014; Sin et al. 2017).

Data Analysis Plan

To examine the influence of both between-person and within-person predictors of children’s daily sadness and irritability, we used generalized linear mixed models (GLMM; Aiken et al. 2015), an extension of multilevel modeling used for nonlinear outcomes. Using the GENLINMIXED procedure in SPSS v. 22, we tested models with various distributions and chose to specify a negative binomial distribution given evidence of overdispersion in the non-continuous dependent variables (Hox 2010). We calculated an incidence rate ratio (IRR) using exponentiation (exponentiated beta = eβ) for each predictor in the model. Of the diaries completed, 2.80% of values were missing; GLMM uses all available data as well as maximum likelihood to estimate conditional parameters and impute missing data (Willett et al. 1998), avoiding listwise deletion and its associated reduction in power.

To assess between-person variation, we used grand-mean centering for which sadness and irritability were each averaged across the 14 diary days for each person. To assess within-person effects, we used person-mean centering for which each child’s daily sadness and irritability was centered on his or her respective mean value (person-centered); this person-centered approach only reflects within-person temporal change and removes all between-person variance (Enders and Tofighi 2007).1 In all models, we allowed intercepts to vary randomly across individuals and specified an autoregressive (AR1) covariance structure for the repeated effect.2 We first estimated unconditional models for daily sadness and daily irritability to assess the proportion of the variance attributable to between-person versus within-person effects. Second, we examined associations between children’s daily sadness and irritability and potential covariates in GLMM models, including child age, sex, race/ethnicity, parent education, parent marital status, and total time spent away from parent. We assessed total time spent away from the parent (in hours) given that this may impact the parent’s reporting of the child’s behaviors. Significant covariates were included in their respective sadness and irritability models. Third, we examined prospective associations in separate models predicting both next day sadness and irritability. For these analyses, we lagged sadness and irritability by one day, such that children’s prior day sadness or irritability (i - 1) predicted their next day sadness or irritability (i). In the first model, we assessed whether prior day sadness predicted next day sadness when controlling for prior day irritability, which provided an estimate of the stability of sadness from one day to the next. Including prior day sadness and irritability in the model allowed for the examination of both homotypic and heterotypic associations. In this same model, we also examined whether prior day irritability predicted next day sadness, controlling for prior day sadness. The equation for this model (other covariates not listed) is:

  • Level 1:
    ηij( Nextdaysadnessi)=β0j+β1j( Priordayperson−centereddailysadnessi−1)+β2j( Priordayperson−centereddailyirritabilityi−1)+rij
  • Level 2:
    β0j=γ00+γ01( Priordaymean−centeredsadnessi−1)+γ02(Priordaymean−centered irritabilityi−1)+u0jβ1j=γ10β2j=γ20

where ij reflects day i for person j, i reflects next day behaviors, i-1 reflects prior day behaviors, and ηij = ln(λij). We also examined whether prior night sleep quality predicted next day sadness in separate models. An identical model was run with irritability as the dependent variable. All predictors were parsed into between-person (mean-centered) and within-person (person-centered) effects in these models.

Lastly, we examined whether children’s average levels of daily sadness and irritability, child sex, baseline depressive and ODD symptoms and functional impairment, and sleep quality moderated the prospective associations of sadness with next day sadness and irritability with next day irritability. Sadness was included as a covariate in models examining irritability and vice versa. Sadness and irritability were lagged such that prior day sadness or irritability predicted next-day sadness or irritability, respectively. Daily sleep quality was parsed into between-person and within-person effects, and both effects were examined as moderators. Child sex and children’s depressive and ODD symptoms, functional impairment, and average levels of daily sadness and irritability were examined as level-2 moderators. Moderators were examined for both the between-person and within-person levels of sadness and irritability, with the exception of average daily sadness and irritability, which were examined as moderators of the within-person associations only. Significant interactions were probed using simple slopes analyses (Aiken and West 1991; Curran et al. 2006).

Results

Associations between primary study variables are presented in Table 2.

Table 2.

Correlations among all major study variables

1 2 3 4 5 6 7 8 9 10 11 12
1. Sadnessa –
2. Irritabilitya 0.34*** –
3. Sleep qualitya −0.07*** −0.05** –
4. Baseline impairment 0.07*** 0.21*** −0.06*** –
5. Baseline ODD symptoms 0.09*** 0.32*** −0.09*** 0.45*** –
6. Baseline depressive symptoms 0.11*** 0.15*** −0.06*** 0.09*** 0.12*** –
7. Child sex 0.06*** 0.04* 0.01 0.002 −0.05** 0.09*** –
8. Child age −0.10*** −0.06** 0.12*** 0.16*** 0.05** 0.01 −0.02 –
9. Child race/ethnicity 0.07*** −0.001 0.07*** 0.04** 0.06*** 0.03 −0.04* −0.03 –
10. Parent education 0.06*** 0.01 −0.01 −0.03 0.11*** 0.10*** −0.03 0.12*** 0.24*** –
11. Parent marital status −0.001 0.07*** −0.01 −0.09*** 0.13*** 0.06*** −0.13*** −0.08*** 0.27*** 0.36*** –
12. Total time spent away from the parenta −0.11*** −0.12*** 0.05** −0.02 −0.10*** −0.06*** 0.02 0.10*** −0.07*** 0.06*** −0.18*** –
Mean 1.59 1.49 4.01 1.36 14.19 15.79 – 4.20 – – – 3.01
SD 2.55 2.09 0.79 0.62 3.84 1.73 – 0.80 – – – 4.00
Open in a new tab

N = 291 participants. Sadness reflects combined sadness and tearfulness variables. Irritability reflects combined irritability and tantrum variables.

a

Reflects mean value across all 14 diary days; IRS = Impairment Rating Scale; ODD = oppositional defiant disorder. Child sex: 0 = male, 1 = female. Child race/ethnicity: 0 = non-White or Hispanic, 1 = non-Hispanic White. Parent education: 0 = both parents have less than a 4-year college degree, 1 = at least one parent with a 4-year college degree. Parent marital status: 0 = not married or living together, 1 = married or living together

*

p < 0.05,

**

p < 0.01,

***

p < 0.001

Daily Sadness and Irritability

Unconditional models for both children’s daily sadness and irritability indicated that 36.0% of the variance in daily sadness was attributable to within-person factors, and 32.4% of the variance in daily irritability was attributable to within-person factors. Variance and covariance results demonstrated significant within-person variation across days (AR1 diagonal) as well as significant between-person variation (intercept variance) (ps < 0.001) for both models.

We next examined whether potential covariates were associated with children’s daily sadness and irritability. Higher levels of children’s daily sadness were significantly associated with younger child age, having at least one parent with a four-year college degree, and less time the child spent away from the parent (ps < 0.002). Higher levels of children’s daily irritability were significantly associated with younger child age, having parents who were married/living together, and less time the child spent away from the parent (ps < 0.02). Significant covariates were included in their respective models.

Prospective Associations with Sadness and Irritability

We observed differential between-person and within-person associations of sadness and irritability from one day to the next (see Table 3). We found positive between-person associations between children’s sadness and irritability on the prior day and their next day sadness and irritability, respectively, indicating between-person stability in these depressive behaviors across diary days. We also observed that within-person sadness and irritability the prior day were inversely related to sadness and irritability, respectively, the following day, such that lower sadness and irritability relative to the child’s own mean on one day was associated with higher sadness and irritability, respectively, relative to the child’s own mean the following day and vice versa (hereinafter referred to as “variability”). Between-person, but not within-person, irritability positively predicted next day sadness, whereas neither between-person nor within-person sadness predicted next day irritability. Neither between-person nor within-person child sleep quality the prior night predicted children’s next day sadness or irritability (ps > 0.11).

Table 3.

Prospective next day associations between sadness and irritability

Next day sadness Next day irritability
b (SE) Exp(b) b (SE) Exp(b)
Intercept 0.15 (0.04) 1.16*** 0.08 (0.03) 1.08*
Prior day sadness (between-person) 0.69 (0.03) 1.99*** 0.03 (0.03) 1.03
Prior day sadness (within-person) −0.09 (0.02) 0.92*** 0.03 (0.02) 1.03
Prior day irritability (between-person) 0.10 (0.03) 1.11** 0.74 (0.03) 2.10***
Prior day irritability (within-person) 0.03 (0.02) 1.03 −0.09 (0.02) 0.91***
Child age −0.07 (0.04) 0.94 −0.04 (0.03) 0.96
Time spent away from parent 0.01 (0.02) 1.01 −0.02 (0.02) 0.99
Parent education −0.30 (0.08) 0.74*** – –
Parent marital status – – −0.40 (0.10) 0.67***
b (SE) P b (SE) p
Level 2, τ00 0.18 (0.03) <0.001*** 0.12 (0.03) <0.001***
Level 1, σ00 0.84 (0.02) <0.001*** 0.92 (0.03) <0.001***
Level 1, ρ 0.13 (0.03) <0.001*** 0.15 (0.03) <0.001***
Open in a new tab

b unstandardized beta; Exp(b) exponentiated beta. Exponentiated beta estimates are only provided for significant predictors

*

p < 0.05;

**

p < 0.01;

***

p < 0.001

Moderators of Day-to-Day Variability of Sadness and Irritability

Average Daily Sadness and Irritability

Children’s average daily sadness did not significantly moderate the within-person association of sadness from one day to the next (b = 0.001, SE = 0.003, p = 0.72).3 However, children’s average daily irritability significantly moderated the within-person association of sadness from one day to the next (b = 0.04, SE = 0.01, Exp(b) = 1.04, p < 0.001). Greater sadness the prior day relative to the child’s own mean was associated with less sadness the following day, and this association was stronger for children with low average daily irritability (b = −0.19, SE = 0.02, Exp(b) = 0.83, p < 0.001) than for children with high average daily irritability (b = −0.12, SE = 0.01, Exp(b) = 0.89, p < 0.001). The within-person association of irritability from one day to the next was not significantly moderated by children’s average daily sadness (b = 0.01, SE = 0.02, p = 0.79), but was significantly moderated by children’s average daily irritability (b = 0.06, SE = 0.01, Exp(b) = 1.06, p < 0.001). Greater irritability the prior day relative to the child’s own mean was associated with less irritability the following day, and this association was stronger for children with low average daily irritability (b = −0.29, SE = 0.04, Exp(b) = 0.75, p < 0.001) than for children with high average daily irritability (b = −0.18, SE = 0.03, Exp(b) = 0.84, p < 0.001).

Child Sex

Child sex significantly moderated both the between-person (b = −0.48, SE = 0.15, Exp(b) = 0.62, p = 0.002) and the within-person (b = 0.09, SE = 0.04, Exp(b) = 1.10, p = 0.03) associations of sadness from one day to the next. Greater sadness the prior day relative to the sample mean was associated with greater sadness the following day, and this association was stronger for males (b = 1.02, SE = 0.09, Exp(b) = 2.78, p < 0.001) than for females (b = 0.54, SE = 0.14, Exp(b) = 1.72, p < 0.001). However, greater sadness the prior day relative to the child’s own mean was associated with less sadness the following day, and this association was stronger for males (b = −0.15, SE = 0.04, Exp(b) = 0.86, p < 0.001) than for females (b = −0.06, SE = 0.02, Exp(b) = 0.95, p = 0.004).

In addition, child sex significantly moderated the between-person (b = −0.32, SE = 0.11, Exp(b) = 0.73, p = 0.01), but not within-person (b = 0.03, SE = 0.04, p = 0.46), association of irritability from one day to the next. Greater irritability the prior day relative to the sample mean was associated with greater irritability the following day, and this association was stronger for males (b = 0.96, SE = 0.08, Exp(b) = 2.60, p < 0.001) than for females (b = 0.64, SE = 0.08, Exp(b) = 1.90, p < 0.001).

Baseline Psychiatric Symptoms and Functional Impairment

The prospective between-person association between prior day sadness and sadness the following day was significantly moderated by children’s baseline depressive symptoms (b = −0.21, SE = 0.05, Exp(b) = 0.81, p < 0.001). Greater sadness the prior day relative to the sample mean was associated with greater sadness the following day, but this association was stronger for children low in baseline depressive symptoms (b = 0.97, SE = 0.10, Exp(b) = 2.63, p < 0.001) than children high in baseline depressive symptoms (b = 0.54, SE = 0.15, Exp(b) = 1.72, p < 0.001). The prospective between-person association between prior day sadness and next day sadness was not significantly moderated by children’s baseline ODD symptoms or functional impairment (ps > 0.45).

The prospective within-person association between prior day sadness and next day sadness was significantly moderated by children’s baseline depressive symptoms (b = 0.07, SE = 0.02, Exp(b) = 1.07, p = 0.003), ODD symptoms (b = 0.05, SE = 0.02, Exp(b) = 1.05, p = 0.02), and functional impairment (b = 0.04, SE = 0.01, Exp(b) = 1.04, p < 0.001). Greater sadness the prior day relative to the child’s own mean was associated with less sadness the following day for children low in baseline depressive (b = −0.08, SE = 0.03, Exp(b) = 0.92, p = 0.01) and ODD symptoms (b = −0.15, SE = 0.02, Exp(b) = 0.86, p < 0.001), but this association was not significant for children high in depressive (b < 0.001, SE = 0.01, p = 0.98) and ODD symptoms (b = −0.05, SE = 0.03, p = 0.08). Similarly, greater sadness relative to one’s own mean the prior day was associated with less sadness the following day, but this association was stronger for children with lower baseline functional impairment (b = −0.16, SE = 0.02, Exp(b) = 0.85, p < 0.001) than for children with higher baseline functional impairment (b = −0.08, SE = 0.02, Exp(b) = 0.92, p < 0.001).

The between-person association between prior day irritability and next day irritability was significantly moderated by children’s baseline depressive symptoms (b = −0.22 SE = 0.04, Exp(b) = 0.80, p < 0.001), ODD symptoms (b = −0.17, SE = 0.02, Exp(b) = 0.84, p < 0.001), and functional impairment (b = −0.14, SE = 0.02, Exp(b) = 0.87, p < 0.001). Specifically, greater irritability relative to the sample mean was associated with greater prospective irritability the following day, and this association was stronger for children with low baseline depressive symptoms (b = 1.05, SE = 0.08, Exp(b) = 2.86, p < 0.001) than for children with high baseline depressive symptoms (b = 0.61, SE = 0.04, Exp(b) = 1.83, p < 0.001), for children with low baseline ODD symptoms (b = 0.98, SE = 0.06, Exp(b) = 2.67, p < 0.001) than for children with high baseline ODD symptoms (b = 0.64, SE = 0.03, Exp(b) = 1.89, p < 0.001), and for children with low baseline functional impairment (b = 0.94, SE = 0.08, Exp(b) = 2.55, p < 0.001) than for children with high baseline functional impairment (b = 0.65, SE = 0.08, Exp(b) = 1.92, p < 0.001).

The prospective within-person association between prior day irritability and next day irritability was also significantly moderated by children’s baseline ODD symptoms (b = 0.05, SE = 0.02, Exp(b) = 1.05, p = 0.01). Greater irritability the prior day relative to the child’s own mean was associated with less irritability the following day, and this association was stronger for children with low baseline ODD symptoms (b = −0.20, SE = 0.04, Exp(b) = 0.82, p < 0.001) than for children with high baseline ODD symptoms (b = −0.10, SE = 0.02, Exp(b) = 0.90, p < 0.001). The prospective within-person association between prior day irritability and next day irritability was not significantly moderated by children’s baseline depressive symptoms (p = 0.92).4

Daily Sleep Quality

Between-person sleep quality of the prior night significantly moderated the between-person (b = 0.21, SE = 0.10, Exp(b) = 1.23, p = 0.03), but not the within-person (b = 0.01, SE = 0.03, p = 0.84), association between children’s prior day sadness and next day sadness. Greater sadness the prior day relative to the sample mean was associated with greater sadness the following day, and this association was stronger for children with better sleep quality the prior night relative to the sample mean (b = 0.96, SE = 0.09, Exp(b) = 2.61, p < 0.001) than for children with poorer sleep quality the prior night relative to the sample mean (b = 0.54, SE = 0.19, Exp(b) = 1.71, p = 0.004). Within-person sleep quality of the prior night did not significantly moderate the between-person (b = 0.03, SE = 0.01, p = 0.052) or within-person (b = 0.01, SE = 0.02, p = 0.57) associations between children’s prior day sadness and next day sadness.

In addition, between-person sleep quality the prior night did not significantly moderate the between-person (b = −0.02, SE = 0.03, p = 0.54) or within-person (b = −0.01, SE = 0.03, p = 0.72) associations between children’s prior day irritability and next day irritability. However, within-person sleep quality of the prior night significantly moderated the between-person (b = 0.02, SE = 0.01, Exp(b) = 1.02, p = 0.05), but not within-person (b = −0.01, SE = 0.01, p = 0.67), association between prior day irritability and next day irritability. Greater irritability the prior day relative to the sample mean was associated with greater irritability the following day, but this association was stronger for children with better sleep quality the prior night relative to the child’s own mean (b = 0.77, SE = 0.09, Exp(b) = 2.15, p < 0.001) than for children with poorer sleep quality the prior night relative to the child’s own mean (b = 0.73, SE = 0.09, Exp(b) = 2.08, p < 0.001).

Discussion

The present study examined prospective associations between children’s daily sadness and irritability and child sleep quality and examined moderators of variability in children’s daily sadness and irritability over a 14-day diary period. To our knowledge, this is the first study to examine daily patterns in sadness and irritability and factors related to their variability across days in order to gain an understanding of the phenomenology of emerging depressve behaviors in young children. We found that between-person and within-person sadness positively predicted next day irritability. Likewise, between-person and within-person irritability positively predicted next day sadness. Results also demonstrated between-person stability and within-person variability in children’s daily sadness and irritability. Sleep quality did not prospectively predict sadness or irritability. We found that children’s average daily sadness and irritability, child sex, baseline depressive and ODD symptoms, functional impairment, and sleep quality moderated the between-person and within-person associations of sadness and irritability from day to day.

Prospective Associations of Sadness and Irritability

The observation of between-person stability in young children’s sadness and irritability demonstrates that children maintain their standing relative to their peers across a two-week period. Thus, children who demonstrate greater sadness or irritability on a given day relative to their peers continue to demonstrate greater sadness or irritability compared to their peers on subsequent days. The between-person approach to understanding depressive behaviors provides useful information regarding how children compare to one another. Further, gaining an understanding of between-person patterns in daily sadness and irritability reflects an important first step in developing norms for depressive behaviors in young children, knowledge critical to determining whether behavior may be typical or atypical (Bufferd et al. 2017).

In addition, we observed within-person variability in preschoolers’ sadness and irritability across diary days, consistent with prior work documenting similar effects in adults (Blaxton et al. 2017; Sin et al. 2017) and older youth (Könen et al. 2015). Our findings suggest that variability in daily sadness and irritability is normative during early childhood. Although preschool-aged children demonstrate the ability to engage in emotion regulation (Carlson and Wang 2007), cognitive capacities to regulate emotion become more sophisticated as children get older (López-Pérez et al. 2016) and thus preschoolers’ self-initiated emotion regulations strategies may be less effective and more contingent on parental responses and environmental factors than older children (Fabes et al. 2001). During early childhood, children rely on external sources for emotion regulation, including parents, teachers, and other attachment figures who can serve as “emotion coaches” (Zimmer-Gembeck and Skinner 2011). This reliance on external sources for emotion regulation may thus contribute to variability in child depressive behaviors during the preschool years when “emotion coaches” are not always available or able to help children regulate their emotions. In addition, young children do not have control over many aspects of their environment, which may increase their emotional sensitivity to context. Preschoolers are also less likely than older individuals to engage in cognitive processes that prolong negative affective states, such as rumination, resulting in greater variability in depressive behaviors.

Prospective Associations with Sleep Quality

Surprisingly, we did not observe prospective associations between preschoolers’ daily sadness and irritability and sleep quality. Our findings are consistent with some prior work demonstrating that sleep quality did not prospectively predict next-day affect in adults (Kalmbach et al. 2014), though other work has found significant links with sleep quality (Sin et al. 2017). It is possible that, despite good average split-half reliability, our single-item measure of sleep quality was not sensitive enough to detect significant associations, and a more comprehensive measure of sleep (naps, nighttime waking, sleep onset latency) may yield different results.

Moderators of Stability/Variability of Children’s Sadness and Irritability

We observed that preschoolers’ sex, baseline psychiatric symptoms and impairment, and sleep quality significantly moderated the between-person associations of sadness and irritability from one day to the next, indicating that these factors increased affective stability relative to one’s peers. Overall, findings provide evidence for greater between-person stability in sadness and irritability for males and for preschoolers with better functioning (fewer depressive and ODD symptoms, less baseline impairment, and better sleep quality). Regardless of whether a child’s daily sadness or irritability was higher or lower than that of his or her peers on a given day, children with higher functioning were more likely to maintain this position (higher/lower) relative to their peers over the course of fourteen days. It is possible that higher-functioning children live in more stable environments (Fiese and Winter 2010), allowing for greater stability relative to peers who may live in less stable environments.

We found a similar pattern for moderators of the within-person variability of sadness and irritability from one day to the next. Specifically, within-person variability in daily sadness and irritability was stronger for males and for children with fewer baseline depressive and ODD symptoms, lower baseline impairment, and lower average daily irritability. Although this pattern of findings is inconsistent with some prior work linking variability with poor outcomes in adolescents (Silk et al. 2003) and adults (Peeters et al. 2006), our findings are consistent with typical developmental patterns of affective variability, which provide support for the adaptive nature of affective variability in young children compared to older youth and adults. Preschoolers with better functioning demonstrate affective stability relative to their peers but are more variable within their own stable range of emotions. Better emotion regulation, either via internal processes or external regulators such as parents and teachers, might allow these children to return to baseline after experiencing a day with negative emotion, in turn explaining the observed variability in negative affect.

Variability in negative affect is particularly relevant in the DSM taxonomy and thus it may be assumed that those with a depressive disorder have more consistently high depressive mood. Although between-person findings reflect this point (i.e., depressed children have greater negative affect compared to their non-depressed peers; Luby et al. 2003), the patterns of depressive symptoms within depressed preschool-aged children across a two-week period are unknown. Based on the current findings, we might expect that preschoolers with clinical depression demonstrate greater within-person stability of negative mood from day to day and that children with better functioning demonstrate less chronic activation of negative mood, allowing for emotional flexibility and rebound. Moreover, given that we observed that within-person change from day to day was normative during this developmental period, our findings may support Luby et al. (2002) proposed DSM criteria modification that depressive symptoms may be less persistent in young children than adults, underscoring the need for developmental sensitivity in assessing preschool depression. Finally, although no sex differences have been observed in mean levels of irritability (Wakschlag et al. 2012) or in depressive symptoms or diagnoses in early childhood (Bufferd et al. 2011, 2017), examining sex differences in the stability and variability of these behaviors in early childhood may provide more nuanced information and may shed light on the preponderance of girls developing depression in adolescence. Future work is needed to examine day to day mood variability in young clinical samples of males and females with depression.

Strengths and Limitations

The present study had several strengths. First, this is the first study to our knowledge to examine the day-to-day phenomenology of sadness and irritability in a preschool-aged sample. Our findings are a critical next step to identifying normative patterns of common depressive behaviors in preschool-age children and set the foundation for future work examining differences between low-risk community samples and clinical samples. Second, we examined children’s depressive behaviors across a 14-day diary. By leveraging this daily tool, we were able to utilize generalized multilevel modeling to disaggregate between-person and within-person effects. Third, our study used behavioral indicators to assess internal mood states. Externalizing problems demonstrate greater cross-informant agreement than internalizing problems (De Los Reyes et al. 2015), and thus asking parents to report on behavioral indicators of depression may have improved parental identification of markers of children’s internalizing affect.

This study also had limitations that future research should address. First, we asked parents to report on their child’s daily behaviors and only collected this data from one reporter, the majority (93.5%) of whom were mothers. Reliance on a single caregiver’s report may have masked important reporter discrepancies useful for assessment, classification, and treatment. Events occurring outside the context of parental supervision are likely underrepresented and children may show certain behaviors with one caregiver rather than another. Moreover, parents often differ in the amount of time spent with their child each day (Craig 2006) and their report of child symptoms and behaviors (De Los Reyes et al. 2015). Future work should assess behaviors in other contexts (i.e., school/daycare) and involve multi-informant assessment, including co-parent and teacher-reports. Second, we relied on a single-item measure to assess children’s daily sleep quality in order to avoid overburdening parents. More detailed and objective information related to children’s sleep should be obtained using a multi-method approach or the use of multidimensional scales. Third, despite our racially and ethnically diverse sample, limited diversity was observed in socioeconomic status and family structure. Further, we recruited a community sample and findings cannot be generalized to clinical samples. Future work is needed to examine these questions in more socioeconomically diverse populations as well as clinical or higher-risk populations to further establish differences in the phenomenology of depressive behaviors between normative and clinically depressed samples.

Fourth, future studies should consider ways to make the diary more accessible to all populations (lower income families without daily internet access) and randomize question order to reduce possible bias in participant responses. Fifth, we assessed children’s depressive behaviors once daily to increase the feasibility of the study, though it is possible that the day-level information collected may not have been sufficiently fine-grained to understand the complex emotion dynamics occurring in early childhood. It will be important for future studies to incorporate network modeling of intra-individual time-series data to learn more about causal pathways (e.g., Epskamp et al. 2018) and ecological momentary assessment to gain a better understanding of these nuanced affective dynamics and more specific antecedents and consequences to behaviors. Lastly, although parents were asked to complete a diary each day after 6 PM, they were permitted to complete the diary about the prior day’s behaviors up until 12 PM the following day. While we implemented this cutoff time to decrease the likelihood of recall bias, it is still possible that the current day’s behaviors impacted parental recall of the prior day’s behaviors for diary entries completed before 12 PM.

Conclusion

The current study elucidated the daily patterns of these common depressive behaviors in a sample of preschool-aged children and underscores the importance of developmentally-sensitive assessment. We found between-person stability but within-person variability of sadness and irritability in a community sample of preschoolers assessed over 14 consecutive days. Males and children with better functioning demonstrated greater stability relative to their peers and greater variability relative to their own mean sadness and irritability. Our findings provide initial evidence for affective variability as a potentially adaptive response to one’s environment in early childhood. Future research must examine these dynamics in a clinical population to further elucidate differences between normative and atypical patterns of sadness and irritability in this developmental period and determine whether affective variability predicts later psychopathology or impairment. Our findings hold important clinical implications for future prevention and intervention. Knowledge of how a child’s behavior compares to both his/her peers and also his/her own average reflects a critical step in developing medical and psychiatric guidelines for the early identification and screening of children at risk for depression, and can equip primary care providers with the knowledge to normalize parental concerns or suggest additional mental health referrals.

Acknowledgements

This research was supported by the University of Maryland (UMD) College of Behavioral and Social Sciences Dean’s Research Initiative Award (LRD), the UMD Research and Scholars Award (LRD), the California State University San Marcos (CSUSM) Grant Proposal Seed Money Award (SJB), and the CSUSM University Professional Development Award (SJB). We are indebted to the families and staff who made this study possible.

Footnotes

Conflict of Interest All authors declare that they have no conflict of interest to report.

Ethical Approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors.

Informed Consent Informed consent was obtained from all individual participants included in the study.

1

One participant completed only one daily diary, and thus within-person effects were able to be examined for 290 participants, whereas between-person effects were able to be examined for 291 participants.

2

We examined multiple covariance structure specifications by comparing Akaike’s information criterion (AIC) and Bayesian information criterion (BIC) estimates; they had negligible effect on the fixed and random parameter estimates or levels of significance. Covariances were modeled as independent by using the scaled identity covariance structure when models did not converge.

3

The model examining average daily sadness as a moderator of the within-person variability of children’s sadness did not converge when using the AR1 covariance structure; thus, a scaled identity covariance structure was applied to make covariances between days independent and to constrain variances to be constant.

4

The dependent variables of sadness and irritability overlapped with items in the ECI depressive and ODD symptom scales (e.g., loses temper, easily annoyed, is angry/resentful, is irritable). However, results remained similar when overlapping items were removed from the ECI symptom scales.

References

  1. Aiken LS, & West SG (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage. [Google Scholar]
  2. Aiken LS, Mistler SA, Coxe S, & West SG (2015). Analyzing count variables in individuals and groups: Single level and multilevel models. Group Processes & Intergroup Relations, 18(3), 290–314. [Google Scholar]
  3. Aunola K, Tolvanen A, Viljaranta J, & Nurmi JE (2013). Psychological control in daily parent–child interactions increases children’s negative emotions. Journal of Family Psychology, 27(3), 453–462. [DOI] [PubMed] [Google Scholar]
  4. Blaxton JM, Bergeman CS, Whitehead BR, Braun ME, & Payne JD (2017). Relationships among nightly sleep quality, daily stress, and daily affect. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 72(3), 363–372. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bouwmans ME, Bos EH, Hoenders HR, Oldehinkel AJ, & de Jonge P (2017). Sleep quality predicts positive and negative affect but not vice versa. An electronic diary study in depressed and healthy individuals. Journal of Affective Disorders, 207, 260–267. [DOI] [PubMed] [Google Scholar]
  6. Bufferd SJ, Dougherty LR, Carlson GA, & Klein DN (2011). Parent-reported mental health in preschoolers: Findings using a diagnostic interview. Comprehensive Psychiatry, 52(4), 359–369. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bufferd SJ, Dyson M, Hernandez I, & Wakschlag L (2016). Explicating the “developmental” in preschool psychopathology In Cicchetti D (Ed.), Developmental psychopathology (3rdedition) (pp. 152–186). Hoboken, NJ: John Wiley & Sons, Inc. [Google Scholar]
  8. Bufferd SJ, Dougherty LR, & Olino TM (2017). Mapping the frequency and severity of depressive behaviors in preschool-aged children. Child Psychiatry & Human Development, 48(6), 934–943. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Carlson SM, & Wang TS (2007). Inhibitory control and emotion regulation in preschool children. Cognitive Development, 22(4), 489–510. [Google Scholar]
  10. Craig L (2006). Does father care mean fathers share? A comparison of how mothers and fathers in intact families spend time with children. Gender & Society, 20(2), 259–281. [Google Scholar]
  11. Curran PJ, Bauer DJ, & Willoughby MT (2006). Testing and probing interactions in hierarchical linear growth models In Bergeman CS, Boker SM (Eds.), Methodological issues in aging research. Notre Dame series on quantitative methods (pp. 99–129). Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers. [Google Scholar]
  12. De Los Reyes A, Augenstein TM, Wang M, Thomas SA, Drabick DA, Burgers DE, & Rabinowitz J (2015). The validity of the multi-informant approach to assessing child and adolescent mental health. Psychological Bulletin, 141(4), 858–900. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dougherty LR, Klein DN, Olino TM, & Laptook RS (2008). Depression in children and adolescents In Hunsley J & Nash EJ (Eds.), A guide to assessments that work (pp. 69–95). New York, NY: Oxford University Press. [Google Scholar]
  14. Dougherty LR, Leppert KA, Merwin SM, Smith VC, Bufferd SJ, & Kushner MR (2015). Advances and directions in preschool mental health research. Child Development Perspectives, 9(1), 14–19. [Google Scholar]
  15. Egger HL, Ascher BH, & Angold A (1999). The preschool age psychiatric assessment: Version 1.1 Durham, NC: Center for Developmental Epidemiology, Department of Psychiatry and Behavioral Sciences, Duke University Medical Center. [Google Scholar]
  16. El-Sheikh M, Kelly RJ, Buckhalt JA, & Benjamin Hinnant J (2010). Children’s sleep and adjustment over time: The role of socioeconomic context. Child Development, 81(3), 870–883. [DOI] [PubMed] [Google Scholar]
  17. Enders CK, & Tofighi D (2007). Centering predictor variables in cross-sectional multilevel models: A new look at an old issue. Psychological Methods, 12(2), 121–138. [DOI] [PubMed] [Google Scholar]
  18. Epskamp S, van Borkulo CD, van der Veen DC, Servaas MN, Isvoranu AM, Riese H, & Cramer AO (2018). Personalized network modeling in psychopathology: The importance of contem-poraneous and temporal connections. Clinical Psychological Science, 6(3), 416–427. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fabes RA, Leonard SA, Kupanoff K, & Martin CL (2001). Parental coping with children’s negative emotions: Relations with children’s emotional and social responding. Child Development, 72(3), 907–920. [DOI] [PubMed] [Google Scholar]
  20. Fabiano GA, Pelham WE Jr., Waschbusch DA, Gnagy EM, Lahey BB, Chronis AM, et al. (2006). A practical measure of impairment: Psychometric properties of the impairment rating scale in samples of children with attention deficit hyperactivity disorder and two school-based samples. Journal of Clinical Child and Adolescent Psychology, 35(3), 369–385. [DOI] [PubMed] [Google Scholar]
  21. Fiese BH, & Winter MA (2010). The dynamics of family chaos and its relation to children’s socioemotional well-being In Evans GW & Wachs TD (Eds.), Decade of behavior (science conference). Chaos and its influence on children’s development: An ecological perspective (pp. 49–66). Washington, DC, US: American Psychological Association. [Google Scholar]
  22. Finsaas MC, Bufferd SJ, Dougherty LR, Carlson GA, & Klein DN (2018). Preschool psychiatric disorders: Homotypic and heterotypic continuity through middle childhood and early adolescence. Psychological Medicine, 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Fung CHL, Nguyen M, Moineddin R, Colantonio A, & Wiseman-Hakes C (2014). Reliability and validity of the daily cognitive-communication and sleep profile: A new instrument for monitoring sleep, wakefulness and daytime function. International Journal of Methods in Psychiatric Research, 23(2), 217–228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Gruber J, Kogan A, Quoidbach J, & Mauss IB (2013). Happiness is best kept stable: Positive emotion variability is associated with poorer psychological health. Emotion, 13(1), 1–6. [DOI] [PubMed] [Google Scholar]
  25. Houben M, Van Den Noortgate W, & Kuppens P (2015). The relation between short-term emotion dynamics and psychological well-being: A meta-analysis. Psychological Bulletin, 141(4), 901–930. [DOI] [PubMed] [Google Scholar]
  26. Hox JJ (2010). Multilevel analysis. Techniques and applications (2nd ed.). New York: Routledge. [Google Scholar]
  27. Kalmbach DA, Pillai V, Roth T, & Drake CL (2014). The interplay between daily affect and sleep: A 2-week study of young women. Journal of Sleep Research, 23(6), 636–645. [DOI] [PubMed] [Google Scholar]
  28. Könen T, Dirk J, & Schmiedek F (2015). Cognitive benefits of last night’s sleep: Daily variations in children’s sleep behavior are related to working memory fluctuations. Journal of Child Psychology and Psychiatry, 56(2), 171–182. [DOI] [PubMed] [Google Scholar]
  29. Könen T, Dirk J, Leonhardt A, & Schmiedek F (2016). The interplay between sleep behavior and affect in elementary school children’s daily life. Journal of Experimental Child Psychology, 150, 1–15. [DOI] [PubMed] [Google Scholar]
  30. López-Pérez B, Wilson EL, Dellaria G, & Gummerum M (2016). Developmental differences in children’s interpersonal emotion regulation. Motivation and Emotion, 40(5), 767–780. [Google Scholar]
  31. Luby JL, Heffelfinger AK, Mrakotsky C, Hessler MJ, Brown KM, & Hildebrand T (2002). Preschool major depressive disorder: Preliminary validation for developmentally modified DSM-IV criteria. Journal of the American Academy of Child & Adolescent Psychiatry, 41(8), 928–937. [DOI] [PubMed] [Google Scholar]
  32. Luby JL, Heffelfinger AK, Mrakotsky C, Brown KM, Hessler MJ, Wallis JM, & Spitznagel EL (2003). The clinical picture of depression in preschool children. Journal of the American Academy of Child & Adolescent Psychiatry, 42(3), 340–348. [DOI] [PubMed] [Google Scholar]
  33. Peeters F, Berkhof J, Delespaul P, Rottenberg J, & Nicolson NA (2006). Diurnal mood variation in major depressive disorder. Emotion, 6(3), 383–391. [DOI] [PubMed] [Google Scholar]
  34. Röcke C, & Brose A (2013). Intraindividual variability and stability of affect and well-being. GeroPsych, 26(3), 185–199. [Google Scholar]
  35. Silk JS, Steinberg L, & Morris AS (2003). Adolescents’ emotion regulation in daily life: Links to depressive symptoms and problem behavior. Child Development, 74(6), 1869–1880. [DOI] [PubMed] [Google Scholar]
  36. Sin NL, Almeida DM, Crain TL, Kossek EE, Berkman LF, & Buxton OM (2017). Bidirectional, temporal associations of sleep with positive events, affect, and stressors in daily life across a week. Annals of Behavioral Medicine, 51(3), 402–415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Sprafkin J, & Gadow KD (1996). Early Childhood Symptom Inventories. Stony Brook, NY: Checkmate Plus. [Google Scholar]
  38. Sprafkin J, Volpe RJ, Gadow KD, Nolan EE, & Kelly K (2002). A DSM-IV-referenced screening instrument for preschool children: The early childhood Inventory-4. Journal of the American Academy of Child & Adolescent Psychiatry, 41(5), 604–612. [DOI] [PubMed] [Google Scholar]
  39. Stringaris A (2011). Irritability in children and adolescents: A challenge for DSM-5. European Child & Adolescent Psychiatry, 20(2), 61–66. [DOI] [PubMed] [Google Scholar]
  40. Thompson RJ, Mata J, Jaeggi SM, Buschkuehl M, Jonides J, & Gotlib IH (2012). The everyday emotional experience of adults with major depressive disorder: Examining emotional instability, inertia, and reactivity. Journal of Abnormal Psychology, 121(4), 819–829. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Tolan PH, & Dodge KA (2005). Children’s mental health as a primary care and concern: A system for comprehensive support and service. American Psychologist, 60(6), 601–614. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Wakschlag LS, Choi SW, Carter AS, Hullsiek H, Burns J, McCarthy K, Leibenluft E, & Briggs-Gowan MJ (2012). Defining the developmental parameters of temper loss in early childhood: Implications for developmental psychopathology. Journal of Child Psychology and Psychiatry, 53(11), 1099–1108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Wichers M (2014). The dynamic nature of depression: A new micro-level perspective of mental disorder that meets current challenges. Psychological Medicine, 44(7), 1349–1360. [DOI] [PubMed] [Google Scholar]
  44. Wiggins JL, Briggs-Gowan MJ, Estabrook R, Brotman MA, Pine DS, Leibenluft E, & Wakschlag LS (2018). Identifying clinically significant irritability in early childhood. Journal of the American Academy of Child & Adolescent Psychiatry, 57(3), 191–199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Willett JB, Singer JD, & Martin NC (1998). The design and analysis of longitudinal studies of development and psychopathology in context: Statistical models and methodological recommendations. Development and Psychopathology, 10(2), 395–426. [DOI] [PubMed] [Google Scholar]
  46. Zimmer-Gembeck MJ, & Skinner EA (2011). The development of coping across childhood and adolescence: An integrative review and critique of research. International Journal of Behavioral Development, 35(1), 1–17. [Google Scholar]

RESOURCES