Maternal Parenting Stress Based on Employment Status: Testing Measurement Invariance Using Multi-Group Latent Class Analysis

Abstract

The primary objective of this study was to test measurement invariance using a multi-group latent class analysis to examine parenting stress across groups of employed and unemployed mothers. This study analyzed data from the third to eighth waves (ages 3–8) of the Panel Study on Korean Children, including a final sample of 1,796 participants with verifiable maternal employment status at the third wave. This analysis identified latent class classification patterns of maternal parenting stress in two groups and assessed similarities and differences in latent classes at both the latent and indicator levels. The key findings were as follows. First, the number of latent classes for parenting stress was consistently determined as three across both groups, indicating a similar model structure, regardless of employment status. Second, the examination of measurement invariance at the latent level revealed a structural invariant model, suggesting that mothers, regardless of their employment status, experienced parenting stress in fundamentally similar ways. This was evidenced in their comparable response patterns to stress-related indicators and aligned fluctuations in stress levels over time. Third, an item-level analysis revealed a notable difference in responses between employed and non-employed mothers when their children were 3 years old, suggesting that employment status may influence how maternal parenting stress is perceived during early childhood. This study provides significant insights into the relationship between maternal employment status and parenting stress by applying the multi-group latent class analysis, contributing to a deeper understanding of maternal parenting stress patterns in different employment contexts.

Keywords

maternal parenting stress maternal employment status multi-group latent class analysis (MG-LCA)measurement invariance Panel Study on Korean Children (PSKC)

Introduction

The transition to parenthood is a significant life event that introduces new responsibilities and roles (Flaherty & Sadler, 2022), often leading to stress due to the demands of child rearing. As parental responsibilities increase, along with societal expectations, the pressures of parenting can become overwhelming, potentially undermining the quality of parenting. Parenting stress is defined as a negative psychological and physical response to the challenges of parenting (Deater-Deckard, 2008), and it tends to intensify when there is a discrepancy between the expected and available resources. Parents may feel overwhelmed when they perceive themselves as underinformed, less competent, or lacking emotional and practical support from partners, family, or friends (Deater-Deckard, 1998). Furthermore, modern social factors such as evolving values, increased workforce participation by women, and diverse family structures further exacerbate this stress, compounding the challenges new parents face.

Maternal parenting stress is particularly common, and failure to manage effectively can negatively affect child development (Rajgariah et al., 2021). Elevated maternal parenting stress is linked to negative outcomes, such as increased internalizing and externalizing problem behaviors in children (de Maat et al., 2021; Dey et al., 2021; Jiang et al., 2023; Neece et al., 2012; Owen et al., 2006) and impairments in executive functioning and problem-solving skills (Park et al., 2023; Tan et al., 2017). Additionally, maternal parenting stress can adversely affect children’s social–emotional development (Anthony et al., 2005; Carapito et al., 2020; Colletti et al., 2008; Ostberg & Hagekull, 2013) and their adjustment to school (Conger et al., 1995; Rogers et al., 2009). These findings underscore the importance of implementing effective strategies to reduce maternal parenting stress in order to support children’s adaptive behavior and supporting their cognitive, social–emotional, and academic development during critical growth stages.

While it is well established that maternal parenting stress plays a significant role in shaping child development, much of the existing research has relied on cross-sectional data, providing a static view rather than a developmental perspective over time. Although some studies have explored long-term patterns of maternal parenting stress, their findings have been inconsistent (Chang & Fine, 2007; Park et al., 2023; Stone et al., 2016; Williford et al., 2007), pointing to meaningful variation in how mothers experience and manage parenting stress. These inconsistencies suggest that maternal parenting stress may not follow a single, uniform pattern across individuals. To address this gap, the current study adopts a longitudinal approach, using latent class analysis to identify distinct patterns of maternal parenting stress. Moreover, considering the dynamic nature of the mother–child relationship in the context of maternal employment, this study explored how these stress trajectories differ by maternal employment status.

Maternal Parenting Stress of Employed and Unemployed Mothers

Since maternal parenting stress is heavily shaped by the conditions and contexts surrounding mothers, maternal employment plays a critical role in influencing it. Like many industrialized countries, South Korea has experienced a significant rise in maternal employment in recent decades, influenced by both growing female labor force participation and economic demands that necessitate continued employment after childbirth. However, several South Korean mothers lack access to extended family support, such as assistance from grandparents or other relatives, which exacerbates the challenges of child rearing. This absence of support has contributed to South Korea’s notably low fertility rate, which has decreased from 4.53 children per woman in 1970 to a projected 0.72 in 2023, marking the sharpest decline among Organization for Economic Cooperation and Development (OECD) countries (Statistics Korea, 2024b).

Despite a significant drop in fertility, the proportion of working mothers living with children aged under 18 in South Korea reached 62.4% in 2024, showing a gradual upward trend (Statistics Korea, 2024a). However, a national gender equality survey revealed that 68.9% of wives primarily assume responsibility for household chores, and 32.2% of mothers reported difficulties in balancing work and childcare obligations (Ministry of Gender Equality and Family, 2021). These data suggest that maternal employment significantly influences maternal parenting stress, indicating a need for national policies and support systems that can better address the unique challenges faced by working mothers.

Employed mothers often struggle to balance multiple responsibilities, such as household duties, childcare, and professional obligations (Yeo & Teo, 2013). Their employment typically leaves them with limited time for their children, which increases the strain of managing both their work and family lives (Kim et al., 2013). The need to divide psychological and physical energy between these competing roles can lead to exhaustion (Forgays & Forgays, 1993; Kim et al., 2013). Moreover, in South Korea, the Confucian ideal that mothers should primarily act as caregivers intensifies feelings of guilt among working mothers who feel that they are not fully engaging with their children (Hong & Lee, 2019).

Contrary to the common assumption that stay-at-home mothers enjoy a more relaxed lifestyle than their employed counterparts, their daily routines are often equally demanding. Unemployed mothers frequently set higher expectations for their children, leading to a unique set of stressors that differ from those experienced by working mothers (Crnic & Greenberg, 1990). By dedicating themselves fully to child rearing, unemployed mothers may struggle to obtain adequate support because domestic responsibilities are often undervalued compared with external employment (Aslam & Adams, 2022; Dillaway & Paré, 2008). Additionally, the continuous presence of unemployed mothers with their children can limit mothers’ opportunities for adult interaction, thereby intensifying the stresses associated with parenting (Crowley, 2014).

Research on maternal parenting stress in relation to maternal employment status presents a complex and varied picture. Some studies indicate that working mothers face higher levels of parenting stress than their stay-at-home counterparts, suggesting that the combined demands of professional and domestic responsibilities contribute to this increase in stress (Dey et al., 2021; Rajgariah et al., 2021). Conversely, other research shows that stay-at-home mothers may experience greater parenting stress owing to the monotonous nature of household tasks, the absence of external rewards, and social isolation from friends and family (Crnic & Greenberg, 1990; Kahn & Cuthbertson, 1998; Song et al., 2014; Yoon et al., 2009). Additionally, some studies have found no significant differences in parenting stress levels between employed and stay-at-home mothers (Forgays & Forgays, 1993; McCarten, 2003; Yeo & Teo, 2013). These varied findings suggest that employed and stay-at-home mothers may experience parenting stress in fundamentally different ways, indicating that these groups should not be considered a single, homogeneous category in studies of maternal parenting stress. Employed and stay-at-home mothers are distinct groups with unique stress-related characteristics. Therefore, this study examined the potential differences between these groups by conducting a measurement invariance test to assess whether such distinctions exist within the latent structure.

Purpose of This Study

Several previous studies have shown that maternal parenting stress generally decreases over time following childbirth (Berryhill & Durtschi, 2017; Chang & Fine, 2007; Park et al., 2023; Pope et al., 2005; Williford et al., 2007). However, these studies also demonstrated significant variance, suggesting that the longitudinal patterns of maternal parenting stress can vary widely among individuals (Chang & Fine, 2007; Park et al., 2023; Stone et al., 2016; Williford et al., 2007). Therefore, this study identified distinct trajectories of maternal parenting stress among different subgroups using a latent class analysis.

Furthermore, this study used a multi-group analysis to investigate whether the latent classes differ by employment status. Although previous research has examined differences in maternal parenting stress patterns based on employment status, it has not explored these differences within the context of latent classes. For example, Park and Park (2016) showed that stay-at-home mothers experienced higher levels of parenting stress than employed mothers immediately after childbirth, with significant increases in stress until the child reached the age of three. Meanwhile, employed mothers’ stress levels showed no significant changes over the same period.

Our study aimed to address this gap in the literature by identifying latent classes in the longitudinal patterns of maternal parenting stress and determining whether these classes and their characteristics vary based on maternal employment status. In the presence of such differences, the study explored their possible underlying causes. Our research questions were as follows:

Research Question 1: What are the patterns of the latent class classification of maternal parenting stress among groups based on maternal employment status?

Research Question 2: At what level of similarity and difference do the classified latent classes of maternal parenting stress show measurement invariance across groups based on maternal employment status?

Research Question 3: In which timepoints do the levels of maternal parenting stress show similarities and differences across groups based on maternal employment status?

Method

Data

In this study, we used data from the Panel Study on Korean Children, conducted by the Korea Institute of Child Care and Education. We compared the classification of the latent classes for maternal parenting stress among groups of mothers with young children, based on maternal employment status. The nationally representative panel data set collects annual surveys from a cohort of children born in 2008 and tracks their personal characteristics and the environmental factors related to their parents, households, and schools as they grow. The longitudinal panel data are ideal for identifying latent classes of maternal parenting stress at multiple time points and assessing the consistency of these classifications between employed and unemployed mothers. This is because the data include repeated annual surveys of the same children and households.

To classify maternal parenting stress based on changes over time, more than three time points of longitudinal data are required (Curran et al., 2010). Accordingly, this study utilized data from the third to the eighth waves of the Panel Study on Korean Children (PSKC), tracking children from age three to eight. The analysis began at the third wave, corresponding to when the children were 3 years old—a key developmental and social transition point in South Korea. At this stage, many children begin attending daycare centers or preschools, typically coinciding with the conclusion of parental leave. Therefore, we began collecting data from the third wave, when the children were 3 years old—a common age for mothers to return to work after parental leave and for children to start preschool or daycare programs (Blum et al., 2023).

According to the Ministry of Education (2023), as of 2020, the enrollment rate in early childhood education—including kindergartens and childcare centers—for children aged 3 to 5 was 94.0%. This widespread participation reflects the norm of institutional early education beginning around age three, which in turn facilitates maternal return to the workforce by providing structured childcare. Entry into elementary school usually occurs at age six or seven (first grade), bringing additional changes in family routines and educational expectations. These shifts in the caregiving and educational landscape may influence the trajectory of maternal parenting stress throughout early childhood.

The final sample consisted of 1,796 mothers, including only those with verifiable maternal employment status at the third wave. At that time, the average age of mothers was 32.8 years (SD = 3.7), with 33.4% employed and 66.6% unemployed. In terms of residential area, 38.0% lived in metropolitan cities, 41.7% in mid-sized or small cities, and 20.3% in rural areas. Mothers reported an average of 1,726.7 min of childcare per week on weekdays and 901.4 min on weekends. Most mothers had completed at least high school, with over 70% having received a college or higher education. The outcome variables—children’s academic performance, school adaptation, and internalizing and externalizing problem behaviors—were measured at the eighth wave, when the children were 8 years old and had entered elementary school.

Measures

Response Variable: Maternal Parenting Stress

Maternal parenting stress was assessed using a scale designed to measure the burden and distress experienced by mothers in their parental role. We calculated the mean score of the 11 items on the scale and used it as an indicator variable across 6 time points. Each item was rated on a 5-point Likert scale, ranging from 1 (Not at all) to 5 (Very much), with higher scores indicating a greater level of burden and distress related to the parental role. The contents of each item used to measure maternal parenting stress, along with the descriptive statistics for each time point categorized by maternal employment status, are presented in Table 1. To ensure the reliability of the scale, Cronbach’s alpha was calculated separately for each group.

Table 1.

Details of Items Related to Maternal Parenting Stress.

Item	Time points^a	Employed			Unemployed
Item	Time points^a	Mean	SD	α ^b	Mean	SD	α ^b
1. I am not confident that I can become a good parent.	Third wave	2.701	0.604	0.858	2.834	0.636	0.866
2. I am unsure whether I can raise my child well.	Third wave	2.701	0.604	0.858	2.834	0.636	0.866
3. There are times when I feel that my child is falling behind other children because I am not fulfilling my parental role adequately.	Fourth wave	2.748	0.630	0.873	2.823	0.647	0.874
4. There are moments when I want to escape from my child.	Fourth wave	2.748	0.630	0.873	2.823	0.647	0.874
5. It feels difficult to treat my child in a more affectionate and warm manner.	Fifth wave	2.681	0.657	0.883	2.770	0.647	0.878
6. These days, there is such a wide range of parenting information available that I feel confused about which to choose.	Fifth wave	2.681	0.657	0.883	2.770	0.647	0.878
7. Since the birth of my child, I no longer enjoy my life as much as before.	Sixth wave	2.595	0.606	0.871	2.684	0.625	0.876
8. The cost of raising my child feels burdensome.	Sixth wave	2.595	0.606	0.871	2.684	0.625	0.876
9. When my child appears emotionally unstable, I feel distressed, as if it is my fault.	Seventh wave	2.538	0.618	0.883	2.606	0.623	0.875
10. When I am tired and my child asks me to play, I feel irritated.	Seventh wave	2.538	0.618	0.883	2.606	0.623	0.875
11. Even when I go to social gatherings, I expect that I will not enjoy myself because of my child.	Eighth wave	2.353	0.650	0.892	2.438	0.690	0.901

The Third to Eighth waves correspond to child ages 3 to 8.

Cronbach’s alpha.

Data Analysis

This study used a multiple-group latent class analysis (MG-LCA) to explore whether classifications of latent classes for parenting stress types remain consistent across groups of mothers differentiated by employment status. As an extension of latent class analysis, MG-LCA enables comparisons of latent class structures across different groups. Following the method proposed by Kankaraš et al. (2010), we analyzed the measurement invariance of parenting stress between employed and unemployed mothers at the latent and indicator levels via four key steps.

First, we identified the optimal number of latent classes for each group based on employment status. We used various information criteria (ICs) to determine the best-fitting model, including the Akaike information criterion (AIC; Akaike, 1974), Bayesian information criterion (BIC; Schwarz, 1978), consistent AIC (CAIC; Bozdogan, 1987), and sample-size adjusted BIC (SABIC; Sclove, 1987), with lower values indicating a better model fit. Additionally, we employed model comparison tests, such as the Lo–Mendell–Rubin adjusted likelihood ratio test (LMR-LRT; Lo et al., 2001) and bootstrapped likelihood ratio test (BLRT; Peel & McLachlan, 2000), to verify the significance by comparing models with k latent classes against those with k– 1 classes. If the p-value for the k-class model was less than .05, the k-class model was selected as the final model (Lo et al., 2001). We also examined the entropy values, which indicate classification accuracy, considering values above .80 as reflecting a well-fitting model (Muthén & Muthén, 2007).

Second, once we confirmed that the number of latent classes was consistent across groups, we applied MG-LCA to test measurement invariance at the latent level using the combined data set. During this process, we included employment status as a categorical covariate in the latent class model to represent the groups being compared. This method is considered the most straightforward for incorporating group effects in latent class models (Kankaraš et al., 2010). The group variable can have both direct and interaction effects on each indicator and latent variable (McCutcheon & Hagenaars, 1997). Direct effects are the group variable’s immediate influence on the indicator variable, whereas interaction effects arise when the group variable moderates the latent variable’s effect on the indicator variable. By progressively constraining these direct and interaction effects, we created three types of nested models. First, the non-invariance model allows both direct and interaction effects. Second, the partial invariance model allows direct effects but constrains interaction effects across groups. Third, the structural invariance model constrains both direct and interaction effects. We sequentially compared these models to determine whether the latent class structure of maternal parenting stress was equivalent across the two groups at the latent level based on ICs and model comparison tests, as in the first step. In this hierarchical model comparison, if a more constrained model was supported, we considered it to be a more parsimonious and statistically sound model theoretically and statistically (Eid et al., 2003; Sohn, 2016).

Third, when the non-invariance or partial invariance models were supported, we tested measurement invariance at the indicator level to identify which indicators contributed to the lack of invariance. As in the first step, we evaluated model fit using ICs and model comparison tests. We also examined the values of bivariate residuals, calculated based on the Pearson chi-square statistic, to compare the chi-square statistic corresponding to the difference in degrees of freedom between the baseline model and the comparison model (Vermunt & Magidson, 2021). Bivariate residuals help assess model fit by comparing models that include or exclude specific variables and are particularly useful in deciding whether to include direct effects (Moors, 2004). Although bivariate residuals do not directly assess interaction effects, excessively large residuals may suggest the presence of unmodeled interaction effects, thus offering an indirect means of evaluation (Brown, 2015). We used Latent Gold 6.0 software (Statistical Innovations, 2021) for these analyses because it can analyze complex latent class models with hierarchical structures and multiple groups. The research model for this study is shown in Figure 1, in which the red arrows represent the direct effects of the group variable on the indicator variables, yellow arrows indicate the interaction effects, and blue arrows show the effects of the group variable on the latent variable.

Figure 1.

Research model.

Results

Determining the Optimal Number of Latent Classes for Maternal Parenting Stress by Group

To determine the optimal number of latent classes by group for maternal parenting stress, differentiated by maternal employment status, we systematically increased the class count from two to five. For each model, we analyzed the ICs, model comparison tests, entropy, classification error, and proportions of each latent class. The findings from these analyses appear in Table 2.

Table 2.

Model Comparison Results by Number of Latent Classes for Each Group.

No. of profiles	BIC	AIC	CAIC	SABIC	Entropy	LMR-LRTp-value	BLRTp-value	Latent class proportions (%)
No. of profiles	BIC	AIC	CAIC	SABIC	Entropy	LMR-LRTp-value	BLRTp-value	1	2	3	4	5
Employed
2	3,945.811	3,845.164	3,970.811	3,866.480	0.857	<.001	<.001	53.1	46.9
3 ^a	3,738.513	3,585.530	3,776.513	3,617.930	0.825	<.001	<.001	47.8	30.9	21.2
4	3,659.223	3,453.903	3,710.223	3,497.388	0.828	<.001	<.001	44.2	26.3	15.2	14.3
5	3,651.856	3,394.200	3,715.856	3,448.769	0.786	<.001	<.001	28.5	26.2	16.2	15.7	13.5
Unemployed
2	8,499.469	8,380.028	8,524.469	8,420.074	0.847	<.001	<.001	54.8	45.2
3 ^a	7,805.012	7,623.461	7,843.012	7,684.332	0.839	<.001	<.001	40.4	34.6	25.0
4	7,539.118	7,295.458	7,590.118	7,377.154	0.825	<.001	<.001	39.6	34.8	14.8	10.8
5	7,453.073	7,147.303	7,517.073	7,249.823	0.808	<.001	<.001	33.5	29.7	22.6	.08	.06

Bolded values represent the final selected latent class solution for each group.

In terms of ICs (AIC, BIC, CAIC, and SABIC), all groups exhibited decreasing values with the addition of more latent classes. When the values decrease continuously, the optimal number of latent classes is typically found at the “elbow point,” where the rate of decrease is the sharpest (Nylund et al., 2007). In our study, the largest reduction in ICs occurred when the number of latent classes rose from two to three, as shown in Figure 2. For the classification quality indicators, the highest entropy values for both groups were observed with two latent classes. However, the model comparison tests (BLRT and LMR-LRT) indicated significance for all class solutions, demonstrating that these tests alone were not adequate to determine the optimal number of classes. Therefore, we needed additional criteria to make an informed decision (Nylund-Gibson et al., 2014). Moreover, in the three-class model, the smallest latent class accounted for 21.2% of employed mothers and 25.0% of unemployed mothers, satisfying the minimum 5% threshold for latent class proportions (Berlin et al., 2014).

Figure 2.

Changes in information criteria with increasing number of latent classes by group.

After considering both statistical indicators and interpretability, we selected a three-class model as the most appropriate for both groups. This model demonstrated substantial decreases in ICs and high classification quality beyond the two-class solution.

Before testing measurement invariance for the identified latent classes across groups, we examined the characteristics of each class separately for employed and unemployed mothers based on the coefficients for maternal parenting stress. Figure 3 shows the latent class profiles for each group. Both groups exhibited a similar classification pattern across three classes: high, moderate, and low levels of parenting stress. Specifically, the first latent class, which represents the highest level of parenting stress, accounted for the largest proportion in both groups (employed mothers: 47.8%, unemployed mothers: 40.4%). The second class showed a moderate level of stress (employed mothers: 30.9%, unemployed mothers: 36.6%), and the third class exhibited the lowest level of parenting stress (employed mothers: 21.2%, unemployed mothers: 25.0%).

Figure 3.

Latent class profiles by group.

Quantitative differences were observed between the groups in terms of their initial stress levels and rate of change over time. Initial stress levels for unemployed mothers were slightly higher across all classes compared with those of employed mothers. Additionally, although both groups showed an overall decrease in stress levels over time, the rate of decrease varied. Employed mothers in the high-stress class experienced a steeper reduction in stress levels after the fifth wave, but the decline for unemployed mothers was more gradual. This suggests that stress levels among high-stress employed mothers tended to stabilize more quickly over time, whereas stress reduction for unemployed mothers occurred at a slower pace.

Results of the Multi-Group Latent Class Analysis

All groups were confirmed to have three latent classes in the classification of maternal parenting stress. Based on this finding, we conducted an MG-LCA using the combined data set, with employment status as a grouping variable. We applied the method for testing measurement invariance proposed by Kankaraš et al. (2010), and the results at the latent level are presented in Table 3. By selecting the model that best fits the data, we can determine the degree of measurement invariance. This model selection helps clarify whether the similarity between groups is due to the types of latent classes themselves or the conditional response probabilities within each class.

Table 3.

Model Comparison Results by Number of Latent Classes for Each Group.

No.	Model	BIC	AIC	CAIC	SABIC	LMR-LRTp-value	BLRTp-value
1	Non-invariance	11,523.657	11,286.116	11,569.657	11,377.538	-	-
2	Partial invariance	11,443.871	11,268.297	11,477.871	11,335.870	.107	.006
3 ^a	Structural invariance	11,415.498	11,270.907	11,443.498	11,326.556	.722	.002

Bolded values represent the final determined measurement invariance model.

A comparison between the non-invariance and partial invariance models revealed that all ICs were the smallest for the partial invariance model. Although the result of the BLRT was significant, that of the LMR-LRT was not. Based on these results, the partial invariance model, which constrains the interaction effects between the group and latent variables to be equal across groups, was deemed more suitable than the non-invariance model. Subsequently, a comparison between the partial and structural invariance models showed that, except for the AIC and LMR-LRT, all ICs were lower in the structural invariance model, and the result of the BLRT was significant. This indicates that the direct effects of the group variable on indicator variables were also absent across groups. Accordingly, the structural invariance model, which constrains both direct and interaction effects to be equal across groups, was selected as the final model. This final model showed that the latent class types and response patterns for indicator variables were equally consistent across groups.

Although we selected the structural invariance model, we also conducted an indicator-level measurement invariance test to identify potential sources of non-invariance among the indicator variables. To accomplish this, we compared the structural invariance model (Model 3) with a series of alternative models (Models 3-1 to 3-6), allowing the direct effects of the group variable on each indicator variable to vary across groups.

Table 4 shows that allowing the direct effect of the group variable on the third-wave indicator variable to differ across groups resulted in lower AIC and SABIC values compared with those of the structural invariance model. Additionally, the BLRT p-value was statistically significant, indicating an improvement in model fit with this adjustment. A further assessment was conducted using a chi-square value of 3.841, which corresponded to a difference of 1 degree of freedom between the structural invariance model and models that allow direct effects for each indicator. The bivariate residuals for indicator variables with allowed direct effects were compared against this threshold. The findings revealed that the bivariate residuals for models permitting a direct effect of the group variable on the third-wave indicator exceeded the threshold, suggesting that this direct effect may vary between groups.

Table 4.

Results of the Indicator-Level Measurement Invariance Analysis for Maternal Parenting Stress Classification Across Groups.

No.	Model	BIC	AIC	CAIC	SABIC	LMR-LRTp-value	BLRTp-value	Bivariateresiduals
3	Structural invariance	11,415.498	11,270.907	11,443.498	11,326.556	-	-	-
3-1	Third wave ^a	11,416.850	11,267.096	11,445.850	11,324.732	.014	.022	7.301
3-2	Fourth wave	11,420.412	11,270.658	11,449.412	11,328.294	.240	.158	1.943
3-3	Fifth wave	11,422.645	11,272.890	11,451.645	11,330.526	.857	.890	0.000
3-4	Sixth wave	11,421.232	11,271.477	11,450.232	11,329.113	.216	.260	1.595
3-5	Seventh wave	11,422.661	11,272.907	11,451.661	11,330.542	.974	.982	0.008
3-6	Eighth wave	11,422.553	11,272.798	11,451.553	11,330.434	.733	.718	0.060

Bolded values represent the final selected indicator variables causing latent non-invariance.

In summary, these results demonstrated that the direct effect of the group variable on the third-wave indicator variable was not consistent across groups. This finding suggests that the group variable, which represents employment status, may affect this specific indicator independently of the latent variable.

Based on the results that support the invariance of latent classes for maternal parenting stress across groups, we analyzed the differences in the response categories for each indicator variable, as shown in Figure 4. Greater discrepancies in the graph’s shape suggest potential differences in the perceptions of indicator variables between groups. One indicator variable (third wave), identified in the model comparisons as having a non-invariant direct effect across groups, exhibited larger differences in response categories—particularly in categories 3 and 4—compared with the invariant indicators (fourth to eighth waves). This finding suggests that differences in response categories for the single non-invariant indicator cannot be fully explained by differences in latent class membership alone, implying the likelihood of specific group differences in the response patterns for this particular indicator variable.

Figure 4.

Group differences in responses to levels of maternal parenting stress.

Conclusion and Discussion

This study examined the latent class patterns of maternal parenting stress over a 6-year period among mothers with 3-year-old children, focusing on classification differences based on maternal employment status and testing for measurement invariance. We conducted an MG-LCA to assess measurement invariance at both the latent and indicator levels and to identify indicator variables that contribute to non-invariance. Below, we summarize the detailed findings, discuss implications, and outline limitations and directions for future research.

Summary of Findings

First, to verify measurement invariance in the latent class classification of maternal parenting stress across groups, we conducted a latent class analysis separately for each group to identify the optimal number of latent classes. The analysis showed that both groups exhibited the same optimal solution of three latent classes (high, moderate, and low), suggesting that the latent class structure for assessing maternal parenting stress was consistent across groups. Although the classification patterns of the latent classes were similar, quantitative differences were observed between the groups, particularly in how stress levels changed over time.

Across all classes, unemployed mothers showed slightly higher initial stress levels, which may be attributed to their limited access to external support and a lack of structured daily routines—factors known to contribute to prolonged parenting stress (Gallie & Russell, 2009; Nomaguchi & Milkie, 2003). In contrast, employed mothers showed a more gradual reduction in stress levels over time, especially in the high-stress class, suggesting that employment may provide access to resources, social networks, and structured schedules that help manage stress more effectively (Nomaguchi & Milkie, 2020; Parkes et al., 2015). Additionally, the slower decline in stress levels observed among unemployed mothers, particularly in the high-stress class, may reflect the restricted access to structured supports and social resources, which are more accessible to employed mothers (Rizzo et al., 2013).

Given the higher initial stress levels observed among unemployed mothers, these findings emphasized the need for early, targeted interventions, such as community-based support programs, mental health resources, and structured social networks, to effectively alleviate long-term stress from the outset. For employed mothers, policies that promote work–life balance, including flexible hours and accessible childcare, may further aid in reducing parenting stress by reinforcing the stability provided by employment. Future research should examine these dynamics in greater detail, focusing on how specific support mechanisms differentially affect parenting stress trajectories in employed versus unemployed mothers.

Second, because both groups were confirmed to have the same number of latent classes, a measurement invariance analysis conducted at the latent level to determine whether the latent class structures for maternal parenting stress were consistent across employed and unemployed mothers. The results supported the structural invariance model as the final model. This finding indicates that the interaction effects between the latent variable (maternal parenting stress) and the group variable (employment status), as well as the direct effects of employment status on indicator variables, were equivalent across groups. In other words, regardless of employment status, mothers showed similar response patterns and comparable stress trajectories over time (Kankaraš et al., 2010; Morin et al., 2016).

The observed structural consistency suggests that maternal parenting stress is shaped by a common psychological framework that persists across employment contexts. This suggests that although employment status may shape how intensely stress is experienced, it does not appear to change the fundamental structure of maternal parenting stress. Therefore, interventions that address core elements—such as emotional strain and the burden of multiple roles—may be broadly effective across mothers in varied socioeconomic contexts. Furthermore, the consistency in direct effects across groups suggests that mothers interpreted and responded to stress indicators at a similar baseline level, reinforcing that the foundational aspects of stress perception and response are unaffected by employment status (Hair et al., 2006; Steenkamp & Baumgartner, 1998). This invariance model showed that the maternal parenting stress structure was inherently robust across employment groups, which may simplify the development of standardized support programs. However, further exploration could help identify any subtle, context-specific differences that may require targeted adaptations for each group, ensuring that the support strategies address individuals’ unique needs effectively.

Finally, although structural invariance was supported overall, a further analysis of measurement invariance at the indicator level showed a direct effect of the group variable at the third time point, when the child was 3 years old. This finding suggests that the likelihood of mothers’ responses at this specific age might differ between groups owing to additional factors specific to each group that are unrelated to the identified latent class structure. While the overall classification of maternal parenting stress levels remained consistent across groups, unique characteristics associated with the child being 3 years old might affect how employed and unemployed mothers perceived and reported stress.

These findings underscore the complex interplay between maternal roles and external demands. The consistent latent class structures across groups suggest that foundational aspects of maternal parenting stress are not merely situational, but instead reflect broader psychological processes common to mothers regardless of employment status. This extends previous research (e.g., Crnic & Greenberg, 1990; Forgays & Forgays, 1993), which often emphasized dichotomous group comparisons, by revealing nuanced patterns within groups. Moreover, the non-invariance detected at the child’s age of three may point to a developmental inflection point, where external obligations (for employed mothers) and intensive caregiving demands (for unemployed mothers) diverge most starkly. Thus, this study contributes to the literature by showing not only the structural similarity in maternal parenting stress but also identifying periods of heightened divergence that warrant targeted policy and psychological interventions.

The age of three represents a significant developmental milestone, characterized by increased demands related to physical activity, early socialization, and growing communication skills, all of which can add to parental stress (Deater-Deckard, 2008). At this stage, children develop greater physical mobility and curiosity, necessitating closer supervision, and they begin to interact more with peers, testing social boundaries (Siegler et al., 2010). Additionally, their emerging communication skills lead to more frequent expressions of independence and assertiveness, both verbally and physically, which can further contribute to elevated maternal parenting stress (Cole et al., 2004; Keane & Calkins, 2004). For employed mothers, balancing these demands with work responsibilities may heighten stress owing to time constraints and fewer opportunities for hands-on caregiving, potentially amplifying work–family conflict during this stage (Chung & van der Horst, 2018; Cooklin et al., 2015). In contrast, unemployed mothers may experience different stressors tied to full-time caregiving, potentially feeling the effects of limited structured routines or external support, both of which are known to contribute to sustained stress (Berry & Jones, 1995; Nomaguchi & Milkie, 2020).

Implications

These findings indicate that the structural invariance of maternal parenting stress remained consistent across groups during the developmental period spanning ages three to eight. The absence of significant differences in parenting stress levels between employed and non-employed mothers challenges existing theoretical expectations. On one hand, theories such as the multiple stressor theory (McBride, 1990) or role strain theory (Goode, 1960) suggest that employed mothers would experience higher stress due to occupying multiple roles. On the other hand, the role accumulation theory (Sieber, 1974) posits that employment may buffer stress by providing additional rewards. Instead, the results in this study align more closely with theoretical perspectives such as the role context approach (Moen et al., 1989). This framework suggests that maternal parenting stress is influenced not merely by employment status or the number of roles occupied, but by qualitative aspects of these roles—for instance, individual differences in mothers’ adaptation and engagement with their work or the overall quality of their job (Forgays & Forgays, 1993; Gómez, 2006; Haggag et al., 2011).

Taken together, these findings suggest that while the sources of maternal parenting stress may differ qualitatively, the overall magnitude of stress may not vary significantly between groups. This highlights the importance of developing tailored interventions that account for individual contextual factors, rather than relying on overly simplified or uniform approaches. Supporting both employed and unemployed mothers may require approaches tailored to the distinct stressors each group encounters. Future studies should consider contextual factors—particularly those arising at key stages of child development—to better identify when and how interventions can be most effective in reducing maternal parenting stress. Programs such as parenting stress management workshops, structured peer support groups, or evidence-based cognitive–behavioral interventions for parents have the potential to help maintain balanced stress levels, enhance adaptive parenting practices, and ultimately improve child outcomes. Additionally, exploring how maternal parenting stress interacts with child development and contextual factors, such as socioeconomic status and social support networks, could provide valuable insights for refining these interventions. Such research would be instrumental in developing optimized support strategies tailored to the distinct needs of employed and unemployed mothers, leading to healthier and more resilient family dynamics.

Limitations and Future Directions

While this study significantly advances our understanding of the relationship between maternal parenting stress and employment status through the application of MG-LCA, several limitations warrant attention. First, although mothers were categorized dichotomously as employed or unemployed, this classification does not capture the variation in employment conditions, such as part-time versus full-time work, job flexibility, or the mothers’ preferences and motivations for employment. Prior research suggests that the subjective fit between mothers’ work preferences and actual employment status plays a critical role in shaping parenting stress (Chung & van der Horst, 2018; Nomaguchi & Milkie, 2003). Future studies should incorporate these nuanced employment-related factors to more comprehensively assess how work contexts influence maternal well-being. Second, while grouping mothers as either employed or unemployed offered a useful starting point, it inevitably glosses over the wide variation in lived experiences within each category. A mother balancing a part-time job with strong family support likely navigates stress very differently from one in a full-time role with limited help. Factors such as income, education, and access to support systems can deeply influence how stress unfolds over time. Future research that takes these differences into account will offer a fuller, more accurate picture of maternal parenting stress. Third, this study focused exclusively on Korean mothers raising children aged three to eight. While the findings offer meaningful contributions within the Korean context and during early childhood, they may not generalize to other cultural settings or developmental stages. In Korea, it’s still common for mothers to carry most of the responsibility for raising children, even when they’re working full-time. Social expectations often push them toward a style of parenting that values constant involvement and personal sacrifice. These pressures—combined with limited support from fathers and strong societal views about how mothers should behave—can shape how parenting stress is experienced and expressed. Because of this, the findings from this study may reflect uniquely Korean patterns that don’t necessarily apply elsewhere. To better understand how maternal parenting stress changes over time and across different cultures, future research should include cross-national comparisons and cover a wider range of developmental periods.

Footnotes

ORCID iDs

JiYoon Kim

Sojin Yoon

Sehee Hong

Ethical Considerations

This study used a secondary data from the Panel Study on Korean Children (PSKC) without identifying information and was exempted from Institutional Review Board approval.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

References

Akaike

(1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control, 19(6), 716–723. https://doi.org/10.1109/tac.1974.1100705

Anthony

L. G.

Anthony

B. J.

Glanville

D. N.

Naiman

D. Q.

Waanders

Shaffer

(2005). The relationships between parenting stress, parenting behaviour and preschoolers’ social competence and behaviour problems in the classroom. Infant and Child Development, 14(2), 133–154. https://doi.org/10.1002/icd.385

Aslam

Adams

T. L.

(2022). “The workload is staggering”: Changing working conditions of stay-at-home mothers under COVID-19 lockdowns. Gender Work and Organization, 29(6), 1764–1778. https://doi.org/10.1111/gwao.12870

Berlin

K. S.

Williams

N. A.

Parra

G. R.

(2014). An introduction to latent variable mixture modeling (Part 1): Overview and cross-sectional latent class and latent profile analyses. Journal of Pediatric Psychology, 39(2), 174–187. https://doi.org/10.1093/jpepsy/jst084

Berryhill

M. B.

Durtschi

J. A.

(2017). Understanding single mothers’ parenting stress trajectories. Marriage & Family Review, 53(3), 227–245. https://doi.org/10.1080/01494929.2016.1204406

Berry

J. O.

Jones

W. H.

(1995). The Parental Stress Scale: Initial psychometric evidence. Journal of Social and Personal Relationships, 12(3), 463–472. https://doi.org/10.1177/0265407595123009

Blum

Dobrotić

Kaufman

Koslowski

Moss

(2023). International review of leave policies and research 2023. http://www.leavenetwork.org/lp_and_r_reports/.

Bozdogan

(1987). Model selection and Akaike’s information criterion (AIC): The general theory and its analytical extensions. Psychometrika, 52(3), 345–370. https://doi.org/10.1007/bf02294361

Brown

T. A.

(Ed.). (2015). Confirmatory factor analysis for applied research (2nd ed.). The Guilford Press.

10.

Carapito

Ribeiro

M. T.

Pereira

A. I.

Roberto

M. S.

(2020). Parenting stress and preschoolers’ socio-emotional adjustment: The mediating role of parenting styles in parent–child dyads. Journal of Family Studies, 26(4), 594–610. https://doi.org/10.1080/13229400.2018.1442737

11.

Chang

Fine

M. A.

(2007). Modeling parenting stress trajectories among low-income young mothers across the child’s second and third years: Factors accounting for stability and change. Journal of Family Psychology, 21(4), 584–594. https://doi.org/10.1037/0893-3200.21.4.584

12.

Chung

van der Horst

(2018). Women’s employment patterns after childbirth and the perceived access to and use of flexitime and teleworking. Human Relations, 71(1), 47–72. https://doi.org/10.1177/0018726717713828

13.

Cole

P. M.

Martin

S. E.

Dennis

T. A.

(2004). Emotion regulation as a scientific construct: Methodological challenges and directions for child development research. Child Development, 75(2), 317–333. https://doi.org/10.1111/j.1467-8624.2004.00673.x

14.

Colletti

C. J.

Wolfe-Christensen

Carpentier

M. Y.

Page

M. C.

McNall-Knapp

R. Y.

Meyer

W. H.

Chaney

J. M.

Mullins

L. L.

(2008). The relationship of parental overprotection, perceived vulnerability, and parenting stress to behavioral, emotional, and social adjustment in children with cancer. Pediatric Blood & Cancer, 51(2), 269–274. https://doi.org/10.1002/pbc.21577

15.

Conger

R. D.

Patterson

G. R.

(1995). It takes two to replicate: A mediational model for the impact of parents’ stress on adolescent adjustment. Child Development, 66(1), 80–97. https://doi.org/10.1111/j.1467-8624.1995.tb00857.x

16.

Cooklin

A. R.

Westrupp

Strazdins

Giallo

Martin

Nicholson

J. M.

(2015). Mothers’ work-family conflict and enrichment: Associations with parenting quality and couple relationship. Child Care Health and Development, 41(2), 266–277. https://doi.org/10.1111/cch.12137

17.

Crnic

K. A.

Greenberg

M. T.

(1990). Minor parenting stresses with young children. Child Development, 61(5), 1628–1637. https://doi.org/10.1111/j.1467-8624.1990.tb02889.x

18.

Crowley

J. E.

(2014). Staying at home or working for pay? Attachment to modern mothering identities. Sociological Spectrum, 34(2), 114–135. https://doi.org/10.1080/02732173.2014.878605

19.

Curran

P. J.

Obeidat

Losardo

(2010). Twelve frequently asked questions about growth curve modeling. Journal of Cognition and Development, 11(2), 121–136. https://doi.org/10.1080/15248371003699969

20.

Deater-Deckard

(1998). Parenting stress and child adjustment: Some old hypotheses and new questions. Clinical Psychology Science and Practice, 5(3), 314–332. https://doi.org/10.1111/j.1468-2850.1998.tb00152.x

21.

Deater-Deckard

(2008). Parenting stress. Yale University Press.

22.

de Maat

D. A.

Jansen

P. W.

Prinzie

Keizer

Franken

I. H.

Lucassen

(2021). Examining longitudinal relations between mothers’ and fathers’ parenting stress, parenting behaviors, and adolescents’ behavior problems. Journal of Child and Family Studies, 30, 771–783. https://doi.org/10.1007/s10826-020-01885-0

23.

Dey

B. K.

Roy

Reshmi

T. H.

(2021). Parenting stress of mothers and behavior problems of children. Jagannath University Journal of Life & Earth Sciences, 6(2), 161–170.

24.

Dillaway

Paré

(2008). Locating mothers: How cultural debates about stay-at-home versus working mothers define women and home. Journal of Family Issues, 29(4), 437–464. https://doi.org/10.1177/0192513X07310309

25.

Eid

Langeheine

Diener

(2003). Comparing typological structures across cultures by Multigroup Latent Class analysis. Journal of Cross-Cultural Psychology, 34(2), 195–210. https://doi.org/10.1177/0022022102250427

26.

Flaherty

S. C.

Sadler

L. S.

(2022). Parenting stress among adolescent mothers: An integrative literature review. Western Journal of Nursing Research, 44(7), 701–719. https://doi.org/10.1177/01939459211014241

27.

Forgays

D. K.

Forgays

D. G.

(1993). Personal and environmental factors contributing to parenting stress among employed and nonemployed women. European Journal of Personality, 7(2), 107–118. https://doi.org/10.1002/per.2410070203

28.

Gallie

Russell

(2009). Work-family conflict and working conditions in Western Europe. Social Indicators Research, 93(3), 445–467. https://doi.org/10.1007/s11205-008-9435-0

29.

Gómez

(2006). Quality of family and work roles and its relationship with health indicators in men and women. Sex Roles, 55(11–12), 787–799. https://doi.org/10.1007/s11199-006-9132-x

30.

Goode

W. J.

(1960). A theory of role strain. American Sociological Review, 25, 483–496. https://doi.org/10.2307/2092933

31.

Haggag

A. K.

Geser

Ostermann

Schusterschitz

(2011). Depressive symptoms in mothers: The role of employment and role quality. Journal of Workplace Behavioral Health, 26(4), 313–333.

32.

Hair

J. F.

Jr Black

W. C.

Babin

B. J.

Anderson

R. E.

Tatham

R. L.

(Eds.). (2006). Multivariate data analysis (6th ed.). Pearson Education.

33.

Hong

Y.-J.

Lee

(2019). The effect of parenting stress on social interactive parenting with a focus on Korean employed mothers’ parenting support from ecological contexts. Children and Youth Services Review, 96, 308–315. https://doi.org/10.1016/j.childyouth.2018.10.038

34.

Jiang

Wang

Yang

Choi

J. K.

(2023). Bidirectional relationships between parenting stress and child behavior problems in multi-stressed, single-mother families: A cross-lagged panel model. Family Process, 62(2), 671–686. https://doi.org/10.1111/famp.12796

35.

Kahn

Cuthbertson

(1998). A comparison of the self-reported mental and physical stress of working and full-time homemaker mothers—A UK pilot study. Stress Medicine, 14(3), 149–154. https://doi.org/10.1002/(sici)1099-1700(199807)14:3<3.0.co;2-7

36.

Kankaraš

Moors

Vermunt

J. K.

(2010). Testing for measurement invariance with latent class analysis. In Davidov

Schmidt

Billiet

(Eds.), Cross-cultural analysis: Methods and applications (pp. 359–384). Routledge.

37.

Keane

S. P.

Calkins

S. D.

(2004). Predicting kindergarten peer social status from toddler and preschool problem behavior. Journal of Abnormal Child Psychology, 32(4), 409–423. https://doi.org/10.1023/b:jacp.0000030294.11443.41

38.

Kim

N. H.

Lee

E. J.

Kwak

S. Y.

Park

M. R.

(2013). A phenomenological study on the experiences of parenting burden of working mother with young children in Korea. Korean Journal of Women Health Nursing, 19(3), 188–200. https://doi.org/10.4069/kjwhn.2013.19.3.188

39.

Mendell

N. R.

Rubin

D. B.

(2001). Testing the number of components in a normal mixture. Biometrika, 88(3), 767–778. https://doi.org/10.1093/biomet/88.3.767

40.

McBride

A. B.

(1990). Mental health effects of women's multiple roles. Journal of Addictions Nursing, 2(3), 4–14. https://doi.org/10.3109/10884609009149669

41.

McCarten

(2003). Differences in depression, anxiety, self-esteem, and parenting stress between employed and stay-at-home mothers: A meta-analysis. Clemson University.

42.

McCutcheon

A. L.

Hagenaars

J. A.

(1997). Comparative social research with multi-sample latent class models. In Rost

Langeheine

(Eds.), Applications of latent trait and latent class models in the social sciences (pp. 266–277). Waxmann.

43.

Ministry of Education. (2023). 2023 Education in Korea. https://english.moe.go.kr/sub/infoRenewal.do?m=0302&page=0302&s=english

44.

Ministry of Gender Equality and Family. (2021). Analysis of the gender equality status survey (Vol. 2021). Ministry of Gender Equality and Family.

45.

Moen

Dempster-McClain

Williams

R. M.

(1989). Social integration and longevity: An event history analysis of women’s roles and resilience. American Sociological Review, 54, 635–647. https://doi.org/10.2307/2095884

46.

Moors

(2004). Facts and artefacts in the comparison of attitudes among ethnic minorities. A multigroup latent class structure model with adjustment for response style behavior. European Sociological Review, 20(4), 303–320. https://doi.org/10.1093/esr/jch026

47.

Morin

A. J. S.

Meyer

J. P.

Creusier

Biétry

(2016). Multiple-group analysis of similarity in latent profile solutions. Organizational Research Methods, 19(2), 231–254. https://doi.org/10.1177/1094428115621148

48.

Muthén

L. K.

Muthén

B. O.

(2007). What is a good value of entropy. http://www.statmodel.com/discussion/messages/13/2562.html?1237580237.

49.

Neece

C. L.

Green

S. A.

Baker

B. L.

(2012). Parenting stress and child behavior problems: A transactional relationship across time. American Journal on Intellectual and Developmental Disabilities, 117(1), 48–66. https://doi.org/10.1352/1944-7558-117.1.48

50.

Nomaguchi

K. M.

Milkie

M. A.

(2003). Costs and rewards of children: The effects of becoming a parent on adults’ lives. Journal of Marriage and Family, 65(2), 356–374. https://doi.org/10.1111/j.1741-3737.2003.00356.x

51.

Nomaguchi

K. M.

Milkie

M. A.

(2020). Parenthood and well-being: A decade in review. Journal of Marriage and Family, 82(1), 198–223. https://doi.org/10.1111/jomf.12646

52.

Nylund-Gibson

Grimm

Quirk

Furlong

(2014). A latent transition mixture model using the three-step specification. Structural Equation Modeling A Multidisciplinary Journal, 21(3), 439–454. https://doi.org/10.1080/10705511.2014.915375

53.

Nylund

K. L.

Asparouhov

Muthén

B. O.

(2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling A Multidisciplinary Journal, 14(4), 535–569. https://doi.org/10.1080/10705510701575396

54.

Ostberg

Hagekull

(2013). Parenting stress and external stressors as predictors of maternal ratings of child adjustment. Scandinavian Journal of Psychology, 54(3), 213–221. https://doi.org/10.1111/sjop.12045

55.

Owen

A. E.

Thompson

M. P.

Kaslow

N. J.

(2006). The mediating role of parenting stress in the relation between intimate partner violence and child adjustment. Journal of Family Psychology, 20(3), 505–513. https://doi.org/10.1037/0893-3200.20.3.505

56.

Parkes

Sweeting

Wight

(2015). Parenting stress and parent support among mothers with high and low education. Journal of Family Psychology, 29(6), 907–918. https://doi.org/10.1037/fam0000129

57.

Park

Kim

H. K.

Mastergeorge

A. M.

(2023). Longitudinal trajectories of maternal parenting stress in Korean families: Children’s executive function and school adjustment. Social Development, 32(3), 813–829. https://doi.org/10.1111/sode.12661

58.

Park

H. J.

(2016). The effects of maternal parenting stress trajectory on affective parenting and socioemotional development of preschooler: Differences between employed and unemployed mothers. Korean Journal of Child Studies, 37(5), 33–52. https://doi.org/10.5723/kjcs.2016.37.5.33

59.

Peel

McLachlan

G. J.

(2000). Robust mixture modelling using the t distribution. Statistics and Computing, 10(4), 339–348. https://doi.org/10.1023/a:1008981510081

60.

Pope

A. W.

Tillman

Snyder

H. T.

(2005). Parenting stress in infancy and psychosocial adjustment in toddlerhood: A longitudinal study of children with craniofacial anomalies. Cleft Palate-Craniofacial Journal, 42(5), 556–559. https://doi.org/10.1597/04-066r.1

61.

Rajgariah

Malenahalli Chandrashekarappa

Venkatesh Babu

D. K.

Gopi

Murthy Mysore Ramaiha

Kumar

(2021). Parenting stress and coping strategies adopted among working and non-working mothers and its association with socio-demographic variables: A cross-sectional study. Clinical Epidemiology and Global Health, 9, 191–195. https://doi.org/10.1016/j.cegh.2020.08.013

62.

Rizzo

K. M.

Schiffrin

H. H.

Liss

(2013). Insight into the parenthood paradox: Mental health outcomes of intensive mothering. Journal of Child and Family Studies, 22(5), 614–620. https://doi.org/10.1007/s10826-012-9615-z

63.

Rogers

M. A.

Wiener

Marton

Tannock

(2009). Supportive and controlling parental involvement as predictors of children’s academic achievement: Relations to children’s ADHD symptoms and parenting stress. School Mental Health, 1(2), 89–102. https://doi.org/10.1007/s12310-009-9010-0

64.

Schwarz

(1978). Estimating the dimension of a model. Annals of Statistics, 6(2), 461–464. https://doi.org/10.1214/aos/1176344136

65.

Sclove

S. L.

(1987). Application of model-selection criteria to some problems in multivariate analysis. Psychometrika, 52(3), 333–343. https://doi.org/10.1007/bf02294360

66.

Sieber

S. D.

(1974). Toward a theory of role accumulation. American Sociological Review, 39, 567–578. https://doi.org/10.2307/2094422

67.

Siegler

DeLoache

Eisenberg

(2010). Theories of social development: Psychoanalytic, learning, social cognition, and ecological. In Siegler

R. S.

DeLoache

J. S.

Eisenberg

(Eds.), How children develop (3rd ed.). Worth Publishers.

68.

Sohn

(2016). Effects of mild response style on psychological measurement: Application of a multi-group latent-class factor analysis. Korean Journal of Educational Research, 54(1), 105–125.

69.

Song

Y. J.

Lee

M. R.

Chun

H. Y.

(2014). Parenting stress changes in both of continuous working and non-working mothers after the birth of their first child: A focus on the effects of the values, knowledge and expectations about their children. Korean Journal of Child Studies, 35(5), 15–35. https://doi.org/10.5723/kjcs.2014.35.5.15

70.

Statistical Innovations. (2021). Latent GOLD 6.0 [Software]. Statistical Innovations, Incorp.

71.

Statistics Korea. (2024a). 2024 regional employment survey: Employment status of married women. https://kostat.go.kr/board.es?mid=a10301010000&bid=211&act=view&list_no=433804

72.

Statistics Korea. (2024b). Provisional birth and death statistics of 2023. https://kostat.go.kr/board.es?mid=a10301010000&bid=204&act=view&list_no=429586.

73.

Steenkamp

Baumgartner

(1998). Assessing measurement invariance in cross-national consumer research. Journal of Consumer Research, 25(1), 78–107. https://doi.org/10.1086/209528

74.

Stone

L. L.

Mares

S. H.

Otten

Engels

R. C.

Janssens

J. M.

(2016). The co-development of parenting stress and childhood internalizing and externalizing problems. Journal of Psychopathology and Behavioral Assessment, 38, 76–86. https://doi.org/10.1007/s10862-015-9500-3

75.

Tan

T. X.

Wang

Ruggerio

A. D.

(2017). Childhood adversity and children’s academic functioning: Roles of parenting stress and neighborhood support. Journal of Child and Family Studies, 26(10), 2742–2752. https://doi.org/10.1007/s10826-017-0775-8

76.

Vermunt

J. K.

Magidson

(2021). How to perform three-step latent class analysis in the presence of measurement non-invariance or differential item functioning. Structural Equation Modeling: A Multidisciplinary Journal, 28(3), 356–364. https://doi.org/10.1080/10705511.2020.1818084

77.

Williford

A. P.

Calkins

S. D.

Keane

S. P.

(2007). Predicting change in parenting stress across early childhood: Child and maternal factors. Journal of Abnormal Child Psychology, 35(2), 251–263. https://doi.org/10.1007/s10802-006-9082-3

78.

Yeo

K. J.

Teo

S. L.

(2013). Child behavior and parenting stress between employed mothers and at home mothers of preschool children. Procedia – Social and Behavioral Sciences, 90, 895–903. https://doi.org/10.1016/j.sbspro.2013.07.166

79.

Yoon

J.-W.

Hwang

R. I.

Cho

H. H.

(2009). Parenting stress and related factors of employed and non-employed mothers with preschool children. Korean Journal of Women Health Nursing, 15(4), 294–302. https://doi.org/10.4069/kjwhn.2009.15.4.294