Descriptive statistics were reported as means and standard deviations of the groups. Multivariate analysis was carried out to analyse the effect of social isolation and loneliness on the overall psychological well-being as well as the individual constructs of psychological well-being. Further, mediation analysis was carried out to examine the role of important socio-demographic characteristics like gender, education, and marital status in the impact of social isolation and loneliness on the psychological well-being of the older adults.
The participants have been categorised based on various dimensions like gender, education status, marital status, social isolation, and loneliness (Table 1). Males comprise 47.19 percent of the total sample, whereas 52.81 percent of the participants are females. Further, almost 46.25% (N = 148) participants are educated, whereas 53.75 percent participants are uneducated. Based on marital status participants have been categorised as ‘currently married’ (31.88 percent), ‘widowed’ (52.19 percent), ‘divorced/separated/deserted’ (5.9%), Moreover, based on the scores of social isolation, and loneliness, 84.38%(N = 270) participants are socially isolated, and 86.88 percent (N = 278) are lonely.
Variables
|
N (%)
|
Mean (SD)
|
|
Table 1
Descriptive Statistics of Categorical Variables
Gender
|
|
1.52 (.499)
|
|
Males
|
151 (47.19)
|
|
|
Females
|
169 (52.81)
|
|
|
Education
|
|
1.46 (.499)
|
|
Educated
|
148(46.25)
|
|
|
Uneducated
|
172(53.75)
|
|
|
Marital Status
|
|
2.30 (1.74)
|
|
Currently Married
|
102(31.88)
|
|
|
Widowed
|
167(52.19)
|
|
|
Divorced/ Separated/ Deserted
|
19 (5.94)
|
|
|
Never Married
|
32(10.00)
|
|
|
Social Isolation
|
|
1.15(.363)
|
|
High Social Isolation
|
270(84.38)
|
|
|
Low Social Isolation
|
50(15.63)
|
|
|
Loneliness
|
|
1.13(.338)
|
|
High Loneliness
|
278(86.88)
|
|
|
Low Loneliness
|
42(13.13)
|
|
|
Source: Authors’ Calculation |
Table 2 presents the mean and standard deviations of the various groups of participants on the dimension of psychological well-being as well as the individual components of psychological well-being. Older adults who had high social isolation and high levels of loneliness had the least scores on overall psychological well-being (M = 109.21, SD = 24.97), as well as for all the components of psychological well-being. Contrarily, participants who reported low on social isolation and loneliness had the best overall psychological well-being (M = 144.57, SD = 28.14). They also exhibited better results in individual components of psychological well-being as compared to their socially isolated and lonely counterparts.
Table 2
Descriptive Statistics of Group of Participants on Dimension of Components of Psychological Well-Being
Group
|
Mean(SD)
|
Social Isolation
|
Loneliness
|
N
|
Psychological Well-Being
|
Autonomy
|
Environmental Mastery
|
Personal Growth
|
Positive Relations
|
Purpose in Life
|
Self-Acceptance
|
High Social Isolation
|
High Loneliness
|
242
|
109.21(24.97)
|
16.69(6.24)
|
17.93(3.82)
|
19.27(4.86)
|
16.87(6.87)
|
23.07(3.32)
|
15.38(5.46)
|
Low Loneliness
|
28
|
129.18(22.30)
|
19.18(5.49)
|
20.43(3.69)
|
22.18(4.33)
|
25.00(6.60)
|
24.50(3.49)
|
17.89(4.88)
|
Low Social Isolation
|
High Loneliness
|
36
|
128.67(37.16)
|
18.94(6.49)
|
20.08(4.44)
|
22.08(6.90)
|
24.44(10.96)
|
24.86(4.07)
|
18.25(7.32)
|
Low Loneliness
|
14
|
144.57(28.14)
|
23.93(7.39)
|
20.79(4.93)
|
23.93(5.59)
|
27.29(7.21)
|
26.64(4.36)
|
22.00(8.09)
|
Source: Authors’ Calculation |
Furthermore, the independent variables, namely social isolation and loneliness exhibited a significant negative correlation (r=-.505**), (Fig. 1) suggesting that with lower scores of social isolation, which is indicative of heightened levels of social isolation, loneliness tends to increase. Hence to further understand the co-existence of social isolation and loneliness, linear regression analysis was performed (Table 3). A significant regression equation was obtained (F= (1,318) = 109.231, p < .01), with an R2 of .256. Therefore, social isolation was a significant predictor of loneliness and explained 25.6% of the variance.
Table 3
Association Between Social Isolation and Loneliness
Model
|
R2
|
B
|
β
|
t-value
|
F
|
Loneliness
|
.256
|
− .995
|
− .506**
|
-10.451
|
109.23
|
Note: ** represent 1% level of significance. |
Source: Authors’ Calculation |
Multivariate analysis (Table 4) was carried out to examine the impact of social isolation and loneliness on psychological well-being of the participants. It was conducted to assess if there are significant differences in psychological well-being among the participants experiencing varying levels of social isolation and loneliness. Wilk’s Lambda test statistic was used for the analysis. A statistically significant MANOVA effect was obtained for social isolation (F = 3.836, p < .01), with the multivariate effect size being approximately .07, suggesting that almost seven percent of the variance was explained by social isolation while determining the attributing factors of psychological well-being among older adults. Furthermore, loneliness (F = 3.782, p < .01) also had a significant effect on psychological well-being and played a vital role in determining psychological well-being among the participants, with partial η2 being .068, suggesting almost seven percent of the variance was explained by loneliness. As the interaction effect of the independent variables – that is, social isolation and loneliness – were not statistically significant; the main effects of social isolation, and loneliness were considered for the analysis.
Table 4
Summary of Multivariate Analysis of Variance
Effect
|
Value
|
F
|
Hypothesis df
|
Error df
|
Partial η2
|
Intercept
|
0.042
|
1191.569**
|
6
|
311
|
0.958
|
Social Isolation
Loneliness
Social Isolation*Loneliness
|
0.931
0.932
0.962
|
3.836**
3.782**
2.050
|
6
6
6
|
311
311
311
|
0.069
0.068
0.038
|
Note: ** represent 1% level of significance. |
Source: Authors’ Calculation |
Furthermore, Test of Between-Subject Effects (Table 5) were analysed to determine how social isolation and loneliness affect the overall psychological well-being, as well as the individual components of psychological well-being of the participants. Results showed that social isolation (F = 12.412, p < .01, partial η2 = .038) had a significant impact on the overall psychological well-being, explaining almost four percent variance. Further, social isolation had a significant impact on various components of psychological well-being, namely autonomy, personal growth, positive relations, purpose in life, self-acceptance. However, it did not impact environmental mastery significantly, suggesting that social isolation does not act as deciding factor in determining older adults’ ability to manage and modify the environmental factors and activities in accordance with their needs and benefits. Moreover, loneliness had a significant impact on the overall psychological well-being (F = 13.152, p < .01, partial η2 = .040) as well as the deconstructed factors of psychological well-being. It was observed that older adults who are not socially isolated and experience lower levels of loneliness exhibit best psychological well-being (M = 144.57, SD = 28.14), encompassing the individual components.
Table 5
Summary of Effect of Independent Variables on Dependent Variables
Source
|
Dependent Variable
|
df
|
F
|
Partial η2
|
Social Isolation
|
Psychological Well-Being
|
1
|
12.412**
|
.038
|
Autonomy
|
1
|
8.997**
|
.028
|
Environmental Mastery
|
1
|
2.927
|
.009
|
Personal Growth
|
1
|
5.707**
|
.018
|
Positive Relations
|
1
|
12.659**
|
.039
|
Purpose in Life
|
1
|
9.190**
|
.028
|
Self-Acceptance
|
1
|
10.482**
|
.032
|
Loneliness
|
Psychological Well-Being
|
1
|
13.152**
|
.040
|
Autonomy
|
1
|
10.240**
|
.031
|
Environmental Mastery
|
1
|
4.755**
|
.015
|
Personal Growth
|
1
|
6.194**
|
.019
|
Positive Relations
|
1
|
15.673**
|
.047
|
Purpose in Life
|
1
|
6.125**
|
.019
|
Self-Acceptance
|
1
|
8.447**
|
.026
|
Social Isolation* Loneliness
|
Psychological Well-Being
|
1
|
.169
|
.001
|
Autonomy
|
1
|
1.142
|
.004
|
Environmental Mastery
|
1
|
1.500
|
.005
|
Personal Growth
|
1
|
.309
|
.001
|
Positive Relations
|
1
|
3.641
|
.011
|
Purpose in Life
|
1
|
.074
|
.000
|
Self-Acceptance
|
1
|
.327
|
.001
|
Note: ** represent 1% level of significance. |
Source: Author's Calculation |
Mediation analysis was performed to assess the mediating role of factors like gender, education, and marital status, in the relationship between social isolation, loneliness, and psychological well-being. The results have been presented in Table 6. Further, the path diagrams (Fig. 2,3, and 4) illustrate direct effects as well the total effects in the form of regression coefficients of the independent variables and mediators on the dependent variable, i.e., psychological well-being.
The results (Table 6) revealed that direct effect of both social isolation, and loneliness on psychological well-being were statistically significant, suggesting that social isolation and loneliness can be considered as determining factors of psychological well-being. However, both the factors namely social isolation (β1 = .012, z1 = .62, p1 > .05) and loneliness (β2 = − .003, z2 = − .48, p2 > .05.) had a non-significant indirect impact while assessing the role of gender as a mediating factor. This implies that gender does not impact the effect that social isolation and loneliness have on the psychological well-being of the older adults. Furthermore, education had a significant impact on the psychological well-being of the older adults (e = 3.44, z = 5.19, p < .01). This is indicative of the fact that education as an independent factor has a role to play in determining the psychological well-being of the older adults. However, it does not mediate the impact of social isolation and loneliness on the psychological well-being. This is reflected in the non-significant indirect effect of social isolation (β1 = .013, z1 = − .20, p1 > .05) and loneliness (β2 = − .010, z2 = − .30, p2 > .05.) on the psychological well-being with education as a mediating factor. It was observed that marital status as an independent factor did not have significant impact on the psychological well-being of the older adults. However, both social isolation (β1=-.16, z1= -0.62, p1 < .05) and loneliness (β2 = .051, z2 = .51, p2 < .05) exhibit statistically significant indirect effect on psychological well-being with marital status as the mediating factor. This suggests that marital status partially mediates the association of social isolation and loneliness with the psychological well-being of the older adults. Here, β1, z1, and p1 denote the regression coefficient, z score, and probability values for social isolation, and β2, z2, p2 are indicative of the regression coefficient, z score, and probability values for loneliness. Further, the total effect of the afore-mentioned independent variables on psychological well-being was reported to be significant, suggesting the importance of assessing social isolation and loneliness as determinants of psychological well-being.