Who Are the Young Adults in Portugal? Daily Usage of Social Media and Mobile Phones, in a No-Kids and No-Independent Housing Context—Results from a Representative Online Survey
Round 1
Reviewer 1 Report
The text focus on a issue socially relevant and brings a contribution to the extant literature. Although, the use of "average young adult" does not seem the most appropriate. The text present a portrait of the young adults in the Portuguese context, and more than considering the "average", the data brings much more relevant information concerning several social issues.
I would advise a change in the title as well in the first sentence of the text that is not totally clear.
Author Response
Dear reviewer of Revision 1,
We want to thank you for the constructive comments. We have tried to consider them, revising the manuscript. Revisions have been highlighted in yellow in the main body of the article text and a new proof-reading was done to eliminate possible grammatical errors.
We have reconsidered the usage of the term “average” as suggested, including in the title of the article.
If you have any questions, please let us know.
Best regards.
Reviewer 2 Report
Methods: please describe the questionnaire, the questions so that if somebody wants to replicate your research, following the methodology you wrote to be possible. Table 1 is results, please move. Please provide a thorough description of the statistical analysis.
Results: all percentages should be followed by confidence intervals. For means: is the data normally distributed or better to address the median and quartiles?
Discussions: would be interesting to provide a comparison with the situation in other countries that are mentioned in the introduction.
Good english
Author Response
Dear reviewer of Revision 2,
We want to thank you for the constructive comments. We have tried to consider them, revising the manuscript. Revisions have been highlighted in yellow in the main body of the article text and a new proof-reading was done to eliminate possible grammatical errors.
As suggested, and in order to enable others to replicate this research in different environments, a questionnaire description was added, highlighting the statistical procedures and the characteristics of the sample and its distribution on this representative sample study (sample with a normal distribution regarding respondents age, with p-value=0,95 on Shapiro-Wilk test). Although means may be affected by extreme values, due to dealing with big samples, we understand means as important indicators of trends on data centrality, nonetheless recognizing the validity of using medians for the same purpose.
Regarding the confidence intervals for the percentages, we have decided not to present them in the article to not generate confusion to readers. That decision is based, particularly, on the fact that the statistical analysis of correlation with z-Tests and t-Tests already shows sufficient confidence in the values. In fact, our test examples reveal that, for a confidence level of 95%, every time a statistically significant difference occurs, it’s p-value<0,05 and also that such confidence interval does not include 0. Furthermore, if the confidence interval included a 0, such difference would not be statistically significant. Take for example the case of the correlation with statistically significant differences of Table 2 between “Single (A)” or “Married/Non-marital partnership (B)” and the “Heterosexual” option of Sexual Orientation. Such confidence interval is [-0,104; -0.019] (does not include 0) with p-value=0,008 (smaller than a p-value of 0,05). As an opposite example, we may look at the case without statistically significant differences of Table 3, between “Have kids (A)” or “Do not have kids (B)” and the “Heterosexual” option of Sexual Orientation. Such confidence interval is [-0,007; 0.091] (which includes 0 in its interval) with p-value=0,125 (which is higher than a p-value of 0,05, thus confidently [i.e., with a 95% level of confidence] rejecting this specific correlation).
Our study intended to take other countries as important theoretical context, although understanding those as not totally comparable to the context of Portugal, due to nuances of social, cultural, demographic, and economic nature. Therefore, we highlight the contribution of our study, being the first in Portugal with a representative sample of such age group of young adults (between 18 and 30 years old). Nonetheless, some of the patterns identified in this article concerning the Portuguese context, are enhanced when other contexts are taken into consideration, in case of cross-national trends and movements, like in the case of the Discussion’s segment between lines 558 and 562.
If you have any questions, please let us know.
Best regards.
