Acceptance of COVID-19 vaccines among college students: a study of the attitudes, knowledge, and willingness of students to vaccinate

ABSTRACT

Universities are considered high risk areas for COVID-19 outbreaks given the crowded environment of campuses with high mobility and limited space. As such, vaccination is considered an essential intervention that could greatly reduce the incidence and spread of this deadly infectious disease. However, the willingness of college students to receive the COVID-19 vaccine varies significantly. Therefore, a study on the acceptance of COVID-19 vaccines in college students that explores the attitudes, knowledge, willingness, and key factors influencing vaccination acceptance is of great significance to improve vaccine coverage and control the pandemic. A cross-sectional survey was conducted on students from three universities in China. Descriptive statistics, independent sample t tests/one-way ANOVA (normal distribution), Mann-Whitney U tests/Kruskal-Wallis H tests (skewness distribution) and multivariate linear regression were performed. As a result, a total of 3,256 students participated in the survey. Students’ willingness to receive the COVID-19 vaccine was high (86%), and they had good knowledge of the vaccine (77.9%). However, they had a low-risk perception of COVID-19 and less positive attitudes toward vaccination (69.8%). The main influencing factors were sex, age, specialty, grades, living environment, spending level, traveling to risk areas, and family members’ vaccination experiences. We believed that to increase vaccination coverage among college students, more attention should be paid for students majoring in Science and Engineering, male students, those in the lower age group, students with low or very high economic levels, living in remote or rural areas, and family members having not received the COVID-19 vaccine.

KEYWORDS:

• Attitudes
• knowledge
• willingness
• COVID-19 vaccines
• college students

Introduction

COVID-19, caused by SARS-CoV-2, is characterized by its highly infectious nature and general susceptibility.¹ As a result, the spread of COVID-19 has been rampant and prevention of the disease has proven extremely difficult, with no specific prevention measures.² However, with the successful development of COVID-19 vaccines and large-scale vaccination campaigns, inoculation appears capable of facilitating prevention and control of the disease.³ According to the World Health Organization (WHO), 182 COVID-19 vaccines have entered clinical/preclinical trial phases. The Chinese government has placed great importance on the research and development of COVID vaccines. In July 2020, China officially launched emergency vaccination programs for people at high risk of exposure to COVID-19. Additionally, China pledged to vaccinate the entire Chinese population free of charge.⁴ While vaccination is one of the most conventional and effective strategies to limit the spread of infectious diseases,^5,6 increased vaccine hesitancy continues to be observed despite strict vaccination efforts.⁷ Underlying reasons include concern about the safety of vaccines, distrust of the government, and stigmatization, among others.^8,9 Willingness of the population to be vaccinated is essential to achieving higher prevention rates and control of the disease. Although the public is aware of the health threats posed by the spread of COVID-19, many people remain skeptical of the vaccines.¹⁰

Universities are considered high risk areas for COVID-19 outbreaks given the crowded environment of campuses with high mobility and limited space. Extensive vaccination of college students is important in the context of vaccine safety assessment and in improving vaccination coverage and control of the disease.⁸ Therefore, it is essential to evaluate the attitudes, knowledge, and willingness of college students to be vaccinated, as well as the factors that influence these aspects. Studies show that the willingness of college students to receive COVID vaccines varies from 53% to 94.7%.^8,11–14 Positive influencing factors include trust in public health experts,¹¹ lack of concern about side effects,¹¹ perception of the severity and fear of the pandemic,^8,15 scientific sources of information,^16,17 women, prosocial behavior,⁸ previous experience with influenza vaccination,¹⁴ and fear of transmitting the virus to family members.¹⁸ The predominant factors leading to hesitancy include fear of side effects,¹¹ negative attitude toward other vaccines,¹⁵ vaccine conspiracy theories,^18,19 insufficient data on vaccines and adverse reactions,²⁰ social stigma, and low economic status.⁸ Meanwhile, vaccine knowledge is positively correlated with vaccine acceptance.²¹ Some studies on COVID-19 vaccine knowledge have shown that college students across different countries and regions have different levels of vaccine knowledge, ranging from 20.8% to 65%.^13,22,23 However, most COVID-19 vaccines studies focus on vaccination willingness and ignore the influence of knowledge on vaccine acceptance. Sources of health information are also critical to vaccine acceptance.¹⁶ For college students in the digital age exposed to multiple sources of information (e.g., healthcare organizations, mass media, and social media), the information source of COVID-19 vaccines and the trust placed in those sources may influence students’ acceptance of vaccines. Vaccine acceptance was positively associated with scientists as information sources but negatively associated with pharmaceutical companies as sources,¹⁶ which is consistent with Sallam’s findings,¹⁹ but not with Superio’s.²⁴ There are also differences in vaccine willingness among students in different majors. Barello et al. found no significant difference in vaccination willingness between Italian medical students and non-medical students. Additionally, medical students had the same high vaccination willingness as non-medical students.¹² However, research by Punsalan showed that American medical students were reluctant to be vaccinated, which may be related to the hesitancy of medical professionals toward vaccination.²⁵ Szmyd et al. found that Polish medical students had higher vaccination willingness than non-medical students (92% vs. 59.4%).¹⁸ Moreover, most previous studies have been centered on fee-based vaccines; considering that China is implementing universal free vaccination, differences in student willingness to undergo vaccines under different policies warrants attention and study. While existing studies show vaccination willingness of college students at varying levels, it is clear that factors affecting vaccination acceptance needs to be explored further. In consideration of these, the purpose of this study was to investigate the attitudes, knowledge, and willingness of college students to be vaccinated against COVID-19 which will help develop and implement effective strategies to improve vaccine coverage in this population.

Methods

Study design and participants

This study adopted the stratified sampling method, snowball sampling method, and cluster sampling method to select college students from three universities in Mainland China. The estimated sample size was derived from the online Raosoft sample size calculator,²⁶ which was calculated based on the 5% error amplitude, 99% confidence, 50% response rate, and 1000,000 students (i.e., all college students in Shandong province in China). The minimum sample size of this study was 664 participants. The inclusion criteria were college students, age ≥18 years old, and informed consent. The exclusion criteria were part-time college students. A total of 3,256 college students participated in the survey.

Questionnaire

Based on COVID-19 vaccination guidelines from the National Health Commission of China,²⁷ WHO,²⁸ New York State Department of Health,²⁹ U.S. Food and Drug Administration,³⁰ and related literature,^15,31 we developed the “Attitudes, Knowledge and vaccination Willingness for the COVID-19 vaccine (AKW)” questionnaire which consisted of following sections.

1. Introductions of the questionnaire: including the purpose of this study, anonymity, confidentiality, guidelines for filling in the questionnaire, and contact information.

2. Baseline information (15 items): sex, age, nationality, specialty, grade, academic background in high school, parental family living environment, political affiliations, family economic conditions, monthly consumption, history of travel to high-risk areas, underlying diseases, vaccine status of family members, and side effects developed after receiving other vaccines.

3. Attitudes toward COVID-19 vaccines (11 items): influences of COVID-19, risk perception, vaccine acceptance, and concerns about COVID-19 vaccines. Each item was scored on a 5-point Likert scale, and the total score was between 11 and 55, where a higher score indicated a more positive attitude.

4. Knowledge of COVID-19 vaccines (9 items): (a) priority groups for vaccination, recommended age group for vaccination, correct methods, contraindications, adverse reactions, matters needing attention, and herd immunity. The knowledge dimension included single-choice and multiple-choice questions which was scored based on the correct rate. 5/0 points were assigned for each single-choice question and 1 point was assigned for each correct answer in multiple-choice questions. The total score was between 1 and 46, where a higher score indicated better knowledge. (b) The sources of acquired knowledge included mobile phone, TV, radio, computer, newspaper, school, community, relatives/friends, professionals, and others.

5. Vaccination willingness (8 items): vaccine selection, vaccination form, duration of protection, willingness, reasons, and vaccine prices. Two of these items were scored to determine the level of vaccination willingness, with scores ranging from 2 to 8; the higher the score, the stronger the willingness to be vaccinated. The other items were objective indicators expressed as percentages.

Validity and reliability

Four experts (i.e., two public health professors, one community nursing expert responsible for vaccination, and one nursing education expert) were invited to review the questionnaires and the scale was revised based on their opinions. The content validity of the questionnaire, as assessed by experts, was 0.98 (Supplemental file I-CVI data). A per-survey of 30 students of different majors was conducted, the outcome of which demonstrated that students believed items in the questionnaires to be easy to understand and acceptable. The Cronbach’s alpha coefficient of the questionnaire was 0.73 (Supplemental file Original data).

Data collection

Data were collected between April 1, 2021, and May 21, 2021. We selected two cities from the west and east Shandong province, and selected three universities from the top, middle and bottom rankings of universities in Shandong Province. We selected students by stratified sampling, snowball sampling, and cluster sampling. First, students were divided by major into four categories, including Humanities and Social Science, Natural Science, Science and Engineering, and Medicine (according to the guideline of China’s discipline classification list). Next, we sampled the second-level majors (total 89 majors) below the four categories and selected 25 majors (at a proportion of 28%). Under Humanities and Social Science, 10 majors were selected, including Applied Economics, Sociology, Education, Physical Education, Foreign Languages and Literature, Art, History, Management, Law, and Informatics. Under Natural Science, 5 majors including Mathematics, Physics, Geography, Biology, and Chemistry were selected. Under Science and Engineering, 6 majors were selected, including Mechanical Engineering, Information and Communication Engineering, Computer Science and Technology, Civil Engineering, Surveying and Mapping Science, and Environmental Science. Under Medicine, 4 majors were selected, including Basic Medicine, Clinical Medicine, Public Health, and Pharmacy. One class was randomly sampled from each grade.

We first communicated with the university staff to explain the purpose of this study and identify the majors and grades of the university students. Then the staff introduced us to qualified college counselors (i.e., full-time teachers in China who are responsible for managing the daily life and learning of college students) and we explained the purpose, content, and information required for the study. Counselors held class meetings with students to explain the survey, and students participated in the study after providing informed consent. Questionnaires were uploaded to the platform Questionnaire Star, a popular online survey website in China, and a QR code poster was generated. Posters were sent to each counselor by e-mail and counselors uploaded the poster to class groups created on the social media/messaging application WeChat. Finally, students scanned the QR code to complete the questionnaire on the Questionnaire Star platform. After the sampling of one major was completed, the same method was utilized to sample students from other schools and majors.

Quality control

The research team for this study consisted of five members. After the target class was selected, counselors held a class meeting and researchers explained the significance of this survey to the students, introduced the method of questionnaire completion, and addressed matters needing attention to ensure the attention of students. After students began to complete the questionnaire, researchers tracked the number of questionnaires in the network background. When the number of questionnaire submissions declined, counselors would issue reminders to students in the class WeChat group. Student cadres played an important role in assisting counselors, as their active cooperation provided an example to other students. Researchers and counselors communicated with student cadres, who encouraged students to participate in the survey from a peer-to-peer aspect. IP address restriction technology was used to ensure that students with the same IP address could complete the questionnaires once.

Ethics approval

This study was approved by the Ethics Committee of Shandong First Medical University (registration number: R202105170156). Informed consent was obtained from each student online. Informed consent included research purposes, voluntary participation, anonymity, confidentiality, and an informed consent option. Only students who agreed to participate could proceed to complete the questionnaire. Students were permitted to withdraw from the survey at any time.

Statistical analysis

Statistical analysis was performed using SPSS Statistics for Windows, Version 19.0 (SPSS Inc; Armonk, NY: IBM Corp.). Frequency distribution was used for count data, which were described as mean ± SD (standard deviation). As the Kolmogorov-Smirnov test for normality is sensitive to a large sample size, and p < .05 is prone to occur, we used a P-P plot and histogram to analyze data normality.³² T-test or variance analysis was conducted if data were normally distributed. In terms of skewed data, Mann-Whitney U test or Kruskal Wallis H test was employed to detect the social-demographic differences. Demographic data were used as independent variables for multiple linear regression to identify the influencing factors of attitudes, knowledge and vaccination willingness. p < .05 indicated statistical significance (double tail).

Results

A total of 3,277 questionnaires were collected, of which 12 were incomplete or lacking basic information and 9 had the same scores in all items. After elimination, a total of 3,256 valid questionnaires were collected, with an effective rate of 99.4%. Among them, 1378 students were Humanities and Social Science majors, 566 students were Natural Science majors, 821 students studied Medicine, and 491 students studied Science and Engineering. The total proportion of science to liberal arts students (Science vs. Liberal arts: 58.5% vs. 41.6%) conformed to the characteristics of Chinese students’ majors.

Basic information of students

In this study, the male to female ratio was approximately 1:1.6, and most participants were aged 18–21 years old (84.7%). Most students were of Han nationality (98.5%). The proportion of students majoring in Science and Engineering, Humanities and Social Science, Medicine, and Natural Science was approximately 1:2.8:1.7:1.2. More than half of students had a science background (58.5%), which is consistent with the proportion of existing majors in China. Most of the students’ families lived in rural areas (47.9%) and new urban communities (39.5%). A majority of students were from medium-income families (75.8%) and most spent approximately 800–1,500 yuan per month (66.4%). Nearly all students had no underlying health conditions (98%) and had not visited high and medium-risk COVID-19 areas in the past 6 months (99.2%). Most students had not yet received the COVID-19 vaccine (81.6%). Nearly half of students’ family members had received vaccines (41.4%). Further details are included in Table 1.

Table 1. Demographics, univariate and multivariate analyses of factors associated with students’ attitudes, knowledge, and willingness to vaccinate (N = 3256)

			Attitudes				Knowledge				Vaccination willingness
Variable	Content	N (%)	M ± SD	t/F	p	β	Scoring rate	Z/H	p	β	Scoring rate	Z/H	p	β
Sex
	Male	1230(37.8)	38.26 ± 5.31	−0.97	.33		73%	−12.70	<.001	−0.38**	84.7%	−2.56	.01	−0.13**
	Female	2026(62.2)	38.44 ± 4.38				80.9%				86.8%
Age(years)
	18~	519(15.9)	38.54 ± 4.73	2.30	.04		75.7%	55.56	<.001	0.03*	86.1%	18.30	.003	0.04**
	19~	978(30)	38.22 ± 4.69				76.4%				85%
	20~	742(22.8)	38.56 ± 4.82				77.4%				85.4%
	21~	520(16)	37.93 ± 4.71				80.6%				86.9%
	22~	314(9.6)	38.41 ± 4.72				81.7%				86.7%
	23~	183(5.6)	39.09 ± 4.98				80.3%				89.2%
Nationality
	Han	3206(98.5)	38.36 ± 4.74	−1.27	.20		78%	−1.84	.07		86%	−1.21	.23
	Others	50(1.5)	39.22 ± 5.72				74.1%				88%
Specialty
	Science and Engineering	491(15.1)	38.58 ± 4.97	1.43	.23	0.59*	74%	81.09	<.001		82.1%	142.52	<.001
	Humanities and Social Science	1378(42.3)	38.47 ± 4.77				77.6%				84.7%
	Medicine	821(25.2)	38.09 ± 4.65				81.2%			0.2**	87.2%
	Natural Science	566(17.4)	38.36 ± 4.66				77.4%				90.7%			0.1*
Grade
	Class of 2017	216(6.6)	38.15 ± 4.88	0.40	.75		81.5%	54.85	<.001	−0.05*	86.6%	2.61	.46	0.05**
	Class of 2018	522(16)	38.23 ± 4.71				80.7%				86.4%
	Class of 2019	789(24.2)	38.41 ± 4.87				78.7%				86.6%
	Class of 2020	1729(53.1)	38.42 ± 4.70				76.3%				85.5%
Academic background in high school
	Liberal Arts	1353(41.6)	38.69 ± 4.63	3.19	.001	0.67**	79%	−3.38	.001		85.8%	−1.86	.06
	Science	1903(58.5)	38.15 ± 4.82				77.1%				86.1%
Parental family living environment
	New community	1286(39.5)	38.52 ± 4.78	1.08	.36		79%	17.64	.001		86.9%	14.39	.006
	Old community over 30 years	220(6.8)	38.62 ± 4.84				77.4%				84.8%
	Village in the city	156(4.8)	38.36 ± 4.53				76.4%			−0.2*	87.4%
	Public rental housing	36(1.1)	38.92 ± 5.38				72.9%				85.8%
	Rural area	1558(47.9)	38.20 ± 4.73			−0.5**	77.3%			−0.16**	85.3%
Politics affiliations
	Party member	142(4.4)	38.17 ± 6.21	1.73	.18		78.1%	27.94	<.001		88.1%	17.78	<.001
	League member	2560(78.6)	38.45 ± 4.65				78.7%				86.3%
	Other	554(17)	38.05 ± 4.78				74.2%				84%
Family economic conditions
	Good	158(4.9)	37.80 ± 6.02	2.73	.07	−0.45*	75.6%	4.07	.13		85.4%	4.00	.14
	Medium	2469(75.8)	38.32 ± 4.54				78.2%				86.3%
	Poor	629(19.3)	38.70 ± 5.18				77.3%				84.9%
Monthly spending level (RMB)
	<800	462(14.2)	38.75 ± 5.13	1.88	.11	−0.31*	76.1%	25.94	<.001	−0.07*	84.4%	9.30	.05	−0.05*
	800–1500	2161(66.4)	38.38 ± 4.60				78.5%				86.3%
	1500–2500	541(16.6)	38.16 ± 4.54				78.5%				86.7%
	2500–3500	61(1.9)	37.34 ± 4.99				74.4%				84.4%
	>3500	31(1)	37.65 ± 9.92				62.3%				79.4%
Travel to above medium risk areas in the past six months
	Yes	25(0.8)	40.52 ± 10.08	1.07	.29	2.07*	63.1%	−3.14	.002	−0.75**	80%	−1.23	.22
	No	3231(99.2)	38.35 ± 4.69				78%				86%
Underlying diseases
	Yes	66(2)	37.42 ± 5.90	−1.63	.10		79.8%	−0.95	.34		82.6%	−1.45	.15
	No	3190(98)	38.39 ± 4.72				77.9%				86.1%
The COVID-19 vaccine was inoculated
	Yes	598(18.4)	38.27 ± 4.97	−0.56	.57		79%	−2.70	.007		94.7%	−19.06	<.001	0.65**
	No	2658(81.6)	38.39 ± 4.70				77.7%				84%
Family members received the COVID-19 vaccine
	Yes	1347(41.4)	38.58 ± 4.69	2.15	.03	0.45*	79.3%	−5.03	<.001	0.15**	88.6%	−9.67	<.001	0.14**
	No	1909(58.6)	38.22 ± 4.79				76.9%				84.1%
You or your family members had side effects after receiving other vaccines
	Yes	44(1.4)	37.91 ± 5.49	−0.65	.52		77.9%	−0.54	.59		77.3%	−3.44	.001	−0.57**
	No	3212(98.7)	38.38 ± 4.74				77.9%				86.1%

(Linear regression analysis was carried out after normalizing knowledge and willingness scores.)

*p < .05, **p < .01.

Score of attitudes, knowledge and vaccination willingness to vaccinate

A normal analysis was conducted on a P-P plot and a histogram of attitudes, knowledge, willingness, and total scale. The results showed that dimensions of knowledge and willingness had skewed distributions, while dimensions of attitudes and total scale had normal distributions. Therefore, the median of knowledge and willingness was used to describe the data distribution, and other dimensions were represented as the Mean ± SD. The score of the attitude dimension was 38.37 ± 4.75 and the scoring rate was 69.8%. In the attitude dimension, “being infected with COVID-19 has a greater impact on the surrounding people or environment” had the highest scoring rate (89%), while the lowest scoring rate was “the risk of getting COVID-19 is high” (46.2%). The scoring rate of the knowledge dimension was 77.9%. The item “the priority population for vaccination” had the highest scoring rate (87.9%), while “correct vaccination method” had the lowest scoring rate (62.3%). The scoring rate of the willingness dimension was 86%. The leading reasons for students’ willingness to get vaccinated were “support national strategies” (89%), “belief in the vaccine” (57.5%) and “organizing group vaccinations at school” (55%). The top three reasons for students’ unwillingness/hesitation to receive vaccines were “worrying about side effects” (50.8%), “uncertainty about vaccines” (42.3%), and “wish to observe the vaccination effect on other people” (38.1%). In terms of vaccine types, students were more willing to choose domestic vaccines (59.7%). Most students expressed hope that schools could organize the student to get vaccinated together (88%) and many hoped that the protection given by vaccines would last more than 10 years (38.1%). Moreover, 76.9% of students were even willing to pay for vaccination, within an acceptable price range of <100 yuan (81.6%) (Table 2 and Figure 1).

Table 2. Scores of attitudes, knowledge and willingness to vaccinate (N = 3256)

Dimension	Item	Score range	Mean ± SD	P50(P25, P75)	Scoringrate (%)
Attitudes		11–55	38.37 ± 4.75		69.8%
	Do you think that contracting COVID-19 has a significant impact on your health?	1–5	3.91 ± 1.22		78.2%
	Do you think it will affect the people around you or the environment if you get COVID-19?	1–5	4.45 ± 0.90		89%
	Do you think that the current pandemic is serious?	1–5	3.12 ± 0.84		62.4%
	Do you think that the pandemic will recur in China?	1–5	3.08 ± 1.01		61.6%
	How much has the pandemic affected your life in the past 6 months?	1–5	3.55 ± 0.93		71%
	How much will the pandemic affect your life in the next 6 months?	1–5	2.89 ± 0.85		57.8%
	Do you think that you are at high risk of contracting COVID-19?	1–5	2.31 ± 0.98		46.2%
	Do you think that you can get prevention from COVID-19 by vaccination?	1–5	3.83 ± 0.61		76.6%
	Do you think that COVID-19 vaccines available on the market are safe?	1–5	3.65 ± 0.69		73%
	Do you think that the vaccine is effective?	1–5	3.74 ± 0.65		74.8%
	How much do you care about vaccine-related information?	1–5	3.85 ± 0.82		77%
Knowledge		1–46		38(32, 41)	77.9%
	Which are the priority groups for vaccination?	0–8		8(7,8)	87.9%
	What is the recommended age group for vaccination?	0/5		4(4, 4)	76.2%
	What are the correct methods of vaccination?	0–8		5(3, 7)	62.3%
	What are the contraindications for the vaccine?	0–5		5(4,5)	85.8%
	Regarding adverse reactions to vaccination, which of the following statements are true?	0–5		5(4,5)	83.2%
	Matters needing attention	0–5		5(4,5)	84%
	How can herd immunity be achieved through vaccination?	0/5		5(5, 5)	79.8%
	In general, how familiar are you with the COVID-19 vaccine?	1–5		3(3, 4)	67.6%
Vaccination willingness		2–8		7(6, 8)	86%
	Would you like to receive the COVID-19 vaccine?	1–6		5(5, 6)	85.2%
	Would you be willing to receive the COVID-19 vaccine if you are charged for it in the future?	1–2		2(2, 2)	88.5%
Total scale		14–109	81.09 ± 9.55		74.4%

Figure 1. Color-coded bars from left to right represent seven different vaccination intentions: “Information sources”; “Vaccine choice”; “Vaccination form”; “Desired vaccine protection periods”; “Reasons for willingness to be vaccinated”; “Reasons for unwillingness/hesitation to be vaccinated”; and “Vaccine prices.” The different options are described below the same color bar chart. The number of students choosing one item/total number of students (3256) = the proportion of this item. The higher the proportion (i.e., more students choosing this item), the higher the willingness of students to choose this option.

Differences between groups and influencing factors

Demographic characteristics of participants were treated as the grouping variables and the differences between groups were analyzed. Since scores of knowledge and willingness were skewed, a U test or H test was used. Attitudes and the total scale were normally distributed, hence a t-test or analysis of variance was used. According to the statistical results, in terms of attitudes, differences in the scores of ages, academic background in high school, and vaccinated family members all had statistical significance. The attitudes score of Liberal Arts students in high school was higher than that of Science students (Liberal Arts vs. Science: 38.69 vs. 38.15, p = .001). Attitudes of students whose families have been vaccinated were more positive (yes vs. no: 38.58 vs. 38.22, p = .03). With regards to knowledge, several factors were significantly different between groups (with the exception of nationalities, family economic status, underlying health conditions, and side effects of other vaccines experienced by family members). Scores of female students were significantly higher than those of male students (female students vs. male students: 80.9% vs. 73%, p < .001), and scores gradually increased with age (18~, 19~, 20~, 21~, 22~, 23~: 75.7%, 76.4%, 77.4%, 80.6%, 81.7%, 80.3%, p < .001). Students who had not visited high risk areas within the past 6 months had significantly higher knowledge scores (yes vs. no: 63.1% vs. 78%, p = .002).

In terms of willingness, grouping variables included sex, age, specialty, family-living environment, political affiliations, whether students or their family members have been vaccinated, and side effects of other vaccines. The score differences between groups were statistically significant. Female students were significantly more willing to be vaccinated than male students (female students vs. male students: 86.8% vs. 84.7%, p = .01). For students >19 years old, the willingness to receive vaccines increased with age (19~, 20~, 21~, 22~, 23~: 85%, 85.4%, 86.9%, 86.7%, 89.2%, p = .003). The knowledge and willingness of medical students were significantly higher than those of students in other majors. Party members were more willing to get vaccinated than League members and the masses (Party members vs. League members and the masses: 88.1% vs. 86.3% and 84%, p < .001). Students whose families experienced no side effects of other vaccines showed a stronger willingness to be vaccinated (no vs. yes: 86.1% vs. 77.3%, p = .001). Compared with other classes, students in the class of 2020 had the lowest score of knowledge and willingness to be vaccinated. Students whose monthly spending level was between 800 yuan and 2500 yuan scored higher in knowledge and willingness than those below 800 yuan and above 2500 yuan. Overall, groups with different sexes, specialties, grades, academic backgrounds, environments, political affiliations, monthly spending levels, and vaccination experiences were significantly different in terms of acceptance of COVID vaccines.

Regression analysis showed that significant influencing factors of attitudes included specialty, academic background in high school, family-living environment, family economic status, monthly spending level, travel to high-risk areas within the previous 6 months, and family vaccinations. Among these, specialty, academic background in high school, travel to high-risk areas, and family vaccinations were positive predictive factors of attitudes. Influencing factors of knowledge included sex, age, specialty, grade, family-living environment, monthly spending level, travel to high-risk areas within the past 6 months, and family vaccinations. In this case, age, specialty, and family vaccinations were the positive predictive factors of knowledge. Significant influencing factors of vaccination willingness included sex, age, specialty, grade, monthly spending level, student or family COVID-19 vaccinations, and side effects of the vaccine. Among these factors, age, specialty, grade, student or family vaccination were positive predictive factors for vaccination willingness. The results are shown in Table 1.

Discussion

Universities are facing great challenges and difficulties in preventing and controlling COVID-19 due to its complexity and unpredictability. As such, vaccination is considered an essential intervention that could greatly reduce the incidence and spread of this deadly infectious disease.^33,34 Therefore, a study on the acceptance of COVID-19 vaccines in college students is of great significance to improve vaccine coverage and control the pandemic. The questionnaire used in this study was self-designed. The questionnaire framework was established by interviewing stakeholders and the initial version of the questionnaire was developed by referencing to existing literature. Items and dimensions of the survey were improved by the Delphi method, and the reliability and validity of the questionnaire were tested by practical measurement. The final questionnaire content validity I-CVI was 0.98, Cronbach’s alpha coefficient was 0.73, and the correlation coefficients between dimensions and the total questionnaire were 0.43–0.85. We believe that the questionnaire is reliable and valuable for use in this research.

In this study, we found that attitudes of university students had the lowest scoring rate. Due to the implementation of strict measures, the COVID-19 epidemic in China has begun to slow down. As a result, many students hold the belief that another wave of the pandemic is unlikely, and the risk of infection is low. However, as a highly infectious, rapid and wide spreading disease, COVID-19 has become a global problem of epic proportions.³⁵ With the longevity of the pandemic, rapid spread of COVID-19 could return to China at any time if measures are relaxed. The pandemic has also been aggravated by novel variants, particularly in India. Reiter et al. found that individuals concerned about COVID-19 were more willing to be vaccinated.³⁶ Although college students did not appear to be aware of the current severity and had decreased risk vigilance, the majority (89%) understood that “If they are infected with COVID-19, they will have a great impact on the surrounding people and the environment.” Students’ recognition of the safety and effectiveness of vaccines was satisfying, like existing research findings.¹⁷

This study showed that attitudes were significantly influenced by students’ academic background in high school. Students with a Liberal Arts background in high school had more positive attitudes toward vaccines than students of Science. Odriozola-González et al. proposed a similar conclusion.³⁷ As students of Liberal Arts experienced greater anxiety and depression caused by the pandemic than students of Science, it can be understood that Liberal Arts students were more concerned about the vaccine and therefore had higher vaccine recognition.³⁶ Hence, school educators should emphasize the severity of the pandemic to improve vigilance among all students. However, in consideration of the differences in psychology and learning style between students of Liberal Arts and Science, targeted measures by educators to improve students’ attitudes toward the pandemic and vaccines is recommended.

Drawing upon the Knowledge-Attitude-Practice theory (KAP), this innovative study integrated the knowledge of COVID-19 vaccines into the questionnaire, which was rare in existing researches. The results demonstrated that knowledge was positively correlated with attitudes and willingness (r = 0.13**, 0.24**; Table 3), indicating that increased vaccine knowledge could effectively improve attitudes and willingness. Overall, most students had good knowledge of vaccines, and questions measuring students’ knowledge about “the key population for vaccination” (87.9%), “vaccine contraindications” (85.8%), and “vaccination precautions” (84%) scored the highest. At present, China is promoting vaccination nationwide. According to the national policy, universities are engaging in collective vaccination, whereby vaccination is followed by lectures on vaccines. Therefore, students had a good grasp of vaccine knowledge. However, low scores were reflected in students’ knowledge regarding the “correct vaccination methods” (62.3%), “self-evaluation of familiarity with vaccine knowledge” (67.6%), and “the age group suitable for vaccination” (76.2%). The correct method of vaccination covered the injection type, vaccination times, and injection choices. Only 38.5% of the students knew that COVID-19 Nucleic Acid PCR Tests were not required before vaccination, and most believed that they were required to get a COVID-19 Nucleic Acid PCR test before receiving a vaccine. The fear of testing may also affect students’ willingness to receive vaccines.³⁸ In terms of age and grade levels, senior students were better at acquiring and absorbing new knowledge, leading to higher scores than juniors.³⁹ In follow-up interviews, we found that the first-year students, having just entered college from high school, were still adjusting their learning style from exam-oriented learning to interest-based learning. As such, although most first-year students have a more positive attitude toward vaccines, their thinking patterns have not yet changed and they do not know how to obtain vaccine knowledge. This is an important factor educators should consider in collective vaccination programs. Interestingly, although individuals who had visited high-risk areas should naturally have shown more interest in vaccines and an increased willingness to vaccinate,⁴⁰ we found that students who had traveled to high-risk areas had less knowledge than those who had not. One possible reason for this finding was that they only traveled to these areas (as opposed to residing in the high-risk areas) and the short duration of the experience did not impart sufficient concern. Moreover, the lack of professional knowledge guidance might also contribute to the low knowledge scores. According to these research findings, schools should carry out vaccine knowledge training in various forms (e.g., lectures, leaflets and knowledge contests), aiming to fill the knowledge gap and improve students’ willingness to vaccinate.

Table 3. The correlation betweenattitudes, knowledge, willingness and total scale

	Attitudes	Knowledge	Willingness
Attitudes	1
Knowledge	0.13**	1
Willingness	0.25**	0.24**	1
Total scale	0.62**	0.85**	0.43**

*p < .05, **p < .01.

Among the 3,256 students surveyed, 81.6% had not been vaccinated and just 18.4% had. However, this vaccination rate was found to be higher than the average rate in China (16.9%).⁴¹ The willingness of non-vaccinated students was 82.7%, an upper level in the existing researches,^8,15,42,43 indicating a strong willingness to be vaccinated. Considering the requirement of the second and third doses, we also investigated the willingness of vaccinated students to receive the second dose. The results showed a strong willingness (95.8%), indicating the integrity of the follow-up vaccination. Approximately two-thirds of students indicated a preference for vaccines that had been developed in China, demonstrating a high trust in domestic vaccines. China currently has three vaccination forms, including company or school-organized vaccination, community-organized vaccination, and individual vaccination by visiting the hospital independently for vaccination. In this study, 88% of students chose to receive vaccines on campus for convenience. On-site collective vaccination was shown to provide encouragement for students who were hesitant about vaccination. However, if students who had vaccine hesitancy were passively vaccinated, they would experience discomfort or mental issues.^44,45 Therefore, educators should understand students’ willingness to vaccinate, and identify reasons for vaccine hesitancy from different perspectives, such as social psychology, family environment, cultural background, and physical condition,⁴⁵ to perform a targeted intervention on students. In this study, the monthly spending level (representing the economic status) was another critical influencing factor of AKW. Students with medium economic levels received higher scores in AKW, while students with either too high or too low economic levels obtained unsatisfactory scores. This finding was not consistent with existing research.⁸ Disadvantaged people with low economic levels may receive more attention from society, while economically advantaged people are essentially ignored. This study suggests that besides focusing only on individuals with low economic levels, the vaccination willingness of the economically advantaged should also be considered. Moreover, 76.9% of the students expressed their willingness to pay for vaccination, with an acceptable price point considered to be <100 yuan (81.6%). Most students were from medium-income families that spend 800–1,500 yuan each month; hence, 100 yuan was an affordable vaccine price. This finding had policy implications for the follow-up pricing strategy, indicating that a COVID-19 vaccine should be reasonably priced to ensure equity in healthcare access.

Specialty was another important factor influencing willingness to vaccinate. The willingness of students majoring in Medical and Humanities were significantly higher than that of students from other majors. Healthcare providers were the most trusted source of information, and medical students played a key role in providing information and consultation on COVID-19.¹³ For other students, the sources of vaccine information and their trust level in these sources was found to affect their willingness.¹³ Students’ willingness was significantly affected if they or their family members had side effects in response to other vaccines, consistent with previous studies.⁴⁶ In line with this, students who had families experiencing no side effects showed a stronger willingness to vaccinate. Adverse events of family members related to vaccines led students to have negative feelings toward the COVID-19 vaccine, leading to vaccine hesitancy and affecting the vaccination rate. With regards to the school-living students, the influence of classmates’ vaccination was also important. In communication with students, it was found that junior students tended to listen to the opinions of their families, while senior students were more influenced by their peers. Class cadres are a group of students with good academic performance and credibility. Educators can improve vaccine acceptance among class cadres to promote the willingness of other students through class cadres’ role as models. At the same time, we also found that in majors with significant sex imbalance (e.g., nursing, mechanical engineering), minority group students, even seniors, were more likely to take advice from family members than peers. The reason may be that in traditionally female/male dominated occupations, minority groups of the opposite sex sometimes felt isolated and unwilling to communicate with classmates.⁴⁷

Sex was an important influencing factor of AKW, and female students were better than male students in attitudes, knowledge, and willingness. During the pandemic, women bore more pressure and had a higher physiological burden than men, and they might face a higher risk of COVID-19 infection.⁴⁸ However, women also had a higher risk awareness and a stronger willingness to vaccinate, enabling them to obtain vaccine information. The living environment also affected AKW, and students living in urban areas had higher vaccine acceptance than those in villages, public rental housing, and rural areas, which is consistent with the findings of previous research.⁴⁹ Residents in urban areas had more access to information and resources during the pandemic. Thus, the government should pay more attention to people living in remote and rural areas to ensure the equity of access to public resources.

There were three limitations to this study. First, the influencing factors need to be further explored and the follow-up study could include more social factors, such as campus environment, cultural backgrounds, vaccine publicity efforts, and government credibility. Second, with the increase of grade level, students’ participation in the survey gradually decreased, which resulted in a low proportion of senior students. Third, this study was conducted just three months after the launch of COVID-19 vaccines, and therefore, results may only reflect the initial willingness of students. Future research should be conducted at a later stage in the vaccination campaign, to further examine willingness and its relationship with the vaccination behavior.

Conclusions

Through this study, we measured the attitude, knowledge, and willingness to receive COVID-19 vaccines among college students. Although strong willingness and good knowledge were reported, students had low-risk perception of COVID-19 and less positive attitudes toward vaccination. Sex, age, specialty, grades, family-living environment, monthly spending level, travel to risk areas, and family members inoculation experiences were the predominant influencing factors. Significant differences were observed among groups with different majors, academic background in high school, sex, age, and economic level. These results suggest that targeted intervention should be conducted on students in the fields of Science and Engineering, male students, low age groups, students with low or very high economic level, students living in remote or rural areas, and family members having not received COVID-19 vaccines. This study offers implications and suggestions for schools to conduct greater education and publicity of COVID-19 vaccines. Future studies should expand the scope of research to include individuals from different countries and regions, which could provide a more comprehensive and objective theoretical basis for improving the vaccination rate.

Contributions

Ning Jiang: Conceptualization, methodology, investigation, acquisition and analysis of data, writing - original draft. Pengfei Gu: acquisition and analysis of data, revising the the article. Ke Liu: investigation, acquisition and analysis of data, revising the the article. Na Song: conceptualization, methodology, acquisition and analysis of data, revising the the article. Xiaolian Jiang: conceptualization, methodology, acquisition and analysis of data, revising the the article. All authors have seen and approved the final version of the manuscript being submitted. They warrant that the article is the authors’ original work, hasn’t received prior publication and isn’t under consideration for publication elsewhere.

Acknowledgments

Our thanks should go to all students in this survey for their time and sharing their experience.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2021.2013077.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.