Sociodemographic characteristics of the study population and descriptive analysis of the variables
A total of 468 participants were recruited for this study. Their ages ranged from 22 to 49 years (mean of 28.14 ± 3.84 years). Table 1 shows the other descriptive statistics of the sample and summarizes the results of the analysis comparing variables on factors associated with mental health problems. The findings showed that gender is linked to anxiety (p < 0.05). Marital status, education, and age were all significantly associated with depression, anxiety, stress, and suicidal ideation (p < 0.05). Therefore, demographic variables were selected for adjustment when testing the moderation model.
Table 1
Sociodemographic characteristics of the study population
Variable | N (%) | Depression (mean ± SD) | p | Anxiety (mean ± SD) | p | Stress (mean ± SD) | p | Suicidal (mean ± SD) | p |
Gender | | | .211 | | .048 | | .507 | | .989 |
Male | 236 (50.4%) | 11.19 ± 5.01 | | 7.46 ± 4.37 | | 11.41 ± 3.78 | | .15 ± .36 | |
Female | 194 (41.5%) | 10.60 ± 4.42 | | 7.76 ± 3.97 | | 11.06 ± 3.56 | | .15 ± .35 | |
Others | 38 (8.1%) | 9.89 ± 5.87 | | 5.89 ± 4.79 | | 10.84 ± 4.46 | | .16 ± .37 | |
Marital status | | | .000 | | .000 | | .000 | | .038 |
Single | 278 (59.4%) | 9.57 ± 4.70 | | 6.63 ± 4.07 | | 10.48 ± 3.71 | | .12 ± .32 | |
Married | 165 (35.3%) | 12.32 ± 4.55 | | 8.39 ± 4.11 | | 12.08 ± 3.53 | | .21 ± .42 | |
Divorce | 25 (5.3%) | 15.16 ± 3.11 | | 10.56 ± 4.85 | | 13.76 ± 3.29 | | .16 ± .37 | |
Education | | | .000 | | .000 | | .000 | | .007 |
High school | 21 (4.5%) | 6.14 ± 4.36 | | 4.00 ± 3.28 | | 8.00 ± 3.47 | | .00 ± .00 | |
Undergraduate | 401 (85.7%) | 11.00 ± 4.59 | | 7.52 ± 4.12 | | 11.35 ± 3.51 | | .14 ± .35 | |
Postgraduate | 46 (9.8%) | 11.54 ± 6.16 | | 8.52 ± 5.13 | | 11.50 ± 5.08 | | .28 ± .45 | |
Age | | | | | | | | | |
Under 30 years old | 380 (81.2%) | 10.66 ± 4.68 | .000 | 7.41 ± 4.06 | .000 | 11.07 ± 3.56 | .002 | .16 ± .37 | .004 |
Over 30 years old | 88 (18.8%) | 11.59 ± 5.54 | | 7.67 ± 5.08 | | 11.85 ± 4.45 | | .10 ± .30 | |
Total | 468 (100%) | 10.84 ± 4.86 | | 7.46 ± 4.27 | | 11.22 ± 3.75 | | 0.15 ± .36 | |
Table 1: Sociodemographic characteristics of the study population
Table 2: Levels of stress, anxiety, and depression among unemployed people
Table 2
Levels of stress, anxiety, and depression among unemployed people
Level | Depression N (%) | Anxiety N (%) | Stress N (%) |
Normal | 60 (12.8) | 108 (23.1%) | 75 (16.0%) |
Mild | 20 (4.3%) | 25 (5.3%) | 71 (15.2%) |
Moderate | 133 (28.4%) | 97 (20.7%) | 140 (29.9%) |
Severe | 130 (27.8%) | 112 (23.9%) | 174 (37.2%) |
Extremely Severe | 125 (26.7%) | 126 (26.9%) | 8 (1.7%) |
Total | 468 (100%) | 468 (100%) | 468 (100%) |
Table 2 is statistically based on the average scores of the 468 samples and compares levels of depression, anxiety, and stress according to DASS-21 cut-off points [32]. Generally, the percentage of laid-off employees experiencing severe or extremely severe symptoms of depression, anxiety, and stress is remarkably high. Specifically, 54.5%, 50.8%, and 38.9% of unemployed people have “severe” to “extremely severe” symptoms of depression, anxiety, and stress, respectively.
Descriptive statistics and Pearson correlation
Table 3 presents the correlations between psychological capital, self-esteem, and mental health problems (depression, anxiety, stress) as well as their means and standard deviations.
Table 3
Correlations, means, and standard deviations among variables
Measure | M | SD | PSY | SELF | DE | AN | ST | SI |
PSY | 3.48 | 0.76 | - | | | | | |
SELF | 24.70 | 5.40 | .809** | - | | | | |
DE | 10.84 | 4.86 | ˗.668** | ˗.700** | - | | | |
AN | 7.46 | 4.27 | ˗.551** | ˗.540** | .819** | - | | |
ST | 11.22 | 3.75 | ˗.574** | ˗.562** | .824** | .800** | - | |
SI | 0.15 | 0.36 | ˗.371** | ˗.339** | .317** | .367** | .273** | - |
Note: PSY = psychological capital, SELF = self-esteem, DE = depression, AN = anxiety, ST = stress, SI = suicidal ideation |
** p < 0.01 |
Table 3: Correlations, means, and standard deviations among variables
Table 3 shows a significant positive correlation between psychological capital and self-esteem (r = .809, p < .01). The results also indicate that psychological capital had significant negative correlations with depression (r = ˗.668, p < .01), anxiety (r = ˗.551, p < .01), and stress (r = ˗.574, p < .01).
Moderating effect analyses
Table 4: Moderation analyses of self-esteem between mental health problems and psychological capital
Table 4
Moderation analyses of self-esteem between mental health problems and psychological capital
Stress |
| Variables | R2 | R2 Change | p | SE | β | t | p |
Step 1 | PSY | .383 | .305 | .000 | .682 | ˗.367 | ˗5.717 | .000 |
| SELF | .314 | ˗.238 | ˗3.719 | .000 |
Step 2 | PSY | .402 | .018 | .000 | .313 | ˗1.611 | ˗5.149 | .000 |
| SELF | .044 | ˗.147 | ˗3.326 | .001 |
| PSY × SELF | .027 | ˗.102 | ˗3.746 | .000 |
Anxiety |
| Variables | R2 | R2 Change | p | SE | β | t | p |
Step 1 | PSY | .360 | .270 | .000 | .364 | ˗.329 | ˗5.030 | .000 |
| SELF | .051 | ˗.240 | ˗3.690 | .000 |
Step 2 | PSY | .360 | .000 | .736 | .368 | ˗1.848 | ˗5.017 | .000 |
| SELF | .052 | ˗.192 | ˗3.701 | .000 |
| PSY × SELF | .032 | .011 | .337 | .736 |
Depression |
| Variables | R2 | R2 Change | p | SE | β | t | p |
Step 1 | PSY | .567 | .430 | .000 | .341 | ˗.311 | ˗5.766 | .000 |
| SELF | .048 | ˗.408 | ˗7.614 | .000 |
Step 2 | PSY | .577 | .010 | .001 | .341 | ˗1.787 | ˗5.241 | .000 |
| SELF | .048 | ˗.349 | ˗7.271 | .000 |
| PSY × SELF | .030 | ˗.099 | ˗3.353 | .001 |
Suicidal Ideation |
| Variables | R2 | R2 Change | p | SE | β | t | p |
Step 1 | PSY | .154 | .120 | .000 | .036 | ˗.248 | ˗3.296 | .001 |
| SELF | .005 | ˗.131 | ˗1.751 | .081 |
Step 2 | PSY | .185 | .031 | .000 | .035 | ˗.140 | ˗3.964 | .000 |
| SELF | .005 | ˗.011 | ˗2.244 | .025 |
| PSY × SELF | .003 | .013 | 4.197 | .000 |
Model 1: Stress as an independent variable
The results showed that psychological capital (β = ˗.367, p < 0.001) and self-esteem negatively predicted stress (β = ˗.238, p < 0.001). In addition, the moderation model was implemented through the SPSS PROCESS macro with Model 1, and the moderating effect was determined based on 5,000 bootstrap samples in generating 95% bias-corrected bootstrap confidence intervals. The interaction (psychological capital × self-esteem) also negatively predicted stress (β = ˗.102, p < 0.05). Overall, these results suggest that self-esteem can moderate the relation between psychological capital and stress among unemployed people.
To examine how self-esteem moderates the relation between psychological capital and stress, simple slopes were plotted for values of high self-esteem (1 SD above the mean) and low self-esteem (1 SD below the mean), as shown in Fig. 1. The results indicated that, for the group with low self-esteem, a significant negative relation was observed between psychological capital and stress (Β = ˗1.063, SE = 0.365, t = ˗2.908, p < 0.05, 95% CIs: ˗1.781 ~ ˗.345). This relation was also significant for the group with high self-esteem (Β = ˗2.160, SE = 0.324, t = ˗6.659, p < 0.001, 95% CIs: ˗2.798 ~ ˗1.523).
Model 2: Anxiety as an independent variable
The results showed that psychological capital negatively predicted anxiety (β = ˗.311, p < 0.001). Self-esteem also negatively predicted anxiety (β = ˗.240, p < 0.001). The interaction (psychological capital × self-esteem) did not negatively predict anxiety (p > 0.05). Overall, these results indicate that self-esteem cannot moderate the relation between psychological capital and anxiety among unemployed people.
Model 3: Depression as an independent variable
The results showed that psychological capital negatively predicted depression (β = ˗.292, p < 0.001). Self-esteem negatively predicted depression as well (β = ˗.408, p < 0.001). The interaction (psychological capital × self-esteem) also negatively predicted depression (β = ˗.099, p < 0.05). Overall, these results suggest that self-esteem can moderate the relation between psychological capital and depression among unemployed people.
To examine how self-esteem moderates the relation between psychological capital and depression, simple slopes were plotted for values of high self-esteem (1 SD above the mean) and low self-esteem (1 SD below the mean), as shown in Fig. 2. The results indicated that, for the low-self-esteem group, a significant negative relation was found between psychological capital and depression (Β = ˗1.252, SE = 0.398, t = ˗3.145, p < 0.05, 95% CIs: ˗2.034 ~ ˗.470). Also, this relation was significant for the high-self-esteem group (Β = ˗2.322, SE = 0.353, t = ˗6.570, p < 0.001, 95% CIs: ˗3.016 ~ ˗1.627).
Model 4: Suicidal ideation as an independent variable
The results showed that psychological capital negatively predicted suicidal ideation (β = ˗.248, p < 0.001). Self-esteem did not predict suicidal ideation (p > 0.05). However, the interaction (psychological capital × self-esteem) positively predicted suicidal ideation (β = .013, p < 0.05). Overall, these results suggest that low self-esteem can moderate the relation between psychological capital and depression among unemployed people.
To examine how self-esteem moderates the relation between psychological capital and suicidal ideation, simple slopes were plotted for values of high self-esteem (1 SD above the mean) and low self-esteem (1 SD below the mean), as shown in Fig. 3. The results indicated that, for the group with low self-esteem, a significant negative relation was observed between psychological capital and suicidal ideation (Β = ˗.209, SE = 0.041, t = ˗5.076, p < 0.001, 95% CIs: ˗2.208 ~ ˗.618), but this relation was not significant for the group with high self-esteem (p > 0.05).
Figure 1
The plot of simple slopes for self-esteem in the relation between stress and psychological capital
Figure 2
The plot of simple slopes for self-esteem in the relation between depression and psychological capital
Figure 3
The plot of simple slopes for self-esteem in the relation between suicidal ideation and psychological capital