Who Looks on the Bright Side? Expectations of Low-Income Parents with a Disruptive Young Child

Abstract

Low-income parents are more likely to have a child with disruptive behaviors. Furthermore, these parents are likely to struggle with low expectations about the future, which can interfere with treatment response to even the most effective treatments. The purpose of this study was to explore correlates of low-income parents’ expectations, specifically hopelessness and optimism, in families with clinic-referred disruptive young children. Using baseline data from an intervention trial (Jones et al., 2021), we used a multi-informant approach to test whether parenting behaviors, child misbehavior, and parental daily stressors were related to parent hopelessness and optimism. Results based on the Everyday Stress Index indicated that only daily stressors, particularly those related to relationships and responsibilities, were related to parent feelings of hopelessness and optimism. Results were interpreted through transactional and ecological systems theories, particularly the Family Stress Model (Conger & Conger, 2002), as well as a low-income parent’s self-efficacy as a caregiver. Clinical implications are discussed.

Across the large, diverse population of low-income families, parents and young children are at-risk of poor physical and psychological outcomes (Brooks-Gunn & Duncan, 1997; Chaudry & Wimer, 2016; Scaramella et al., 2008; Sullivan et al., 2019, 2021), including child disruptive behaviors (Jones et al., 2016). Importantly, parent coping skills and their children’s developmental trajectories vary widely across families with limited resources (Hustedt et al., 2017). Offering insight into this variability are parent expectations. Some parents appear to have more positive expectations, including elevated levels of optimism and hope, and these positive expectations appear to translate to better parenting (Baker et al., 2005; Kashdan et al., 2002; Loh et al., 2017) and child well-being (Jones et al., 2002). Accordingly, parents who endorse higher levels of pessimism and hopelessness appear to be at-risk of poor parent and child-level problems. A systems-wide view of the complex transactions between parenting, child behavior, and stress may offer insight into what contributes to parent expectations.

The purpose of this study is to identify potential factors that are related to parent expectations in an at-risk, treatment-seeking sample of low-income families in a secondary analysis of baseline data originally collected in the context of an intervention trial (Jones et al., 2021). First, we outline how behaviors specific to parents, children, and their dyadic relationship are related to parental expectations, drawing from Bronfenbrenner’s (1977) ecological and Sameroff’s (1975) transactional theories. Then, we discuss specific expectations pertinent to low income, treatment-seeking families, as well as how such expectations may promote or impair treatment response. We hypothesized that constructs at each proposed system level (i.e., parent, child, and parent-child) will relate to parent expectations in a low income sample of parents seeking treatment for disruptive behavior in their young children.

A Transactional and Ecological Perspective on Parent Expectations

Parental expectations are likely contingent on activity across multiple different systems; parenting practices, child behavior, and external stress all play a role in the degree to which a parent is hopeful or optimistic. Bronfenbrenner (1977) originally proposed an ecological systems model wherein child development is a function of a series of interlocking contexts that vary from proximal to increasingly distal. Evolution of this theory, however, focused on the importance of “enduring forms of interaction” in shaping a child’s developmental trajectory (Bronfenbrenner & Morris, 1998, p 996). Sameroff’s (1975) seminal work complements Bronfenbrenner’s, arguing that actors in each of these systems influence the others. Taken together, processes that spur development forward are transactional and reciprocal, shaping both parent and child as they unfold. Parent-child interactions constitute one such process, and these interactions are particularly important for young children. For example, a parent may praise a child for being calm, resulting in a more even-keeled parent and a child who is more likely to engage in future calm behavior. In disruptive young children, two ways of measuring the quality of parent-child interactions are child misbehavior and parenting skills. Therefore, it is important to consider both child and parent characteristics separately, as well as behaviors and factors that link the two within a family context (e.g., stress, parenting).

In addition to parent-child interactions, Belsky (1984) highlights the critical role contextual stress plays in shaping parenting and parent expectations. Low-income parents raising a child with disruptive behavior problems are doubly at-risk. First, parents cope with stress specifically related to financial strains, including poorer living conditions and higher degrees of family instability (Evans, 2004). Second, regardless of socioeconomic status, parents of children with disruptive behavior problems report higher levels of parenting stress and family conflict than parents of non-clinic referred children (Deault, 2010; Theule et al., 2013). Coupled together, caring for a child with disruptive behaviors in the context of financial strain and associated stressors likely influences the way parents think about themselves, the world, and their parenting, translating to negative expectations.

Parental Expectations in Treatment-Seeking Families

Parental expectations are particularly critical in the context of treatment-seeking families. Behavioral Parent Training (BPT), which involves both the child and the parent as the agents of treatment, is one of the most effective evidence-based treatments for promoting child and family well-being; Comer et al., 2013; Evans et al., 2013; Higa-McMillan et al., 2016; Stoll et al., 2020). Further, recent evidence suggests BPT may be just as effective for low income families relative to families with more resources (e.g., Dedousis-Wallace et al., 2020). This effect is heartening, given that children from low income families are more likely to struggle with disruptive behavior disorders (Jones et al., 2016). However, families who are socially disadvantaged are more likely to encounter obstacles (e.g., difficulty accessing a clinic, concerns about stigma) resulting in attrition and poor engagement (Chacko et al., 2016; Jones et al., 2021; Nock & Kazdin, 2005), which may result in poor treatment response (Lundahl et al., 2006; Leijten et al., 2013; Reyno & McGrath, 2006). Accordingly, better understanding factors related to parent engagement is critical in enrolling and retaining underserved families in BPT programs–a necessary step along the path toward equitable access to evidence-based psychological services.

Research supports the role of parental expectations in explaining some of the variability in therapy outcome (e.g., Jiang et al., 2016; Kaiser et al., 2010). For example, a parent with negative expectations may view their parenting and their child’s behavior as uncontrollable and immutable, disadvantaging them from a positive outcome, before even beginning their first treatment session (Durand et al., 2009). Current treatment engagement literature focuses on parental attributions about child behavior, despite theoretical models highlighting the importance of more global parent cognitions (Pereira & Barros, 2019). Accordingly, understanding what contributes to low-income parents’ global positive and negative expectations may translate to more effective intervention among this group of underserved families.

Two contrasting, albeit related, indicators of parental expectations are hopelessness and optimism (Chang et al., 1994; Kashdan et al., 2002). Parent hopelessness, which is the anticipation of bad outcomes, engenders a sense of helplessness to control outcomes (Abramson et al., 1989; Cohen et al., 2017). In contrast to hopelessness, optimism describes a positive outlook on the future. Recent research suggests optimism is socially patterned, with people of higher socioeconomic status being more likely to report more optimistic frameworks (Boehm et al., 2015). Parental optimism is associated with more effective parenting (Jones et al., 2002), particularly when parents contend with children with disruptive behaviors (Koenig et al., 2010), and with improved long-term child behavior (Durand et al., 2009). When parenting in multiply-stressed families, such as those with low incomes and disruptive children, hopelessness may prevent, while optimism may promote, engaging with and benefiting from evidence-based treatment.

The Current Study

In a sample of families with clinic-referred disruptive young children, the purpose of this study was to explore correlates of low-income parents’ expectations, specifically hopelessness and optimism. The current study drew from Bronfenbrenner’s (i.e., Bronfenbrenner & Morris, 2006) bioecological systems model to theorize how three constructs central to child and parent well-being relate to parenting expectations: child misbehavior, parenting, and daily stressors. We hypothesized that, after accounting for child and parent demographics, parenting, child behavior, and daily stressors would each explain variability in parent expectations hopelessness and optimism. Specifically, we expected that higher levels of negative parenting, child problem behaviors, and daily stressors, as well as lower levels of positive parenting, would relate to more hopelessness and less optimism.

Our operationalization of study constructs stems from the following rationale. As parents were seeking support for child disruptive behavior, we specifically measured this type of child behavior (McMahon & Forehand, 2003). We assessed both positive and negative parenting practices, as parents responded to recruitment materials to enhance their caregiving skills, and, in turn, their child’s behavior. We used both parent self-report and behavioral observations of parent-child interactions to comprehensively assess parenting, given data suggesting observations of parenting are particularly important when studying low-income families (Herbers et al., 2017). We assessed daily stressors using parent perceptions of stress, which have been reported to be better indicators than more objective measures (e.g., income) (Hustedt et al., 2017). As low-income families cope with a wide variety of stressors, we drew from previous research to evaluate two specific facets of daily stress: 1) coping with hardship (i.e., material and financial) and 2) managing relationships and responsibilities.

Method

The current study is a secondary data analysis of baseline data from a project examining strategies to increase engagement in BPT (Jones et al., 2021). The IRB approved all study procedures.

Participants

One hundred and one low-income (100–250% of Federal Poverty Guidelines) parents and their 3-to-8 year old child with clinically significant disruptive behaviors (Eyberg Child Behavior Inventory Problem score > 15 or Intensity score > 131; Eyberg & Pincus, 1999) constituted the original sample (Jones et al., 2021). Approximately 80% of these families had applied for government assistance (e.g., Medicaid, unemployment benefits, Food Stamps) in the last year. Exclusion criteria included the presence of parental child abuse/neglect, significant developmental delays of the child, and current parental psychotic or substance use disorders. Parents were also excluded if they met diagnostic criteria for a current depressive episode as defined by meeting criteria for a Major Depressive Episode on the Mini International Neuropsychiatric Interview (Sheehan et al., 1998) and had a score indicating a severe depressive episode (score > 27) on the Beck Depression Inventory (Beck et al., 1996).¹ The demographic characteristics of the sample are presented under Child and Parent Characteristics in Table 1 (see Results section).

Table 1.

How Demographics and Processes Relate to Parent Expectations: Hopelessness and Optimism

				Correlating With
Systems	M / %	SD	Range	Hopelessness	Optimism
Child Characteristics
% female (1 =girl; 0=boy)	44%			−.07	.03
Age (years)	4.61	1.27	2.92–7.83	.03	.08
Parent Characteristics
Age (years)	31.91	6.02	20–48	−.08	.15
% female	93.8%			−.08	.04
% Underrepresented	34%			.02	.15
Race:
Black or African American	20%
White	69.8%
American Indian or Alaskan Native	1%
More than one race	8.3%
Ethnicity:
Not Hispanic/Latino	93.8%
Hispanic/Latino	6.3%
Education:
High school or less	8%			.03	−.14
Some college or vocational training	28%			−.13	.01
College	47%			.08	.03
Advanced degree	17%			.03	.06
Parenting
Positive
Observed attends & rewards per min.	1.06	.86	0–4.14	.18†	−.09
Parent report1	49.72	5.70	36–60	−.11	−.03
Negative
Observed questions & instructions per min.	3.97	1.95	.79–10.22	.02	−.06
Parent report1	17.38	3.89	9–27	.05	−.04
Child Behavior
Number of problems2	22.94	6.19	4–36	.04	.11
Intensity of problems2	153.14	28.80	103–232	.06	.08
Daily Stressors			21–63	.52*	−.31**
Hardship				.46**	−.23*
Responsibilities/Relationships				.52**	−.28**

Note:

^*p < .001;

^**p = .002;

^†p = .08;

¹= Alabama Parenting Scale (Shelton et al., 1996);

²= Eyberg Child Behavior Inventory (Eyberg & Pincus, 1999);

³= Everyday Stressors Index (Hall et al., 1985, Hall, 1987; Hall et al., 1985)

Procedure

Families were recruited via word-of-mouth, local schools, and advertisements targeting areas (e.g., workplaces, retail outlets, & social media) with an overrepresentation of low-income parents. In addition, social service, health care, and other agencies that serve low-income families (see Jones et al., 2021, for details) were targeted for recruitment. Advertisements used phrasing such as “does your child test your limits?”, “are you having trouble finding or paying for services?”, and “get free help managing your child’s behavior” to attract low-income families with children with elevated disruptive behaviors.

Interested parents contacted a project research assistant. The research assistant then conducted a phone screen to assess key eligibility criteria. Eligible families were then invited to an in-person interview to obtain consent, confirm eligibility, and complete self-report measures. The behavioral observation task was conducted at the beginning of the next session.

Measures

Parent & Child Demographics

The parent completed a measure of demographic information about themselves (e.g., age, gender, education), the participating child (e.g., age, gender), and their family (e.g., household income).

Parenting Practices

We used behavioral observations and self-report data to assess parenting. Behavioral observations of both positive and negative parenting occurred during a parent-child clinic observation (i.e., Child’s Game) which averaged five minutes (see McMahon & Forehand, 2003). In the context of child-directed play (i.e., the Child’s Game), parental verbalizations, such as questions and instructions, detract from the focus on the child and developing positive, reinforcing interactions, whereas praise, rewards, and close attention to the child’s appropriate behavior promote the development of positive parent-child interactions (McMahon & Forehand, 2003). To capture this dynamic, four behaviors were recorded: (a) Attends (positive attention in which the parent provides an ongoing verbal description of what the child is doing; e.g., “You are stacking the blocks”), (b) Rewards (positive attention that is provided following the child’s appropriate behavior; e.g., “Good job!” “Thank you for picking up your toys!”), (c) Questions (an interrogation to which the only appropriate response is verbal), and (d) Instructions (parent-issued command toward the child). Attends and Rewards are considered “Do” Behaviors (positive behaviors to increase) and thus a facet of positive parenting, whereas Questions and Instructions in this context are labeled as “Don’t” Behaviors (negative behaviors to decrease) and associated with unhelpful (i.e., negative) parenting (Bell & Eyberg, 2002).

Undergraduate research assistants assessed parenting behaviors from videotapes of the parent-child interactions. The coders received approximately 50 hours of training. Prior to beginning coding, undergraduate research assistants reached at least 80% agreement with expert coders on a series of training parent-child observation videos.

More than half (51%) of the videos were double coded for fidelity. When two coders failed to reach 80% agreement, they met in person and jointly coded the observation, resolving outstanding discrepancies. This occurred in 45% of the observations (agreement prior to resolving discrepancies was 70% for Attends + Rewards and 80% for Instructions + Questions). We used the resolved discrepancy scores in these analyses, effectively resulting in data with 100% agreement across coders. Behaviors are reported as a rate per minute to account for variability in interaction length.

Parents completed an abbreviated version of the Alabama Parenting Questionnaire (APQ), a self-reported measure of positive and negative parenting practices (Shelton et al., 1996). Numerous studies have since examined the APQ (e.g., Clerkin et al., 2007; Essau et al., 2006), and reported adequate reliability and validity data. Of relevance, Clerkin et al. (2007) examined the psychometric properties of the APQ in a preschool population. A three-factor solution emerged with the following factors: positive parenting, negative/inconsistent parenting, and punitive parenting. In the current study, we used the 12-item Positive Parenting subscale (e.g., You play games and do other fun things with your child; α =.81) and the seven-item Negative Parenting subscale (e.g., You threaten to punish your child and then do not actually punish him/her; α = .63). Items pertaining to punitive parenting were left out of the current study due to IRB restrictions.

Child Behavior Problems

We operationalized disruptive behavior using the quantity and intensity of problem behaviors parents endorsed on the Eyberg Child Behavior Inventory (ECBI) (Eyberg & Pincus, 1999). On this form, parents indicate whether a behavior (e.g., whining, yelling or screaming) is currently a problem, and, if so, the intensity of the behavior or how often it occurs using a Likert scale ranging from 1 (never) to 7 (always). Problem scores range from 0 to 36 and intensity scores range from 36 to 252. The ECBI has adequate reliability and validity data (Eyberg & Pincus, 1999). In this sample, reliability estimates were α = .85 for the problem subscale and α = .91 for the intensity scale.

Daily Stressors

Parents reported on their experiences of stress using the Everyday Stressors Index (ESI) (Hall, 1983, 1987; Hall et al., 1985). This instrument consists of 20 items assessing how much (ranging from [1] not at all to [4] bothered a great deal) a variety of stressors bother each parent (example item: “How much are you bothered by having too many responsibilities?”). Scores range from 20 to 80 with higher scores indicating more stress. The measure has adequate reliability and validity data (e.g., Hall et al., 1985). The alpha coefficient for the current sample was α = .85.

Parent Expectations

We operationalized expectations using two scales: the Beck Hopelessness Scale (BHS; Beck & Steer, 1990; Beck et al., 1974) and the Life Orientation Test – Revised (LOTR; Scheier et al., 1994). The BHS uses 20 items to assess hopelessness. Each item presents a statement (e.g., I don’t expect to get what I really want) to which a parent indicates True or False. The BHS has substantial reliability and validity data (e.g., Beck et al., 1974; Innamorati et al., 2013). In this study, we generated a total score, with a possible range of 0 to 20 (α = .86). Higher scores indicated higher levels of hopelessness.

The LOT-R assessed how optimistic parents felt. Parents rate the extent to which, from 0 (strongly disagree) to 4 (strongly agree), they agree with each of 6 statements (e.g., In uncertain times, I usually expect the best). This measure has substantial reliability and validity data (e.g., Scheier et al., 1994; Solberg Nes et al., 2009) (for current sample α = .74). Higher scores represent more optimism.

Results

Data Analytic Plan

First, we conducted preliminary analyses to evaluate normality, examine the extent and mechanism of missing data, and characterize the sample of parents. We then ran bivariate correlations of each independent variable with the two indicators of expectations to assess which variables were related to optimism and hopelessness. Next, we conducted two, five-block hierarchical regression analyses in SPSS to examine the relationship of parenting, child disruptive behavior, and daily stressors with caregiver expectations after controlling for child and parent demographics. We bootstrapped results 5000 times and report robust standard errors to account for skew. In the first block, we entered the child characteristics age and gender. In the second block, we created four different binary variables to represent parent education level (high school or less, some college, college, and advanced degree). We then entered dummy-coded parent education level and a binary underrepresented status variable which indicated whether or not the parent identified as Hispanic, Latino, White, Black, Native American or Alaskan Native, or more than one race. We created a binary variable (White vs. Underrepresented Minority) given that the overall prevalence of each unique ethnic or racial category was small. In the third block, we entered self-reported and observations of parenting behavior. In the fourth block, we entered parent reports of the number and intensity of child problem behaviors. Finally, in the fifth block, we included factor scores (i.e., hardship and relationships/responsibilities) representing parent reports of stress derived from the ESI (Hall, 1983, 1987; Hall et al., 1985).

Preliminary Analyses

Data inspection indicated one outlier which we eliminated from analyses.² From the remaining sample (n = 100), one parent did not report race or ethnicity, and three were missing behavioral observation data due to video recording malfunctioning, resulting in 96 complete cases. All four missing cases identified as female and Non-Hispanic. Two identified as Black, one opted out of reporting race, and one was White. We elected to use listwise deletion, given research suggesting that cases representing fewer than 5 % of the sample are unlikely to bias findings (Graham, 2009).

The means (or percentages for categorical variables), standard deviations, and observed range of each independent variable for complete cases are presented in Table 1. Of note, slightly less than half of the children were female. About one-third of parents identified as an underrepresented minority with the great majority of participating parents being female. Behavioral observations of parenting indicated the “Don’t” (negative) behaviors occurred at a rate over 3.7 times higher than “Do” (positive) behaviors. In contrast, parents self-reported that positive parenting behaviors occurred substantially more frequently, with a positive parenting item-level mean equaling 4.14 on a one to five scale, than negative parenting behavior, with a negative parenting item-level mean equaling 2.48. On average, parents reported the occurrence of almost 23 of the possible 36 child problem behaviors, with average intensity levels well surpassing clinical cutoffs. On average, parents reported being “a little bothered” by everyday stressors (mean item score = 1.75).

Regarding our outcome constructs, on average, parents reported normal levels of hopelessness (M = 2.63, SD = 3.27), with 4 % of the sample endorsing moderate to severe levels of hopelessness (Beck & Steer, 1993). Parents endorsed levels of optimism (M = 21.56, SD = 5.27) consistent with another sample of parents with children with cancer (Fayed et al., 2011). Hopelessness and optimism were moderately negatively correlated (r[94] = −.50, p < .001).

Recent research with the ESI suggests reliable and valid factors exist within this measure (Damron, 2016; Gissandaner et al., 2021). Thus, building from factors previous researchers have identified (i.e., Damron, 2016; Gissandaner et al., 2021), we conducted a confirmatory factor analysis to identify factors which may differentially relate to our two outcomes.

The factor measurement model of the ESI indicated two factors, moderately correlated, with six and eight items loading on the first and second factors, respectively (χ² [74] = 76.54, p = .40, RMSEA = .019, 90% CI [.000, .061], CFI = .991, SRMR = .065, r [94] = .68, p < .001). Like previous research (Damron, 2016; Gissandaner et al., 2021), we labeled these factors as Hardship and Relationship/Responsibility, respectively. The Hardship factor assesses the extent to which parents lack resources to meet their basic needs (e.g., I have problems with having not enough money for basic necessities such as clothing, housing, food, and health care), and the Relationships/Responsibilities factor assesses how much stress results from interpersonal challenges and overwhelming daily responsibilities (e.g., I have too many responsibilities). Standardized item loadings ranged from .55 to .85 for the Hardship factor and .29 to .64 for the Relationships/Responsibilities factor.

What Constructs Relate to Parent Expectations?

Correlations between each independent and dependent variable (hopelessness & optimism) are also reported in Table 1. Daily stressors was the only variable correlated with hopelessness and optimism. In partial support of study hypotheses, as parents reported increased levels of stress, both in relation to material hardship as well as stress due to responsibilities and relationships, hopelessness increased and optimism decreased. Neither child characteristics, parent characteristics, parenting, nor child behavior was related to hopelessness or optimism.

Fig. 1 summarizes the results of the two hierarchical regression analyses for the three constructs of interest: parent behavior, child behavior, and daily stress. Congruent with the correlational analyses, regression steps including child characteristics,³ parent characteristics³, child behavior, and parenting behavior did not explain variance in hopelessness or optimism (p values for all F Δ statistics > .20). Only the fifth block, in which we entered factor scores for the Hardship and Relationships/Responsibilities factors (McNeish & Wolf, 2020), explained variance in hopelessness (R² = .35, R² Δ = .28, F Δ [2, 81] = 17.24, p < .001) and optimism (R² = .20, R² Δ = .13, F Δ [2, 81] = 6.64, p = .002).⁴ As presented in Table 2, parents’ self-reported stress falling under the Hardship factor was unrelated to either hopelessness or optimism; however, increased stress falling under the Relationship/Responsibilities factor was related to significantly higher levels of hopelessness and lower levels of optimism.

Fig. 1.

The Relationship of Each Block of Independent Variables with Hopelessness and Optimism after Controlling for Child and Parent Demographic Characteristics. Note: Solid lines indicate that R² Δ was significant at p ≤ .002. Under the Daily Stress block, only the Relationships/Responsibilities factor was significantly related to outcomes

Table 2.

Final block of hierarchical regressions for outcomes hopelessness and optimism

Outcome	Predictor	b	BCa 95% CI
Hopelessness	Child age	.28	[−.21, .82]
	Child gender	−.62	[−2.08, .64]
	Parent education
	High school or less	−.93	[−3.37, 1.22]
	Some college or vocational school	−1.34	[−2.86, .09]
	College or further	−1.12	[−3.24, 1.18]
	Parent minority status	−.45	[−1.80, 1.07]
	Positive parenting
	Observed attends & rewards per min.	.47	[−.50, 1.38]
	Parent report	.04	[−.06, .13]
	Negative parenting
	Observed questions & instructions per min.	.01	[−.33, .36]
	Parent report	.003	[−.12, .13]
	Child behavior
	Number of problems	−.11	[−.25, .03]
	Intensity of problems	.003	[−.02, .03]
	Daily Stressors
	Hardship	.39	[−1.01, 2.13]
	Relationships/responsibilities	5.66	[2.36, 9.19]
Optimism	Child age	−.06	[−1.04, .92]
	Child gender	.98	[−1.18, 3.19]
	Parent education
	High school or less	−1.42	[−5.91, 3.59]
	Some college or vocational school	.34	[−2.71, 3.21]
	College or further	1.94	[−.94, 4.51]
	Parent minority status	2.15	[−.29, 4.30]
	Positive parenting
	Observed attends & rewards per min.	−.40	[−1.74, 1.09]
	Parent report	−.12	[−.35, .08]
	Negative parenting
	Observed questions & instructions per min.	−.05	[−.76, .69]
	Parent report	−.05	[−.36, .23]
	Child behavior
	Number of problems	.20	[−.18, .55]
	Intensity of problems	−.001	[−.07, .06]
	Daily Stressors
	Hardship	.28	[−2.40, 2.91]
	Relationships/responsibilities	−7.62	[−13.10, −2.55]

NOTE: bs are unstandardized; BCa = bias-corrected and accelerated; bolded text indicates p < .01

Discussion

In partial support of study hypotheses, findings suggested that daily stressors, specifically those pertaining to parent responsibilities and relationships, were related to parental expectations of hopelessness and optimism. Unexpectedly, after accounting for child and parent demographic factors, neither parenting skills nor child misbehavior, indicators of parent-child interactions, were associated with parent expectations. Further, stress related to financial hardship was unrelated to parent expectations.

We were surprised to find neither parenting nor child misbehavior was associated with parenting expectations, particularly as our sample consisted of parents seeking skills to better manage their child’s misbehavior. At the time of seeking treatment, it may be the case parents were less concerned about their parenting skills and severity of child misbehavior as they were actively involved in seeking support for ameliorating these specific factors. Parents are more likely to consider their child’s misbehavior as less problematic when they are receiving treatment (Johnston & Leung, 2001). Whereas parents may feel more in control of their parenting and child’s behavior, they likely still feel powerless over other stressors outside of the scope of treatment.

In line with hypotheses, parental day-to-day stress was related to parent expectations. Given the current sample of financially-strained families, we were surprised to find that only stress attributed to having too many responsibilities and too few or strained relationships, and not financial hardship, was related to hopelessness and optimism. However, for low-income families, stress may be the largest factor in determining parental hope and optimism. In contrast, parenting and child behavior may be secondary to the daily stressors of too much responsibility and insufficiently supportive relationships.

Transactional and ecological systems theories (e.g., Family Stress Model; Conger & Conger, 2002) undergird the well-established link between stress, parenting, and child behavior problems, particularly for low-income families; however, in theories such as the Family Stress Model, economic pressure and poor parental well-being are first in the chain of factors predicting poor child well-being. Whereas our use of cross-sectional data precludes examination of the sequential contributions of the independent variables we examined, evidence from the Family Stress Model provides a theoretical framework for conceptualizing their respective order. Accordingly, parenting and child problem behaviors may only contribute to expectations in the context of daily stress.

Still unclear, however, is why does stress attributed to many roles, responsibilities, and strained relationships account for variance in parental expectations, where stress attributed to final hardship does not? One aspect of this factor, strained relationships, may indicate that limited social support among low income parents of children with disruptive behaviors contributes to low parental expectations. Perceived social support is a well-documented resilience promoting factor among low-income mothers (Taylor & Conger, 2017). When systems of welfare underperform, informal social networks may provide material support when unexpected financial challenges arise (e.g., being able to borrow a friend’s car if unexpected car trouble arises, watching a friend’s child in the event of a last-minute work issue; Radey, 2018; Wadsworth, 2012). In addition to tangible aid, supportive relationships function as a coping resource for all persons, particularly those coping with stress associated with living near poverty. For instance, romantic partners can provide comfort in the context of economic pressure (Conger et al., 1999). Given the critical role supportive social networks play in the well-being of low-income families, stress in this domain may be particularly disheartening.

The literature on a construct closely related to hopelessness and optimism, parenting self-efficacy (i.e., a parent’s belief in their ability to effectively serve as a caregiver), may also offer some insight (refer to Rand, 2018 for a detailed explanation on the similarities and differences between these three constructs). Conceptually, parent self-efficacy appears to be closely tied to managing many responsibilities and having positive adult relationships. When responsibilities and relationships become overwhelming, parent self-efficacy decreases, potentially leading to poor parental and child adjustment (Albanese et al., 2019). Parents of children with disruptive behavior problems report higher levels of parenting stress and family conflict than parents of non-clinic referred children (Deault, 2010; Theule et al., 2013), and this higher level of stress relates to lower parenting self-efficacy (Jones & Prinz, 2005). As such, parents coping with elevated stress who perceive themselves as less able to fulfill their caregiver role are likely to feel less optimistic and more hopeless, and the stress associated with these concerns may outweigh stress of hardship. Finally, it is important to note that we assessed financial strain in low income families only; financial strain may emerge as a significant contributor to parental expectations in a sample of families spanning a broader range of incomes.

Limitations and Strengths

Several limitations characterized this study. Overall, hopelessness scores were low, and thus identifying a signal between potential factors and hopelessness may have been more difficult. Importantly, however, such low scores are promising from a clinical perspective, indicating that parents seeking treatment had some hope that future well-being could be better. Indeed, we anticipate the scale of study findings to only increase in an at-risk but non-treatment seeking sample, as treatment-seeking families may inherently evidence higher levels of hope, motivation, and optimism. Cross-sectional data preclude forming firm conclusions on causality and direction of effects. Additionally, one construct that we did not assess that may help explain the findings: parenting self-efficacy. Further, whereas families in this sample were low income, few met criteria for being in poverty (<100% Federal Poverty Line) and a relatively large percent reported college-level educations. Future research with a higher-risk sample of families (e.g., meet federal guidelines for poverty) should attempt to replicate these findings. Finally, there was a high degree of racial and ethnic variability both between and within families, which, in conjunction with our modest sample size, provided insufficient power to analyze whether race or ethnicity was related to parent expectations. As experiences of discrimination may be related to increased levels of hopelessness and decreased feelings of optimism for parents in various contexts, it may be the case that families who are Black, Indigenous, and people of color experienced higher levels of negative expectations than other families from non-marginalized racial and ethnic groups. Future research focusing on the relationship between discrimination and parent expectations is necessary.

This study also has several strengths. First, it is one of few studies to examine hopelessness and optimism in a sample of treatment seeking low-income families who are at-risk for child problem behavior yet underserved. Extant research on parent expectations largely consists of middle-class, White samples. As low socioeconomic status is related to negative parent expectations (e.g., Nock & Kazdin, 2001), diminished treatment engagement (Chacko et al., 2016; Jones et al., 2021; Nock & Kazdin, 2005), and possibly poor treatment outcomes (e.g., Leijten et al., 2013), an in-depth examination of these questions in a low-income sample is necessary. Extrapolating from relatively small percentages of low-income families in extant studies is insufficient for understanding the role of parental expectations in these families. Second, we examined multiple potential contributors (parent behavior, child behavior, stress) to parent expectations. Third, we used a multi-informant framework, drawing from self-reported data as well as behavioral observations.

Clinical Implications

Our findings indicate that relationship strain and stress from overwhelming responsibilities link with parent expectations, above and beyond other factors such as parenting and child misbehavior. Especially in a sample of low-income families, parent expectations are an attractive intervention avenue: they are likely much more malleable than other, external factors (i.e., parent income, living conditions; Hoza et al., 2006). As such, sensitive and responsive clinicians attuned to considerations specific to low-income families may benefit from including parental perceptions of stress, particularly those related to their relationships and responsibilities, in their case conceptualizations of disruptive child behavior. Problem solving ways to enhance relationships, and, if possible, reduce responsibilities may be beneficial. Further, parent-based interventions in populations of caregivers dealing with elevated stress may need to include a larger focus on parent coping (see Chronis-Tuscano et al., 2020). For instance, employing cognitive restructuring techniques on unhelpful expectations (e.g., my child’s behavior will never change) may facilitate increased helpful cognitions that build a better basis for connection with the child and effective discipline (e.g., my child has the ability to behave better) (Durand et al., 2013). Parenting treatment may also benefit from including caregiver affect regulation skills, drawing from treatments such as Dialectical Behavioral Therapy (Linehan, 2014) or Acceptance and Commitment Therapy (Hayes et al., 2016). Cognitive restructuring, emotional regulation, and acceptance will all aid in promoting parent cognitive flexibility and openness to change, even while coping with elevated stress levels. Finally, interventions specifically targeting parenting optimism (i.e., Durand’s, 2011 Optimistic Parenting; Durand et al., 2013) may be particularly helpful when individualizing parent training protocols for low-income populations.

Data Availability:

Work with data is ongoing and have thus not yet been deposited. Data are available upon request with approval.