The background statistics of the respondents are presented in Table 2. There are relatively few respondents from the youngest and oldest age groups, with a majority representing 30–39-year age group. The gender balance varies among the agencies. Generally, more females than males answered the questionnaire, except in the Public Employment Service, where the majority of respondents are men. Since the distribution of the questionnaire was conducted through a contact person at each agency, there is no information regarding the age and gender of the non-respondents who received the questionnaire and thus no information on how representative this sample is for office workers at the agencies. However, in comparing the gender distribution of our sample with that of each agency based on their annual reports (presented in Table 1), there is a general congruence. (There is, however, an overrepresentation of women in the sample of the Transport Administration and an overrepresentation of men in the sample of the Public Employment Service.) In the case of the Transport Administration, the higher share of women in our survey is likely due to women being overrepresented among the office workers (versus other roles) in that agency. The difference in gender share for the Public Employment Service is likely due to the gender balance at the division that received the questionnaire.
Commuting – telework
In Table 3, the responses to the question regarding the pandemic’s effects on the general work situation are presented. Based on the responses, it is clear that the pandemic has greatly affected the work situation in Sweden though the country (compared to many other countries) appealed to voluntary behavioural adjustments. Only a few respondents state that they are unaffected.
The responses regarding travel to work indicate that, though liberal use of restrictions, the appeals expressed had a major impact on whether the studied office workers commuted to work. Figure 1 shows the average number of days per week the respondents commuted to work before and during the time when the study was conducted. The average number of commuting days per week dropped from 4.4 to 0.5 for women and from 4.5 to 0.8 for men. Further analyses show that of those commuting 5 days a week prior to the pandemic, 66% did not commute at all, while 16% reduced their commuting to only 1–2 days a week. These results can be interpreted as a massive increase in the number of persons working from home or another location outside of the workplace. At the same time and for both genders, there is a statistically significant difference between the agencies in the number of commuting days during the pandemic. As Fig. 1 indicates, this variation corresponds well to the variation in number of commuting days prior to the pandemic, i.e. the commuting pattern seems to have remained though on a very low level during the pandemic.
The variation in number of commuting days before the COVID-19 pandemic may be explained by the share of respondents working part-time. No question relating to working hours was included in the questionnaire, but the co-variation between number of commuting days and the share of women and respondents in ages of 20–39 years, see Table 2, indicates that this could be the case as these factors correlate with the propensity of working part time, .
In the survey, the respondents were asked to rate (scale 1–100) how well they thought teleworking works during the COVID-19 pandemic. In Tables 4, 5, 6, the responses are separated by organisation, gender and age, and the number of days per week commuting before the pandemic. Teleworking seems work reasonably well, with an average rating of 74. Respondents from the Public Employment Service and the Environmental Protection Agency were the most pleased with how teleworking worked. These agencies had also the largest reduction in the number of commuting days per week compared to the situation before the COVID-19 outbreak (and, correspondingly, highest increase in the level of telework).
These results also reveal that the respondents who previously commuted 2–3 days a week thought that working from home worked most well. This may be because those who teleworked a few days a week before COVID-19 likely already have the technical and practical setup ready at home, but it could also be an effect of this survey’s inclusion of part-time workers, whose number of commuting days did not change extensively. The results also show that the more days the respondent commuted prior to the pandemic, the lower he or she ranked how well teleworking worked during COVID19.
Generally, women in the data set were more pleased with teleworking than men. Further, how well teleworking is considered to work appears to increase with age. This may be due to factors connected to the situation at work but also at home, e.g. younger employees more often need support and guidance from more senior colleagues and engage more often socially with their co-workers, both of which are more difficult to do when working from home. But younger people are more likely live in smaller homes and to have children (and disturbances) at home. Another possible explanation relates to COVID-19 in that persons in higher age groups are more concerned about contracting COVID-19 since they risk more severe health outcomes.
The respondents were also asked to reflect on the adjustment to working from home. Many expressed that there was not much of a difference: e.g. ‘I work from home but do overall the same job, albeit in a different way’ and ‘I have a good workspace with all the equipment I need, so it works well’. On the other hand, many others commented on the need for technical and physical support in order to cope with the new situation: e.g. ‘I had to furnish part of a room in the apartment as a place to work’ and ‘Among other things, I needed to buy a height-adjustable desk in order to work at home without getting (serious) trouble with my neck’. Some even expressed concerns regarding the quality of the work situation and the work carried out: e.g. ‘More work at home with poorer technical equipment (especially number of screens) and more disruptive moments (children)’ and ‘It is difficult to delegate and collaborate effectively via Skype. We are recommended to work at home as much of the time as possible. Assignments and projects require physical presence to be done well.’ Though most of the respondents worked at home all week, the results in Fig. 1 indicate that some respondents spent 1–2 days a week at the office, and their worries were captured as well: e.g. ‘We sit in an activity-based office. You feel incredibly worried when you don’t have your own room.’
Business trips and virtual collaboration
The result in Table 7 reveals that, on average, more than 75% of the respondents made business trips often or sometimes before the COVID-19 pandemic, and among those that made business trips before, this dropped to almost 2% during the pandemic. The business travel thus almost disappeared. The lack of business travel seems, however, not to have affected the working situation gravely. On the question how meetings and other events that previously would have required travel were being carried out, presented in Table 8, an average of 3% of respondents answered that all or almost all meetings/errands had been cancelled and that 90% had managed to complete most of their business tasks using virtual collaboration. The result thus indicates that the business of the five public agencies was maintained. The Public Employment Agency was least affected, but that agency also had the lowest level of business trips for the beginning.
In Table 9, the average rating regarding how well the virtual collaboration functioned is somewhat lower than the rating of telework, at 73.17 compared to 74.26. Analyses of the material also show that there is a correlation, though at a low degree, between the rated function of teleworking and that of virtual collaboration (Pearson correlation coefficient 0.60, p = 0.000). According to Table 10, younger respondents ranked how well virtual collaboration worked higher than older respondents, and women higher than men. The differences are not statistically significant though.
In the same way as for telework, the respondents were asked to reflect on virtual collaboration. Many comments from respondents were connected to virtual collaboration and the social dimension, e.g. ‘I have a fear of getting caught up in the mobile working methods, resulting in less social contact’ and ‘This adds greater importance and value to the social dimension once you meet physically in the office’. On the other hand, some respondents viewed the increased use of virtual collaboration as nothing extraordinary, e.g. ‘We were already doing a lot of virtual meetings, so the change won’t be too big, especially in international cooperation’. This type of comment corresponds well with the responses given to the question dealing with thoughts about virtual collaboration in the future (see Table 11). More than 90% of respondents believe themselves and colleagues will become much better at collaborating virtually; however, only 5% believe in a massive change in how they and their colleagues will work in the future.