On March 13, 2020, Wyoming Governor Mark Gordon issued a statewide directive on self-quarantining based upon travel (Gordon, 2020). On March 30th, 2020 Teton County Public Health officials ordered all residents to follow social distancing guidelines (Riddell, 2020). Social distancing guidelines and travel restrictions have forced many individuals to disconnect from normal social networks built around a multitude of factors, one being proximity. Social networks have been shown to be positively associated with mental health, particular for those individuals with high stress (Achat, et al., 1998). Increased psychological impact has been shown in the initial stages of the Covid-19 pandemic leading to strongly negative effects (Wang, et al., 2020). A key finding that geographic proximity is effective in moderating the effort needed to maintain relationships (Martin, et al., 2006) has led us to consider how social distancing guidelines and the subsequent stress associated with Covid-19 and those guidelines has led to changes in support networks, specifically the changes in parental relationships with their adult children. Our exploration of secondary infection rates of Covid-19 led us to explore the social and personal impacts of the many changes to lifestyles and relationships that are occurring in response to the pandemic.
Our hypothesis is that adult children who live more than 500 miles from their parents will experience a greater improvement in the quality of their relationship in comparison to adult children who live closer than 500 miles from their parents. The primary mechanisms for this difference is the reduced opportunity for interaction which has been shown to maintain the quality of nearby relations in comparison to far relations (Rivera, et al., 2010). We believe the difference in frequency of interaction leads to a greater gap in the quality of parent/child relationship for far relations. With social distancing, the mechanisms for interaction are balanced through the use of technology, the need for support increases, and therefore the relationship gap for far relations will be closed more significantly than for nearby relations.
We chose an observational study to retroactively evaluate the impact social distancing treatments have had on the parent child relationship based upon geographic distance. We used an online survey due to the constraints of in person interviews and the goal of collecting large numbers of responses, to control for confounds, quickly. All participants completed an online consent form and results were anonymous. We used a test survey with 12 initial respondents and then modified our survey based upon our initial results. Our survey was broken into the following sections: demographic information, questions about personal outlook, parent/child relationship, and questions about changes due to social distancing. If respondents did not have parents who were alive, then the survey automatically ended. We sought reliability in our responses through consistent formatting, clear wording, and clear question expectations. Questions for the parent/child relationship section were sourced from the Network of Relationship Inventory (Furman and Buhrmester, 2009) to help evaluate the nature of the relationship prior to social distancing.
We used a non-probabilistic convenience sampling technique of individuals who the author had email access to. This sample all lived in Teton County, Wyoming or Teton County, Idaho. The combined population of these communities is approximately 32,000. With 72 respondents(~33% response rate), we preserved independence by measuring <10% of our population. We sent survey requests to approximately 210 individuals who span ages from 20-60 years old, economic backgrounds (with a slight skew towards middle/upper income brackets), educational background (with a slight skew towards college educated), and geographic locations of where their parents currently live. Our population was limited to rural individuals, nearly all of whom were white. We received 72 responses with 8 responses omitted, 3 due to deceased parents, and 5 because those respondents were living with their parents for a total sample size of 64 individuals. Our population were individuals connected to Teton County Search and Rescue, Mountain Academy Middle and Upper School families, and friends of the author. Our hope was that the geographic diversity of upbringing would have mitigated the bias associated with these specific groups.
Modifications from Field Testing:
Based upon the results from our initial field testing (n=6 of far parents and n=6 for near parents), we removed questions that explored the dynamics of home, focused our medium of communication to just phone and video, and increased the reliability of our questions by improving the structure of our questions and the wording. We researched and included questions based upon best practices in parent/child relationship evaluation - Network of Relationship Inventory. We focused our final survey on the relationship between adults and their parents based upon geographic distance.
We evaluated how much a relationship has changed since social distancing guidelines have been implemented. A strong positive change was scored as a 2, a positive score as 1, no change 0, negative change scored -1, and a strong negative change was scored as -2. 27 out of 46(58.7%) far relations respondents reported a positive change in their relationship whereas only 8 out of the 18(44.4%) near relations recorded a positive change. See Table 1 for more analysis on this relationship. Relatedly, of near relation respondents, (7/18 or 39%) had a positive change in outlook versus (6/18 or 33%)) had a negative outlook. This compares to far relations where (25/46 54%) had a negative change in outlook and (6/46 or 13%) had a positive change in outlook. These results point to an interesting possible relationship that an increase in distance from has led to an increase in negative outlook on current events - a possible confound worth further exploration.
Table 1: Shows the descriptive stats of how relations change for our two groups (near & far). Note standard deviations are both larger than their respective means. Far relations do have a larger median and mean.
We chose to conduct a difference of means comparison to determine if there was a statistically significant difference between our means. Our null hypothesis (H0) is that there is no difference in relationship change between near and far parents.
H0 : xf - xn = 0.
Our alternative hypothesis (HA) is that far relations will experience a greater positive change in the relationship.
HA : xf - xn > 0
From looking at the basic descriptive statistics, we see that far parents (𝜇=0.65, SD = 0.67) show a slight greater positive change in the relationship in comparison to near parents (𝜇=0.50, SD = 0.79). Upon conducting a difference of means test, we determined that the significance of this result (p=0.239, Cohen's d=0.215) shows there is little significance (p>0.05) and with minimal effect size (Cohen's d < 0.5) in this result. At the 𝛼=0.05, we fail to reject our null hypothesis and were not able to find that there was a statistically significant difference in relationship change between parents that lived far and parents that lived near their children.
Figure 1a & 1b: Relationship change is measured as strongly negative(-2), negative(-1), no change(0), positive(1), strongly positive(2). Both histograms are nearly normally distributed with 1a having greater % frequency of responses (~59%) showing positive change compared to the % frequency of positive change (~44%) seen in 1b.
We further explored our dataset by evaluating for collinearity. We compared the health of the relationship prior to social distancing, the change in personal outlook, the change in communication and the imbalance in relational support to see if any of those factors had correlation with the change in relationship. We found that there was a small positive linear correlation with a change in the relationship if communication increased (r = 0.3). We found the health of the relationship prior had nearly no effect on a change in the relationship (r=0.04). We found that a change in one’s personal outlook on life had a very small positive correlation with relationship change (r= 0.15). And lastly, found that the primary supporter in the relationship (parent or child) had little correlation with relationship change (r=0.03).
This study yielded three insights. First, our hypothesis was not supported by the data. Adult children with parents who live far away did not have a significant improvement in their relationship in comparison to adult children whose parents live close by. The second was that all participants who increased their communication with their parents reported an improvement in their relationship. The third was that a greater percentage of children who lived far from their parents had a negative outlook on life compared to those who lived near their parents.
In conclusion, we sought to find a change in the quality of relationship during the social distancing guidelines for adults in Teton County based upon their distance from their parents. The data did not support this finding.
We would have improved our sampling technique to be a probabilistic simple random sample of residents within Teton County - this is feasible given the geographically constrained region. This would have increased the representativeness of the results by increasing ethnic, economic, and professional diversity of our sample. We would have increased the sample size to increase the significance of our results and to reach minimum thresholds for independence. The increased sample size would also have allowed us to control for collinearity of confounding factors. A further improvement would be to test these results over time as social distancing continues to persist. Lastly, we would have continued to refine the questions asked to better control for confounds within our sample.
We recommend partial replication studies to further validate and improve upon our results of no difference. We recommend future studies have a sample population of sufficient size and that a baseline evaluation of the role distance plays in the quality of a relationship outside the constraint of Covid-19 restrictions be established through research or further studies. We also recommend that this study be replicated using stratified sampling based upon building two populations of adults whose parents either live far or near and then randomly sampling from each of those populations for a comparison.
Achat, H., Kawachi, I., Levine, S. et al. (1998) Social networks, stress and health-related quality of life. Qual Life Res 7, 735–750. https://doi.org/10.1023/A:1008837002431
Furman, W., & Buhrmester, D. (2009). The Network of Relationships Inventory: Behavioral Systems Version. International journal of behavioral development, 33(5), 470–478. https://doi.org/10.1177/0165025409342634
Gordon, M. (2020). Executive Order 2020-2.pdf. Retrieved from https://drive.google.com/file/d/1FTEUxYXwwbIjcnwDl18w0pTLtM8ivTab/view
Martin, J. L., & Yeung, K. T. (2006). Persistence of close personal ties over a 12-year period. Social Networks, 28(4), 331-362.
Wang, C.; Pan, R.; Wan, X.; Tan, Y.; Xu, L.; Ho, C.S.; Ho, R.C. (2020) Immediate Psychological Responses and Associated Factors during the Initial Stage of the 2019 Coronavirus Disease (COVID-19) Epidemic among the General Population in China. Int. J. Environ. Res. Public Health 17, 1729.
Riddell, Travis. (2020, March 30). Retrieved from https://www.tetoncountywy.gov/DocumentCenter/View/13649/Teton-Health-Order-20-4-gatherings-with-only-your-household
Rivera, M. T., Soderstrom, S. B., & Uzzi, B. (2010). Dynamics of dyads in social networks: Assortative, relational, and proximity mechanisms. annual Review of Sociology, 36, 91-115.