**3. Results**

Table 1 shows the descriptive statistics of variables used in the current study and gender differences in the dependent variable. There is a difference in protective behavior changes since COVID-19 between middle school boys and girls. Middle school girls had greater behavioral changes (17.20) than middle school boys (16.40) since the COVID-19 pandemic. Girls showed more decreased outdoor activities compared to boys (4.19 vs. 3.94), more reduced frequency of social gatherings (4.34 vs. 4.03), and more increased time at home (4.68 vs. 4.49). The average age of the middle school students was 14.4 years old. Slightly more than one-fourth of the total sample reported that they subjectively perceived being in poverty. The average scores for health status, family satisfaction, levels of sanitation since COVID-19, and perceptions regarding the risk of COVID-19 were 16.23, 38.77, 16.42, and 18.06, respectively.

Table 2 indicates which factors significantly influenced protective behavior changes since COVID-19 among middle school students. Model 1 includes individual factors and reveals that gender was associated with protective behavior changes since COVID-19 (*β* = 0.77, *p* < 0.05). When family factors were entered into model 2, girls were still more likely to have greater protective behavior changes since COVID-19 as compared to boys (*β* = 0.76, *p* < 0.05). Health status was negatively related to protective behavior changes since COVID-19 (*β* = −0.12, *p* < 0.01). Regarding family factors, family satisfaction was positively associated with protective behavior changes (*β* = 0.12, *p* < 0.001). Once COVID-19 environmental factors were included in model 3, gender, health status, and family satisfaction remained significant (β = 0.68, *p* < 0.05; *β* = −0.14, *p* < 0.001; *β* = 0.06, *p* < 0.01). For COVID-19 environmental factors, both levels of sanitation since COVID-19 and perceptions regarding the risk of COVID-19 were significantly related to protective behavior changes since COVID-19 (*β* = 0.27, *p* < 0.001; *β* = 0.35, *p* < 0.001).



Note: \* *p* < 0.05; Significant difference between males and females.

**Table 2.** Regression Results of Unstandardized Coefficients (standard error), Predicting Protective Behavior Changes Since COVID-19.


Note. \* *p* < 0.05. \*\* *p* < 0.01. \*\*\* *p* < 0.001.

### **4. Discussion**

The principal purpose of the current study was to examine determinants of protective behavior changes since COVID-19 in the context of three dimensions: individual, family, and environmental. This study revealed how individual, family, and environmental determinants influenced protective behavior changes among middle school students in South Korea since the emergence of the COVID-19 pandemic. Even though all people, regardless of their demographics or socioeconomic status, have been greatly influenced by the unprecedented spread of COVID-19, middle school students are at particular risk because of the importance of the adolescent years for development [33]. Given that little is known about protective behavior changes since COVID-19 among middle school students, the present study contributes to understanding the determinants of COVID-19 protective behavior changes in this population. This study indicated that both individual and family factors were important to understand protective behavior changes. Further, levels of sanitation since COVID-19 and perception of risk regarding COVID-19 also influenced protective behavior changes among middle school students.

Individual factors that influenced protective behavior changes since COVID-19 among middle school students included gender and perceived health status. This study indicated that there is a gender difference in the protective behavior changes since the coronavirus outbreak. Middle school girls showed that they were more inactive since COVID-19 compared to middle school boys. More specifically, girls reported decreased outdoor activities, reduced frequency of social gatherings, and increased time at home. Generally, being inactive may be perceived in a negative way, but the meaning during the coronavirus

pandemic could be understood to be more careful and cautious so as not to be infected by the virus. In general, adolescent boys are more likely than adolescent girls to engage in risky behaviors [34], which in the time of COVID-19, may manifest as engaging in social activities. Greater behavior changes during the coronavirus pandemic means that people are less likely to be infected, indicating that middle school boys should be educated to be more careful to avoid the spread of COVID-19. That greater health status was related to fewer protective behavior changes since COVID-19 demonstrated that healthy individuals may underestimate the risks of coronavirus and therefore continue to behave as they did before the pandemic. That is, healthy people may think that they are able to overcome the virus even if they are infected and think that the coronavirus is similar to the seasonal flu. However, given higher death rates around the world due to the virus and the unprecedented influence on the global economy and individuals' lives [35], everyone should be very careful not to be infected, regardless of health status. Seemingly healthy people are not immune from coronavirus infections and negative consequences, but even if they have a mild infection, they can be a vector for transmitting the virus to the most vulnerable, such as older adults or those with underlying conditions that make them more at risk. Thus, even healthy middle school students must participate in preventive measures, such as social distancing, in order to protect themselves and other people.

This study confirmed that some family factors were also related to protective behavior changes since coronavirus emerged. Households' economic resources were not related to such behavioral changes, while family satisfaction was associated with the behavioral changes. The specific protective behavior changes include reduced outdoor time with friends and increased time at home due to social distancing. If people have negative family situations, spending more time together at home due to COVID-19-related policies may lead to conflict and other negative outcomes [36]. On the other hand, middle school students who have good relationships with their family members may be more accepting of spending more time at home. That is, individuals with higher family satisfaction may be more likely to engage in such COVID-19 preventive practices by staying home with their family members, leading them to be more protected from coronavirus disease. However, middle school students who have more conflicts and disharmony with parents or siblings may be exposed to higher risks of COVID-19 through behaviors which may expose them to the virus. That is, improved relationships between children and parents or siblings can result in protective behavior changes related to social distancing, leading to decreased risks of coronavirus infection among middle school students.

It is also important to pay attention to environmental factors related to COVID-19 to understand protective behavior changes since the coronavirus disease emerged. This study revealed that higher levels of sanitation since COVID-19 and greater perception of risk related to COVID-19 results in greater protective behavior changes among middle school students. In other words, environmental changes due to the virus have an impact on middle school students' behaviors. Middle school students who increase their sanitation behaviors may also be more likely to follow social distancing guidelines and therefore be at decreased risk for coronavirus infection. Thus, it is critical for middle school students to be educated about the importance of sanitary behaviors during the COVID-19 pandemic. This education on sanitary behaviors can be helpful to prevent the spread of coronavirus as middle school students change their behaviors to protect themselves from COVID-19. We recommend that this education includes the following: the importance of frequent hand washing; not touching one's face; not sharing personal items; and avoiding going to crowded places. In addition, middle school students who better perceived the risks of COVID-19 showed greater changes in their daily behaviors, thereby lowering their potential exposure to coronavirus. As middle school students are still maturing and spend most of their time at home and school, the role of parents and teachers is important to increase their perceptions regarding the risk of COVID-19. That is, adults who are responsible for the health of middle school children should first understand the risks of COVID-19, because they are a primary source to teach the children about the risks of COVID-19 [23]. However, along with receiving information from parents and other adults, children are likely to access and retrieve information from social media such as Instagram, Facebook, Twitter, etc. [19]. Given that social media provides both true and false information [37], it is imperative to teach middle school children how to distinguish correct information about COVID-19 from inaccurate information.

#### **5. Conclusions**

This study indicated the importance of environmental factors to influence behaviors that decrease the risk of COVID-19 infection. After a year, the COVID-19 pandemic is still affecting people around the world; however, little is known about determinants of protective behavior changes among middle school students since COVID-19, which helps middle school students be at lower risk of being infected by COVID-19. This study contributes to literature related to COVID-19, particularly among middle school students, and suggests to consider three factors–individual, family, and environmental—in order to prevent middle school students from contracting and spreading the virus.

Even though this study sheds light middle school students' protective behavior changes since the beginning of the COVID-19 pandemic, findings should be interpreted in the context of limitations. First, this study was conducted in South Korea, so behavioral changes among middle school students in other countries might be different from those in South Korea. Thus, cultural differences should be considered if the findings are to be applied to other settings. Second, additional factors can influence protective behavior changes since COVID-19, and we recommend that extra variables should be included in future studies. Third, this study focuses on middle school students regardless of disability status. Children with disabilities may have different determinants of protective behavior changes since the COVID-19 to consider. Therefore, we suggest that future studies focus on children with disabilities to identify if they differ from children without disabilities in terms of factors influencing protective behavior changes since the COVID-19 pandemic. Fourth, we calculated the sample size using a sample size calculator [38] with 5.24 confidence interval and 95% confidence level. The sample size needed in this study was 349; however, we recommend that an advanced power analysis or sample size estimation should be considered in future studies. Fifth, this survey needed to be completed during September and October 2020. Due to limited time and budget, we chose a convenience sample.

**Author Contributions:** Conceptualization, J.L., and H.L.; Data curation, H.L.; Formal analysis, H.L.; Investigation, J.L.; Methodology, J.L.; Project administration, H.L., and G.C.; Resources, G.C.; Writing—original draft, J.L., and J.A.; Writing—review and editing, J.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of Inha University (#200810-1A).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author. The data are not publicly available due to ethical reasons.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

