Transcending Quarantine: Analyzing Character Traits and Mental Health During Covid-19 Lockdowns

By Aileen Tang, Charisse Hung, Josh Li, Sivaji Turimella, Giselle Kurniawan

Source: Everyday Health

Table of Contents:

Section I: Introduction

Section II: The Dataset

Section III: Relationship between DASS-21, GHQ-12, and SEC

Section IV: Relationship Between Three Psychological Measures and Four Factors

Section V: Relationship Between Three Psychological Measure and Sub-Factors

Section VI: Relationship between DASS-21 and Age

Section VII: Relationship between DASS-21 and Student Status

Section VIII: Strategies for Better Health: Improving “Transcendence” Strengths?

Section IX: Conclusion

Appendix: Dataset

Section I: Introduction

As we pass the one year anniversary of the COVID-19 lockdown in the United States, many of us are intimately aware of the price that it has come with. This lockdown, in the beginning, felt temporary. It was, as we were told, supposed to last two weeks. But two weeks became two months and two months became (what feels like) an eternity. Unsurprisingly, our collective mental health has taken a toll.

In this blog, we seek to better understand how COVID-19 (the restrictions, the disease itself, and how it affected our society) affected the mental health of our society. While this dataset focuses on a questionnaire of ~900 people from Italy, it offers insights that may help us better understand general trends of how society has been affected at large. Why read on? Learn what character and personality traits are most correlated to improved mental health and how to potentially improve them in yourself!

However, as a disclaimer, it is important to note that there may be cultural and country specific factors that may have affected the outcome of this dataset in particular. Thus, it’s important to remember that these conclusions may or may not be extrapolatable to populations that have a different economic status, cultural values, or country-specific developments. Additionally, in this dataset, there were 703 female respondents, and only 241 male respondents. Thus, it is important to keep in mind that this dataset may reflect trends or conclusions that may be less or more applicable across genders. Without a larger dataset, it is difficult to ensure reproducibility of these results and conclude which trends may transcend genders, cultures, and countries.

Section II: The Dataset

This particular dataset utilizes three different health tests and questionnaires to evaluate the overall health and well-being of the patient. These three measures are as follows: Depression, Anxiety, and Stress Scale (DASS-21), General Health Questionnaire (GHQ-12), and Self-Efficacy for Covid-19 (SEC).

  • DASS-21 measures the level of depression, anxiety and stress in subscores and sums these subscores into a final score. A higher DASS-21 score reflects a more poor well-being.
  • GHQ-12 assesses the individual’s habits and routines and asks if the individual usually acts in that manner. The test checks for minor psychotic disorders, and a higher score denotes a higher likelihood of having these disorders.
  • SEC is an assessment that the researchers developed to gauge people’s perceived self-efficacy during quarantine. 5 questions were created, and the response is based on a Likert scale from 1 to 5, for a maximum total score of 25. Generally a higher score reflects higher perceived self-efficacy.

Section III: Relationship between DASS-21, GHQ-12, and SEC

We decided to look into the relationship between these three measures to see if using all of them would be beneficial to our analysis. In the visualization, it is visible through the regression plane fitted onto the plot that there is a significant relationship between all three variables. This consistency between all three tests allows for us to look at all the tests with a degree of reliability. Furthermore the following plot clearly shows that as the DASS-21 and GHQ-12 scores increase, SEC scores decrease.

Each participant completed the Values in Action Inventory of Strengths-120 (VIA-IS-120) which measures different character strengths. This resulted in each participant having a numerical score for 24 individual character strengths. In order to simplify the data, the researchers performed a Principal Component Analysis (PCA) to reduce the dimensions of the data set and find the factors that contribute to the most variation. Four factors were extracted, and labelled transcendence, interpersonal, openness, and restraint. The character strengths that contribute to each factor are as follows:

Section IV: Relationship Between Three Psychological Measures and Four Factors

(DASS-21, GHQ-12, SEC and Openness, Restraint, Transcendence, Interpersonal)

To accomplish our project’s objective, we ran linear regression models to examine the associations of the four factors (Openness, Restraint, Transcendence, and Interpersonal) with the three psychological measures (DASS-21, GHQ-12, and SEC). We conceptualized that the larger the absolute value of the slope, the stronger the relationship between a factor and its psychological measure.

Key Takeaways: Transcendence has the strongest negative relationship with GHQ-12. Restraint has the weakest negative relationship with GHQ-12. A negative relationship means that an increase in this particular factor will lead to a decrease in the GHQ-12 score. A lower GHQ-12 score is associated with better health.
Key Takeaways: Transcendence has the strongest negative relationship with DASS-21. Restraint has the weakest negative relationship with DASS-21. A negative relationship means that an increase in this particular factor will lead to a decrease in the DASS-21 score. A lower DASS-21 score is associated with better health.
Key Takeaways: Transcendence has the strongest positive relationship with SEC. Interpersonal has the weakest positive relationship with SEC. A positive relationship means that an increase in this particular factor will lead to a higher SEC score. A higher SEC score is associated with better health.

Based on our observations, the Transcendence factor seems to have the strongest relationship with all three psychological measures. In particular, all of its linear models produced the slope with the greatest absolute value (in comparison to other strengths) of -0.2332 for DASS-21 (Depression, Anxiety and Stress Scales), -0.08729 for GHQ-12 (General Health Questionnaire) , and 0.1084 for SEC (Self-Efficacy measure for COVID-19). This may therefore imply that people well-endowed with the transcendence strengths (e.g hope, perseverance, zest) scored lower for psychological distress (lower levels of depression, anxiety and stress scale), higher for general mental health (fewer symptoms of worsening mental health) and higher for self-efficacy during the COVID-19 lockdown situation.

Section V: Relationship Between Three Psychological Measure and Sub-Factors

Upon discovering that across all of the four factors (Openness, Transcendence, Restraint, and Interpersonal), the Transcendence character factor had the strongest impact on DASS-21, GHQ-12, and SEC scores, we wanted to dig deeper to identify which of the sub-factors, or the characteristics that made up each of the overarching factors, had the highest impact on improving health. Unsurprisingly, many of the character strengths that fall under the Transcendence character factor were most strongly related with improved health. These factors, with the most impactful listed first, included (1) Hope, (2) Zest, and (3) Gratitude for a lower DASS-21 score, (1) Zest, (2) Hope, and (3) Curiosity for a lower GHQ-12 score, and (1) Zest, (2) Hope, and (3) Curiosity for a higher SEC score.

Section VI: Relationship between DASS-21 and Age/Age Groups

When evaluating DASS-21 scores with respect to the age of participants, the visualization above shows that DASS-21 scores from roughly 0–30 are present in all age groups. Since a lower DASS-21 score is desirable and corresponds to less depression, anxiety, and stress, this shows that despite the global pandemic, healthy mental health was present in people of all life stages. This visualization also shows that high DASS-21 scores seem to only be from the younger age groups. Almost all individuals with DASS-21 scores above 30 are less than 40 years old and furthermore those with DASS-21 scores above 40 tend to all be under 30 years old. This suggests that the highest levels of depression, anxiety, and stress one month into COVID-19 lockdowns were mostly found in young adults.

The three histograms above show the frequency of DASS-21 scores among people separated by an age range (i.e. 18–40, 41–60, and 61–80). The shape of the distribution is right-skewed for all 3 histograms, which is encouraging, as this implies that a majority of the participants in all of the age groups tend to have lower DASS-21 scores. By looking at the sum of all of the frequencies, we can see that a majority of the participants are within the 18–40 range. The maximum DASS-21 score gets smaller with older generations. The mean DASS-21 score similarly reflects a decreasing trend, which implies that people in older generations tend to have lower DASS-21 scores on average.

It is important to note that this dataset utilizes cross-sectional data as opposed to longitudinal data. This is key in determining how we are to interpret these histograms against each other. We cannot make any generalizations about age correlating to a lower DASS-21 score, as each data entry is a different person, not a group of people studied over time. Instead, we can suggest a generational difference when it comes to DASS-21 scores, meaning that younger generations might have higher DASS-21 scores on average compared to older generations.

Section VII: Relationship between DASS-21 and Student Status

Another factor that we wanted to analyze was the relationship between DASS-21 scores and whether an individual is a student. With the rise of online learning and lockdown of in-person classes, which results in a decline of student life, it begs the question whether this has an effect on a student’s mental health. Through segmenting the data between student and non-student status, it can be seen that on average, being a student does lead students to have higher DASS-21 scores, and thus, poorer mental health during Covid-19 lockdown. This may also have a relationship with age, as many students are often younger as well. This is another potential avenue to investigate with this dataset.

Section VIII: Strategies for Better Health: Improving “Transcendence” Strengths?

With Transcendence being most strongly correlated with improved health, it is important to highlight the character traits most related to this factor: Hope, Spirituality, Zest, Gratitude, Perseverance, Self-regulation, Love, Forgiveness, and Curiosity. Unsurprisingly, all of the traits are also the 3 sub-factors, or character strengths, that are most related to improved psychological measures.

With a quick Google search, it is not difficult to find insightful articles and strategies that can help you improve these traits within yourself. Some of them are as follows:

Article Recommendations on How To Improve Zest:

Article Recommendations on How To Improve Hope:

Article Recommendations on How To Improve Curiosity:

Article Recommendations on How To Improve Gratitude:

Section IX: Conclusion

The data presented that the highest levels of depression, anxiety, and stress one month into COVID-19 lockdowns were mostly found in young adults. The data also showed us that transcendence strength is one of the keys for a stronger mental health, lower levels of psychological distress and a higher self-efficacy during the COVID-19 pandemic. This conclusion was also supported by the fact that zest, hope, curiosity, and gratitude (which all make up parts of transcendence strength) are associated with better wellbeing during the COVID-19 pandemic.

So, what’s the key to “surviving the lockdown?” The hard truth is that there is no hard and fast answer to that. This data has specific cultural, country specific, and gender factors that may have affected the outcome of this dataset in particular. In addition, the size of this dataset makes it difficult to generalize these outcomes to the general population. However, it seems like being grateful for what you have in life, being hopeful for a better future, approaching every day with zest, and keeping your mind engaged with pursuits that you are curious about are some habits that are keeping people going in lockdown. With that, we wish you the best in becoming a more grateful, curious, hopeful, and zestful person.

Section X:

The dataset utilized for this article can be found here. The associated paper can be found here.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store