Correlates of loneliness and dissatisfaction following SARS pandemic lockdown: Child and parent perspective

Inclusion and exclusion criteria

Children between 10 and 18 years of age, studying in either government or private schools of Chandigarh, and their parents who gave consent for themselves and their children were included. Children and parents with pre-existing diagnosed psychological comorbid conditions were excluded from the study.

Children and their parents were enrolled through multistage stratified random sampling to increase the trustworthiness of matched rate estimate. The study population was grouped into subclusters with a goal to minimize the individual stratum variance, and the sample was chosen from all clusters/strata. One cluster of government and private schools each was chosen from north, south, east and west of Chandigarh. The study was conducted in two stages.

Stage I. The cluster in charge (school principal) was contacted through email, WhatsApp and telephonically to explain the objectives of the study. The participants were informed in detail about the research objectives and methodology. Since the children were <18 years of age, consent was obtained from the parents. Google forms were sent through email and WhatsApp to the cluster in charge for circulation among the students and their parents in their respective clusters. Parents were requested to allow their children to fill their forms independently.

To ensure maximum participation, the principle investigator’s contact number was also circulated for any further queries to the cluster incharge, teachers, parents and children. Google forms were sent to 5176 students and their parents.

Stage II: After taking electronic consent, telephonic interviews were done and an online survey was completed using Google forms. The Google form had two sections. The first part had sociodemographic details of the children and parents (name, age, gender, education); and work profile, income, type and size of family, religion, etc. of parents.

The second part had questions from parallel forms (children and parents) of all three scales used in the study. In addition to the three scales, parental distress was assessed using DASS- 21 in parents Google form. Bilingual versions of all the scales were made available to all the participants of the study. Both children and parents were assessed on the following scales.

Loneliness and social dissatisfaction questionnaire (LSDQ; Asher et al.)14

It measures loneliness of the child and has child and parent versions. It has 24 items rated on a 5-point Likert scale (definitely yes to definitely no). The total scores of the scale range from 16 to 80. Of the 24 items, 16 measure feelings of loneliness, social adequacy and subjective perception of peer relation and 8 items are filter/tricky items, used to disguise the actual purpose of assessment. Item numbers 6, 9, 12, 17, 20 and 21 are scored reverse, and higher scores are indicative of higher feeling of loneliness, social inadequacy and poor perceived peer relationship. The Cronbach alpha for the child version is 0.89 and for the parent version is 0.96. The scale has also been reported to have excellent internal consistency.¹⁵ The Hindi adaptation was done for both the child and parent versions and the Cronbach alpha (child 0.93; parent 0.98) and intraclass correlation coefficient (ICC; child 0.87 and parent 0.88) suggested good consistency.

Parent–child relationship (PCR) questionnaire (Furman and Giberson, 1995)¹⁶

This scale measures quality of PCR and has two parallel forms for the child and parents. It measures five factors: warmth, personal relationship, disciplinary warmth, power assertion and possessiveness. We used the personal relationship domain with 10 items as it has been found to be strongly associated with other domains of the scale.¹⁴ Items were rated on a 5-point Likert scale (hardly at all to extremely much) and the score range is 10–50. Higher scores indicate more intimate relationship and togetherness. The Cronbach alpha has been reported to be 0.76 for parent version and 0.91 for child version. The scale has also been reported to have good to excellent internal consistency.¹⁷ Cross-cultural adaptation was done and adapted Hindi version showed good to excellent internal consistency: Cronbach alpha (child 0.97; parent 0.94) and ICC (child 0.93; parent 0.90).

Conflict behaviour questionnaire (CBQ) short form (Robin and Foster, 1984)¹⁸

This self-report questionnaire is a short form of CBQ 44 by Prinz, Foster, Kent and O’Leary. It has 20 items with a true/false rating and measures the perceived conflict between child and parent interaction at home. It has two parallel forms for parent and child. The scale is reported to have good internal consistency for children 0.94 and for parents 0.95 (Prinz, Foster, Kent and O’Leary, 1979). High scores are indicative of a negative communication between child and parent. We used Hindi versions of both parent and child CBQ with cross-validation: ICC (child 0.80 and parent 0.88) and Cronbach alpha (child 0.89 and parent 0.92).

Depression, Anxiety and Stress Scale-21 (Lovinbond and Lovinbond, 1995)¹⁹

This scale was used to assess distress among parents. This self- report scale has three subscales: depression, anxiety and stress. Each subscale has 7 items and items have to be rated on a 4-point scale (0 not at all; 1 some degree; 2 considerable degree and 4 very much). Cut-off scores for three subscales are also given from normal to extremely severe stress, anxiety and depression. It has good psychometric properties. Cronbach alpha between 0.86 and 0.90 has been reported with good convergent validity (0.84 depression; 0.75 anxiety; 0.90 stress).

Sample size calculation

We estimated that about 23 000 students in the 10–18 years age group were studying in the four clusters of schools chosen for this study. Sample size was determined by Cochran formula for finite population. The estimated sample for our study was 647 to have a confidence level of 99% with real value within 5% of the surveyed value. To ensure better stability of variance and covariance, Google forms were sent to 8-fold of the estimated sample size.20 Hence, 5176 Google forms were distributed.

Statistical analysis

Obtained data were analysed using IBM SPSS version 17.0 Statistics for Windows (IBM Corp., Armonk, NY, USA). Skewness of data was examined to see if the values were normally distributed. Descriptive statistics were used (frequency, percentage, mean and standard deviation) for demographic variables of the children and parents. Independent sample t-test was used to assess differences between the level of loneliness, group comparison on the basis of age, gender of child and parent along with other demographic variables. Bivariate correlation (Pearson product moment) was computed and used for analyses of the association between clinical and sociodemographic variables.

In line with a previous study, scores from child appraisal of loneliness and parent perception of child loneliness on LSDQ, loneliness scores were further divided into two half categories (Children: low loner 22–44; high loner 45–66; Parents: low loner 26–45; high loner 46–64).¹⁵

Sociodemographic profile

Among children, 1873 (61.9%) were girls, slightly less than half (1403; 46.3%) were matriculate and about one-fifth (600; 19.8%) had intermediate level education. Among parents, 1735 (57.3%) were fathers, their education levels were 1139 (39%) intermediate level and 401 (13.2%) graduate and postgraduate level; 1482 (48.9%) were employed in ministerial jobs and 629 (20.8%) were house-makers; 1857 (61.3%) parents were earning ≤₹39 032 per month. About half (1641; 54.2%) belonged to lower middle and 880 (29.1%) to upper middle socioeconomic status; 1661 (54.8%) were living in joint families and 1687 (55.7%) were residents of urban areas. The majority (2422; 80%) were Hindu by religion (Table I).

The mean (SD) scores on loneliness and social dissatisfaction measure for children were 52.62 (5.47) (LSDQ-C) and for parents 48.41 (4.46) (LSDQ-P). The mean score of child-reported PCR (PCR-C) was 35.65 (7.50) and for PCR-P 35.11 (7.43) and on conflict behaviour dimension 10.28 (1.91) (CBQ-C) and 9.52 (1.59) (CBQ-P; Table II).

Level of loneliness and dissatisfaction (LSDQ-C and LSDQ-P)

Some children may experience more loneliness than others and level of loneliness may affect their social, emotional and behavioural functioning. We compared high loners and low loner groups to see its effect on PCR and conflict behaviour. In our sample, the range of child reported loneliness was 22–66 and parent reported child loneliness was 26–64. The original measure range for child and parent version is 16–80. The majority of children reported high on loneliness and dissatisfaction (2581; 85.2%) and the parents (2268; 74.9%) also reported that their children were feeling lonely staying at home during the Covid- 19 pandemic (Fig. 1).

Close

FIG 1.

Level of loneliness among children and parents. *Children (score range 22–66; low loner 22–44; high loner 45–66). *Parent (score range 26–64; low loner 26–45; high loner 46–64)

Export to PPT

Association of high and low loneliness and dissatisfaction (LSDQ) with PCR, conflict behaviour questionnaire

We found a significant difference between low and high loneliness groups and its association with child reported PCR and conflict behaviour. The only negative but significant correlation was between level of loneliness and dissatisfaction (LSDQ-C) and CBQ-C (r= –0.137, n=3028, p<0.05). Children who scored low on loneliness and dissatisfaction, their loneliness level decreased as their PCR quality increased. These findings suggest that high loners scores were related to the quality of their PCR. For children, loneliness and dissatisfaction were inversely associated with conflict behaviour. Those who experienced more loneliness and dissatisfaction did not have much interaction with their parents, and thus had fewer conflicts. No significant association was found between level of loneliness (LSDQ-P) and PCR-P and CBQ-P.

Gender and age-wise difference on high and low level of loneliness

Gender-wise significant difference was seen between levels of loneliness (LSDQ-C) and PCR-C. Boys who scored high on loneliness significantly also reported poor PCR (low loneliness 35.77 [7.59]; high loneliness 36.85 [7.00]; p<0.001) and more conflicts with their parents (low loneliness 10.10 [1.90]; high loneliness 10.31 [1.91]; p<0.005) than girls. High loner children aged 14–18 years reported more conflicts with their parents (low loneliness 10.07 [1.94]; high loneliness 10.33 [1.92]; p<0.005) than low loner children of the same age range.

Distress among parents and its association with other variables

Distress among parents was assessed using DASS-21 in terms of stress, anxiety and depression. Obtained scores on DASS- 21 were multiplied by two to make them comparable with the original scale (DASS-42). The mean score on stress was 22.78 (5.40), anxiety 19.46 (6.69), and depression subscale was 23.86 (5.35). Severity of distress was also assessed on the basis of the cut-off scores given by the developer of the scale. Approximately one-third of parents reported moderate 1155 (38.1%) to severe 1059 (35%) levels of stress. About 1324 (43.7%) reported severe anxiety and 515 (17%) extremely severe anxiety. Severe depression was reported by 1067 (35.3%) of the parents (Fig. 2).

Close

FIG 2.

Severity of distress among parents

Export to PPT

Association between child’s appraisal and parent’s perception of loneliness, PCR and conflict behaviour

To explore the association between the children and parent reported loneliness, PCR and conflict behaviour dimensions from children and parent’s perspective, Pearson product– moment correlation was used.

A significant positive association was found between child reported loneliness (LSDQ-C) and (LSDQ-P) parent perception of child’s loneliness (r = –0.088, n=3028, p<0.01), PCR-C and PCR-P (r = –0.159, n=3028, p<0.01). Although non-significant, child reported CBQ-C was found negatively associated with parent reported CBQ-P.

Among children, significant negative association was found between child appraisal of his loneliness (LSDQ-C) with depression among parents (r= –0.036, n=3028, p<0.05) and non- significant negative association was found with child reported PCR-C, CBQ-C, stress-P and anxiety-P. Among parents, parent perception of child’s loneliness had significant positive correlation with PCR (r= –0.044, n=3028, p<0.05) and also with anxiety (r = –0.039, n=3028, p<0.05) reported by them. No other significant association was found between LSDQ-P with CBQ- P, stress-P and depression-P.

Association between demographic and clinical variables

Age-wise children were categorized into two groups (age 10–13 years: n=1005; age 14–18 years: n=2023). Younger children reported better PCR than older children (p<0.05), and although non-significant, more conflicts and higher loneliness were reported by older children. Age-wise parent group division was also done (age 31–45 years: n=1303, age 45–60 years: n=1725). Younger parents reported more intimate relationship with their children than older parents (p<0.01). No other age-wise signi- ficant differences were observed among parents (Table III).

Gender-wise association was also explored for both children and parents groups (children: boys n=1155; girls n=1873; parents: fathers n=1735, mothers n=1293). Among children, significant difference was observed on PCR-C between boys and girls, girls have reported better quality of relationship with their parents (p<0.05), and similarly, among parents (PCR-P), mothers reported more intimate relationship and closeness to their children than fathers (p<0.05). Girls and mothers reported better PCR than boys and fathers. Boys scored significantly higher on loneliness measure (LSDQ-C) than girls (p<0.05). Father’s perception of their children loneliness (LSDQ-P) was significantly higher than mothers (p<0.05). On the loneliness scale, boys felt more lonely than girls (p<0.05) and fathers also reported that their sons were feeling more lonely than their daughters (p<0.05). Girls and mothers reported better PCR than boys and fathers. No significant difference was observed on conflict behaviour measures among them (Table III).

Means were also compared by grade level of the children. Post-hoc analysis revealed children of higher secondary level felt more lonely than matriculate and middle-grade children (middle 52.29 [5.73]; matric 52.65 [5.50]; higher secondary 53.02 [5.77]; p<0.05) and no significant difference was observed on PCR and conflict behaviour dimension. Among parents, educational status-wise comparison revealed that parents who were graduate and above had more intimate relations and were more close to their children than parents who had studied till higher secondary (graduate and above 35.74 [7.30]; higher secondary 34.42 [7.52]; p<0.05) and less educated (up to higher secondary) were more stressed (graduate and above 12.01 [2.65]; higher secondary 11.29 [2.67]; p<0.05) than parents who were graduate and above.

Parents living in urban locality had more intimate relationships with their children in comparison to parents living in rural areas (urban 35.36 [7.34]; rural 34.54 [7.64]; p<0.01). Participants living in rural areas seemed to be more difficult and parents exhibited more negative communication with their children than those in urban areas during the Covid-19 lockdown (urban 32.58 [4.44]; rural 32.96 [4.40]; p<0.05).

Strengths of the study

Our study has the advantage of a large sample size and a heterogeneous population representative of students from different socioeconomic backgrounds.

Implications and future directions

Our study showed that children who share cordial relations with their parents had fewer conflicts and also scored low on loneliness. This emphasizes the need for parental education to prevent the effects of loneliness on the future mental health of children. Our study emphasizes the need to plan guidance strategies with a joint effort of schools and families to strengthen within family relations of children. Families are the first centres of social life and the primary cores of education, and there is a need to make strategies and action plans to reduce loneliness among children. Our study suggests providing psychological and emotional support and strategies to prevent loneliness, dissatisfaction among children and its effect on PCR. Resolving parent–child conflict behaviour and technology-based inter- vention during prolonged period of home confinement may also be helpful. Loneliness and dissatisfaction among children may have long-term consequences, thus must be addressed.

Future studies should explore positive experiences during the Covid-19 pandemic along with loneliness and dissatisfaction and positive aspect of PCR from India and other countries as well.