Remote Learning Experience and Adolescents’ Well-Being during the COVID-19 Pandemic: What Does the Future Hold?

Background. Major shifts within the education system have taken place during the COVID-19 pandemic; frontal teaching was often replaced with remote learning, which has affected students in many ways. We investigated the associations and predictors of perceptions of the remote learning experience on well-being (life satisfaction, self-rated health, psychosomatic, and psychological symptoms). Methods. We conducted a cross-sectional research study consisting of 1019 school students in Israel aged 11–18 (53.5% girls, 46.7% boys). Questionnaires were distributed from May–July 2021 during school time. The percentages of participants with various levels of well-being (WB) and remote learning experience were compared. Multiple regression procedures were used to analyze factors predicting wellbeing. Results. All of the remote learning items had statistically significant positive correlations with life satisfaction and self-rated health (i.e., better overall WB was associated with a more positive perception of the remote learning experience). Male gender, high socioeconomic status, greater involvement in lessons in the past year, and connection to the pedagogical team/school and peers predicted better overall WB (F-ratio = 14.03; p < 0.01; adjusted R2 = 0.08). Conclusions. Our results highlight the need for schools to target youths’ coping skills, which may lead to better remote learning experiences. These findings also provide several implications for the need to support children and adolescents through positive activities, relaxation/mindfulness, and cognitive coping to deal with the psychosomatic symptoms during remote learning periods.


COVID-19 was declared as a global pandemic by the World Health Organization in
March 2020 [1]. Restrictions were enacted around the world, most of which caused disruption to the daily life of children and adults alike, many affecting education systems [2].
For children and adolescents, this has perhaps been most notable in the area of education. In March 2020, nearly 1.87 million Israeli youth from kindergarten through 12th grade transitioned from in-person education to remote learning platforms [3]. Previous studies have identified this phenomenon as emergency remote teaching, describing a temporary change of instruction as a result of a crisis or traumatic situation [4,5]. Emergency remote teaching has many technological, pedagogical, and social challenges for both teachers and students [6][7][8]. According to a report by the Organization for Economic Cooperation and Development, children's daily lives have been impacted by the COVID-19 pandemic. School and leisure activities for children have been affected, many activities being offered online instead of in-person. Children were expected to spend more time on digital devices in order to stay in school and connected with others [9].
Several theoretical articles have postulated the impact of remote learning on both typical and vulnerable populations regarding social difficulties, compromised learning, and

Participants
The target population of the study was students in grades 5-12 in Israeli schools. Students were given the option to participate or not. The sampling method was probabilistic random sampling within layers. Participating students were asked to fill out a questionnaire during one of their classes during remote sessions. The classes were sampled from the most updated list from the Israeli Ministry of Education by age group, geographical distribution by sector, and type of school. The sample included a total of 1019 Israeli students aged 11 to 18 (53.3% girls, 46.7% boys; chi-squared = 21.47; p < 0.001). The representation of students from 12th grade was the lowest (n = 56, 5.5% of the sample; chi-squared = 78.96; p < 0.001; Table 1).

Data Collection
The average response rate was 25.5%. The research tool was an online anonymous self-completion questionnaire. The questionnaire includes the HBSC core questionnaire and new questions on the effects of COVID-19. The HBSC is a large cross-national study that has been used to examine the health and lifestyle determinants of school age children in 52 countries and regions for over 30 years. The methodology was consistent with that of the International HBSC Survey. The sampling error was ±3.1% at a 95% confidence level [27].
This study received approval from the Israeli Ministry of Education, as well as the Ethics Committee of Israel's Chief Scientist Office.

Independent Variables
The HBSC includes items describing participants' sociodemographic characteristics [27,28]. More specifically, study participants reported their self-identified gender (male or female), grade (5th through 12th), and change in socioeconomic status. Change in socioeconomic status was evaluated using the question: "according to your opinion, in comparison to before corona virus 2019 pandemic, in the past 30 days, what is the socioeconomic status of the family." The question refers to the last 30 days in order to learn about the average socioeconomic status. Answers consisted of the following options: "much worse", "less good", "no change", "better", and "much better". Answers were combined to create the following scores: 1 point-"much worse or less good, 2 points-"no change", and 3 points-"better or much better".

Remote Learning Experience
Two broad categories of remote learning experience were evaluated: attitudes toward remote learning, and remote learning and social connectedness.
Attitudes toward remote learning consisted of the following questions: (1) In the past 12 months, how much were you involved in lessons conducted remotely (via Zoom). Answer options were: "I did not attend lessons", "I connected with the camera off and did not listen to the lesson", "I connected to the lesson with camera off and listened", "I connected to the lesson with camera on and listened without being active", and "I connected to lesson with camera on and was active". Answers were coded in the following way: "I did not attend lessons" or "I connected with camera off and did not listen to the lesson" (1 point); "I connected to the lesson with camera off and listened" (2 points); "I connected to the lesson with camera on and listened without being active" (3 points); and "I connected to lesson with camera on and was active" (4 points). The second question was: "How do you feel about the lessons being conducted remotely?" Answers included: "dislike very much" (1 point); "dislike slightly" (2 points), "like slightly" (3 points); "like very much" (4 points).

Remote Learning and Social Connectedness
Remote learning and social connectedness was assessed via the following two questions: (1) "During COVID-19, while studying remotely, how much did you feel connected or disconnected from the pedagogical team and school?" and (2) "During COVID-19, while studying remotely, how much did you feel connected or disconnected from your peers?" For both questions, the following scale was used: "very much disconnected" (1 point); "slightly disconnected" (2 points); "slightly connected" (3 points); "very connected" (4 points); and "very much connected" (5 points) [28].

Dependent Variables
The dependent variables consisted of well-being outcomes (life satisfaction, self-rated health, and psychosomatic and psychological symptoms). All measures were adopted from the HBSC survey mandatory questions protocol.
Life satisfaction was measured by the Cantril's ladder [27], an efficient, global measure with high construct validity [28]. Young people were presented a picture of a ladder with steps ranging from 0 to 10 (where 10 represented the best possible life and 0 the worst).
They were asked to indicate where on the ladder they would place their life at present. This scale was used as a continuous measure.

Self-Rated Health
SRH is a standardized indicator that has been used extensively in various health research. For the current analysis, self-rated health was assessed using the question: "How would you describe your health in the past year?" Answers were scored on a 4-point scale ranging from 1 (excellent) to 4 (poor) and were then recoded to 1 (excellent), 2 (good), and 3 (poor and fair) [27].

Psychosomatic and Psychological Symptoms
Psychosomatic and psychological symptoms were assessed using eight questions pertaining to pain (headache, abdominal, back), feelings (bad mood, anger, irritability), sleeping (difficulties sleeping or falling asleep) and dizziness. Each question was scored on a 5-point scale ranging from "almost every day" (1) to "rarely or never" (5). A total score was computed. In accordance with the HBSC methodology protocol, two independent indexes were created: (1) somatic complaints (abdominal pain, headache, back pain, dizziness); and (2) psychological complaints (nervousness, depression, bad mood/irritability, difficulty sleeping). In both indexes, lower scores represent more psychosomatic complaints [27,28].

Data Analysis
Demographic characteristics, well-being, and attitudes toward remote learning are described using descriptive statistics (n, percentage, and median). This analysis was conducted for the entire group based on the sociodemographic characteristics.
None of the dependent variables met the normal distribution assumption. Therefore, a non-parametric statistic was used. More specifically, sociodemographic differences in prevalence of the three self-rated health categories (poor or fair, good, excellent) were compared using the chi-squared test. Differences in sociodemographic characteristics in the other two well-being measures (life satisfaction, and psychosomatic and psychological symptoms) were compared using the Mann-Whitney U test for dichotomized variables (gender) and the Kruskal-Wallis test for multilevel categorical variables (grade and change in socioeconomic status). For the psychosomatic and psychological symptoms variable, differences between females and males were presented using a box-plot figure. In the figure, the central box is representative of the values from the lower quartile to the upper quartile. The vertical line extends from the minimum value to the maximum value, while excluding the outside values (as they are displayed as separate points). An outside value was defined as either: (1) a value lower than the lower quartile minus 1.5 times the interquartile range, or (2) a value higher than the upper quartile plus 1.5 times the interquartile range. The middle line is representative of the median.
Associations between participants' attitudes toward the remote learning experience and sociodemographic characteristics and well-being measures were examined using Spearman's rank correlations. Subsequently, four separate multiple regression procedures to analyze factors predicting each of the well-being measures were conducted. All independent variables were checked for multicollinearity by using the variance of inflation factor (variance of inflation factor > 10). For continuous variables, only variables that were statistically significantly correlated with the well-being measures were entered into the model. Similarly, for categorical variables, only variables that statistically significantly differed between participants with different well-being measures were entered into the model.
For all statistical analyses, SPSS Statistics for Windows, version 23 (SPSS Inc., Chicago, IL, USA), was used; the level of significance was set to p < 0.05 (2-tailed).

Ethical Considerations
The study protocol was approved by the Ethics Committee of Ariel University, confirmation number: RO10,203; 21 May 2021. A preliminary letter regarding the survey was sent to the parents of the students. They were asked to confirm their children's participation. On the day of the survey, it was made clear to the students that the questionnaire was anonymous, and that their names should not be written on it.

Sociodemographic Characteristics of the Study Participants
The total number of participants was 1019 students in the 5th (n = 121, 11.9% of the sample) to 11th and 12th grades (n = 177, 17.3% of the sample). Most study participants were female (n = 544, 53.3% of the sample) and Jewish (n = 795, 78.0% of the sample; Table 1).
Participants' well-being measures according to sociodemographic characteristics. Regarding self-rated health, statistically significant differences were found in all sociodemographic characteristics ( Table 2). More specifically, among males, the percentage of participants reporting "poor and fair" overall health was greater than in females (52.31% of males vs. 42.09% of females; Chi-square = 10.61; p = 0.01). Compared to the 5th grade, the percentages of participants with "good" health in the other grades were statistically Children 2022, 9, 1346 6 of 13 significantly lower (chi-squared = 9.69, p = 0.001). The percentage of participants with "excellent" health belonging to the "better/much better" and "the same" socioeconomic status (51.90 and 49.09%, respectively) was statistically significantly greater than the percentage observed in "the same" socioeconomic status group (35.12%; chi-squared = 12.08; p = 0.0005).

Association between Participants' Remote Learning Experience and Sociodemographic Characteristics
In all remote learning items, statically significant correlations were found with grade (i.e., the higher the grade, the more negative the attitudes toward remote learning, except for connectedness with peers; Table 4).

Association between Participants' Remote Learning Experience and Well-Being Measures
In all remote learning items, statistically significant correlations were found with well-being measures, i.e., better well-being was associated with better remote learning experience (r ranges from 0.06 to 0.28: p = 0.0001; Table 5).

Prediction of Well-Being
According to the multiple regression analysis conducted, both sociodemographic characteristics and remote learning experience predicted all the well-being outcomes. More specifically, in the sociodemographic variables, being a female and reporting an adverse change in socioeconomic status predicted less favorable well-being outcomes in three indexes. In the remote learning variables, greater involvement in lessons and better connectedness with peers predicted better well-being in all four indexes assessed. Lower connectedness to the pedagogical team predicted lower well-being in three indexes. The four models' adjusted R 2 ranged from 0.07 (psychosomatic symptoms) to 0.14 (psychological symptoms and life satisfaction) with p < 0.0001. For additional information, refer to Table 6. Notes: In all well-being variables, except for "self-rated health", a higher score represents better well-being.

Discussion
The COVID-19 pandemic is revolutionizing education globally, and not only in Israel [7][8][9]. Schools have been forced to use technology and media to continue their educational activities [10].
This study was the first to examine remote learning experiences during COVID-19 among a sample of Israeli students. We found that over 50% of the sample disliked remote learning, and more than a third reported that they did not actively participate in the remote lessons.
Male students had slightly more favorable attitudes toward remote learning, with higher percentages reporting that they "like very much" remote learning compared to female students. Female students tended to participate with their camera off compared to male students. These gender differences are consistent with other studies that have reported that males have overall more favorable attitudes toward technology when compared to women [29,30]. However, a lower prevalence of females participating with the camera off may not be directly related to negative attitudes toward remote learning. It may be related to other factors, such as, for example, possible concerns regarding their appearance on camera. Therefore, gender differences in camera usage should be further explored.
Surprisingly, we found that children in lower grades had more favorable attitudes toward remote learning compared to older students. This finding somewhat contradicts other studies reporting greater difficulties with remote learning among younger children [14,31].
However, such studies examined how parents perceived the remote learning of their children [10,32], whereas our study investigated the children's perceptions and attitudes. In addition, children attending elementary schools in Israel had overall more in-person learning opportunities compared to those in higher grades, who continued to learn remotely when the younger children had already returned to school [26], which may explain why their overall remote learning experience was more negative.
Our second objective was to examine factors predicting adolescents' well-being. The present study found that overall, remote learning experience predicted well-being. More specifically, a higher level of involvement in remote learning and greater connectedness to school pedagogical teams and friends predicted greater life satisfaction, better overall health, and fewer psychosomatic symptoms. Peer support and teacher support have been often linked to better health outcomes among adolescents [33,34]. It is possible that active participation in lessons also increased the level of connectedness with both the school and schoolmates, which, in turn, was associated with the level of support from peers and teachers.
While school grade, a change in SES, and feelings toward remote learning were associated with the well-being measures, they did not emerge as significant predictors. Younger students are generally more protected by their parents, which could be the reason why their answers regarding the SES change questions were neutral or more positive compared to those given by the older students. A decline in SES could reflect job or business loss experienced by one or both parents. Recent studies conducted during the pandemic have found that parental stress associated with SES worries could affect the child's well-being [35].
It is also important to note the gender differences in well-being. Apart from selfrated health, in which female students reported better overall health, they scored significantly lower on all other well-being measures (life satisfaction, and psychological and somatic symptoms) compared to male students. This finding is consistent with other studies showing that females consistently tend to report more mental health problems than males [27,[36][37][38] The males' poorer self-rated health could reflect the specific conditions of the pandemic in which social distancing restrictions have lessened their physical activity levels and increased their sedentary behavior, all which are associated with poorer health [39][40][41][42].

Strengths, Limitations, and Directions for Future Study
Our study had several strengths. This was the first study in Israel, with findings that can be used to generate suggestions within the field of education and online well-being. It sheds light on students' attitudes and emotional reactions to remote learning during a time of crisis. The present study also has a few limitations. First, the study used a cross-sectional design, which does not allow inferences on causal relationships. A future longitudinal design may be able to better reflect the direction of the associations established in the present study. Second, the study variables were all self-report measures, which may be subject to common method biases [43]. In addition, due to the pandemic, some of the data were collected while the children were attending remote zoom lessons, without the active presence of the teacher or research assistant. This may have impaired participation rates and skewed our results toward more positive outcomes. Another limitation was that we did not include additional variables in the study (e.g., nutrition levels, accommodation), which could have enriched the findings, specifically for low SES participants.

Conclusions
The present study demonstrated that negative attitudes and experiences with remote learning, along with decreasing connectedness with the school framework and classmates, may have severe consequences for adolescents' well-being. The results of this study highlight the need for a better education policy, and interventions that emphasize student connectedness and engagement with the school educational teams, as well as with peer students in times of crises, when there is an immediate need for a transition to remote learning. It is imperative that coping strategies are taught to children in order to help them overcome any issues or barriers associated with the COVID-19 pandemic. In addition, special attention should be given to girls, to students in higher grades, and to those whose families suffered additional trauma (such as loss of income).

Data Availability Statement:
The author is the sole person who conceived the study, conducted the research, and wrote the article. The data presented in this study are available on request from the author.