The Impact of Technostress on Teacher Educators’ Work–Family Conflict and Life Satisfaction While Working Remotely during COVID-19 in Pakistan
Anna Dabrowski
Joseph M. Kush
Round 1
Reviewer 1 Report
This study provides a much needed consideration on the impact of remote learning in education institutions. The complexities of technology, family demands, and overall life satisfaction are articulated and discussed in this paper, with acknowledgement of the additional challenges facing female teacher educators. The research was well considered and contextualised in terms of the pandemic and the nation of Pakistan, and relevant research on the topic was well linked. This is a timely and relevant piece of work for teacher educators and tertiary education providers, not only during the pandemic, but beyond.
Author Response
Thank you for the opportunity to revise our manuscript entitled: “The Impact of Technostress on Teacher Educators’ Work-Family Conflict and Life Satisfaction While Working Remotely During COVID-19 in Pakistan” submitted for editorial review and possible publication in the Educational Sciences. The reviews were very helpful, combining some valued encouragement about the value of our work with constructive suggestions on how to enhance the paper. We list in point form below the modifications made in response to the suggestions offered.
We believe the paper is significantly enhanced by the revisions, and look forward to the response to the revised paper.
Reviewer 2
1) There are a few instances in which ‘COVID-19’ is written as ‘Covid-19’; please be consistent with the capitalization throughout.”
We have corrected the COVID-19 throughout manuscript. It is consistent now.
2) On page 13 lines 137-139, some readers may find it confusing to disentangle a positive relationship among two negative attributes. Consider specifically spelling out this relationship for the reader, revising this to something along the lines of, “Prior research has also found that those with a greater degree of work-family conflict had weakened individual, physical, and psychological well-being and contentment with life [25].”
Corrected and mentioned
Prior research has also found that those with a greater degree of work-family conflict had weakened individual, physical, and psychological well-being and contentment with life [25].
3) In Table 1, is the column labeled ‘F’ supposed to be labeled ‘N’ for sample size?
Done as suggested
4) How was the age categorization decided on? Is there prior literature demonstrating that above or below age 30 represents a meaningful cutpoint for how individuals relate to technostress? This decision also has implications for Table 5 (e.g., perhaps a cutpoint of age 40 results in no significant differences, while a cutpoint of age 25 results in even larger differences). Overall, this decision should be described in more detail.
Pakistani universities follow some criteria to recruit faculty, less than thirty years for the entry position of lecturer and above thirty for the senior academic positions. Age categories were decided according to this age criteria for recruiting faculty in the Pakistani Universities.
5) Consider providing three correlations in Table 2:
Cor(TS, WFCS) = .381
Cor(LS, WFCS) = .449
Cor(TS, LS) = .xxx?
Done and reported in text description of the data analysis and result section
6) Please ensure abbreviations are consistent throughout the paper (e.g., the Work-Family Conflict Scale is referred to as ‘WFCS’, ‘WFC’, and ‘WAFCS’).
We have corrected this throughout the manuscript as WFCS.
7) The reliability estimates on page 5 of TSS (lines 190-191), WFCS (line 205) and LS (line 211) are somewhat low and reflective of only moderate reliability. It is relatively accepted psychometric practice that coefficient alpha values less than 0.5 are indicative of poor internal consistency, values between 0.5 and 0.75 are indicative of moderate reliability, values between 0.75 and 0.9 are indicative of good reliability, and values greater than 0.9 are indicative of excellent reliability. This should be addressed as a limitation.
As suggested by the reviewer low Cronbach value scale is addressed in the limitation.
Major comments
8) The correlations among the dependent variables presented in Table 2 could be moved to in-text. I would also discourage the use of significance testing for correlations (as this can be driven in part by sample size), and would simply describe the meaningfulness of the relationship, such as referring to Cohen’s effect size interpretations, as it seems is already done. Relatedly, the results presented in Table 3 and discussed on page 6 lines 234-237 do not describe mediation effects. See Edwards & Konold, 2020, e.g., or your citation [22] for a more detailed overview of mediating variables. These regression results go in tandem to the correlation results (in fact, squaring the Cor(TS, WFCS) = .381 maps directly onto the linear regression R-square estimate of .145). Overall, these results could be presented more concisely.
This is a useful piece of feedback. The table 2 and 3 were deleted and moved into the text.
9) The results presented in Table 4 and discussed on page 6 lines 242-249 accurately demonstrate gender differences on each of the three outcome variables. However, it is possible these differences would disappear in the presence of other variables. To overcome this, a set of regression models, in which each of the three instruments would independently be considered as outcome variable, with all of the demographic variables simultaneously included as predictor variables, would allow for a more nuanced understanding of how various demographic variables are related to each of the outcomes of interest (e.g., TS = b0 + b1*Gender + b2*Age + b3*Qualification + b4*University_type). This would allow for interpretations such as, (as an example) ‘After controlling for the effects of age, qualification level, and university type, gender was a significant predictor of TS (b = x.xx, p = .001), indicating that females had significantly higher levels of technostress than males.’
Thank you for these suggestions. The regression was run and presented in table 3. The analysis was run among the demographic variables and explained in text under data analyses and results section.
10) The Tukey post-hoc results presented in Table 6 should be elaborated on. For example, the LS ANOVA model results in an omnibus significance (F(2, 289) = 22.812, p < .001), indicating that at least one group mean differs from the others. I am assuming the PhD group (M = 23.08, SD = 4.74) is significantly greater than Masters group (M = 17.25, SD = 7.76), but perhaps the PhD group (M = 23.08, SD = 4.74) is not significantly different than M.Phil group (M = 22.69, SD = 6.81). Unfortunately, the in-text description only provides a mean and standard deviation of the two groups; an additional p-value for each of these comparisons is needed to understand which specific groups differ from others.
Changes incorporated in the text according to the new table.
11) The results presented in Tables 5 and 6 could be replaced with the aforementioned linear regression analyses described above.
It has been replaced with regression analysis.
Reviewer 2 Report
Comments to the Authors:
Thank you for the opportunity to review this manuscript entitled, “The impact of technostress on teacher educators’ work-family conflict and life satisfaction while working remotely during COVID-19 in Pakistan”. This research provides a baseline for developing countries to understand how teachers’ professional and personal lives may be impacted by the pivot to remote-work, and has clear and important implications for improving the preparation of teachers’ use of technology.
In general, I found the results to be somewhat limiting, which I elaborate on below. That being said, I believe the authors are more than capable of adequately addressing these issues. Overall, I believe this paper shows great promise and would be an excellent fit in Education Sciences.
Minor comments (in no particular order)
1) There are a few instances in which ‘COVID-19’ is written as ‘Covid-19’; please be consistent with the capitalization throughout.
2) On page 13 lines 137-139, some readers may find it confusing to disentangle a positive relationship among two negative attributes. Consider specifically spelling out this relationship for the reader, revising this to something along the lines of, “Prior research has also found that those with a greater degree of work-family conflict had weakened individual, physical, and psychological well-being and contentment with life [25].”
3) In Table 1, is the column labeled ‘F’ supposed to be labeled ‘N’ for sample size?
4) How was the age categorization decided on? Is there prior literature demonstrating that above or below age 30 represents a meaningful cutpoint for how individuals relate to technostress? This decision also has implications for Table 5 (e.g., perhaps a cutpoint of age 40 results in no significant differences, while a cutpoint of age 25 results in even larger differences). Overall, this decision should be described in more detail.
5) Consider providing three correlations in Table 2:
Cor(TS, WFCS) = .381
Cor(LS, WFCS) = .449
Cor(TS, LS) = .xxx?
6) Please ensure abbreviations are consistent throughout the paper (e.g., the Work-Family Conflict Scale is referred to as ‘WFCS’, ‘WFC’, and ‘WAFCS’).
7) The reliability estimates on page 5 of TSS (lines 190-191), WFCS (line 205) and LS (line 211) are somewhat low and reflective of only moderate reliability. It is relatively accepted psychometric practice that coefficient alpha values less than 0.5 are indicative of poor internal consistency, values between 0.5 and 0.75 are indicative of moderate reliability, values between 0.75 and 0.9 are indicative of good reliability, and values greater than 0.9 are indicative of excellent reliability. This should be addressed as a limitation.
Major comments (in no particular order)
Overall, I believe the study would be greatly improved with a more concentrated set of analyses, such as estimating a linear regression model for each of the three instruments as outcome / dependent variables, with all of the demographic variables included as predictor / independent variables. Ultimately, this would allow for more nuanced interpretations of findings. This is elaborated on in the comments below.
8) The correlations among the dependent variables presented in Table 2 could be moved to in-text. I would also discourage the use of significance testing for correlations (as this can be driven in part by sample size), and would simply describe the meaningfulness of the relationship, such as referring to Cohen’s effect size interpretations, as it seems is already done. Relatedly, the results presented in Table 3 and discussed on page 6 lines 234-237 do not describe mediation effects. See Edwards & Konold, 2020, e.g., or your citation [22] for a more detailed overview of mediating variables. These regression results go in tandem to the correlation results (in fact, squaring the Cor(TS, WFCS) = .381 maps directly onto the linear regression R-square estimate of .145). Overall, these results could be presented more concisely.
9) The results presented in Table 4 and discussed on page 6 lines 242-249 accurately demonstrate gender differences on each of the three outcome variables. However, it is possible these differences would disappear in the presence of other variables. To overcome this, a set of regression models, in which each of the three instruments would independently be considered as outcome variable, with all of the demographic variables simultaneously included as predictor variables, would allow for a more nuanced understanding of how various demographic variables are related to each of the outcomes of interest (e.g., TS = b0 + b1*Gender + b2*Age + b3*Qualification + b4*University_type). This would allow for interpretations such as, (as an example) ‘After controlling for the effects of age, qualification level, and university type, gender was a significant predictor of TS (b = x.xx, p = .001), indicating that females had significantly higher levels of technostress than males.’
10) The Tukey post-hoc results presented in Table 6 should be elaborated on. For example, the LS ANOVA model results in an omnibus significance (F(2, 289) = 22.812, p < .001), indicating that at least one group mean differs from the others. I am assuming the PhD group (M = 23.08, SD = 4.74) is significantly greater than Masters group (M = 17.25, SD = 7.76), but perhaps the PhD group (M = 23.08, SD = 4.74) is not significantly different than M.Phil group (M = 22.69, SD = 6.81). Unfortunately, the in-text description only provides a mean and standard deviation of the two groups; an additional p-value for each of these comparisons is needed to understand which specific groups differ from others.
11) The results presented in Tables 5 and 6 could be replaced with the aforementioned linear regression analyses described above.
Author Response
August 30th , 2022
Dr. Jerome Pan
Assistant Editor
Educational Sciences
Dear Dr. Jerome,
Thank you for the opportunity to revise our manuscript entitled: “The Impact of Technostress on Teacher Educators’ Work-Family Conflict and Life Satisfaction While Working Remotely During COVID-19 in Pakistan” submitted for editorial review and possible publication in the Educational Sciences. The reviews were very helpful, combining some valued encouragement about the value of our work with constructive suggestions on how to enhance the paper. We list in point form below the modifications made in response to the suggestions offered.
We believe the paper is significantly enhanced by the revisions, and look forward to the response to the revised paper.
Reviewer 2
1) There are a few instances in which ‘COVID-19’ is written as ‘Covid-19’; please be consistent with the capitalization throughout.”
We have corrected the COVID-19 throughout manuscript. It is consistent now.
2) On page 13 lines 137-139, some readers may find it confusing to disentangle a positive relationship among two negative attributes. Consider specifically spelling out this relationship for the reader, revising this to something along the lines of, “Prior research has also found that those with a greater degree of work-family conflict had weakened individual, physical, and psychological well-being and contentment with life [25].”
Corrected and mentioned
Prior research has also found that those with a greater degree of work-family conflict had weakened individual, physical, and psychological well-being and contentment with life [25].
3) In Table 1, is the column labeled ‘F’ supposed to be labeled ‘N’ for sample size?
Done as suggested
4) How was the age categorization decided on? Is there prior literature demonstrating that above or below age 30 represents a meaningful cutpoint for how individuals relate to technostress? This decision also has implications for Table 5 (e.g., perhaps a cutpoint of age 40 results in no significant differences, while a cutpoint of age 25 results in even larger differences). Overall, this decision should be described in more detail.
Pakistani universities follow some criteria to recruit faculty, less than thirty years for the entry position of lecturer and above thirty for the senior academic positions. Age categories were decided according to this age criteria for recruiting faculty in the Pakistani Universities.
5) Consider providing three correlations in Table 2:
Cor(TS, WFCS) = .381
Cor(LS, WFCS) = .449
Cor(TS, LS) = .xxx?
Done and reported in text description of the data analysis and result section
6) Please ensure abbreviations are consistent throughout the paper (e.g., the Work-Family Conflict Scale is referred to as ‘WFCS’, ‘WFC’, and ‘WAFCS’).
We have corrected this throughout the manuscript as WFCS.
7) The reliability estimates on page 5 of TSS (lines 190-191), WFCS (line 205) and LS (line 211) are somewhat low and reflective of only moderate reliability. It is relatively accepted psychometric practice that coefficient alpha values less than 0.5 are indicative of poor internal consistency, values between 0.5 and 0.75 are indicative of moderate reliability, values between 0.75 and 0.9 are indicative of good reliability, and values greater than 0.9 are indicative of excellent reliability. This should be addressed as a limitation.
As suggested by the reviewer low Cronbach value scale is addressed in the limitation.
Major comments
8) The correlations among the dependent variables presented in Table 2 could be moved to in-text. I would also discourage the use of significance testing for correlations (as this can be driven in part by sample size), and would simply describe the meaningfulness of the relationship, such as referring to Cohen’s effect size interpretations, as it seems is already done. Relatedly, the results presented in Table 3 and discussed on page 6 lines 234-237 do not describe mediation effects. See Edwards & Konold, 2020, e.g., or your citation [22] for a more detailed overview of mediating variables. These regression results go in tandem to the correlation results (in fact, squaring the Cor(TS, WFCS) = .381 maps directly onto the linear regression R-square estimate of .145). Overall, these results could be presented more concisely.
This is a useful piece of feedback. The table 2 and 3 were deleted and moved into the text.
9) The results presented in Table 4 and discussed on page 6 lines 242-249 accurately demonstrate gender differences on each of the three outcome variables. However, it is possible these differences would disappear in the presence of other variables. To overcome this, a set of regression models, in which each of the three instruments would independently be considered as outcome variable, with all of the demographic variables simultaneously included as predictor variables, would allow for a more nuanced understanding of how various demographic variables are related to each of the outcomes of interest (e.g., TS = b0 + b1*Gender + b2*Age + b3*Qualification + b4*University_type). This would allow for interpretations such as, (as an example) ‘After controlling for the effects of age, qualification level, and university type, gender was a significant predictor of TS (b = x.xx, p = .001), indicating that females had significantly higher levels of technostress than males.’
Thank you for these suggestions. The regression was run and presented in table 3. The analysis was run among the demographic variables and explained in text under data analyses and results section.
10) The Tukey post-hoc results presented in Table 6 should be elaborated on. For example, the LS ANOVA model results in an omnibus significance (F(2, 289) = 22.812, p < .001), indicating that at least one group mean differs from the others. I am assuming the PhD group (M = 23.08, SD = 4.74) is significantly greater than Masters group (M = 17.25, SD = 7.76), but perhaps the PhD group (M = 23.08, SD = 4.74) is not significantly different than M.Phil group (M = 22.69, SD = 6.81). Unfortunately, the in-text description only provides a mean and standard deviation of the two groups; an additional p-value for each of these comparisons is needed to understand which specific groups differ from others.
Changes incorporated in the text according to the new table.
11) The results presented in Tables 5 and 6 could be replaced with the aforementioned linear regression analyses described above.
It has been replaced with regression analysis.
Author Response File: 
Author Response.docx
Round 2
Reviewer 2 Report
Comments to the Authors:
The authors were responsive to our reviews. I believe the paper is now more streamlined with the analyses and results. Overall, I have no major hesitations about this paper, and believe it should be accepted for publication in Education Sciences.
