The demand for a COVID-19 vaccine in Kenya

Figures & data

Table 1. Characteristics of Survey Respondents

Characteristic	Respondents mean (SD)
Mean (SD) Age (years)	30.05 (7.94)
Mean (SD) Household size (people)	4.11 (2.64)
Income (KSh)^a
Less than 3,000	0.13 (0.34)
3,000 to 6,000	0.10 (0.30)
6,001 to 8,000	0.06 (0.25)
8,001 to 10,000	0.10 (0.30)
10,001 to 17,000	0.16 (0.37)
17,001 to 25,000	0.17 (0.37)
25,001 to 40,000	0.13 (0.34)
More than 40,000	0.14 (0.35)
Female	0.37 (0.48)
College educated	0.87 (0.34)
Possess health insurance ^b	0.72 (0.45)
Employed ^a	0.73 (0.44)
Rural residence	0.24 (0.43)
Religious believe
Christian	0.95 (0.23)
Other	0.05 (0.23)
Region of residence
Central	0.11 (0.32)
Coast	0.08 (0.27)
Eastern	0.07 (0.25)
Nairobi	0.39 (0.49)
Nyanza	0.08 (0.28)
Rift Valley	0.18 (0.39)
Western	0.05 (0.22)

*Sample size 963.

Table 2. Turnbull willingness to pay distribution functions (n = 963)

		Turnbull CDF
Duration (years)	1	20	1	20
Efficiency (%)	70	70	98	98
Bid Range ($)
0–1.49	0.148	0.166	0.099	0.139
1.49–3.12	0.169	0.237	0.099	0.139
3.12–12.09	0.365	0.330	0.296	0.294
12.09–30.23	0.527	0.429	0.427	0.389
30.23–60.45	0.647	0.608	0.574	0.509
60.45–120.90	0.785	0.695	0.681	0.651
120.90–181.35	0.846	0.749	0.781	0.798
181.35+	1.000	1.000	1.000	1.000

Table 3. Estimation results of willingness to pay models for a COVID-19 vaccine in Kenya

	Model 1		Model 2		Model 3
Variable	Coefficient		Coefficient		Coefficient
Constant	−0.516***	(0.137)	−1.398***	(0.501)	−2.066***	(0.727)
Duration of protection	0.218*	(0.163)	0.012*	(0.009)	0.258*	(0.167)
Efficiency of protection	0.348**	(0.163)	0.012**	(0.006)	0.404***	(0.168)
Contracting probability					0.001	(0.004)
Hospitalized probability					0.006**	(0.007)
Death probability					0.007	(0.007)
Age					−0.024**	(0.011)
Household size					0.118***	(0.037)
Income (USD)					0.002**	(0.001)
Gender (Female = 1, Male = 0)					−0.413**	(0.178)
College education (Yes = 1, No = 0)					0.475**	(0.243)
Health insurance (Yes = 1, No = 0)					0.052	(0.192)
Employment status (Employed = 1, Other = 0)					0.023	(0.058)
Knowledge score					0.272	(0.221)
Household location (Rural = 1, Urban = 0)					0.264*	(0.204)
Religion (Other = 1, Christian = 0)					0.070	(0.396)
Region of Residence
Nairobi					0.460**	(0.238)
Rift Valley					0.357**	(0.204)
Sample size	963		963		902
Alpha parameter	0.519		0.519		0.533
Wald Chi^2	6.412		6.412		58.230
P-value (Prob>Chi^2)	0.040		0.040		<0.001
Log likelihood	−1169.990		−1169.990		−1071.104

The parametric model used is a Weibull distribution with parameters ɑ ˄ σ, $\sigma =X_i\prime$β. Standard error in parenthesis. Dummy variables used as base are Western for region and Christian for religion. *p < 0.10 **p < 0.05 ***p < 0.01. Model 1 uses dummy explanatory variables for duration of protection and efficiency (higher levels take a value of 1). Model 2 uses the actual values of efficiency (70 and 98) and duration (1 and 20) as explanatory variables. 

Table 4. Mean willingness to pay (USD) for hypothetical COVID-19 vaccine

Attribute	Parametric mean WTP (n = 963)		Nonparametric lower bound mean WTP (n = 963)
1 year protection, 70% efficacy	102.00	(17.36)	49.81	(5.20)
20 years protection, 70% efficacy	131.96	(23.85)	64.37	(6.27)
1 year protection, 98% efficacy	152.83	(29.31)	64.90	(5.87)
20 years protection, 98% efficacy	197.72	(36.60)	68.25	(5.73)

Standard errors in parenthesis.

Figure 1. Reasons for rejection of bid scenarios for a COVID-19 vaccine (n = 158)

Figure 2. Aggregate demand curves for the COVID-19 vaccine

Table A1. Respondents COVID-19 perceptions, knowledge, and vaccine acceptability

Variable	Mean	Standard deviation
What do you think is your probability of contracting COVID-19?^a (n = 930)	28.07	24.52
What do you think is your probability of being hospitalize in case of contracting COVID-19? ^a (n = 929)	44.80	35.24
What do you think is the mortality rate in case of contracting COVID-19? ^a(n = 940)	17.63	14.11
Knowledge score with respect to recommendation to reduce spread, forms of transmission and symptoms of COVID-19	10.07	1.59

Notes. Sample size n = 963, except when noted. ^a Sample size as noted because of respondents skips.

Table A2. Responses to double-bounded dichotomous questions (n = 963)

Question			Responses
First discrete choice question		Yes				No
Percentage of respondents		58.36				41.64
Second discrete choice question	Yes		No		Yes		No
Percentage of respondents	25.57		74.43		23.94		76.06