Assessing Parenting Stress

Parental Stress

Parental stress is a steadily growing area of research. A review of the psychological literature revealed a consistent growth in the extant literature over the past 40 years. A PsycINFO® search conducted on February 15, 2013, using search terms “parent” and “stress” evidenced this growth; we depict this in Figure 1. The total publications for articles published in English consistently increased over time: 1971–1980: 301; 1981–1990: 1,533; 1991–2000: 2,672; and 2001–2010: 4,436.

OPEN IN VIEWER

Figure 1.

English-language publications about parent stress over the last four decades.

Parental stress is positively correlated with child behavioral and developmental maladjustment; more parental stress is associated with more child problems. During a child’s infancy, high levels of parental stress can negatively impact the quality of caregiving, which may interfere with parent–child attachment. Likewise, a baby’s sense of security is impacted by parental stress, especially when the parent has difficulty with their own emotion regulation or maintaining familiar routines for the baby (Figner, Mackinlay, Wilkening, & Weber, 2009). Parent–child reactions can act in synchrony such that infants with a diminished sense of security display anger or distress, consequently eliciting higher levels of parental stress. From toddlerhood to adolescence, increased parental stress can create a chaotic family environment that contributes to children’s behavioral problems (Coldwell, Pike, & Dunn, 2008). Regarding emotional development, children in high-stress households may experience feelings of powerlessness that result in children having low self-esteem and high anxiety (Fiese & Winter, 2010).

Parents’ life stress is also associated with poor parent–child relationships (Berry & Jones, 1995). Family circumstances such as poverty have been found to affect child adjustment, and there is evidence to suggest that this is moderated by parenting stress (Conger, Ge, Elder, Lorenz, & Simons, 1994; Steele et al., 2016). Daily stressors can be exacerbated by poverty further taxing one’s emotional resources for effective parenting. For example, the stress of bills and managing finances are greater when one is unable to pay overdue bills, work and getting children to school is more stressful when one lacks reliable transportation, and living in high-crime neighborhoods can tax parental cognitive and emotional resources (Conger & Donnellan, 2007), which can have a secondary negative impact on child development. Furthermore, life stressors can have a cumulative negative effect, which can result in children who are raised in poor, single-parent families being cognitive and emotional disadvantaged relative to peers who live in middle-class families (Berk, 2013).

Individual Differences in Parenting Stress

Family composition and family transitions are also predictive of stress levels within the family. Single parenting is associated with higher levels of stress than is coparenting (Waldfogel, Craigie, & Brooks-Gunn, 2010). Families undergoing transitions, such as two households blending into a single family, is stressful for adults and for children despite the typically positive circumstances under which families become blended (Hetherington, 1992; Maguire, 2015). Responses to family composition, particularly families adjusting to new stepparenting and half-sibling roles, vary by children’s genders, ages, and by types of blended families. Mother–stepfather families are the most common form of blended family. When mothers retain custody and remarry, boys appear to adjust well, whereas girls appear to be more negatively affected by having a new stepfather (Bray, 1999; Hetherington, 1992; Visher, Visher, & Pasley, 2003). In contrast, in heterosexual marriages, when fathers retain custody following remarriage, children of both genders typically have an initial negative reaction to the new arrangement (Buchanan, Maccoby, & Dornbusch, 1996). Regardless of gender or type of blended family arrangement, children school age and older tend to have more difficulty transitioning to a blended family (Hetherington & Stanley-Hagan, 1999). Researchers have also examined families with same-sex couple parents and have found that children thrive at least as well as do children who are raised in homes with heterosexual parents, and family stresses do not appear to be affected by parents’ sexual orientation (Box, Know, Van Rijn-Van, & Gartrell, 2016; Goldberg, 2010; Golombok et al., 2003).

In the United States, there are a growing number of children being parented by grandparents. Grandparenting can be a financial burden if grandparents are retirees; retirees may be ill-prepared for the unanticipated expenses, particularly in cases such as kinship foster care (Chen, Mair, Bao, & Yang, 2015; Ehrle & Green, 2002). National survey data have revealed that kinship foster care, which does not receive financial support from social services, is more likely to occur in single-parent or grandparent households in which the families are living below the poverty line (Ehrle & Green, 2002).

Families that experience frequent transitions or separations experience a unique set of stressors. Military families and families with an incarcerated parent experience myriad challenges and stresses in parenting for both the “at home” and the “away” or distal parent. For example, military parents experience high rates of parental stress as a result of military demands including frequent relocation (Finkel, Kelley, & Ashby, 2003; Gil-Rivas, Kilmer, Larson, & Armonstrong, 2017), long and unpredictable parental duty hours (Willerton, Schwarz, Wadsworth, & Oglesby, 2011), pressure to conform to strict behavioral standards (Burrell, Adams, Durand, & Castro, 2006), parental deployment, and fears related to a deployed parent’s risk of injury or death (Ender, 2006). As a result, military children have been found to show more internalizing (e.g., anxiousness, depression, and withdrawal) and externalizing (e.g., aggression and attentional difficulties) behaviors during parental deployment than their counterparts whose military parent was not deployed (Chartrand, Frank, White, & Shope, 2008). Likewise, for both military and civilian families, times of family transition or stressful situations (e.g., loss of job and marital problems) lead to heightened parental stress that may increase family conflict and result in diminished parenting ability (Rycus & Hughes, 1998).

Interestingly, research does not support the notion that parental divorce negatively affects children. However, parental conflict does negatively impact children, regardless of whether parents are married and together, separated, or divorced (Strohschein, 2005). Furthermore, research suggests that among families with high conflict, marital disruption has a less negative impact on children’s emotional well-being than remaining in a high-conflict family (Jekielek, 1998).

Individual child differences can also relate to varying levels of parenting stress. Children with difficult temperaments can be challenging because they are highly reactive, less flexible, and do not adjust easily (Ramos, Guerin, Gottfried, Bathurst, & Oliver, 2005). Therefore, difficult children are more challenging to parent. For example, it is more aversive to break from difficult children’s routine, and they are more emotionally reactive when they are upset, thus further challenging parents (Carey, 1986).

Child characteristics also relate to parenting stress and to parenting quality. There is evidence to suggest that children with a difficult temperament in stressful family circumstances may be abused at higher rates than easier to manage children (U.S. Department of Health and Human Services, 2003). When children have psychological, physical, or developmental disabilities or special medical needs (e.g., diabetes), their families may be more vulnerable to stresses that can negatively affect the quality of parenting (Berry, 2009; Dumas, Wolf, Fisman, & Culligan, 1991; Woodman, 2014).

Given the continued and growing interest in parental stress in the literature and its relevance to child development, it is important that researchers have reliable tools to measure parenting stress. The current article reviews the development and use of the PSS (Berry & Jones, 1995). Below, we describe the development of the measure and provide a summary of its use in the extant literature.

The PSS

Berry and Jones introduced the PSS in 1995. The theoretical underpinning of the PSS was a transactional model of stress. Berry and Jones viewed parenting stress as stemming from an interaction between parents and children that was a result of the bidirectional process. At the time that Berry and Jones introduced the PSS, a transactional model of parenting stress was novel; it challenged the dominant view in parenting at the time that focused on the impact of parents on children (Abidin, 1986). A unique contribution of Berry and Jones’ work was that the bidirectional theory broadened the focus of parenting stress to emphasize the importance of the child impacting the parent and parenting stress being a dynamic process. This bidirectional or interactional approach has received considerable empirical support in ensuing years (e.g., Sroufe, Egeland, Carlson, & Collins, 2005).

The PSS continues to be a useful measure because it operationalizes the stress of parental experience differently than do other measures. The PSS captures not only the demands of parenting, such as opportunity costs and limitations on personal resources, but also the rewards of parenting such as fulfillment and personal growth. The full-scale score allows rewards to offset costs. The PSS has specificity in that it queries parenting stresses irrespective of marital, financial, or other life stress (Lessenberry & Rehfeldt, 2004).

An important characteristic of the PSS is that it was designed for use with any kind of parent. In Berry and Jones’s (1995) original article, they incorporated data collected from a heterogeneous sample of parents. The norming samples included mothers and fathers of typically developing children as well as of children receiving school or outpatient services for emotional and/or behavioral problems. Furthermore, the PSS successfully differentiated between clinical (mothers of children receiving outpatient services for behavioral problems) and nonclinical samples (mothers of children not receiving treatment; Berry & Jones, 1995).

The PSS is brief and easy to administer. It is an 18-item self-report measure in which parents respond to statements about their typical relationship with their child. For each statement, respondents rate their level of agreement on a 5-point Likert-type scale (1 = strongly disagree, 2 = disagree, 3 = undecided, 4 = agree, and 5 = strongly agree). To compute the parental stress score, the positive Items 1, 2, 5, 6, 7, 8, 17, and 19 are reverse scored, and then, all items are summed. Higher scores reflect more parental stress. The possible range of the PSS is 18 (low stress) to 90 (high stress).

Berry and Jones (1995) reported that the scores of the scale were reliable, with a coefficient α of .83 and a mean interitem correlation of .23. A 6-week test–retest correlation was .81. The scale can be administered in pen/paper format or by computer administration. The PSS has been included in two major measurement books: The Handbook of Psychological Tests (Maltby, Lewis, & Hill, 2001) and Handbook of Family Measurement Techniques (Touliatos, Perlmutter, & Holden, 2001). The scale is widely used and has been readily available from authors at no cost; it is included here in the Appendix. In the past 5 years alone, the scale’s authors reportedly received requests for permissions for use from 250 researchers and clinicians (J. Berry, personal communication, May 15, 2016). Since 2010, requests have come from 42 countries, and researchers have requested permission to translate the scale into the following languages: Albanian, Bahasa Malaysia, Cantonese, Danish, Estonian, Filipino, French, German, Greek, Hebrew, Hindi, Hungarian, Italian, Korean, Kurdish, Lithuanian, Norwegian, Portuguese, Romanian, Spanish, Swedish, Tamil, Thai, Turkish, Urdu, and Vietnamese (J. Berry, personal communication, May 15, 2016).

Parenting Stress Is Cross-Cultural

The balance between the demands and rewards of parenting appears to generalize to other cultures. Research with the PSS has been conducted by clinicians, researchers, and students from around the world. It is of interest that individuals from diverse cultural backgrounds view the PSS as an instrument that would be relevant and useful in their population of interest. Furthermore, few individuals asking permission to use the scale also request permission to adjust or modify the scale in some minor way. By far, the most common modification request is translation of language. Possibly, this is evidence that researchers, clinicians, and students from across the globe similarly conceptualize parenting stress and see face validity in the instrument.

Research Goal

The current project systematically reviewed the articles published in English worldwide, which have used the PSS. The overarching goal was to provide readers with an index by which to compare parenting stress in their unique samples to that of other published samples. We hope that this will provide a basis for interpreting relative differences and similarities in parenting stress scores on the PSS across samples.

Literature Search

We conducted a PsycINFO and Google Scholar search using the paired search terms “PSS” and “Berry.” Additionally, we searched PsycINFO and Google Scholar for the original PSS publication (Berry & Jones, 1995) and used the forward cite option to obtain a list of articles in the database that referenced Berry & Jones’ 1995 article. Additional articles were found during the article screening process, as some studies identified via the original searches referenced articles not included in the initial search results. If those articles utilized the PSS, they were also included.

Article Screening Process

The initial searches yielded 269 articles, which were screened for the following criteria: (1) the study utilized the PSS, (2) the article was published in a peer-reviewed journal, (3) publications were available in English, (4) the articles reported PSS descriptive data (sample means and standard deviations), (5) if the scale was not used in its original form, the authors detailed modifications made to the scale. This screening process yielded 25 articles.

Researchers then reviewed the 25 articles carefully to understand the methodology of PSS utilization. Upon examination, articles were divided into four categories: (1) did not modify the scale (n = 13), (2) modified the wording (e.g., translated into another language or changed parent to “grandparent”; n = 4), (3) modified the number of items (i.e., dropped 1 item; n = 1), (4) modified the number of items and modified the response scale (e.g., dropped 1 item and used a 4-point Likert-type scale instead of a 5-point scale; n = 3), and (5) methods section did not sufficiently describe the measure so we were unable to determine if the authors had used it unaltered (n = 4), and when we attempted to contact these authors, they did not respond to inquiries.

Implications

The current study has broad implications. The information contained in this article will allow researchers and clinicians who are interested in measuring parental stress to make comparisons to a diverse group of peer-reviewed, published data. Furthermore, although it appears that parental stress is cross-cultural, some modifications were made by researchers a priori and during data analysis to accommodate perceived sample differences. The current study allows researchers and clinicians who are interested in using the scale to better understand how other researchers have modified the scale and why.

Limitations and Future Directions

The current study is limited in that descriptive data for relative comparisons across studies are not as definitive for interpretation as would be predictive norms or clinical cutoff levels. Given that Item 2 appears to be psychometrically difficult in some samples and in translations, future studies on the psychometric properties may consider either changing the wording of this item so that it is unambiguous, possibly changing “There is little or nothing I wouldn’t do for my child(ren) if it was necessary” to “There is little or nothing I wouldn’t do for my child(ren) if it was necessary for their health or survival” so that the meaning translates more readily across cultures. We hope that in the future, large-scale studies will be conducted to provide parenting stress norms in order to better inform researchers and service providers about when parents may benefit from additional support in order to keep parenting stress in a typical range.