Navigation – Plan du site

AccueilNuméros3Commuting behaviour in times of t...

Commuting behaviour in times of the COVID-19 pandemic: a comparative first and second wave study from Latvia

Les déplacements en temps de Covid-19: une étude comparative des deux premières vagues de la pandémie en Lettonie
Zaiga Krisjane, Elina Apsite-Berina, Girts Burgmanis, Toms Skadins et Maris Berzins

Résumés

Depuis le début de l’année 2020 l’Europe se trouve confrontée à une détresse économique et sociale inédite en raison de la pandémie de Covid-19. L’expansion incontrôlable du SARS-Cov-2 a en effet profondément affecté les comportements, le bien-être sociétal et les activités quotidiennes des personnes.
Notre étude est consacrée aux changements intervenus dans les déplacements quotidiens entre la première et la deuxième vague de la pandémie, ainsi qu’aux différences territoriales dans les habitudes de mobilité de la population durant la « crise Covid »”.
Nous avons utilisé un questionnaire en ligne (CAWI, Computer-Aided Web Interview) en mars 2021 avec un échantillon total de n=1023 répondants auxquels nous avons demandé de comparer leur situation entre 2020 et 2021, correspondant aux 2 premières vagues de la pandémie en Lettonie.
L’article s’articule autour des 3 questions suivantes : 1) Quels sont les changements intervenus au niveau des déplacements quotidiens entre ces 2 vagues? 2) Quelles sont les différences entre les régions? 3) Quelles sont les caractéristiques socio-démographiques qui impactent les habitudes de mobilité?
Nos résultats montrent qu’environ un tiers de la population s’est mise au télétravail suite aux restrictions imposées et que cette solution représente un facteur crucial dans les inégalités socio-économiques en matière de mobilité. Nous n’avons pas relevé de différences significatives entre les 2 vagues, néanmoins l’hétérogénéité de nos découvertes apparaît lorsqu’on compare régions, groupes d’âge, niveaux d’éducation et types d’emplois. Nous avons trouvé un faible impact de la pandémie sur les déplacements pour des groupes de population moins éduqués, et un impact important voire de grandes inégalités socio-économiques comparativement à des étudiants ou aux télétravailleurs diplômés du supérieur, spécialistes, patrons, dirigeants d’entreprises et employés du tertiaire, vu que ce sont ceux qui font face aux changements les plus évidents dans leur vie quotidienne.

Haut de page

Texte intégral

This work was supported by the National Research Program Project grant number VPP-IZM-2018/1-0015 and ERDF grant 1.1.1.2/VIAA/1/16/184.

Introduction

1Since early 2020 Europe has been coping with previously unfamiliar economic and social distress caused by the COVID -19 pandemic. The virus’s uncontrollable spread (SARS-CoV-2) has thoroughly affected behavioural patterns and well-being, everyday activities, and population groups’ commuting patterns globally.

22020 is signposted as a year of another crisis over-shadowing the global economic, social and political scene. Besides, to control the spread of the virus depending on intra-state decisions population has been put at the ’territorial traps’ (Wang et al. 2020).

3A previous study on daily mobility changes during lockdown measures in Poland confirms visible shifts in the transport system’s exploitation. It indicates occupation as an important factor in travel time changes (Borkowski et al., 2021). This has put us at the ’Great Lockdown’, which has fundamentally changed human behaviour and previously established geographic patterns of the working places (Reuschke, Felstead, 2020; Beck et al., 2020).

4It is also previously acknowledged that human behavioural habits and lifestyles due to the imposed restrictions and lockdown measures are experiencing substantial shifts (Mazzoleni et al., 2020). Lack of daily commuting and work from home arrangements have been recognised as the most evident changes along with the COVID-19 pandemic (Kramer, Kramer, 2020).

5In times of the COVID-19 pandemic, individual well-being has become a hot topic more than ever. The COVID-19 outbreak has become one of the most significant events of our times, dramatically changing people’s daily routines and restricting their movements and interactions. Recent studies show that such vast changes in landscape potentially had significant effects on individual well-being (Brooks et al., 2020; Lima et al., 2020). Socio-spatial restrictions implemented by national governments, such as physical distancing and self-isolation at home, promoted a rapid increase in mental stress due to loneliness and the inability to detach from the work (Wood, Michaelides, 2021).

6At the same time, evidence from macro-environment studies shows that the COVID-19 pandemic also contributes to significant changes in well-being at the regional level. The degree to which the pandemic affects a place and exposes people to various risks is related to the regions’ social and socio-spatial inequalities and inequities (de Haas, 2020).

7It is also questioned the pros and cons for different occupation groups, meaning who the workers are more suitable for remote work and who are present at their workplaces during the national emergency (Kramer, Kramer, 2020). The first COVID-19 outbreak in Latvia is scrutinised through the work-life balance perspective using the population survey method (Krisjane et al., 2020)

8Another way of studying actual changes in population flows is through the mobile phone data to see the shifts in mobility across countries and regions. In France, restrictions have induced decreased mobility, particularly in regions with high COVID-19 expansion (Pullano et al., 2020; Iacus et al., 2020). Declining patterns of mobility according to the mode of transportation and the development of the virus in Poland have been studied using Google COVID19 Community Mobility reports (Wielechowski et al., 2020). Conducted analysis confirms the effectiveness of lockdown measures inducing social distancing as the most effective restrictive measure contributing to lower activity at daily commuting.

9In this case study of Latvia, we exploit a population survey to comparatively explore changes in commuting patterns. We fill in the gap contributing to comparing expected changes before the COVID-19 pandemic, outcomes of the restriction implementation and the most recent changes in the mobility patterns.

10The analysis was constructed around three main research questions: 1) What are the differences in daily commuting patterns during Latvia’s first and second COVID-19 wave? 2) How do commuting patterns translate into regional differences? 3) What sociodemographic characteristics influence commuting habits?

Ongoing emergency in Latvia: years 2020 and 2021

11The Centre for Disease Prevention and Control is the controlling institution in Latvia that accounts for the statistical data and sets the mandatory restrictions to the whole society. This study substantiates the background information data from the Latvian Centre for Disease Prevention and Control (abbreviated as SPKC). These were the number of daily COVID-19 tests, confirmed cases and outcomes for the entire country and the number of tests and confirmed cases by the municipality. Timely information on the governmental decisions was collected from the respective website’s sources in Latvia.

12The first case of COVID-19 in Latvia was recorded on March 2, 2020 (LSM.lv, 2020a). State emergency began on March 13, 2020. The first step of restrictions concerning the education system was when all educational facilities began the remote study process. All public events with more than 200 participants were cancelled and prohibited. No more than two people were allowed to gather in public places, excluding people performing work duties and members of the same family from one household (Likumi.lv, 2020).

13A peak of 48 new confirmed cases was recorded on April 1. The first death was reported on April 3. During the first emergency, 26 people died (SPKC, 2021a).

14Figure 1 highlights that the most significant increase in confirmed new cases was in the capital Riga and the central part of Latvia. Additionally, an increase was observed in the largest cities regionally and municipalities in the northern part of the country. It also clearly shows that the rise in cases for Riga was much higher than in other municipalities (SPKC, 2021b).

Figure 1. The increase in confirmed COVID-19 cases in Latvia’s municipalities from March 19 to June 9, 2020.

Figure 1. The increase in confirmed COVID-19 cases in Latvia’s municipalities from March 19 to June 9, 2020.

Authors’ figure based on SPKC, 2021b data

15After being extended several times, the COVID-19 emergency ended on June 9. However, not all restrictions were lifted at that time, and it was decided that the measures would continue to address the threat and consequences of COVID-19 (TVNET, 2020).

16Over the summer, the number of new cases remained low, never surpassing 18 (SPKC, 2021a). Nonetheless, the limitation of visitors’ flow to public catering was renewed (Cabinet of Ministers, 2020a), and the control of international travellers was strengthened (Cabinet of Ministers, 2020b).

17The number of new cases began to increase in the second half of September (SPKC, 2021a), and the situation continued to worsen, leading to the second emergency, which started on November 9, 2020. Initially, it was set to last until December 6 (LSM.lv, 2020b) but was extended several times, finally ending on April 7, 2021 (Apollo.lv, 2021a). Over that time, restrictions were also tightened, including limiting face-to-face public services and a nighttime curfew during the weekends (LSM.lv, 2020c; LSM.lv, 2020d).

18Despite this, the number of cases still tended to increase to the point that on numerous days in late 2020 and early 2021, the number of new cases was above 1000. The record number of cases (n=1831) was reported on December 31, 2020. The number of deaths (1847) was also much higher than during the first emergency (SPKC, 2021b).

19From February 9 to March 21, the number of new cases tended to decline compared to previous weeks. After that, the situation worsened, as there were spikes in the number of new cases compared to the last week (SPKC, 2021b). These developments happened when the survey was conducted.

20Figure 2 further highlights the second wave, where the spread of the COVID-19 was much more notable. While a significant increase was observed in the country’s central part and cities (especially for Riga), spatial characteristics differed to some extent. Substantial increases were typical to some municipalities in the east and west, but this was no longer the case for the northern part.

Figure 2. Increase in confirmed COVID-19 cases in the municipalities of Latvia during the second COVID-19 emergency.

Figure 2. Increase in confirmed COVID-19 cases in the municipalities of Latvia during the second COVID-19 emergency.

Authors’ figure based on SPKC, 2021b data

21The final day of the second COVID-19 emergency was April 6 (Ministry of Foreign Affairs, 2021). At the same time, almost all restrictions, except trade-related, remained in force (Apollo.lv, 2021b).

Data and methods

22This study explores available statistical data on the COVID-19 pandemic outbreak in Latvia. It is a free access data from the Open data portal accounting for information from the Centre for Disease Prevention and Control of Latvia. This data allows mapping dynamics of registered COVID-19 cases over time. National level decisions were taken from the Cabinet of Ministers’ on preventing the virus from March 2020 to April 2021.

23In March 2021, questionnaires were distributed to 1023 respondents aged 18-64. The questionnaire was conducted using a computer-assisted web interview (CAWI) approach covering all Latvia regions, accounting for gender, age and regional distribution in a sample.

24The sociodemographic characteristics of the sample are given in Table 1. Respondents who completed the questionnaire incorrectly were eliminated from the analysis, which meant that 823 (80%) of all questionnaires were accurately completed by respondents and considered valid.

Table 1. Socio-demographic characteristics of sample (N = 823).

Rīga

Pierīga

Kurzeme

Zemgale

Latgale

Vidzeme

Total

Gender

Male

45.5%

53.4%

64.0%

52.6%

51.0%

47.9%

50.7%

Female

54.5%

46.6%

36.0%

47.4%

49.0%

52.1%

49.3%

Age

18-24

14.0%

6.2%

10.7%

9.7%

14.3%

11.5%

11.3%

25-34

24.4%

17.4%

22.7%

14.9%

16.3%

20.8%

20.2%

35-44

20.4%

24.2%

17.3%

24.6%

19.4%

17.7%

21.0%

45-54

21.8%

24.8%

24.0%

28.9%

29.6%

28.1%

25.3%

55-64

19.4%

27.4%

25.3%

21.9%

20.4%

21.9%

22.2%

Education

Primary and secondary education

23.7%

17.4%

24.0%

16.7%

22.4%

18.8%

20.8%

Vocational education

15.4%

21.1%

28.0%

23.7%

30.6%

29.2%

22.2%

Higher education

60.9%

61.5%

48.0%

59.6%

47.0%

52.1%

57.0%

Occupation

Low-skilled worker

12.7%

24.4%

25.3%

25.4%

29.9%

26.9%

21.6%

Qualified worker

62.7%

53.1%

46.7%

57.0%

55.7%

53.8%

56.7%

Manager or company owner

14.1%

18.1%

17.3%

12.3%

6.2%

9.7%

13.5%

Student

10.5%

4.4%

10.7%

5.3%

8.2%

9.7%

8.2%

Type of settlement

Rural

0.0%

28.0%

40.0%

14.9%

16.3%

19.8%

15.4%

Urban

100.0%

72.0%

60.0%

85.1%

83.7%

80.2%

84.6%

Total sample

N = 279

N = 161

N = 75

N = 114

N = 98

N = 96

N = 823

Materials and procedure

25Participants completed a questionnaire-based survey consisting of third parts – general, demographic-specific and site-specific information about their commuting patterns in the pre-COVID-19 times and during the first and second waves of COVID-19.

26In the first part of the questionnaire, we asked respondents for general sociodemographic information (sex, age, education level, occupation). These questions were essential to reveal changes in respondents’ commuting patterns between the waves of COVID-19 in general and show the effects and differences of implemented restrictions on commuting patterns of different sociodemographic groups in both waves of COVID-19.

27The second part of the survey included questions aimed at capturing information on respondents’ place of residence. Respondents were asked in which region of Latvia (Riga, Pieriga, Kurzeme, Zemgale, Latgale, Vidzeme) and what area (rural or urban) they live in. This set of questions was designed to test the differences in respondents’ commuting patterns between regions and types of settlements after implementing restrictions in both waves of COVID-19.

28The third part of the questionnaire was designed to elicit information about the respondents’ commuting patterns in the pre-COVID-19 era and the first and second wave of COVID-19 when the government implemented various restrictions (frequency of travel to work and/or studies). Respondents were asked to estimate an approximate number of trips to work or educational institution per week or month and check the most appropriate ordinal answer (e.g., every working day, few days per week, at least once in a month, a few times a month, a less often than once per month, never). This part included three questions on commuting patterns, i.e., one for each period (pre-COVID-19 times, first wave of COVID-19, second wave COVID-19).

Data analysis

29In our analysis, we coded eight independent variables (region and settlement type where respondent’s place of residence is located and respondent’s age, gender, occupation, education, economy sector, nationality). All of which we assume hypothetically may demonstrate and explain differences in changes of respondents’ commuting patterns between both waves of COVID-19.

30The statistical analysis of the data was performed three-fold. We compared respondents’ commuting patterns between (1) pre-COVID-19 times and first wave, (2) pre-COVID-19 times and second wave and (3) the first wave of COVID-19 and second wave. Thus we could clarify how respondents’ commuting patterns differed between waves of COVID-19 when the government implemented distinct types of restrictions and how these restrictions changed commuting patterns of different groups in each of the waves. Second, we analysed data for each region separately. The main aim of such an approach was to assess the potential effect of the region and other factors that may have an influence on changes in respondents’ commuting patterns in times of the COVID-19 pandemic.

31The development of six dependent variables included two steps. First, we transformed ordinal answers on questions eliciting an approximate number of trips to work or educational institutions per week or month to an interval scale. Such transformations were performed to overcome the difficulty of determining the changes in commuting patterns between pre-COVID-19 times, first and second waves of COVID-19 using original respondents’ answers. Previous studies suggest that such transformations between ordinal and interval scale variables are possible. They can be made using various techniques, including IRT models (e.g. Rasch model) (Harwell & Gatti, 2001) or Markov chain Monte Carlo scaling method (Granberg-Rademacker, 2010). However, these techniques were not entirely appropriate for our study because they initially were created to transform Likert scale answers, mostly measuring latent constructs. Our study used a different approach because we needed to develop the dependent variables that reflect changes in observable behaviour between different periods. Original answers we converted to an approximate number of days per month each respondent commuted to work in each period. For approximation, we used the following conversion (1) every working day - 21 days, (2) few days per week - 11 days, (3) at least once in a month - 4 days, (4) a few times per month - 3 days, (5) a less often than once per month – 0.75 days, (6) never - 0 days. The purpose of the first step was to develop the dependent variables (see Table 2) used to determine whether commuting patterns differed between the first and second waves of COVID-19 after implementing different restrictions.

32Second, using in the previous step converted answers, we calculated three dependent variables representing the changes in the number of commuting days to work or studies between (1) pre-COVID-19 times and first wave, (2) pre-COVID-19 times and second wave and (3) the first wave of COVID-19 and second wave (see Table 2). For each respondent, we subtracted the number of commuting days to work or studies in pre-COVID-19 times from the number of commuting days to work or studies in the first wave of COVID-19. The similar calculations were also made for two other pairs of periods. The purpose of the second step was to develop dependent variables to determine how restrictions implemented by the government affected changes in commuting patterns across different groups.

33Kruskal-Wallis one-way analysis of variance was used to examine the differences in commuting patterns across different groups in the first and second waves of COVID-19. The Kruskal Wallis H test was selected according to the needs of the analysis. First, it is helpful to determine statistically significant differences between the two or more groups of an independent variable on a continuous or ordinal dependent variable. Second, it is used when the data do not meet the requirements for a parametric test, i.e., are not normally distributed. The normality of dependent variables was determined using a Shapiro-Wilk test which shows that the distribution of all dependent variables departed significantly from normality, i.e., p<0.05 (see Table 2). Based on these outcomes, a nonparametric test – Kruskal Wallis H test – was selected.

Table 2. Results of normality test.

W

df

P-value

Commuting days to work or studies in pre-COVID-19 times

.704

823

.00

Commuting days to work or studies in the first wave of COVID-19

.812

823

.00

Commuting days to work or studies in the second wave of COVID-19

.786

823

.00

Changes in commuting days to work or studies between pre-COVID-19 times and the first wave of COVID-19

.743

823

.00

Changes in commuting days to work or studies between pre-COVID-19 times and second wave of COVID-19

.763

823

.00

Changes in commuting days to work or studies between first COVID-19 times and second waves of COVID-19

.706

823

.00

34However, the Kruskal-Wallis H test is an omnibus test statistic and cannot tell which specific groups of the independent variable are statistically significantly different from each other. Dunn’s post-hoc tests with adjusted Bonferroni correction for multiple comparisons were used to determine which groups experienced statistically significant changes in the commuting patterns in both waves of COVID-19. The Dunn’s post-hoc test was performed only for cases with statistically significant values of the Kruskal-Wallis test.

Results

35Foremost, we tested the hypothesis stating that different restrictions implemented by the government in each of the waves of COVID-19 have a different effect on respondents’ commuting patterns. Kruskal-Wallis test indicated that enforced regulation significantly affects respondents’ commuting patterns H (2) = 173.04, p = 0.000. Dunn’s pairwise tests were carried out for the three pairs of groups. There was strong evidence (p < 0.01, adjusted using the Bonferroni correction) of a difference between commuting patterns in pre-COVID-19 times and both waves of COVID-19. The median number of commuting days for pre-COVID-19 times was 21 days per week compared to 11 days per week in both waves of COVID-19. There was no evidence of a difference between the first and second waves of COVID-19.

Table 3 Changes in respondents’ commuting patterns to work and educational institutions across the first and second wave of COVID-19.

Independent variables

Pre-COVID-19 and the first wave

Pre-COVID-19 and second wave

The first and second wave

M

SD

H

M

SD

H

M

SD

H

Region

Riga

-6.67

7.99

35.27**

-7.85

9.02

44.29**

-1.19

6.87

12.93*

Pieriga

-4.76

7.19

-5.50

8.09

-.75

6.17

Kurzeme

-3.13

6.29

-3.48

6.56

-.35

4.38

Zemgale

-2.66

5.73

-2.46

6.24

.19

4.95

Latgale

-3.89

6.86

-4.81

7.92

-.93

5.14

Vidzeme

-3.50

6.79

-4.25

7.04

-.75

5.69

Type of Settlement

Urban

-4.86

7.39

1.05

-5.73

8.24

2.77

-.88

6.02

.67

Rural

-3.96

6.57

-4.02

7.49

-.08

5.56

Gender

Male

-4.07

6.99

5.66

-5.07

8.13

2.14

-.99

5.78

.22

Female

-5.38

7.52

-5.87

8.16

-.50

6.14

Age

18-24

-6.97

8.65

17.21**

-9.37

9.09

37.16**

-2.39

8.00

19.13**

25-34

-5.45

7.52

-6.28

8.79

-.83

6.50

35-44

-4.86

7.42

-5.64

8.12

-.78

6.04

45-54

-3.49

6.54

-4.01

7.16

-.53

5.05

55-64

-4.18

6.67

-4.23

7.41

-.07

4.95

Education

Secondary education or lower

-3.71

7.12

30.50**

-5.25

8.38

16.11**

-1.53

6.44

4.48

Vocational education

-2.86

6.16

-3.63

6.96

-.78

5.12

Higher education

-5.81

7.55

-6.26

8.38

-.46

6.07

Ethnicity

Latvian

-5.02

7.37

5.15

-5.91

8.27

6.03

-.89

6.04

1.18

Russian

-4.63

7.20

-4.98

8.11

-.36

6.05

Othe

-3.08

6.77

-4.01

7.35

-.94

5.22

Occupation

Low-skilled worker

-1.78

4.97

58.44**

-1.99

5.94

82.16**

-.21

3.70

25.96**

Qualified specialist

-5.35

7.49

-6.12

8.21

-.60

6.46

Manager or company owner

-4.30

7.07

-5.08

8.04

-.77

4.99

Student

-7.81

8.79

-11.10

9.12

-3.28

8.14

Economy sector

Primary sector

-3.13

6.41

43.63**

-2.97

7.40

47.47**

.13

7.27

2.73

Secondary sector

-1.84

4.99

-2.11

6.09

-.27

3.78

Tertiary sector

-5.49

7.49

-6.15

8.17

-.68

5.96

Statistically significant at ** p<.01; * p<.05

36Next, by using dependent variables representing the changes in the number of days with commuting activities to work or studies, we tested the hypothesis stating that restrictions implemented by the government in both waves of COVID-19. The first hypothesis was that commuting patterns of various groups were affected differently – the second hypothesis was determined by geographic variations in commuting patterns in Latvia.

37From Table 3 it can be seen that not all the independent variables that are statistically significant in terms of changes in the number of commuting days to work or studies are appropriate for all comparisons. Commuting patterns of five groups (regions, age, education, occupation, economy sector) were affected differently by restrictions in the first and second wave of COVID-19 compared to pre-COVID-19 times. Comparing the effect of restrictions on commuting patterns between the first and second waves of COVID-19, only three groups (region, age, occupation) show significant differences.

38We find that restrictions implemented in both waves compared to pre-COVID-19 time’s affected geographic variations in decreasing the number of commuting days to work or studies in Latvia. Dunn’s pairwise tests to compare all groups were carried out for the first wave and pre COVID-19 time’s comparison. Strong evidence (p<0.05, adjusted using the Bonferroni correction) showed that the decrease in commuting days to work or studies in Riga was significantly higher than in Zemgale. There was no evidence of other significant changes in commuting patterns across regions in the first wave of COVID-19. Examining changes in commuting patterns between the first and second wave of COVID-19, we found a similar relationship, i.e., the decrease in commuting days was higher in Riga than in Zemgale (p<0.05, adjusted using the Bonferroni correction). In the second wave of COVID-19 compared to pre-COVID-19 time, there was substantial evidence (p<0.01, adjusted using the Bonferroni correction) in a decrease in commuting days which were significantly higher in Riga than in other regions of Latvia except for Pieriga.

39Restrictions implemented by the government changed commuting patterns of age groups differently. This relationship can be seen for all comparisons. We find that in the first wave of COVID-19, the decrease in the number of commuting days for 18-24 years olds was significantly higher than for age groups 45-54 and 55-64 (p<0.01, adjusted using the Bonferroni correction). There was no evidence of a difference between the other pairs. In the second wave of COVID-19, compared to pre-pandemic times, the commuting patterns of 18-24 years olds were affected more than other age groups (comparisons p<0.01, adjusted using the Bonferroni correction). Moreover, the restrictions implemented by the government in the second wave also changed commuting patterns of 18-24 olds more than other groups than restrictions implemented in the first wave (comparisons p<0.01, adjusted using the Bonferroni correction).

40We find significant changes in commuting patterns among different education groups. In the first wave of COVID-19 compared to pre-pandemic times, the highest decrease in the number of commuting days to work or studies experienced respondents with higher education than respondents with vocational, secondary or lower education (comparisons p<0.01, adjusted using the Bonferroni correction). In the second wave compared to pre-pandemic times, there were only significant changes in the number of commuting days between respondents with higher education and those with vocational education (comparisons p<0.01, adjusted using the Bonferroni correction). There was no evidence of age as a discriminating factor of commuting patterns between the first and second waves of COVID-19.

41Another important factor that reflects the influence of restrictions on commuting patterns is the respondent’s occupation. Our results show that being a student significantly increased the risk to be exposed to changes in commuting patterns in both waves of COVID-19. Students compared to other groups had the highest decrease in commuting days (comparisons p<0.01, adjusted using the Bonferroni correction). This pattern can be seen for all comparisons, including changes in the number of commuting days between pre-pandemic times and both waves of COVID-19 and between both waves of COVID-19. These results are not surprising and strengthen previous findings in terms of age groups. Dunn’s pairwise tests also show that qualified specialists, managers or company owners had a more significant decrease in commuting days than low-skilled workers did (comparisons p<0.01, adjusted using the Bonferroni correction). These changes of commuting patterns for qualified specialists, managers or company owners and low-skilled workers can be seen for both waves of COVID-19 compared to pre-pandemic times but not between the first and second waves of COVID-19. There is no evidence for all comparisons of periods that commuting patterns significantly differ between qualified specialists and managers or company owners.

42Finally, data analysis on how restrictions affected different economic sectors reveals that respondents who worked in the tertiary sector were affected more than those who worked in other sectors. Dunn’s pairwise test shows that respondents who worked in the tertiary sector experienced a significant decrease in commuting days in the first and second wave of COVID-19 compared to pre-pandemic times (for both comparisons p<0.01, adjusted using the Bonferroni correction). At the same time, there was no evidence of changes in commuting patterns between the primary and secondary sectors.

Concluding remarks

43This study focused on the COVID-19 pandemic induced changes in commuting patterns. It also synchronously undertakes to shed light on regional differences and socioeconomic disparities.

44The study used a population CAWI survey, which allowed identifying behavioural trends for different sociodemographic groups. 

45The research revealed that Latvia is an interesting case study as it experienced two different scenarios of the COVID-19 outbreak. At first, the epidemic status was not critical, and implemented lockdown measures in early 2020 appeared to be effective. However, the second wave of the virus in late 2020 has severely affected society and sectors of the economy. Such measures as a request for remote work and education sharply affected daily commuting patterns. Similar to other countries due to school closure and decrease in commuting to work, except for workers in vital fields, a stable one-third of the population were working from home (WFH) or the “new normal” in Australia (Beck et al. 2020) in the Netherlands (de Haas, 2020), South America and South Africa (Balbontin et al. 2021), USA (Brough et al. 2021).

46Our study confirmed that the government’s restriction during the first and second waves significantly affected the respondent’s frequency of movements to work or educational institutions by limiting the number of trips. There was also a significant difference in pre-COVID-19 commuting patterns and both waves of the pandemic. However, statistical analysis confirmed no statistical differences when the first and second periods were compared. 

47In the case of Latvia, geographical differences for commuting patterns were present, particularly when comparing respondents who live in Riga and other regions of Latvia. For example, Capital Riga showed a more substantial decrease in commuting days to work or studies than Zemgale. Comparison between the waves displays a similar association as the decrease in commuting days decreased more for Riga. However, the second wave analysis result highlighted close and traditional daily commuting patterns between the capital Riga and Pieriga, that is the suburban area. Assumably population living in the surroundings of Riga were among the ones whose travel and work behaviours have been affected the most. WFH, with the quality and stability of Internet access, is an opportunity for more distant locations, and those working remotely vastly exploit the Internet and available technologies (Brough et al., 2021). 

48Our findings suggest that in comparing the first and second waves, the most significant decrease in commuting days was for young people. Our results show that being a student significantly increased the risk to be exposed to changes in commuting patterns in both waves of COVID-19. Presumably, shift to the remote education was among the most striking elements of the lockdown. 

49More heterogeneous results come to the surface when controlling for educational level; from the “COVID crisis”, respondents with higher education were affected the most compared to the lower level of education. Overallthe ability to do WFH illuminates far-reaching socioeconomic inequalities, which will be relevant both during the “COVID crisis” and beyond (Brough et al., 2021). 

50Furthermore, also occupational differences were examined. From the results, we see that qualified specialists, managers or company owners had a more significant decrease in commuting days than low-skilled workers did. Respondents who worked in the tertiary sector were affected more than those who worked in other sectors.

51Interestingly “COVID crisis” commuting patterns primarily affected young people, people with higher education, employed in the tertiary sector, as they are those coping with the most evident changes in their daily lives. Similar results were found in Australia and the USA, where WFH overall is more common among those with higher income and living in urban areas (Beck et al. 2020; Brough et al. 2021).

52Traditionally mitigation of crisis impacts would come at the expense of those exposed to fewer opportunities, resources and deepening regional disparities. Thus, more evidence is necessary on factors underlying people’s well-being for various population groups, such as those mentioned above.

Haut de page

Bibliographie

APOLLO.lv (2021a), Oficiāli: Ārkārtējo situāciju pagarina līdz 6. aprīlim, par komandantstundas turpināšanu vēl diskutēs, https://www.apollo.lv/7173068/oficiali-arkartejo-situaciju-pagarina-lidz-6-aprilim-par-komandantstundas-turpinasanu-vel-diskutes (accessed 13.04.2021).

APOLLO.lv (2021b), Šodien oficiāli beigusies ārkārtējā situācija. Kādi drošības pasākumi jāievēro no 7. aprīļa, https://www.apollo.lv/7218799/sodien-oficiali-beigusies-arkarteja-situacija-kadi-drosibas-pasakumi-jaievero-no-7-aprila (accessed 13.04.2021).

BALBONTIN C., Hensher D.A., Beck M.J., Giesen R., Basnak P., Vallejo-Borda J.A. & Venter C. (2021), “Impact of COVID-19 on the number of days working from home and commuting travel: A cross-cultural comparison between Australia, South America and South Africa”, Journal of Transport Geography96, 103188.

BECK M.J., HENSHER D.A. & WEI E. (2020), “Slowly coming out of COVID-19 restrictions in Australia: Implications for working from home and commuting trips by car and public transport”Journal of Transport Geography, 88, 102846.

BORKOWSKI P., Jażdżewska-Gutta M. & Szmelter-Jarosz A. (2021), “Lockdowned: Everyday mobility changes in response to COVID-19”, Journal of Transport Geography, 90, 102906.

BROOK S.K., Webster R.K., Smith L.E., Woodland L., Wessely S., Greenberg N. & Rubin G.J. (2020), “The psychological impact of quarantine and how to reduce it: rapid review of the evidence”, The Lancet, 395, pp. 912-920, https://doi.org/10.1016/S0140-6736(20)30460-8

BROUGH R., Freedman M. & Phillips D.C. (2021), “Understanding socioeconomic disparities in travel behavior during the COVID19 pandemic”, Journal of Regional Science61, 4, pp. 753-774.

BURCHELL B., Reuschke D. & Zhang M. (2020), “Spatial and temporal segmenting of urban workplaces: The gendering of multi-locational working”, Urban Studies, 58, 11, https://journals.sagepub.com/doi/10.1177/0042098020903248

CABINET OF MINISTERS (2020a), Ministru kabinets pastiprina atsevišķus Covid-19 izplatības ierobežojumus, https://www.mk.gov.lv/lv/jaunums/ministru-kabinets-pastiprina-atseviskus-covid-19-izplatibas-ierobezojumus (accessed 12.04.2021).

CABINET OF MINISTERS (2020b), Ministru kabinets pastiprina Covid-19 izplatības kontroli ieceļotāju vidū, https://www.mk.gov.lv/lv/jaunums/ministru-kabinets-pastiprina-covid-19-izplatibas-kontroli-iecelotaju-vidu (accessed 12.04.2021).

DE HAAS M., Faber R. & Hamersma M. (2020), “How COVID-19 and the Dutch ‘intelligent lockdown’ change activities, work and travel behaviour: Evidence from longitudinal data in the Netherlands”, Transportation Research Interdisciplinary Perspectives, 6, 100150.

IACUS S.M., Santamaria C., Sermi F., Spyratos S., Tarchi D. & Vespe M. (2020), “Human mobility and COVID-19 initial dynamics”, Nonlinear Dynamics, 101, 3, pp. 1901-1919.

KRAMER A., Kramer K.Z. (2020), “The potential impact of the Covid-19 pandemic on occupational status, work from home, and occupational mobility”, Journal of Vocational Behavior, https://doi.org/10.1016/j.jvb.2020.103442

KRISJANE Z., Apsite-Berina E., Berzins M., Skadins T. & Burgmanis G. (2020), “Work-life balance during the COVID-19 outbreak: the case of Latvia”, Baltic Region, 12, 4, pp. 39-60.

LIKUMI.lv (2020), Par ārkārtējās situācijas izsludināšanu, https://likumi.lv/ta/id/313191-par-arkartejas-situacijas-izsludinasanu (accessed 11.04.2021).

LIMA C.K.T., de Medeiros Carvalho P.M., Lima I.D.A.A.S., de Oliveira Nunes J.V.A., Saraiva J.S., de Souza, R.I., da SILVA C.G.L. & Neto M.L.R. (2020), “The emotional impact of Coronavirus 2019-nCoV (new Coronavirus disease)”, Psychiatry research, 287, 112915.

LSM.lv (2020a), Latvijā Covid-19 dēļ atkal izsludina ārkārtējo situāciju; būs jauni stingrāki ierobežojumi, https://www.lsm.lv/raksts/zinas/latvija/latvija-covid-19-del-atkal-izsludina-arkartejo-situaciju-bus-jauni-stingraki-ierobezojumi.a380750/ (accessed 11.04.2021).

LSM.lv (2020b), Ārkārtējo situāciju Covid-19 dēļ Latvijā pagarina līdz 11. janvārim; atkal spēkā «2+2» (precizēts), https://www.lsm.lv/raksts/zinas/latvija/arkartejo-situaciju-covid-19-del-latvija-pagarina-lidz-11-janvarim-atkal-speka-22-precizets.a383719/ (accessed 13.04.2021).

LSM.lv (2020c), Ārkārtējo situāciju pagarina līdz 7. februārim, https://replay.lsm.lv/lv/ieraksts/ltv/206509/arkartejo-situaciju-pagarina-lidz-7-februarim/ (accessed 13.04.2021).

LSM.lv (2020d), Uz 3 nedēļām ierobežo tirdzniecību, klātienes pakalpojumus un aicina ievērot klusuma period, https://www.lsm.lv/raksts/zinas/latvija/uz-3-nedelam-ierobezo-tirdzniecibu-klatienes-pakalpojumus-un-aicina-ieverot-klusuma-periodu.a385668/ (accessed 13.04.2021).

LSM.lv (2020e), Latvijā miruši vēl 44 slimnieki ar Covid-19; inficējušies 511 cilvēki, https://www.lsm.lv/raksts/zinas/latvija/latvija-mirusi-vel-44-slimnieki-ar-covid-19-inficejusies-511-cilveki.a386816/ (accessed 13.04.2021).

MAZZOLENI S., Turchetti G. & Ambrosino N. (2020), “The COVID-19 outbreak: from ’black swan’ to global challenges and opportunities”, Pulmonology, 26, 3, pp. 117-118.

MINISTRY OF FOREIGN AFFAIRS (2021), Par drošības pasākumiem pēc ārkārtējās situācijas beigām, https://www.mfa.gov.lv/konsulara-informacija/jaunumi/66928-par-arkartejas-situacijas-izsludinasanu-latvija (accessed 12.04.2021).

PULLANO G., Valdano E., Scarpa N., Rubrichi S. & Colizza V. (2020), “Population mobility reductions during COVID-19 epidemic in France under lockdown”, MedRxiv, doi: https://doi.org/10.1101/2020.05.29.20097097

REUSCHKE D., Felstead A. (2020), “Changing workplace geographies in the COVID-19 crisis”, Dialogues in Human Geography, 10, 2, pp. 208-212.

ROSE-REDWOOD R., Kitchin R., Apostolopoulou E., Rickards L., Blackman T., Crampton J., ROSSI U. & Buckley M. (2020), “Geographies of the COVID-19 pandemic”, Dialogues in Human Geography, 10, 2, pp. 97-106.

SPKC (2021a), COVID-19 izmeklējumi, apstiprinātie gadījumi un iznākumi, https://data.gov.lv/dati/lv/dataset/covid-19 (accessed 12.04.2021).

SPKC (2021b), COVID-19 apstiprināto gadījumu skaits un 14 dienu kumulatīvā saslimstība pa administratīvajām teritorijām, https://data.gov.lv/dati/lv/dataset/covid-19-pa-adm-terit (accessed 11.04.2021).

TVNET (2020), Kariņš: Ārkārtējo situāciju Latvijā izbeigs 9.jūnijā, bet visi ierobežojumi atcelti netiks, https://www.tvnet.lv/6989064/karins-arkartejo-situaciju-latvija-izbeigs-9-junija-bet-visi-ierobezojumi-atcelti-netiks (accessed 11.04.2021).

WANG F., Zou S. & Liu Y. (2020), “Territorial traps in controlling the COVID-19 pandemic”, Dialogues in Human Geography, 10, 2, pp. 154-157.

WIELECHOWSKI M., Czech K. & Grzęda Ł. (2020), “Decline in Mobility: Public Transport in Poland in the time of the COVID-19 pandemic”, Economies, 8, 4, https://doi.org/10.3390/economies8040078

WOOD S., Michaelides G. (2021), “Homeworking during the Pandemic”, Work-Life Balance Bulletin: A DOP Publication, 5, 1, pp. 9-12.

Haut de page

Table des illustrations

Titre Figure 1. The increase in confirmed COVID-19 cases in Latvia’s municipalities from March 19 to June 9, 2020.
Crédits Authors’ figure based on SPKC, 2021b data
URL http://journals.openedition.org/belgeo/docannexe/image/55939/img-1.jpg
Fichier image/jpeg, 266k
Titre Figure 2. Increase in confirmed COVID-19 cases in the municipalities of Latvia during the second COVID-19 emergency.
Crédits Authors’ figure based on SPKC, 2021b data
URL http://journals.openedition.org/belgeo/docannexe/image/55939/img-2.jpg
Fichier image/jpeg, 269k
Haut de page

Pour citer cet article

Référence électronique

Zaiga Krisjane, Elina Apsite-Berina, Girts Burgmanis, Toms Skadins et Maris Berzins, « Commuting behaviour in times of the COVID-19 pandemic: a comparative first and second wave study from Latvia »Belgeo [En ligne], 3 | 2022, mis en ligne le 26 octobre 2022, consulté le 28 novembre 2022. URL : http://journals.openedition.org/belgeo/55939 ; DOI : https://doi.org/10.4000/belgeo.55939

Haut de page

Auteurs

Zaiga Krisjane

University of Latvia, zaiga.krisjane@lu.lv

Articles du même auteur

Elina Apsite-Berina

University of Latvia, elina.apsite-berina@lu.lv

Articles du même auteur

Girts Burgmanis

University of Latvia, girts.burgmanis@lu.lv

Toms Skadins

University of Latvia, toms.skadins@lu.lv

Maris Berzins

University of Latvia, maris.berzins@lu.lv

Articles du même auteur

Haut de page

Droits d’auteur

CC-BY-4.0

Creative Commons - Attribution 4.0 International - CC BY 4.0

https://creativecommons.org/licenses/by/4.0/

Haut de page
  • Logo Fondation Universitaire/Universitaire Stichting
  • Logo Fonds de la Recherche Scientifique - FNRS
  • Logo National Comittee of Geography
  • Logo SRBG
  • DOAJ - Directory of Open Access Journals
  • OpenEdition Journals
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search