Socioeconomic disparities in optimism and pessimism among children and adolescents: Does age matter?

doi:10.21203/rs.3.rs-4851024/v1

Download PDF

Article

Socioeconomic disparities in optimism and pessimism among children and adolescents: Does age matter?

https://doi.org/10.21203/rs.3.rs-4851024/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Socioeconomic status (SES) exhibits graded relationships with psychological resources. Dispositional optimism, a crucial psychological asset, encompasses both optimism and pessimism, two related but separated facets considered to be asymmetric in susceptibility to the impact of SES in adults. Little research has explored (a) the effects of SES disparities on both facets of dispositional optimism in children and adolescents and (b) and whether the effects of SES on these facets change with age on the basis of measurement invariance testing. Addressing these gaps, the present study investigated the associations between SES and optimism and pessimism, and whether these relationships change with age in children and adolescents. A sample of 4730 Chinese participants (ages 9–19) from diverse SES backgrounds completed assessments of family SES, optimism, and pessimism. Analyses revealed a positive SES gradient with optimism and a negative gradient with pessimism. While the SES-optimism link was consistent across ages, the SES-pessimism correlation strengthened with age, suggesting a developmental shift in SES's psychological impact. Our findings shed light on the nuanced and complex ways in which SES intersects with optimism and pessimism during critical developmental periods—potentially guiding more targeted interventions aimed at the stimulation of positive changes in both optimism and pessimism.

Biological sciences/Psychology

Biological sciences/Psychology/Human behaviour

socioeconomic status

optimism

pessimism

age

children

adolescents

Numerous studies have illuminated the profound relationship between family SES and individuals’ health, well-being, and psychological resources^1–6. Serving as a significant psychological resource, dispositional optimism has been associated with various adaptive outcomes^7,8. Research indicates that SES can positively predict individuals’ dispositional optimism^2,9,10.

Although most studies consider dispositional optimism as unidimensional, recent evidence^11–14 suggests it consists of two distinct factors: optimism and pessimism, relating to individuals’ expectations of positive and negative outcomes respectively. This seemingly simple distinction might potentially activate two distinct motivational systems—approach and avoidance—which could in turn individually influence a person’s thoughts, emotions, and volitional actions^7,15,16. Recent research has found that optimism and pessimism exhibit differential effects on various indicators of individuals’ health and subjective well-being^12,13,16,17. However, a significant question remains: does the relationship between SES and both optimism and pessimism vary?

Few studies have investigated the specific relationship between SES and both optimism and pessimism, with some suggesting an asymmetric connection between them^3,18,19. For instance, Taylor and Seeman employed data from four adult datasets and found that SES was significantly correlated with pessimism but not optimism¹⁹. Robb et al. obtained similar results in elderly participants¹⁸. Heinonen et al. analysed the relationship between individuals’ childhood family SES and their optimism and pessimism in adulthood³. They again found that after controlling for adult SES, childhood SES was only significantly correlated with adult pessimism but not optimism. These studies consistently found that family SES significantly predicted individuals’ pessimism but not necessarily their optimism. Whether the same pattern holds for optimism and pessimism during childhood and adolescence remains unclear.

Childhood and adolescence are critical periods for the formation and development of personality including optimism and pessimism^14,20. Optimism during childhood and adolescence is not only closely associated with youth’s current academic achievements, interpersonal relationships, and life satisfaction^21–23, but also serves as a predictor of well-being in adulthood^24,25. Furthermore, studies have shown optimism during childhood and adolescence can significantly predict individuals’ income and education levels in adulthood––with individuals possessing higher optimism likely achieving higher SES^26,27. Thus, understanding the relationship between family SES and the levels of optimism and pessimism of children and adolescents can shed light on the dynamics of optimism and pessimism more broadly––providing a basis for targeted interventions to foster positive personality development in children and adolescents and facilitate social mobility.

According to the Reserved Capacity Model²⁸, compared to children and adolescents from high SES families, their counterparts from low SES families are more likely to experience chronic stress in their lives due to the scarcity of material resources²⁹. Such stressors might include factors such as living in overcrowded and noisy environments or facing travel limitations due to subpar public transport amenities. These demanding environments can significantly deplete the cognitive and psychological resources of children and adolescents. On the other hand, low SES also implies that parents often confront greater survival pressures. This may lead them to be less patient in child-rearing and turn to harsh parenting^30,31, causing tension in parent–child relationships³². Furthermore, these parents may be less likely to serve as effective role models for their children or teach them adaptive strategies for coping with stress and setbacks^33,34. Consequently, children and adolescents from low SES families may have fewer opportunities or lack chances to develop their capacities²⁸. As a result, children and adolescents from low SES families may exhibit insufficient reserved capacities, leading to a lack of control over the present and future^28,30 —leading to fewer positive expectations and more negative expectations about the future. Accordingly, we hypothesize that family SES positively predicts the optimism of children and adolescents while negatively predicting their pessimism.

To date, what has been less extensively investigated is whether the links between SES, optimism, and pessimism change with increasing age during childhood and adolescence. Understanding this topic is vital for policy, practice, and research because it can offer insights into when socioeconomic inequalities reproduce and how they can be altered over the life course.

The Reserved Capacity Model²⁸ offers a possible framework for understanding how family SES may relate to children’s and adolescents’ optimism and pessimism. It suggests that the challenging environment faced by children and adolescents from low family SES backgrounds, along with their limited chances to develop their abilities, may persist as they grow older without significant change³⁵. Thus, the relationship between SES and optimism may not vary markedly with age. By contrast, Taylor and Seeman found that irrational beliefs correlate only with individuals’ pessimism, not with optimism¹⁹. Irrational beliefs represent a type of cognition³⁶. Just as family SES has a significant impact on the cognitive development of children and adolescents^37–39, the evolution of irrational beliefs is also influenced by individuals’ educational attainment—an important indicator of SES⁴⁰.

The Cumulative Advantage Theory⁴¹ also appears germane in this context. This framework posits that individuals or groups who have a small advantage at the beginning of a process tend to accumulate more advantages over time, while those who start with a disadvantage face increasing challenges and disadvantages. Thus, children and adolescents from higher SES families are likely to show greater cognitive development compared to those from poorer backgrounds. This difference accumulates and widens over time as youth age³⁷. Even though irrational beliefs are merely a subset of cognition, the broadening gap between SES and cognitive development as age progresses serves as indirect evidence for our predictions. Drawing on both theoretical frameworks, we hypothesize that the relationship between SES and pessimism is likely to become stronger with increasing age.

Before delving into the above hypothesis, it is imperative to address the methodological concerns related to measuring optimism and pessimism. To explore the associations between SES and optimism as well as pessimism, the measures of optimism and pessimism should function equivalently in more socioeconomically diverse samples. However, few studies have done so before presenting substantive associations between SES and optimism and pessimism. Establishing measurement invariance (MI) will not only validate the constructs but also provide a clearer understanding of their relationships^42–44.

In sum, the purpose of this study was to (a) evaluate the MI of the scales to assess optimism and pessimism across different SES groups, (b) investigate the relationship between family SES and optimism as well as pessimism in children and adolescents, and (c) examine if and how this relationship changes with age. As the Chinese version measure of optimism and pessimism has demonstrated similar bi-dimensional structure across age groups¹⁴, we hypothesized that the MI across SES groups would be established. Based on the literature review, the present study tested the following three hypotheses:

(1) The MI of the scales for optimism and pessimism would be established (H1); (2) Family SES positively predicts the optimism of children and adolescents while negatively predicting their pessimism (H2);

(3) The relationship between SES and optimism may not vary markedly with age, while the relationship between SES and pessimism is likely to strengthen with increasing age (H3).

Participants and procedure

In total, 4,820 participants were initially recruited from 12 schools: four primary schools, four junior high schools, and four senior high schools in central and southwest China based on different socioeconomic development of the areas where the schools are located^45,46. The data from 90 participants were excluded because these participants failed to complete either or both questionnaires about socioeconomic status, optimism and pessimism.

Students in Grade 4 through Grade 11 were recruited. Grade 4 was chosen because the lower bound was based on competence: children aged around 9–10 years (Grade 4 in China) have been found to be able to complete personality self-reports⁴⁷. Grade 11 was selected as the upper bound because adolescents in Grade 12 had to prepare for college entrance examinations, and their parents and teachers did not agree to participate in the survey. The final useable sample thus comprised 4,730 children and adolescents (age range: 9–19 years old, M_age = 13.30 years, SD = 2.26; 2,535 boys (53.60%)) in Grades 4–11. The final sample included at least 392 participants at each grade level, and was diverse in terms of socioeconomic status. There were numerous observations at each SES category: the percentages of participants with low, medium and high SES were 16.15%, 42.48% and 41.37% correspondingly. More descriptive information on the sample can be seen in Table 1. Missing data were less than 1% and were estimated with Expectation Maximization procedure in SPSS 25 (SPSS, Chicago, IL, USA).

Initially, we approached the principals of the targeted schools to elucidate the research protocol and sought their endorsement and collaboration. Thereafter, we provided a detailed verbal overview of the study's scope and objectives to students, their guardians, and class teachers. Following this, written informed consent was secured from the guardians and all the participants. Subsequently, the survey was administered in class. Two trained graduate students of psychology in each classroom explained the requirements of the survey with standard instructions emphasizing the authenticity, independence, and integrity of all responses. The study was conducted according to the guidelines of the Declaration of Helsinki and received ethics approval from the Research Ethics Committee of Wuhan Sports University (approval no. 2019001, 14 April 2019).

Table 1

Descriptive information on the sample (n **= 4,730**)
	N	Boys (%)	M	SD
Age
Grade 4	514	56.80	9.66	0.51
Grade 5	566	55.10	10.59	0.50
Grade 6	392	51.30	11.66	0.53
Grade 7	645	54.00	12.70	0.52
Grade 8	709	56.60	13.55	0.51
Grade 9	605	58.00	14.72	0.56
Grade 10	825	50.30	15.53	0.47
Grade 11	474	45.40	16.62	0.60
SES
Low	764	53.50	1.50	0.65
Medium	2009	53.20	4.10	0.81
High	1957	54.00	7.01	1.01
Note. M = Mean, SD = standard deviation, SES = socioeconomic status. Family SES was categorized into three groups: low, medium, and high, based on the cutoff points suggested by Boyce et al.⁴⁹ for classifying SES scores.

Measures

Family socioeconomic status

Family socioeconomic status (SES) was assessed by the Family Affluence Scale (FAS)⁴⁸. It included four items indicating the family affluence: ownership of a family car, van, or truck (0, 1, 2 or more), own bedroom (no = 0, yes = 1), travel times during the past 12 months (0, 1, 2, 3 or more) and ownership of family computer(s) (0, 1, 2, 3 or more). High total scores indicate high family affluence. Following standards of the SES categories⁴⁹, the total scores were divided into three categories: low SES (scores ranging from 0 to 2), middle SES (scores ranging from 3 to 5), and high SES (scores ranging from 6 to 9). Compared to traditional indicators of family socioeconomic status (e.g., family income, parental education level or parental occupation), the FAS generally yield good response rates and is more suitable for children and adolescents to answer than other measures^48,50,51. Moreover, the FAS has been well validated with moderate internal consistency and good validity. It is widely used among children and adolescents both in China and western countries^48,50. The Cronbach’s α coefficient and McDonald’s omega of the FAS in this sample were 0.54 and 0.59 respectively, which was in line with previous studies^14,50 and reflects the relatively small number of items.

Optimism and pessimism scales

The Chinese version of the Life Orientation Test-Revised (CLOT-R) is a measure of optimism and pessimism⁵². The CLOT-R consists of five positively phrased items (e.g., ‘When things are bad, I expect them to go better’) constituting the Optimism subscale, and five negatively phrased items (e.g., ‘I hardly ever expect things to go my way’) constituting the Pessimism subscale. Response options ranged from ‘strongly disagree’ to ‘strongly agree’ for each statement on a 5-point Likert scale scored from 1 to 5. High Optimism subscale scores indicate greater optimism, and high Pessimism subscale scores represent greater pessimism. The CLOT-R has been used in Chinese children and adolescents with good validity and reliability^14,16,53. The bi-dimensional structure of the CLOT-R among Chinese children and adolescents was supported in previous studies^14,52. In this study, the CLOT-R also showed good structural validity of bi-dimensional model: χ²/df = 16.78, RMSEA (Root Mean Square Error of Approximation) = 0.06, CFI (Comparative Fit Index) = 0.95, TLI (Tucker-Lewis Index) = 0.94, SRMR (Standardized Root Mean Square Residual) = 0.03, which was better than that of one-factor model: χ²/df = 97.24, RMSEA = 0.14, CFI = 0.70, TLI = 0.61, SRMR = 0.11. The Optimism and Pessimism subscales had high internal reliability in this study: Cronbach’s α coefficient and the McDonald's omega for the Optimism subscale were both 0.71, and for the Pessimism subscale were 0.80 and 0.81, respectively.

Statistical analysis

Measurement invariance test

As a preliminary step, we sought to ensure the robustness of our measures by testing their measurement invariance across SES groups. Specifically, to examine measurement invariance of optimism and pessimism across SES, we investigated four types of measurement invariance: configural invariance, metric invariance, scalar invariance, and strict invariance using Mplus 8.4⁵⁴. Initially, we assessed configural invariance to examine whether the bi-dimensional structure of optimism and pessimism was equivalent across different SES groups. In this configural model, all items pertaining to optimism and pessimism were used as manifest indicators to define latent variables for optimism and pessimism within each SES group. Subsequently, we examined metric, scalar, and strict invariance by progressively constraining the models and comparing the changes in fit. The metric invariance involved setting the factor loadings equal across SES groups, scalar invariance entailed constraining the intercepts to be equal across SES groups, and strict invariance required fixing the residual variances to be equal across SES groups.

MI of a specific level was identified to be established by comparing the fit indices with those obtained from a less constrained model⁵⁵. Due to the sensitivity of the chi-square difference test to sample size, we used the changes in the RMSEA (ΔRMSEA) and the changes in the CFI (ΔCFI) as additional criteria to assess measurement invariance. Specifically, the establishment of MI at a given level was determined when ΔRMSEA ≤ 0.015 or ΔCFI ≤ 0.010^55,56.

Moderation model test

To test whether there are socioeconomic differences in optimism and pessimism among children and adolescents, and whether the relationship between family SES and optimism and pessimism varies by age, we used the PROCESS macro for SPSS (Model 1) to progress the moderation model test⁵⁷. PROCESS performs computations to derive bias-corrected and accelerated bootstrapped confidence intervals (CI) for the magnitude of each direct or conditional effect using 10,000 re-samples. A confidence interval that does not include zero indicates a significant effect. Prior to conducting the analysis, all variables (except for sex) were transformed into z-scores to obtain standardized coefficients.

Preliminary analyses

As seen in Table 2, the preliminary results indicated no violation in the distribution of primary variables, with the absolute magnitude of skewness and kurtosis values remaining below 2⁵⁸. The correlations of interest were in the expected directions. Specifically, SES exhibited a significant positive relationship with optimism (r = .19, p < .001), indicating that individuals with higher SES tend to expect positive outcomes in the future. Conversely, SES displayed a significant negative association with pessimism (r = − .12, p < .001), suggesting that individuals with greater socioeconomic advantages tend to hold fewer negative expectancies about the future.

Table 2

Univariate and bivariate statistics for all study variables. (n **= 4,730**)
Variables	M	SD	Min	Max	Skew	Kurtosis	1	2	3	4
1. Sex	0.46	0.50	0	1	-	-	-
2. Age	13.30	2.26	9	19	-0.17	-1.03	0.03	-
3. SES	4.88	2.18	0	9	-0.09	-0.74	-0.01	-0.07^***	-
4. Optimism	17.48	4.08	5	25	-0.35	-0.02	-0.02	-0.04^**	0.19^***	-
5. Pessimism	14.42	4.66	5	25	0.19	-0.43	-0.01	-0.20^***	-0.12^***	-0.27^***
Note. Sex was coded: boy = 0, girl = 1. ^ p < .01, ^*p < .001

Relatively consistent with the previous study¹⁴, age exhibited distinct associations with both optimism and pessimism in the present study. There were significant weak negative correlations between age and optimism (r = − .04, p < 0.01), and age and pessimism (r = − .20, p < .001). There was also a weak correlation between optimism and pessimism (r = − .27, p < .001). These results suggest that optimism and pessimism are related but different concepts, and best conceptualised as bi-dimensional constructs in the context of child and adolescent research. Sex had no significant correlation with other variables in this study. The role of sex was therefore not considered in subsequent analyses⁵⁹.

Measurement invariance test

Table 3 displays the model fit indices of the four models examined in the analyses. For the configural model, we tested whether the underlying structure of optimism and pessimism remained consistent across different SES groups. Its fit indices met the psychometric standards, with CFI and TLI exceeding 0.90 and RMSEA below 0.08. The configural invariance was established, indicating that it could serve as the baseline model for subsequent analyses. In the metric model, the factor loadings (i.e., the relationship between the items and latent factors) were constrained to be equal across the SES groups. We tested whether the items had the same meaning and significance in measuring the latent constructs (optimism and pessimism) across different SES groups. According to the fit indices shown in Table 3, metric invariance was established, with the changes of CFI and RMSEA between metric model and configural model below 0.01 and 0.015, respectively (ΔCFI < 0.010, ΔRMSEA < 0.015). In the scalar model, besides constraining the factor loadings, the intercepts (item means) were also constrained to be equal across SES groups. We tested whether the items had the same intercepts, which means the same level of optimism and pessimism across different SES groups. As shown in Table 3, the changes of CFI and RMSEA between scalar model and metric model were below the critical thresholds, indicating the scalar invariance was established. Based on the scalar measurement invariance, residual variances of the observed indicators were further constrained to be equal across SES groups in the strict model. The fit indices of the strict model shown that the strict measurement invariance was also established.

Table 3

Results of measurement invariance of the bi-dimensional structure of optimism and pessimism across SES
Model	χ²	df	CFI	TLI	RMSEA [90%CI]	ΔCFI	ΔRMSEA
Configural	713.19	102	0.944	0.926	0.062 [0.057, 0.066]	–	–
Metric	728.32	118	0.945	0.937	0.057 [0.053, 0.061]	0.001	0.005
Scalar	760.92	134	0.943	0.943	0.054 [0.051, 0.058]	0.002	0.003
Strict	822.43	154	0.939	0.947	0.052 [0.049, 0.056]	0.004	0.002
Note: ΔCFI and ΔRMSEA were obtained by comparing the CFI (Comparative Fit Index) and the RMSEA (Root Mean Square Error of Approximation) of a model at a certain level with the CFI and the RMSEA of a model with fewer constraints at the previous level.

Together, these results suggest that measurement invariance was achieved for each level, which supported the hypothesis H1. The bi-dimensional structure of optimism and pessimism remained stable across different SES backgrounds, indicating that the observed differences in optimism and pessimism scores across SES levels can be attributed to genuine variations in these psychological constructs rather than being influenced by potential measurement biases.

Moderation model test

We first tested the study hypotheses (H2 & H3) for optimism. Results shown in Table 4 revealed that family SES positively predicted optimism of children and adolescents (regression coefficient = 0.19, p < .001, 95%CI [0.16, 0.22]): children and adolescents from higher SES families were more likely to have positive expectations about their future. Moreover, consistent with the previous study¹⁴, age significantly but slightly predicted optimism (regression coefficient = -0. 03, p < .05, 95%CI [-0.06, -0.002]): adolescents were more likely than younger youths to have fewer positive expectations regarding the future. No interaction effect between SES and age was found for optimism (regression coefficient = 0. 003, p > .05, 95%CI [-0.03, 0.03]). Given the significant correlation between optimism and pessimism (r = − .27, p < .001), we further tested the moderation model with pessimism as a covariate. Controlling for pessimism yielded similar results (see Supplementary Table 1).

Table 4

Testing the moderating effect of SES on optimism and pessimism respectively. (n **= 4,730**)
Regression Model		Global Fit Index			Significance of Regression Coefficient
Outcome	Predictors	R²	F	β	SE	95% CI
Optimism	SES			0.19	0.01	[0.16, 0.22]
	Age			-0.03	0.01	[-0.06, -0.002]
	SES×Age	0.04	59.88^***	0.003	0.01	[-0.03, 0.03]
Pessimism	SES			-0.13	0.01	[-0.16, -0.10]
	Age			-0.21	0.01	[-0.24, -0.18]
	SES×Age	0.06	103.68^***	-0.09	0.01	[-0.12, -0.06]
Note. All variables were converted to z-scores prior to analysis. The 95% confidence intervals in bold indicated the regression coefficients were significant.

With respect to pessimism, family SES negatively predicted pessimism of children and adolescents (regression coefficient = -0.13, p < .001, 95%CI [-0.16, -0.10]; see Table 4), such that children and adolescents from lower SES families were likely to have more negative expectations about their future. Moreover, consistent with the previous study¹⁴, age significantly and negatively predicted pessimism (regression coefficient = -0. 21, p < .001, 95%CI [-0.24, -0.18]): adolescents, compared to younger youths, were likely to have fewer negative expectations regarding the future. The moderating effect of age on the association between family SES and children’s and adolescents’ pessimism was statistically significant (regression coefficient = -0. 09, p < .001, 95% CI [-0.12, -0.06]). To interpret this moderation, an interaction plot was drawn following the simple slopes (Fig. 1A). The simple effects analysis indicated that for adolescents, the negative relationship between SES and pessimism was stronger as implied by the high beta (B_simple = -0.22, t = -10.63, p < 0.001), while for younger youths, the beta was still negative but weaker (B_simple = -0.04, t = -2.01, p = 0.04). We further plot the Johnson-Neyman Fig. 6⁰ (Fig. 1B) for the moderating effect of age on the relationship between family SES and pessimism based on the output from PROCESS macro for SPSS (Model 1)⁵⁷. According to Fig. 1B, there was a Johnson-Neyman point of age (z-score of age = -1.01, and corresponding age is 11.02 years) at which age point the effect of family SES on pessimism turned from non-significant region to significant region. As age increased, the negative association between family SES and pessimism became stronger. We also tested the moderation model with optimism as a covariate. Controlling for optimism yielded similar results (see Supplementary Table 1).

Using a socioeconomically diverse sample with a sufficient number of participants (at least n = 392) at each grade (Grades 4–11), the current study explored the effects of SES disparities on both facets of dispositional optimism in children and adolescents and whether the profiles of SES effects vary with age on the basis of measurement invariance testing. Results showed that the sub-scales for optimism and pessimism in children and adolescents demonstrate measurement invariance across SES categories. Building on this, results indicated that the levels of optimism and pessimism in children and adolescents exhibit a gradient based on family SES. Furthermore, the linkage between SES and pessimism strengthens with increasing age.

To our knowledge, the measurement invariance of optimism and pessimism across different categories of SES has not been tested in previous studies focusing on the relationship between SES and optimism and pessimism^3,18. This introduces ambiguity in interpreting the observed relationships between variables. That is, these relationships could potentially reflect the true underlying associations between concepts, or they may arise from individuals from different SES backgrounds responding differently to the measurement items for optimism and pessimism. Therefore, prior to separately examining the relationships between SES and optimism and pessimism, this study assessed whether the measurement of optimism and pessimism demonstrates cross-SES measurement invariance. The results indicated that the measurement of optimism and pessimism exhibits strict factorial invariance across different SES groups. This measurement invariance serves as a prerequisite for a subsequent, non-ambiguous understanding of the relationship between SES and optimism and pessimism, suggesting that the observed relationships can indeed reflect genuine associations between the concepts and are meaningful^42–44.

As predicted, the results of the correlation and regression analyses demonstrate that high SES is associated with higher optimism and lower pessimism among children and adolescents, whereas low SES is associated with lower optimism and higher pessimism. In comparison to children and adolescents from low SES families, those from high SES families tend to have better living environments, increased access to resources, and fewer opportunities to face chronic negative events²⁹. Moreover, parents from high SES families are more likely to adopt positive parenting styles³⁰, establish better parent–child relationships³², guide their children in adopting more adaptive coping strategies when dealing with stress, and serve as more effective role models^33,34. These supportive physical and interpersonal environments help children and adolescents from high SES families develop reserved capacities during childhood and adolescence²⁸. With more internal resources to cope with uncertain futures, they are more likely to have a sense of control over the future. Consequently, they exhibit higher optimism and lower pessimism.

Just as there is a health gradient based on SES in the field of physical health, this study found that the domain of optimism and pessimism also exhibits a gradient influenced by SES. That is, the differences in family SES ultimately appears to be internalised to create disparities in optimism and pessimism. This result differs from previous findings based on adult or elderly participants. In those populations, SES was only related to their pessimism, showing no significant association with optimism^3,18,19. Compared to children and adolescents, the personality development of adults or the elderly is related to the social roles they play at different life stages and their experiences in various fields such as work and marriage^61,62. The transition of multiple social roles and life experiences in different fields for adults or the elderly may potentially attenuate the effects of SES on the development and changes in optimism and pessimism. However, for children and adolescents, the family is their primary environment, and their social roles are relatively singular, potentially making them more susceptible to the influence of family SES.

For children and adolescents, the relationship between family SES and their optimism, as well as SES and their pessimism, did not exhibit asymmetry. However, a difference became apparent in whether these relationships changed with age. Our study found that the positive relationship between SES and optimism was not influenced by age, but the negative relationship between SES and pessimism changed with age. Specifically, as age increased, the relationship between SES and pessimism strengthened. The beneficial effects of higher family SES on pessimism were more salient among adolescents compared to those who were younger in age. Consistent with the findings of Taylor and Seeman¹⁹, cognitive development, including the development of irrational beliefs, was only related to pessimism but not optimism. Cognitive development follows the cumulative advantage effect, where early developmental advantages accumulate with age^37,41. Parents from higher SES families, compared to those from lower SES families, are likely to invest more resources (such as having more books for children at home⁶³) to promote cognitive development in their children. As a result, children and adolescents from different SES backgrounds show disparities in cognitive development, and these disparities are likely to become more prominent with age—and potentially strengthen the link between SES and pessimism with age. It is noteworthy the results of simple effect analysis revealed that the turning age point at which the relationship between SES and pessimism shifted from non-significant to significant occurred at around 11 years of age, which aligns with the turning point of the relationship between SES and cognitive development at around the same age³⁷. This indirectly supports our interpretation of the relationship between SES and pessimism changing with age from a cognitive development perspective.

When interpreting the results of this study, it is important to acknowledge several limitations. Firstly, the study employed a cross-sectional assessment of SES and optimism and pessimism, which precludes any conclusions about causality. Secondly, according to Bradley and Corwyn³⁰, SES can influence individual development from before birth all the way into adulthood. Our study included children in a narrow grade and age band: Grades 4–11 (ages 9–19), covering the late childhood to adolescence period. These results may not hold for younger children. Thirdly, all variables in this study were derived from self-report measures. Although common method bias was addressed, a multi-informant approach would yield more comprehensive and robust results. Future research could employ a longitudinal design covering a wider age range (e.g., 0–18 years) and incorporate multiple methods (e.g., behaviour observation for younger children) and reports from multiple informants (e.g., parents, peers, teachers or childcare centre workers), providing a more detailed and accurate depiction of when and how SES influences the development of children’s optimism and pessimism over time.

In addition, while this study drew on the Reserved Capacity Model²⁸ to elucidate how SES may relate to children’s or adolescents’ optimism and pessimism, and point to a cumulative advantage effect⁴¹ in cognitive development to attempt to explain the strengthening relationship between SES and pessimism as age increases, it did not directly test the underlying mechanisms behind relationships including the linkages of SES–optimism, SES–pessimism, and how SES–pessimism changes with age. This limitation restricts our ability to provide specific recommendations for the design and improvement of educational and social policies. Future research should further explore the pathways underlying these relationships, offering a basis for the development of targeted intervention measures.

Building upon the assessment of measurement invariance of optimism and pessimism across different SES categories, this study examined the relationships between family SES and optimism as well as pessimism among children and adolescents. Furthermore, it delved into whether SES showed differential associations with optimism and pessimism across age. The results indicated that disparities in family SES were associated with differences in optimism and pessimism among children and adolescents. Understanding these differences provides insights for further efforts aimed at reducing or eliminating disparities in the allocation of psychological resources stemming from SES inequality, and promoting social mobility. Moreover, the moderating effects of age contribute to identifying the life period during which the development of optimism and pessimism is sensitive to access to socioeconomic resources. This deepens our comprehension of the complex links between SES and both optimism and pessimism throughout late childhood and adolescence, thus improving the potential for more targeted interventions aimed at the stimulation of positive changes in both optimism and pessimism.

Competing interests

The authors declare that they have no competing interests.

Funding

This work was supported by the Natural Science Foundation of Hubei Province (No. 2022CFB943); Institute of Psychology, University of Chinese Academy of Sciences (No. GJ202001); Educational Science Planning Project of Hubei Province (No.2020GB056); Wuhan Sports University (No. 2021Z01), and the China Scholarship Council (No.202008420167).

Author Contribution

Conceptualization: R.Z., W.H., X.X.C., X.X., X.B.H.; Formal analysis: R.Z., W.H.; Methodology: R.Z., W.H., X.C.X., G.X.W.; Writing-original draft preparation: R.Z.; Writing-draft and editing: B.M.S., X.C.X, X.B.H.; Supervision: B.M.S., G.X.W., X.X. All authors read and approved the final manuscript.

Acknowledgement

We were grateful for the support from the 14th Five-Year-Plan Advantageous and Characteristic Disciplines (Groups) of Colleges and Universities in Hubei Province for Exercise and Brain Science from Hubei Provincial Department of Education. Our gratitude is also extended to each individual who participated in this study.

Data Availability

The datasets generated by the survey research during the current study are available in the Dataverse repository, https://figshare.com/s/7720d64d02196418db47

Cohen, S., Janicki-Deverts, D., Chen, E. & Matthews, K.A. Childhood socioeconomic status and adult health. Ann NY Acad Sci. 1186(1), 37–55 (2010). https://doi.org/10.1111/j.1749-6632.2009.05334.x
Ek, E., Remes, J. & Sovio, U. Social and developmental predictors of optimism from infancy to early adulthood. Soc Indic Res. 69, 219–242 (2004). https://doi.org/10.1023/B:SOCI.0000033591.80716.07
Heinonen, K., et al. Socioeconomic status in childhood and adulthood: Associations with dispositional optimism and pessimism over a 21-year follow-up. J Pers. 74(4), 1111–1126 (2006). https://doi.org/10.1111/j.1467-6494.2006.00404.x
Poulton, R. et al. Association between children’s experience of socioeconomic disadvantage and adult health: A life-course study. Lancet. 360(9346), 1640–1645 (2002). https://doi.org/10.1016/S0140-6736(02)11602-3
Sutin, A.R., Stephan, Y. & Terracciano, A. Parental educational attainment and offspring subjective well-being and self-beliefs in older adulthood. Pers Individ. Dif. 128, 139–145 (2018). https://doi.org/10.1016/j.paid.2018.01.023
Xu, Y. & Yilmazer, T. Childhood socioeconomic status, adulthood obesity and health: The role of parental permanent and transitory income. Soc Sci Med. 283, 114178 (2021). https://doi.org/10.1016/j.socscimed.2021.114178
Carver, C.S. & Scheier, M.F. Dispositional optimism. Trends Cogn Sci. 18(6), 293–299 (2014). https://doi.org/10.1016/j.tics.2014.02.003
Scheier, M.F. & Carver, C.S. Dispositional optimism and physical health: A long look back, a quick look forward. Am Psychol. 73(9), 1082–1094 (2018). https://doi.org/10.1037/amp0000384
Boehm, J.K., Chen, Y., Williams, D.R., Ryff, C. & Kubzansky, L.D. Unequally distributed psychological assets: Are there social disparities in optimism, life satisfaction, and positive affect? PLoS One. 10(2), e0118066 (2015). https://doi.org/10.1371/journal.pone.0118066
Schutte, J.W., Valerio, J.K. & Carrillo, V. Optimism and socioeconomic status: A cross-cultural study. Soc Behav Pers. 24(1), 9–18 (1996). https://doi.org/10.2224/sbp.1996.24.1.9
Bates, T.C. The glass is half full and half empty: A population-representative twin study testing if optimism and pessimism are distinct systems. J Posit Psychol. 10(6), 533–542 (2015). https://doi.org/10.1080/17439760.2015.1015155
Scheier, M.F. et al. Optimism versus pessimism as predictors of physical health: A comprehensive reanalysis of dispositional optimism research. Am Psychol. 76(3), 529–548 (2021). https://doi.org/10.1037/amp0000666
Whitfield, J.B., Zhu, G., Landers, J.G. & Martin, N.G. Pessimism is associated with greater all-cause and cardiovascular mortality, but optimism is not protective. Sci Rep. 10(1), 12609 (2020). https://doi.org/10.1038/s41598-020-69388-y
Zou, R. et al. Cross-sectional age differences in dispositional optimism in Chinese children and adolescents. Pers Individ Dif. 102, 133–138 (2016). https://doi.org/10.1016/j.paid.2016.06.063
Elliot, A.J. The hierarchical model of approach-avoidance motivation. Motiv Emot. 30, 111–116 (2006). https://doi.org/10.1007/s11031-006-9028-7
Zou, R., Hong, X., Wei, G., Xu, X. & Yuan, J. Differential effects of optimism and pessimism on adolescents’ subjective well-being: mediating roles of reappraisal and acceptance. Int J Environ Res Public Health. 19(12), 7067 (2022). https://doi.org/10.3390/ijerph19127067
Serlachius, A. et al. Is dispositional optimism or dispositional pessimism predictive of ideal cardiovascular health? The Young Finns Study. Psychol Health. 30(10), 1221–1239 (2015). https://doi.org/10.1080/08870446.2015.1041394
Robb, K.A., Simon, A.E. & Wardle, J. Socioeconomic disparities in optimism and pessimism. Int J Behav Med. 16, 331–338 (2009). https://doi.org/10.1007/s12529-008-9018-0
Taylor, S.E. & Seeman, T.E. Psychosocial resources and the SES-health relationship. Ann NY Acad Sci. 896(1), 210–225 (1999). https://doi.org/10.1111/j.1749-6632.1999.tb08117.x
Soto, C.J. & Tackett, J.L. Personality traits in childhood and adolescence: Structure, development, and outcomes. Curr Dir Psychol Sci. 24(5), 358–362 (2015). https://doi.org/10.1177/0963721415589345
Ben-Zur, H. Happy adolescents: The link between subjective well-being, internal resources, and parental factors. J Youth Adolesc. 32, 67–79 (2003). https://doi.org/10.1023/A:1021864432505
Oberle, E., Schonert-Reichl, K.A. & Zumbo, B.D. Life satisfaction in early adolescence: Personal, neighborhood, school, family, and peer influences. J Youth Adolesc. 40, 889–901 (2011). https://doi.org/10.1007/s10964-010-9599-1
Tetzner, J. & Becker, M. Think positive? Examining the impact of optimism on academic achievement in early adolescents. J Pers. 86(2), 283–295 (2018). https://doi.org/10.1111/jopy.12312
Daukantaite, D. & Bergman, L.R. Childhood roots of women’s subjective well-being: The role of optimism. Eur Psychol. 10(4), 287–297 (2005). https://doi.org/10.1027/1016-9040.10.4.287
Renaud, J., Barker, E.T., Hendricks, C., Putnick, D.L. & Bornstein, M.H. The developmental origins and future implications of dispositional optimism in the transition to adulthood. Int J Behav Dev. 43(3), 221–230 (2019). https://doi.org/10.1177/01650254188206
Burger, K. & Mortimer, J.T. Socioeconomic origin, future expectations, and educational achievement: A longitudinal three-generation study of the persistence of family advantage. Dev Psychol. 57(9), 1540–1558 (2021). https://doi.org/10.1037/dev0001238
Segerstrom, S.C. Optimism and resources: Effects on each other and on health over 10 years. J Res Pers. 41(4), 772–786 (2007). https://doi.org/10.1016/j.jrp.2006.09.004
Gallo, L.C. & Matthews, K.A. Understanding the association between socioeconomic status and physical health: Do negative emotions play a role? Psychol Bull. 129(1), 10–51 (2003). https://doi.org/10.1037/0033-2909.129.1.10
Crienaard, L., Nicolaou, M., Sawyer, A., Quax, R. & Stronks, K. Understanding the impact of exposure to adverse socioeconomic conditions on chronic stress from a complexity science perspective. BMC Med. 19(1), 1–20 (2021). https://doi.org/10.1186/s12916-021-02106-1
Bradley, R.H. & Corwyn, R.F. Socioeconomic status and child development. Annu Rev Psychol. 53(1), 371–399 (2002). https://doi.org/10.1146/annurev.psych.53.100901.135233
Gülseven, Z. et al. The mediational roles of harsh and responsive parenting in the longitudinal relations between socioeconomic status and Turkish children’s emotional development. Int J Behav Dev. 42(6), 563–573 (2018). https://doi.org/10.1177/01650254187832
Ge, T. Effect of socioeconomic status on children’s psychological well-being in China: The mediating role of family social capital. J Health Psychol. 25(8), 1118–1127 (2020). https://doi.org/10.1177/13591053177504
Chen, E., Lee, W.K., Cavey, L. & Ho, A. Role models and the psychological characteristics that buffer low-socioeconomic-status youth from cardiovascular risk. Child Dev. 84(4), 1241–1252 (2013). https://doi.org/10.1111/cdev.12037
Chen, E. & Miller, G.E. “Shift-and-persist” strategies: Why being low in socioeconomic status isn’t always bad for health. Perspect Psychol Sci. 7, 135–158 (2012). https://doi.org/10.1177/17456916124366
Bradley, R.H. & Corwyn, R.F. Age and ethnic variations in family process mediators of SES. In: Bornstein, M.H., Bradley, R.H., editors. Socioeconomic status, parenting, and child development (ed. Bornstein, M.H. & Bradley, R.H.) 161–188 (Lawrence Erlbaum Associates Publishers, 2003).
Ellis, A., David, D. & Lynn, S.J. Rational and irrational beliefs: A historical and conceptual perspective. In: David, D., Lynn, S.J., Ellis, A., editors. Rational and irrational beliefs: Research, theory, and clinical practice (ed. David, D., Lynn, S.J. & Ellis, A.) 3–22 (Oxford University Press, 2010).
Caro, D.H., McDonald, J.T. & Willms, J.D. Socio-economic status and academic achievement trajectories from childhood to adolescence. Can J Educ. 32(3), 558–590 (2009). https://www.jstor.org/stable/canajeducrevucan.32.3.558
Li, M. et al. Development of cognitive control during adolescence: The integrative effects of family socioeconomic status and parenting behaviors. Dev Cogn Neurosci. 57, 101139 (2022). https://doi.org/10.1016/j.dcn.2022.101139
López Boo, F. Socio-economic status and early childhood cognitive skills: A mediation analysis using the Young Lives panel. Int J Behav Dev. 40(6), 500–508 (2016). https://doi.org/10.1177/0165025416644689
Vassou, C. et al. Irrational beliefs trigger depression and anxiety symptoms, and are associated with increased inflammation and oxidative stress markers in the 10-year diabetes mellitus risk: The ATTICA epidemiological study. J Diabetes Metab Disord. 20(1), 727–739 (2021). https://doi.org/10.1007/s40200-021-00805-3
DiPrete, T.A. & Eirich, G.M. Cumulative advantage as a mechanism for inequality: A review of theoretical and empirical developments. Annu Rev Sociol. 32(1), 271–297 (2006). https://doi.org/10.1146/annurev.soc.32.061604.123127
Flake, J.K., Pek, J. & Hehman, E. Construct validation in social and personality research: Current practice and recommendations. Soc Psychol Pers Sci. 8(4), 370–378 (2017). https://doi.org/10.1177/1948550617693
Hughes, B.T. et al. The Big Five across socioeconomic status: Measurement invariance, relationships, and age trends. Collabra Psychol. 7(1), 24431 (2021). https://doi.org/10.1525/collabra.24431
Maassen ED, van Assen MALM, Nuijten MB, De Roover K, Wicherts JM. The dire disregard of measurement invariance testing in psychological science. Psychol Methods. (2023). https://dx.doi.org/10.1037/met0000624
Wuhan Municipal Bureau of Statistics. Wuhan statistical yearbook – 2019 (ed. China Statistics Press, 2019). (In Chinese)
Tongren Municipal Bureau of Statistics. Tongren Statistical Yearbook – 2019 (ed. China Statistics Press, 2019). (In Chinese)
Soto, C.J., John, O.P., Gosling, S.D. & Potter J. The developmental psychometrics of Big Five self-reports: Acquiescence, factor structure, coherence, and differentiation from ages 10 to 20. J Pers Soc Psychol. 94(4), 718–737 (2008). https://doi.org/10.1037/0022-3514.94.4.718
Currie, C. et al. Researching health inequalities in adolescents: The development of the Health Behaviour in School-Aged Children (HBSC) Family Affluence Scale. Soc Sci Med. 66(6), 1429–1436 (2008). https://doi.org/10.1016/j.socscimed.2007.11.024
Boyce, W., Torsheim, T., Currie, C. & Zambon, A. The Family Affluence Scale as a measure of national wealth: Validation of an adolescent self-report measure. Soc. Indic. Res. 78, 473–487 (2006). https://doi.org/10.1007/s11205-005-1607-6
Liu, Y. et al. Reliability and validity of Family Affluence Scale (FAS II) among adolescents in Beijing, China. Child Indic. Res. 5, 235–251 (2012). https://doi.org/10.1007/s12187-011-9131-5
Molcho, M., Gabhainn, S. N. & Kelleher, C. C. Assessing the use of the Family Affluence Scale (FAS) among Irish schoolchildren. Ir. Med. J. 100, 37–39 (2007).
Liu, Z. J. & Chen, H. C. Chinese revision of life orientation test in junior high school students. Chin. J. Clin. Psychol. 15, 135–137 (2007). (In Chinese)
Chen, W. et al. Socioeconomic status and life satisfaction in Chinese adolescents: Analysis of self-esteem as a mediator and optimism as a moderator. Pers. Individ. Dif. 95, 105–109 (2016). https://doi.org/10.1016/j.paid.2016.01.036
Muthén, L. K. & Muthén, B. O. Mplus User’s Guide. 8th ed. (Muthén & Muthén, 1998–2017).
Cheung, G. W. & Rensvold, R. B. Evaluating goodness-of-fit indexes for testing measurement invariance. Struct. Equ. Model. 9, 233–255 (2002). https://doi.org/10.1207/S15328007SEM0902_5
Chen, F. F. Sensitivity of goodness of fit indexes to lack of measurement invariance. Struct. Equ. Model. 14, 464–504 (2007). https://doi.org/10.1080/10705510701301834
Hayes, A. F. Partial, conditional, and moderated moderated mediation: Quantification, inference, and interpretation. Commun. Monogr. 85, 4–40 (2018). https://doi.org/10.1080/03637751.2017.1352100
West, S. G., Finch, J. F. & Curran, P. J. Structural equation models with nonnormal variables: Problems and remedies. In Hoyle, R. H., Ed. Structural equation modeling: Concepts, issues, and applications (ed. Hoyle, R. H.) 56–75 (Sage Publications, 1995).
Bernerth, J. B. & Aguinis, H. A critical review and best-practice recommendations for control variable usage. Pers. Psychol. 69, 229–283 (2016). https://doi.org/10.1111/peps.12103
Lin, H. Probing two-way moderation effects: A review of software to easily plot Johnson-Neyman figures. Struct. Equ. Model. 27, 494–502 (2020). https://doi.org/10.1080/10705511.2020.1732826
Roberts, B. W. & Nickel, L. B. Personality development across the life course: A neo-socioanalytic perspective. In Handbook of personality: Theory and research, 4th ed. (ed. John, O. P. & Robins, R. W.) 259–283 (Guilford Press, 2021).
Roberts, B. W. & Yoon, H. J. Personality psychology. Annu. Rev. Psychol. 73, 489–516 (2022). https://doi.org/10.1146/annurev-psych-020821-114927
Wolf, S. & McCoy, D. C. Household socioeconomic status and parental investments: Direct and indirect relations with school readiness in Ghana. Child Dev. 90, 260–278 (2019). https://doi.org/10.1111/cdev.12899

No competing interests reported.

Supplementarytable.docx

Download PDF

Reviewers agreed at journal
17 Sep, 2024
Reviews received at journal
07 Sep, 2024
Reviewers agreed at journal
22 Aug, 2024
Reviewers invited by journal
19 Aug, 2024
Editor assigned by journal
14 Aug, 2024
Editor invited by journal
14 Aug, 2024
Submission checks completed at journal
13 Aug, 2024
First submitted to journal
02 Aug, 2024

You are reading this latest preprint version

Socioeconomic disparities in optimism and pessimism among children and adolescents: Does age matter?

Status:

Version 1

Abstract

Figures

Introduction

Methods

Participants and procedure

Measures

Family socioeconomic status

Optimism and pessimism scales

Statistical analysis

Measurement invariance test

Moderation model test

Results

Preliminary analyses

Measurement invariance test

Moderation model test

Discussion

Conclusion

Declarations

Competing interests

Funding

Author Contribution

Acknowledgement

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1