Socio-psychological impact of the Covid-19 pandemic on Young Adults in Germany

This project was carried out as part of the TechLabs “Digital Shaper Program” in Münster (summer term 2021).

Abstract

The Covid-19 pandemic has significantly impacted people’s life across the globe and produced a unique set of challenges for both individuals and society as a whole. Navigating these challenges led to the implementation of a wide range of protective measures to preserve public health in the wake of the Covid-19 pandemic. As a result, socio-psychological interactions between individuals and individuals with society experienced a lasting change and had to adapt to the new challenges and restrictions. Our project aims to gain further insights on how young adults in Germany are socio-psychologically impacted by these changes imposed by the Covid-19 pandemic. To fulfill this research objective, we utilized a data science and machine learning based approach to identify four statistically significant focus areas. The four focus areas include the compliance with legal Covid-19 measures as well as the reasoning behind compiling or failing to comply with these measures. Furthermore, the perceived political representation and impact of the Covid-19 pandemic on political beliefs are analyzed. Our findings can be used to derive improved measures to address the issues of young adults in Germany in context of the Covid-19 pandemic and aim to inspire a more fact-based and engaging public discourse on these issues.

Introduction

Since the end of 2019, the Covid-19 pandemic has significantly impacted all areas of people’s everyday lives. In an effort to combat the spread of the virus, governments all over the world have implemented a wide array of measures to protect public health. These measures usually include far-reaching restrictions and thereby impact the foundation of individual life as well as interpersonal relationships between individuals in the context of socio-psychological interactions.

The goal of our TechLabs project was to gain insights into the socio-psychological impact and perception of the Covid-19 pandemic among citizens and especially among young adults in Germany. We aim to answer questions about how important Young Germans deem the imposed measures to combat the Covid-19 pandemic in Germany and what motivations they have to adjust their behavior accordingly. Furthermore, we analyzed how opinions on the Covid-19 pandemic relate to broader political beliefs and perceptions.

Methodology

To answer the aforementioned questions, we utilized data provided by “The Youth Study of TUI Foundation” in cooperation with the WZB (Berlin Social Science Center). The research for this project was conducted under the lead of Marcus Splittler in September 2020 and questioned 1.011 young adults aged between 16 and 26 about political beliefs and Covid-19 related issues. Participants were recruited on the basis of a representative quota plan and answered the questions through an online questionnaire. The final dataset was provided by the GESIS (Leibniz Institute for the Social Sciences).

After an initial analysis and examination of the data, the first step was to clean and shape the dataset. One challenge was that the data frame included a lot of records that either did not provide a definitive answer (“Do not know/no response”) or that some questions were not asked in the survey wave (“Question not asked”). Hence, we replaced these values (999 or 977 in numeric representation) with the more Python-like label NaN.

In a second step we performed a tool-guided data meta-analysis with the respective Python libraries such as dtale, PandasProfiling and SweetViz to identify focus areas for our further research. The focus areas were chosen based on correlation measures as well as predictive power scores (PPS), which are created by utilizing Decision Tree Regressors to identify linear and nonlinear relationships between features.

In a third step, we then mainly focused on both descriptive and inferential data analysis and the visualization of the identified focus areas and relationships with Python libraries such as numpy, pandas, seaborn and matplotlib.

Results

The previously mentioned meta-analysis led to the identification of four distinct focus areas based on statistical significance. The four focus areas investigate the compliance with Covid-19 measures as well as the impact of Covid-19 on perceived political representation and political beliefs:

Focus Area #1: Perception and compliance with legal Covid-19 measures

Looking at first descriptive results the data shows that young Germans assess the measures taken so far to combat the Covid-19 pandemic in Germany as appropriate. However, we see no linear relationship between the assessment of the measures and participants’ age.

The perceived importance of measurements to combat the Covid-19 pandemic is reflected in Young Germans’ behavior just before the second wave of the pandemic in September. Again, the descriptive data indicate that young Germans mainly follow the measures and recommendations. However, some young Germans only follow the measures and recommendations to a limited extent. Overall participants are more likely to follow Covid-19 measures if they deem these measures and restrictions as necessary and appropriate as shown by a moderately positive correlation between compliance with and evaluation of Covid-19 measures (r = .53, p = .000).

The Covid-19 pandemic has immense social and psychological consequences for Young Germans and their plans for the future. Asked for their assessment of the future and whether they have a rather optimistic or pessimistic view about their personal situation, we find a weak correlation (r = .23, p = .000).

Focus Area #2: Reasons to follow or disregard legal Covid-19 measures

Looking at the reasons why young Germans aim to follow the measurements, we see that the two strongest reasons are to protect their own health and the health of others. These two variables, namely the protection of one’s own health and the health of others, also have a fairly strong correlation with r = .83 and a p-value of .000 as well as the most significant PPS across all features (.94 PPS and .95 PPS respectively). In addition to this, the degree of compliance with legal Covid-19 measures is also a significant predictor for giving the aforementioned arguments as reasons for following the measures with PPS scores of .66 and .63.

Across all participants, protecting the health of others is generally a stronger reason to follow measures than the protection of one’s own health. On the other hand, enforcing penalties for failing to comply, although efficient, is a lesser motivator to follow legal measures when compared to the protection of health. This relationship suggests that emphasizing the protection of public health might be the most efficient method to increase legal compliance with Covid-19 measures among young adults in Germany.

Fig. 1: Visual insights into (1) perception and compliance with legal Covid-19 measures and (2) reasons to follow these measures.

Focus Area #3: Perceived political representation in context of the Covid-19 pandemic

The question of whether young people feel they have sufficient political representation was examined. It becomes clear that most young people do not feel that they are being addressed enough, since it is mainly the interests of the older generation that receive attention.

A correlation between political representation (feature democracy_18) and the assessment of the measures (feature corona_2 and corona_4) as well as the perception of future change (feature corona_1) reveals a weak positive linear relationship. The correlation was determined using Pearson’s method. Thus, there is no strong correlation that the lack of political representation of young citizens is associated with evaluating Corona measures as well as possible future changes. This relationship shows that the overall perceived lack of political representation is not limited to or only associated with certain evaluations of protective measures, but rather an omnipresent trend among all young adults which can be deemed as problematic and might further increase political disenchantment of young adults.

Focus Area #4: Political beliefs and future prospects in context of the Covid-19 pandemic

During the first Covid-19 wave, most young adults in Germany did not fear any impact, on future prospects in general, yet. Rather, the data shows that they saw the situation as an opportunity for globalization, digitalization as well as aspects to stop climate change to come into sharper focus and be driven forward. Only the perceptions on effects on immigration and emigration and on the political parties rejecting the EU were varied as well as fears that this would be influenced by Covid-19. Further data analysis indicates that participants tend to take an either optimistic or pessimistic stance on these issues in the wake of the Covid-19 pandemic, meaning that individuals who perceive threats or opportunities in one topic area are more likely to extrapolate this assessment across all topic areas.

Although perceived future prospects differed to some extent, there were no differences between these groups in terms of opportunities or risks and the perceptions were very similar. The computation of various correlations shows a weak linear relationship between future prospects and the previously mentioned issues. All in all, this leads to the conclusion that young adults remain hopeful about the future and that the Covid-19 pandemic did not fundamentally change political beliefs. Nevertheless, individuals who tend to view the impact of Covid-19 on the future more positively or negatively are more likely to demonstrate this assessment across a range of political issues.

Fig. 2: Visual insights into political beliefs and future prospects of young German adults during the Covid-19 pandemic.

Conclusion

The TUI Foundation’s youth study deals with the psychosocial impact of the Covid-19 pandemic on young Germans and was used as a foundation for our data analysis. Relationships between features were identified based on correlation and regression algorithms and led to the identification of four focus areas. In general, a rather optimistic assessment of the future can be seen across young citizens, for example in the trend development of globalization and digitalization. However, most young citizens do not feel sufficiently represented politically. Nevertheless, this lack of representation does not indicate that the specified measures are being strictly pursued. The majority of all young German citizens follow the legal Covid-19 measures, which can be attributed to the protection of their own and their general health. This results in a rather optimistic psychosocial basic attitude for the future.

In conclusion our findings show that there is a wide range of socio-psychological impacts on young adults in Germany. Measures or campaigns targeting areas of interest for young adults e.g. climate change or digitalization could increase feelings of political representation and thereby decrease the danger of broad political disenchantment among young people. In addition to this, instilling hope in these topics of interest could also lead to a more positive and optimistic outlook on the future in general as optimistic convictions on one topic area usually correlate with an overall more optimistic outlook across the whole spectrum of political issues. To increase compliance with Covid-19 measures among young adults, regulators should primarily opt for messages focusing on the protection of the health of other members of society as this is the strongest motivator for complying with measures to combat the Covid-19 pandemic.

During this journey, we faced a lot of challenges: Developing a concept for our analysis, finding a compatible data set, dealing with Python package errors and collaborating online. The collaborative effort to solve these problems and face these challenges together helped us grow as a team and taught us to solve complex Data Science and Machine Learning problems. Hence, we say thank you to the TechLabs MS team for the opportunity to participate in this project and for the help provided over the last months!

The team

David Kösters Data Science: Python

Hannah Claßen Data Science: Python

Kai Wohlgemuth Data Science: Python

Natascha Löffler Data Science: Python

Mentor

Christian Porschen

Our community Members share their insights into the TechLabs Experience