Social trust during the pandemic: Longitudinal evidence from three waves of the Swiss household panel study

Figures & data

Figure 1. Government stringency index: the case of Switzerland. Note: The Stringency index is a composed measure based on nine response indicators including school closures, workplace closures and travel bans, rescaled to a value range 0–100 (100 = strictest) | Source: Hale et al., (2021).

Table 1. Descriptive statistics (N_total = 5843).

					Mean
	N	Min	Max	SD	2019	COVID	2020	Var
Social Trust	5714	0	10	2.54	6.3	5.8	6.3	6.48
Belonging	5789	0	10	2.51	–	7.8	–	6.30
Health	5822	1	5	0.68	4.0	4.3	4.0	0.46
Gov. Crisis	5709	0	4	0.90	–	3.1	–	0.82
Age	5843	14	99	18.6	49.6	54.2	4.6	34.97
Female	5843	0	1	0.49	0.5	0.5	0.5	0.24
Children	4427	0	1	0.44	–	0.3	–	0.20
Satisfied with …
Life in general	5781	0	10	1.56	–	8.0	–	2.43
Leisure activities	5728	0	10	2.32	–	6.8	–	5.41
Restrictions …
Unproblematic	5649	0	1	0.49	–	0.40	–	0.24
Necessary evil	5649	0	1	0.50	–	0.55	–	0.25
Problematic	5649	0	1	0.20	–	0.05	–	0.04
Partner status …
Living together	5582	0	1	0.46	–	0.70	–	0.20
Living apart	5582	0	1	0.29	–	0.09	–	0.08
Single	5582	0	1	0.40	–	0.30	–	0.16
Occupation …
Employed	5760	0	1	0.50	–	0.56	–	0.24
Unemployed	5760	0	1	0.12	–	0.02	–	0.01
Inactive	5760	0	1	0.49	–	0.42	–	0.24
Worried for …
Personal health	5779	0	10	2.71	–	3.4	–	7.39
Health of close one	5757	0	10	2.77	–	5.2	–	7.71
Lifestyle changes	5748	0	10	2.55	–	2.7	–	6.52
Social relations	5744	0	10	2.56	–	2.9	–	6.57

Note: Do note that a correlation matrix is available in the appendix if interested.

Figure 2. Fixed-effects regression: trust development from 2008 until 2021 with 95% confidence intervals. Note: The results from the extra COVID-19 survey is highlighted. A numerical table of the results can be found in the appendix of the article. The fourth sample, added in 2020, is not included here.

Figure 3. First difference regression: within-individual factors of change in trust between t₁ and t₂. Note: change between the annual 2019 wave and the extra COVID-19 wave, i.e. the ‘decrease’ in social trust. A numerical table of the results can be found in the appendix.

Figure 4. First difference regression: within-individual factors of change in trust between t₂ and t₃. Note: Change beetween the extra COVID-19 wave and the annual 2020 wave, i.e. the ‘increase’ in social trust. A numerical table of the results can be found in the appendix.

Figure 5. Multinominal logistic regression: confidence intervals of 95% (N_valid = 4049). Note: Reference for the restriction variable=‘people who don’t think the restrictions are problematic at all and necessary’ | Reference for the partner variables= ‘people who share a household with their partner’ | Reference for the female variable= ‘male’ | reference for the age groups= ‘14–25’ | Reference for the employment status variable= ‘employed’ | Log likelihood −2330.7528 (iteration 4) | Prob chi2 0.00000 | Lr chi2(42) 11.98 | Pseudo R2 0.0235 | A numerical presentation of the multinominal logistic regression is included in the appendix.

Table A1. First differences regression – within-individual factors for the ‘drop’ in trust between t₁ and t₂.

Social Trust	Coef. (Std. Err)	t (P > |t|)	95% Conf. Interval
			Lower	Upper
Wave	−.519 (.035)	−14.82 ***	−.588	−.451
Age
Age: 14–29	.174 (.477)	0.37	−.760	1.110
Age: 65 plus	−.079 (.258)	−0.31	−.586	.427
Sex	0	(omitted)
Partner
Yes, living apart	−.154 (.158)	−0.94	−.464	.155
Yes, living together	−.102 (.198)	−0.52	−.492	.287
Occupation
Employed	.158 (.098)	1.60	−.035	.352
Unemployed	−.123 (.234)	−0.53	−.492	.287
Health Status	−.006 (0.40)	−0.15	−.085	.073
Satisfaction with …
Financial situation	.014 (.018)	0.82	−.020	.050
Leisure activities	.031 (.012)	2.55 **	.007	.056
Personal relationships	.092 (.017)	5.15 ***	.056	.127
Life in general	.053 (.022)	2.34 **	.008	.097
Constant	14.742 (.753)	19.56 ***	13.263	16.219
Other values …
F test	3.39
Prob > F	0.0000
R-square
Within	0.0941
Between	0.0141
Overall	0.0268
Rho	.697
Nvalid	13637

Notes: *** P(>|t|) < 0.001; ** P(>|t|) < 0.01; * P(>|t|) < 0.05 | Reference category for Age = ‘30–64-year-olds’ | Reference category for partner variables = ‘Single’ | Reference category for occupation = ‘Inactive’.

Table A2. First differences regression – within-individual factors for the ‘rise’ in trust between t₂ and t₃.

Social trust	Coef. (Std. Err)	t (P > | t |)	95% Conf. Interval
			Lower	Upper
Wave	.620 (.034)	17.71 ***	.552	.689
Age groups
Age: 14–29	.901 (.444)	2.03 *	.030	1.772
Age: 65 plus	.335 (.287)	1.17	−.227	.898
Sex	0	(omitted)
Partner
Yes, living apart	−.019 (.172)	−0.11	−.357	.318
Yes, living together	.134 (.223)	0.60	−.304	.572
Occupation
Employed	.123 (.102)	1.20	−.078	.325
Unemployed	−.340 (.250)	−1.36	−.832	.151
Health Status	.046 (.042)	1.11	−.035	.129
Satisfaction with …
Financial situation	.040 (.019)	2.09 *	.002	.078
Leisure activities	.041 (.012)	3.26 ***	.016	.067
Personal relationships	.051 (.019)	2.68 **	.013	.088
Life in general	.071 (.023)	3.05 **	.025	.116
Constant	−8.673 (.787)	−11.01 ***	−10.217	−7.129
Other values …
F test	3.37
Prob > F	0.0000
R-square
Within	0.1201
Between	0.0076
Overall	0.0220
Rho	.7009
Nvalid	13183

Notes: *** P(>|t|) < 0.001; ** P(>|t|) < 0.01; * P(>|t|) < 0.05 | Reference category for Age = ‘30–64-year-olds’ | Reference category for partner variables = ‘Single’ | Reference category for occupation = ‘Inactive’.

Table A3. Patterns of meaningful change in social trust (N_total = 5843).

	Wave
Pattern	2019	Covid	2020	Match (%)	Perspective
1	Stable	Stable	Stable	68,61	Cultural
8	Stable	Decrease	Increase	9,05	Set-Point
2	Stable	Stable	Increase	4,57	Unknown
6	Stable	Decrease	Stable	3,80	Experiential
18	Decrease	Stable	Stable	2,67	Experiential
5	Stable	Stable	Decrease	1,50	Unknown
12	Increase	Stable	Stable	1,44	Experiential
3	Stable	Increase	Stable	1,40	Experiential
22	Decrease	Increase	Stable	1,40	Set-Point
14	Increase	Decrease	Stable	1,25	Set-Point
9	Stable	Increase	Decrease	1,19	Set-Point
23	Decrease	Stable	Increase	1,13	Set-Point
13	Increase	Decrease	Increase	0,75	Set-Point
15	Increase	Stable	Decrease	0,30	Set-Point
20	Decrease	Increase	Decrease	0,26	Set-Point
19	Decrease	Decrease	Increase	0,24	Set-Point
17	Decrease	Stable	Decrease	0,14	Experiential
11	Increase	Stable	Increase	0,10	Experiential
16	Decrease	Decrease	Stable	0,10	Experiential
21	Decrease	Increase	Increase	0,06	Set-Point
4	Stable	Increase	Increase	0,02	Experiential
7	Stable	Decrease	Decrease	0,02	Experiential
10	Increase	Increase	Stable	0,02	Experiential

Notes: Unknown label occurred when changes happened in the last wave. Numerically, Stable = 0; Increase = 1; Decrease = −1. The ‘Pattern’ column is present for reference purposes.

Table A4. Multinominal logistic regression analysis: summer of 2020 data (N_valid = 4049).

Meaningful change	RRR		95% Conf. Interval
Stable Response	(base outcome)	Z	Lower	Upper
Positive Response
Gov. Crisis Management	1.035 (.105)	0.34	.847	1.265
Rest. Civil Rights: Dum 2	.720 (.117)	−2.02	.523	.990
Rest. Civil Rights: Dum 3	.847 (.403)	−0.35	.332	2.156
Partner: Yes, living apart	.903 (.383)	−0.24	.393	2.075
Partner: No, single	1.430 (.455)	1.12	.766	2.670
Children	.725 (.172)	−1.35	.456	1.155
Female	.674 (.109)	−2.41	.490	.928
Life Satisfaction	1.086 (.077)	1.17	.944	1.249
Satis. Leisure Activities	1.084 (.046)	1.90	.997	1.178
Belonging	.946 (.033)	−1.53	.882	1.015
Health	.876 (.118)	−0.98	.672	1.141
Work: Unemployed	1.634 (1.01)	0.79	.481	5.542
Work: Inactive/Retired	1.395 (.285)	1.62	.932	2.082
Age: 26–35	.794 (.342)	−0.53	.341	1.848
Age: 36–55	.869 (.347)	−0.35	.397	1.901
Age: 56–75	.808 (.314)	−0.55	.377	1.731
Age: 76+	1.101 (.488)	0.22	.461	2.627
Worries: Own health	1.059 (.044)	1.38	.976	1.149
Worries: Others’ health	1.014 (.038)	0.39	.942	1.093
Worries: Lifestyle	1.032 (.047)	0.69	.943	1.130
Worries: Social relations	.970 (.044)	−0.65	.886	1.062
Intercept	.041 (.036)	−3.58	.007	.236
Negative Response
Gov. Crisis Management	.731 (.038)	−5.94	.659	.810
Rest. Civil Rights: Dum 2	.782 (.073)	−2.61	.651	.940
Rest. Civil Rights: Dum 3	.650 (.156)	−1.79	.406	1.041
Partner: Yes, living apart	.893 (.206)	−0.49	.568	1.403
Partner: No, single	.784 (.157)	−1.21	.529	1.163
Children	.966 (.118)	−0.28	.760	1.227
Female	.908 (.083)	−1.04	.758	1.087
Life Satisfaction	.976 (.038)	−0.60	.903	1.055
Satis. Leisure Activities	1.033 (.023)	1.42	.987	1.081
Belonging	1.015 (.022)	0.70	.972	1.060
Health	.938 (.073)	−0.81	.805	1.094
Work: Unemployed	2.003 (.603)	2.31	1.109	3.615
Work: Inactive/Retired	1.146 (.132)	1.18	.913	1.439
Age: 26–35	.972 (.243)	−0.11	.595	1.588
Age: 36–55	1.055 (.247)	0.23	.666	1.670
Age: 56–75	1.070 (.249)	0.29	.678	1.689
Age: 76+	.909 (.260)	−0.33	.519	1.594
Worries: Own health	.992 (.023)	−0.29	.947	1.040
Worries: Others’ health	1.039 (.022)	1.81	.996	1.084
Worries: Lifestyle	1.030 (.026)	1.16	.979	1.083
Worries: Social relations	1.003 (.025)	0.15	.954	1.055
Intercept	.477 (.238)	−1.48	.179	1.270
Pseudo R2	0.0235

Notes: *** Pr(>|z|) < 0,001; ** Pr(>|z|) < 0.01; * Pr(>|z|) < 0.05 | Reference for the restriction variables = ‘people who don’t think the restrictions are problematic at all and necessary’ | Reference for the partner variables = ‘people who share a household with their partner’ | Reference for the female variable = ‘male’ | Reference for the age groups = ‘14–25’ | Reference for the employment status variable = ‘employed’ | ‘RRR’ = Relative Risk Ratio | Log likelihood −2330.7528 (iteration 4) | Prob chi2 0.00000 | Lr chi2(42) 111.98.

Figure A1. Wave-to-wave challenge-generating the ‘measningful change’ variable. Note: 2018 = Reference datapoint |1 = 2019; 2 = the extra COVID-19 wave; 3 = 2020.

Table A5. Fixed-effects regression analysis – trust development from 2008 to 2021.

Social Trust	Coef. (Std. Err)	t (P > | t |)	95% Conf. Interval
			Lower	Upper
Wave
2009	.050 (.024)	2.02 *	.001	.098
2010	.052 (.025)	2.09 *	.003	.101
2011	.105 (.025)	4.14 ***	.055	.155
2012	.077 (.025)	3.06 **	.027	.127
2013	−.048 (.026)	−1.85	−.100	.002
2014	.105 (.024)	4.24 ***	.056	.154
2015	.019 (.025)	0.77	−.029	.068
2016	−.013 (.025)	−0.52	−.064	.037
2017	.165 (.025)	6.41 ***	.114	.215
2018	.193 (.025)	7.54 ***	.143	.243
2019	.061 (.026)	2.34 *	.010	.113
COVID-19	−.533 (.032)	−16.59 ***	−.596	−.470
2020	.252 (.026)	9.67 ***	.201	.303
2021	.032 (.026)	1.23	−.019	.084
Constant	6.165 (.020)	304.55 ***	6.125	6.205
Other values …
F	(14.25557 = ) 82.89
Prob > F	0.0000
R-square
Within	0.0130
Between	0.0005
Overall	0.0021
Rho	.6815
Nvalid	138798

NoteS: *** P(>|t|) < 0.001; ** P(>|t|) < 0.01; * P(>|t|) < 0.05 | Reference category for Age = ‘30–64-year-olds’ | Reference category for partner variables = ‘Single’ | Reference category for occupation = ‘Inactive’.

Table A6. Descriptive statistics: correlations matrix.

	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16
1 Social Trust	1
2 Belonging	.112	1
3 Health	.129	.221	1
4 Gov. Crisis	.200	.124	.060	1
5 Age	.104	.090	−.106	.160	1
6 Sex	.007	−.065	−.029	.014	−.022	1
7 Children	−.0125	−.031	.061	−.051	−.388	.032	1
8 Life in general	.180	.463	.359	.155	.171	−.027	−.071	1
9 Leisure activities	.099	.273	.247	.097	.112	−.019	−.111	.452	1
10 Restrictions perception	.014	−.061	−.005	−.320	−.028	−.033	.024	−.084	−076	1
11 Partner status	−.037	−.273	−.087	−.060	-.067	.106	−.017	−.185	−.068	.039	1
12 Occupation	.026	.009	−.096	.060	.489	.036	−.316	.079	.094	−.054	.092	1
13 Worry: Personal health	−.193	−.244	−.422	−.049	.079	.042	−.013	−.279	−.219	−.026	.020	.036	1
14 Worry: Health of close	−.169	−.165	−.211	−.016	−.123	.110	.030	−.201	−.170	−.055	.001	−.043	.584	1
15 Worry: Lifestyle change	−.208	−.324	−.226	−.140	−.129	.011	.019	−.361	−.278	.048	.070	−.057	.478	.403	1
16 Worry: Social relations	−.185	−.363	−.221	−.102	−.106	−.008	.038	−.343	−.271	.036	.061	−.035	.454	.393	.693	1

Note: The numbers correlate with variable names found to the right.

Figure A2. Wave-to-Wave Change in Social Trust of at least 1 Standard Deviation (%). Note: Nvalid in 2019 = 8732| COVID = 5669| 2020 = 15534.

Hale, T., Angrist, N., Goldszmidt, R., Kira, B., Petherick, A., Phillips, T., Webster, S., Cameron-Blake, E., Hallas, L., Majumdar, S., & Tatlow, H. (2021). A global panel database of pandemic policies (Oxford COVID-19 government response tracker). Nature Human Behaviour, 5(4), 529–538. https://doi.org/10.1038/s41562-021-01079-8

 PubMed Web of Science ®Google Scholar