Online Behaviours during the COVID-19 Pandemic and Their Associations with Psychological Factors: An International Exploratory Study

Round 1

Reviewer 1 Report

The submitted manuscript shows the results of a study that aimed to assess cross-cultural differences in time spent online. Overall, the manuscript is well organized. Nevertheless, it needs some important corrections:

First of all, at the end of the introduction, please clearly state the purpose of the study and the hypotheses (or research questions).

In the methodology, please include the eligibility criteria for study participants. Moreover, nowhere is it explained why these countries were chosen. Why not include the U.S., where the tradition of Internet use is the longest? Why were nearby countries (Spain and Portugal) chosen, followed by distant ones such as Japan or Brazil? I don't see the consistency in the sample selection. I understand that it is probably a criterion of availability or returnability of questionnaires, but all this should be described very clearly.

Did the authors use tools to protect the questionnaire from the phenomenon of fake/bot responders? (e.g. did they check the time of filling in the questionnaires, used codes, etc.).

I wish you all the best!

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

1.      The title is not reflective of the variables nor study. Please include more of what was studied in the title. Please re-develop the title.

2.      I encourage the authors to consider decreasing the number of variables within the paper. The current number of variables is overwhelming and difficult to understand and keep track of when considering the relationships. Please ask yourselves: what are the key variables of interest? Maybe some of the other variables could be used to compose another paper.

3.      The introduction needs to be focused on the variables. PUI is a lengthy definition (line 71-77. Can you select the variables you want to study from this definition versus all of them? The literature presented is international to match the international base of the study.

4.      Participants appear to be described. Unfortunately, this was placed before the procedure. I encourage information about the procedure first so readers understand how the study was designed and participants selected.

5.      The instruments are described including reliability.

6.      The procedure needs more information. How were the countries selected and why? How were participants selected? What were the inclusion and exclusion criteria?  How long did it take to complete the instruments? What did you do with incomplete data? For a study from so many countries, the sample size is small. Did you explore statistically the adequate number of participants to recruit?

7.      The statistical analysis section is well written.

8.      Because of the number of variables examined, the results are difficult to follow. It is like you are “fishing”. Table 1 is not a comparison table (line 248) but descriptive information. Line 251 – I do not understand difference within countries. Did you mean among countries?

9.      The discussion section contains a lot of repeat information from the results. Again, please try to limit the variables in your study and focus on key ones in the discussion. Make curtain that you are comparing your findings to the literature versus just repeating it.

10.  The limitations present interesting variables. For the number of countries and variables, the sample size was not large.

11.  Conclusions are shared but again the study had an overwhelming number of variables.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

I have gone through the paper entitled “Online Behaviours During the COVID-19 Pandemic: A Cross-national investigation” with interest. It discusses important issues, especially in the context of the rapidly increasing digital technologization of every aspect of human life. The strong advantage of the manuscript is the relatively large sample and the possibility to show cross-cultural differences in online activities. However, several concerns were raised during the review process. 

Most notably, the study was cross-sectional, the authors did not present the study hypothesis, and the results added little to the existing knowledge.

Next, the authors did not explain why they examined the psychological factors such as anxiety (although some information was mentioned in the study) and self-compassion (this personal characteristic was not mentioned in the introduction, and why they examined cross-cultural differences?

In addition, the authors did not explain why they have chosen a sample consistent with adults (M=33, SD=11), although adolescents and emerging adulthood are mostly affected by PUI. 

The introduction lacks a theoretical explanation of the main studied variable e.g. online activities, the authors only mentioned PUI (without its clear definition) and social media overuse. The paragraph about PUI and addiction, and body dissatisfaction is only briefly relevant to the subject. I can not agree with the statement that currently most studies only report the increase in general online addiction level, as most of them follow the current lines of study-specific online activities, and the use of the PUI questionnaire is rather rare. Thus it is questionable for me to state that there is an existing “gap” in the literature in this field.

The introduction ends with 3 core study questions. In the first one, the authors used the term “this sample” – this is incorrect because it refers to the reader to some indefinite sample e.g. clinical group, and since the above paragraph was about the PUI relation with addiction and body dissatisfaction it suggests that this is the studied clinical sample – which is not true.

The second study question refers to “time spent online”, however, there are no references to cross-cultural differences in this field to justify such a study question. 

The third question construction suggests that appearance anxiety is the opposite characteristic of self-compassion – which is not true.

Next, the authors enigmatically stated that they will also be tested “how these associations differed across demographic, psychological, and behavioural dimensions, including gender, age, self-reported mental disorder, and duration of internet use during the start of the COVID-19 pandemic.”

First, the word dimensions suggest that all of these three categories build one higher-order factor, next

again there is no explanation for including these demographic characteristics and excluding others.

Mostly, the self-reported mental disorder should be more precisely described – what question(s) was(were) presented to the participants, especially as in the result section the authors added this aspect as mental illness e.g. anxiety, depression, anxiety. If we measure such serious mental health problems with only 1 question the results are very questionable.

In addition, what kind of behavioral dimensions do you mean, because there is nothing about it in this sentence. 

In the method section, the authors mentioned additional questions in the survey to assess the presence of mental illness “including and level of physical activity, as well as the IPED consumption.

The sentence suggests that “physical activity” is a kind of mental illness (moreover in the next parts of the manuscript the authors used the term “physical activity time”, or “activity level” or just “generally active” or “any type of sporting activity” or “any type of live sporting activity”, or a specific kind of physical activity “engaged in at least one sports activity “, or the number of physical activities, so what was exactly measured, because all of these terms refers to different findings?). Moreover, as was mentioned above this should be precisely described, also with the information about the reliability of these questions as they were included in the statistical analysis as quantitative measures.

If the person was supposed only to answer YES/NO on the question about depression, addiction, anxiety, etc. the variable gained from the questions was categorical, so why do you use Pearson’s correlation (similar for gender) 

In the result section, the authors describe the results indicated by the label “mental illness” in a new way e.g. “experiencing” depression/ anxiety – this label refers not to a disorder but also to feeling at the moment depressed (in not a clinical manner) which is not the same as depression, etc. Thus I have serious methodological doubts about this aspect of the project. 

The title of table 1 suggests the results for statistical comparison analysis, however only descriptive statistics were shown – this must be added. Also, the authors should at least briefly describe that the rest characteristics were similar across countries.

Moreover, there is an occur label in Table 1,2, namely “Addiction+ smoking”, this suggests that smoking is not a kind of substance use addiction. Which is not true.

The sentence in lines 259-260 is hard to understand and needs language corrections. 

Why you did not report F for the morel in table 3, and the adjusted R2?

Where is the statistic that would show how the associations between online activities and appearance anxiety/self-compassion differed across the country? In multivariable modeling in line 265 you stated that also country was be included in the model, however, it was not. 

Moreover, the question partially suggests different statistical analyses, you have only shown the multiple regression models for each online behavior with included additional demographic, psychological, and behavioral characteristics.  

Discussion

The authors stated in lines 308-9 that “Individuals reported that the COVID-19 pandemic led to them spending more time online” – however, nothing was mentioned about it, the authors only measured time spent online and there was no information about the time spent online before the pandemic. This conclusion is incorrect.

Next, the authors reported cross-cultural differences in the “degree of general internet browsing behavior” observed in the UK, Italy, and Portugal”, however in the result section only duration of the Internet use significantly differed across countries, and the authors did not mention about any other significant difference. 

The explanation of higher Internet time duration 

by linking it with “experiencing highest COVID-19 burden” is questionable.

Especially as these references did not confirm fully such a conclusion e.g. the first reference is based on data collected from France, India, Italy, the UK, and the USA with the highest burden in lower and upper-middle-income countries.

In the next paragraph 

The authors suggest the causal effect by using the expression “The results showed that appearance anxiety, self-compassion, IPEDs use, and engagement in any sporting activities might generally contribute to various forms of online activities.” – this is incorrect as the cross-sectional method does not allow to formulate causal conclusions. 

The next sentence suggests the comparison analysis that was not performed: “Specifically, appearance anxiety was found to be relatively high in the following 318 online activities: general surfing, online shopping, online gambling, social networking, pornography use, digital hoarding, and cyberchondria” – the correlation/regression analysis does not provide information about the higher/lower level in different groups/variables.

Similar comment for the sentence in lines 331-2 “In this study individuals using IPEDs had increased online shopping, pornography, 331 streaming, and cyberchondria: - such analysis was not performed so the conclusion is not relevant to the findings

-The sentences in lines 323-5 are only authors' speculations and need justification by proper references

-In lines 337-8 the authors stated that “The associations found are new, and no other studies explored general physical activity and its association with online behaviours.” This is not true, for review see Goodyear et al. (2021)

- the association between self-compassion and online behaviors should be more deeply explained

-in line 365 the authors stated ” our sample consisted of relatively healthy volunteers (…)” however over 50% of the participants stated that expressed depression or anxiety (although it was measured only by 1 question thus may be biased the word “healthy” is not confirmed by any objective condition) – based on what the authors make such conclusion 

- small editorial errors should be corrected e.g. lack of bracket “)” - line 80; “Person correlation analyses” (line 231, 294), unnecessarily bold words in line 265 and lines278-9; and changing the font or leading in this section; broadened word Cyberchondria in table 2; the different graphics in table 1 and 2, and 3.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 4 Report

Comments and Suggestions for Authors

 Thank you for the opportunity of reviewing this interesting article. This study aimed to  characterize the specific online behaviours among this sample;   explore the country-based differences in time spent online, and identify any potential associations between specific online behaviours and the self-directed negative feelings  of appearance anxiety vs. the positive feelings of self-compassion. Autors also examined how   these associations differed across demographic, psychological, and behavioural dimensions, including gender, age, self-reported mental disorder, and duration of internet use   during the start of the COVID-19 pandemic. This approach seems to be extremely relevant and promising.

Comments and Suggestions for Authors:

·      The authors' remark raises serious doubts: «To date, COVID-19 studies have focused mainly on the increase of general internet   use rather than on defining specific online behaviours and how these might differ across   countries». Many studies of this kind have already been published. For example in Frontier in psychology etc.

·      There are some doubts about the data sampling. Represented heterogeneous data:(Brazil (n = 711; 32.0%)  and  Hungary (n = 47; 2.1 %). Please explain the quantitative characteristics of the data.

·      Please provide characteristics of respondents for each country separately.

·      Please describe the association extraction algorithm in more detail. On what theoretical basis did you base yourself?

·      The authors wrote: «In addition to basic information on age and gender, additional questions were also included in the survey. The authors developed these to assess  the presence of mental illness including and level of physical activity, as well as the IPED consumption.» Please indicate in which repository you can see them? Or place in the Appendix to the article

·      Please edit the title of the article to better match the content of your article.

·      The process of discussing the results can be extended by applying the results and extrapolating them to other similar studies.

·      Please describe in detail how your study fits for aims and scope of International Journal of Environmental Research and Public Health (IJERPH).

·      For theoretical framework and bibliography additional current references should be included to new research 2021-2022.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

The authors made sufficient revision of the manuscript thus in my opinion it may be published,