The Search Between Two Worlds: Motivations for and Consequences of U.S.-Dwelling Chinese’s Use of U.S. and Chinese Media for COVID-19 Information

Abstract

As the COVID-19 pandemic continues to inflict damage throughout the world, some minority groups are bearing a disproportionate share of its impacts. We concentrated on one such group, U.S.-dwelling Chinese, who have had to cope with challenges related to acculturation, health, safety, and racism. Recognizing that health information seeking was an essential step in helping maintain and improve health behaviors, we conducted a two-wave longitudinal study (N = 1,284) to examine the various factors predicting U.S.-dwelling Chinese’s use of U.S. and Chinese media for COVID-19 information as well as the consequences of their information seeking. Overall, we found that acculturation, accuracy (i.e., information insufficiency) and defense (i.e., conspiratorial beliefs) motivations, trust in media, and perceived information gathering capacity played a key role in explaining information seeking from an intercultural viewpoint, and that the use of U.S. and Chinese media was associated with different health behaviors. These findings contribute to theory and practice in a variety of ways.

Health information seeking is a process in which individuals purposively acquire information from selected channels to guide their health-related decision-making (Freimuth et al., 1989). Having sufficient health information is often considered an essential first step that dictates the number and type of actions that will be initiated and maintained when the need for health management arises (Chen et al., 2010). Studies have shown that being able to obtain relevant health information and develop adequate health knowledge leads to positive health outcomes, including better health status, greater self-efficacy, and reduced hospitalization rates (Sørensen et al., 2012).

For immigrants and sojourners (i.e., people living someplace temporarily) in the United States, seeking health information has always been a major challenge. When accessing health information, many immigrants and sojourners face numerous barriers, including language, cultural, economic, technological, and educational barriers, as well as an unfamiliarity with U.S. health care systems (Chen et al., 2010). Difficulties in obtaining health information among immigrants and sojourners translate into worse health status, less access to health care, and lower insurance rates when compared with U.S.-born residents (Zhao, 2010). Despite the existence of recommendations for improving access to health information among immigrants and sojourners in the United States, any understanding of the predictors and consequences of their information seeking is limited.

When it comes to global pandemics, health information seeking can be even more difficult for certain immigrant and sojourner populations because these groups face additional hardships. The COVID-19 pandemic has not only wreaked havoc on people’s health but also stirred up xenophobia and racism (Callaway et al., 2020). The number of verbal and physical assaults and hate crimes against Asians in the United States has soared during the COVID-19 pandemic (Asian Pacific Policy and Planning Council, 2020). This hostility against Asian immigrants and sojourners clearly increases the burden shouldered by these groups, who have to deal with the double threat of the virus and racism, and forces them to look for information related to safety as well as personal health. In addition, uncertainty with regard to the disease and the rapidly changing nature of the ongoing pandemic, along with confusion over what health behaviors are recommended, add to the mounting challenges encountered by immigrants and sojourners who have less-than-desirable access to health information.

In the current study, we focus on the COVID-19 pandemic as a research context in which we investigate the health information-seeking behaviors of immigrant and sojourner populations in the United States. Specifically, we focus on U.S.-dwelling Chinese, including Chinese immigrants and sojourners living in the United States during the COVID-19 pandemic. Since the first reported coronavirus cases allegedly came from China, this group has been a chief target of discrimination during this period. One of the fastest-growing ethnic groups in the United States, there are currently more than four million people of Chinese ancestry living in the United States (2018). Similar to other immigrant and sojourner populations, U.S.-dwelling Chinese encounter a wide range of barriers to their acculturation into U.S. society and lack the knowledge necessary to manage their health well (Chen et al., 2010). In addition, compared with other immigrant populations, U.S.-dwelling Chinese experience higher rates of various illnesses, including stomach and liver cancers and tuberculosis (Zane et al., 1994). For U.S.-dwelling Chinese, health information seeking consists of lifelong experiences coping with two different societies and cultures (Chen et al., 2010). The group’s health information-seeking behaviors during the COVID-19 pandemic are of paramount importance to their health and well-being at this difficult time.

Informed by the literature on intercultural, health, risk, and mass communication, we had two main objectives in this study. The first objective was to examine the various predictors of U.S.-dwelling Chinese’s health information-seeking behaviors during the COVID-19 pandemic by applying the multiple-motive framework adopted from the heuristic-systematic model (HSM; Eagly & Chaiken, 1993) and the risk information seeking and processing model (RISP; Griffin et al., 1999). Specifically, rather than focusing on their general information seeking regarding COVID-19, we examined their use of Chinese and U.S. media for information on COVID-19. In addition, we examined the accuracy, impression, and defense motivations for information seeking by integrating the intercultural aspects relevant to U.S.-dwelling Chinese. The second objective was to examine the consequences of U.S.-dwelling Chinese’s health information-seeking behaviors during the COVID-19 pandemic. In this study, the consequences focused on the adoption of various health behaviors. In particular, we examined how U.S.-dwelling Chinese’s use of Chinese and U.S. media for information on COVID-19 related to four distinct health behaviors recommended by the Centers for Disease Control and Prevention (CDC, 2021): handwashing, wearing face masks, social distancing, and hoarding food and supplies. To achieve these two objectives, we conducted a two-wave longitudinal survey among a sample of U.S.-dwelling Chinese during the COVID-19 pandemic that assessed their reactions to the pandemic, including predictors of information seeking at Wave 1 of the survey and information seeking and health behaviors at Wave 2 of the survey. Notably, the longitudinal nature of our design allowed us to go beyond a focus on behavioral intention and to assess actual behaviors at Wave 2.

Achieving these two objectives means our study contributes to the current literature by extending the existing information-seeking framework into an intercultural health pandemic context. It takes into account intercultural factors, such as acculturation, in the RISP model, compares people’s motivations for their differential use of host versus homeland media for health information, and examines the potential of formally integrating three motivations (i.e., accuracy, impression, and defense) into the RISP model. Finally, it establishes a direct link between information seeking and health behaviors that addresses the underexplored “so what” question in health information seeking (Griffin et al., 2013). Practically speaking, we hope our findings can provide guidance to public health professionals, educators, media practitioners, agencies, and organizations who serve immigrant and sojourner groups in this increasingly globalized world.

Acculturation and Health Information Seeking

Early work on immigrants and sojourners’ information seeking has paid considerable attention to their use of libraries for information essential to them as they adjust to their new culture (Komlodi, 2005). Thus, the role played by acculturation in influencing health behaviors among these groups has been a major topic in this domain (Du & Li, 2015). Acculturation is a process that involves “changes in cultural attitudes, values, and behaviors that occur upon contact between two formerly autonomous population groups, such as a colonized and a colonizer group, or an immigrant group and its host society” (Anderson, 2004, p. 1). Scholars have noted a phenomenon called the immigrant paradox, which relates to acculturation’s effects on health (Lara et al., 2005). On one hand, greater acculturation into the host country is associated with a number of positive health outcomes, including better health-related knowledge and access to health care services (Lara et al., 2005). On the other hand, greater acculturation can also increase risky health behaviors, such as unsafe sex (Du & Li, 2015).

Prior studies have conceptualized and operationalized acculturation differently. Some scholars regarded acculturation as a unidimensional model that delineated one’s movement away from their home culture to the host culture, whereas others agreed on a bidimensional model that treated acculturation as consisting of two independent processes, with one being the maintenance of the home culture and the other the adoption of the host culture (Du & Li, 2015). Acculturation as a multidimensional concept has traditionally been measured using the length of time spent in the host country, proficieny with the host country’s language (e.g., English), and number of friends from the host country (Du & Li, 2015). Perceived distance from the home country and the host country, respectively, has also been employed as indicators of acculturation, which may more accurately represent the bidimensional model (Ahadi & Puente-Díaz, 2011).

When it comes to information seeking, media preference is another principal indicator of acculturation (Ye, 2005). Research has provided general support for the idea that, as individuals become more acculturated into the host culture, their use of media in the host culture language increases, whereas their use of media in their native tongue decreases (Hwang & He, 1999; Ye, 2005). In this study, we focused on two types of media that acculturation likely influenced: host media and homeland media. Host media are media owned by and produced in the host country. Research has shown that host media often portray immigrants negatively, exposure to which can result in negative outcomes (e.g., decreased self-esteem) in immigrants (Correa, 2010; Etchegaray & Correa, 2015). In comparison, homeland media are media owned by and produced in the country of origin. Homeland media used to have only limited, indirect, and delayed influence on overseas emigrants, which has now changed due to the advent of the internet (Yin, 2015). While research on this topic remains scarce, we expected homeland media to depict overseas emigrants more positively than host media, and to abridge the perceived distance from one’s country of origin (Etchegaray & Correa, 2015; Yin, 2015).

This study’s main outcome variables focused on U.S.-dwelling Chinese’s use of U.S. media (i.e., host media) and Chinese media (i.e., homeland media) for information on COVID-19. There is limited research that has examined the simultaneous use of both media types for health information seeking. Consistent with the findings mentioned above on acculturation and media use, Wang and Yu (2015) found that Chinese immigrants who were more acculturated were more likely to use U.S. health websites, whereas those less acculturated tended to seek health information from Chinese websites. This study operationalized acculturation both in the traditional way (i.e., a combination of English proficiency and number of American friends), which we referred to as acculturation, and as perceived distance from both countries, which we referred to as perceived distance, to differentiate it from its traditional operationalization (Ahadi & Puente-Díaz, 2011). We propose the following hypotheses regarding the influence of acculturation and perceived distance on U.S.-dwelling Chinese’s use of U.S. and Chinese media for information on COVID-19:

• H1: More acculturation into the United States (a) and increased perceived distance from China (b) will be related to decreased use of Chinese media for information on COVID-19, whereas increased perceived distance from the United States (c) will be related to increased use of Chinese media for information on COVID-19.

• H2: More acculturation into the United States (a) and increased perceived distance from China (b) will be related to increased use of U.S. media for information on COVID-19, whereas increased perceived distance from the United States (c) will be related to decreased use of U.S. media for information on COVID-19.

Motivations for and Consequences of Information Seeking

Individuals have diverse motivations for information seeking, and motivations vary across individuals. Employing a multiple-motive framework, the HSM proposes three broad motives for information processing: accuracy, impression, and defense motivations (Chaiken et al., 1996). Accuracy motivation is defined as “the desire to hold attitudes and beliefs that are objectively valid” (Chaiken et al., 1996, p. 556). When accuracy motivated, individuals tend to seek and process information in more objective and unbiased ways to look for the truth. Impression motivation is defined as the desire to maintain the attitudes and beliefs that meet social goals (Chaiken et al., 1996). Individuals driven by impression motivation are more likely to seek and process information that will produce desired consequences in their interpersonal relationships (e.g., making a good impression on others). Defense motivation is defined as a desire to form and defend conclusions that are congruent with one’s self-definitional attitudes and beliefs, which include one’s values, social identities, and personal characteristics (Chaiken et al., 1996). Defense-motivated individuals seek and process information selectively, to reinforce their preexisting attitudes and beliefs. We incorporated these three classes of motivations in this study and integrated them with the RISP model.

Having its roots in the HSM (Eagly & Chaiken, 1993), the RISP model proposes various cognitive and affective factors related to information seeking and processing and maps the relationships among these factors (Griffin et al., 1999). The core of the RISP model is information insufficiency, defined as the gap between how much a person currently knows about a risk and the information sufficiency threshold that a person needs to reach to deal with that risk confidently (Griffin et al., 1999). The greater the gap is, the more likely individuals are to seek information. Thus, information insufficiency represents accuracy motivation (Griffin et al., 1999). Because accuracy motivation accentuates obtaining objectively valid information, we propose that accuracy-motivated individuals will go out of their way to seek out such information, regardless of where it came from, as long as it is valid. Therefore, we hypothesize that information insufficiency should motivate increased information seeking from both U.S. and Chinese media:

• H3: Greater information insufficiency will be related to increased use of U.S. (a) and Chinese (b) media for information on COVID-19.

Adopting the concept of subjective norms from the theory of planned behavior (TPB; Ajzen, 1991), the RISP model proposes a factor, informational subjective norms, defined as the amount of social pressure a person feels to stay informed of a risk (Griffin et al., 1999). The more informational subjective norms a person feels, the more likely that person is to seek out information. While informational subjective norms were not originally proposed as an indicator of impression motivation (Griffin et al., 2013), this variable could be reconceptualized as one, because it represents a person’s desire to satisfy others’ expectations about having sufficient knowledge of a risk. If there is social pressure for a person to know enough about a risk, then that person will likely seek out additional information to meet this social expectation, and will care less about the source of the information. Importantly, this study focused on the overall social pressure that people experienced, and did not differentiate between social pressures that may have originated from different sources. We predict that informational subjective norms should motivate increased information seeking from both U.S. and Chinese media:

• H4: More informational subjective norms will be related to increased use of U.S. (a) and Chinese (b) media for information on COVID-19.

The RISP model does not propose any factors related to defense motivation. This study included two conspiratorial beliefs closely related to the COVID-19 pandemic, which one could consider indicators of defense motivation, when juxtaposed with the use of U.S. and Chinese media. In particular, both beliefs were based on real-world conspiracy theories and claimed that the coronavirus was man-made (Allsop, 2020). One belief, labeled a conspiratorial belief against China, claimed that China bioengineered the virus; the other belief, labeled a conspiratorial belief against the United States, claimed that the United States created the virus as a bioweapon. We assumed that individuals who held these conspiratorial beliefs were motivated to defend them by selectively choosing what media content to consume. In other words, individuals were more likely to choose media that supported their conspiratorial beliefs than those that opposed them (Hart et al., 2009). Chinese media were more likely to cover the conspiracy concerning the United States’s involvement and to refute the conspiracy about China’s involvement, whereas U.S. media were likely to do the opposite (Allsop, 2020). Therefore, we propose the following hypotheses based on defense motivation:

• H5: Having a stronger conspiratorial belief against China will be related to increased use of U.S. media (a) and decreased use of Chinese media (b) for information on COVID-19.

• H6: Having a stronger conspiratorial belief against the United States will be related to decreased use of U.S. media (a) and increased use of Chinese media (b) for information on COVID-19.

Apart from motivations, the RISP model also proposes two additional proximate predictors of information seeking. Drawing from the TPB’s (Ajzen, 1991) concept of perceived behavioral control and the HSM’s (Eagly & Chaiken, 1993) concept of capacity, perceived information gathering capacity refers to a person’s ability to seek out information (Griffin et al., 1999). The more that individuals perceive themselves as being capable of looking for information, the more likely they are to seek information. As this study assessed perceived informational gathering capacity more generally without differentiating between U.S. and Chinese media, we assume that this overall perceived informational gathering capacity should motivate the use of both U.S. and Chinese media:

• H7: Stronger perceived information gathering capacity will be related to increased use of U.S. (a) and Chinese (b) media for information on COVID-19.

Relevant channel beliefs are beliefs (e.g., trustworthiness, usefulness) about information channels (Griffin et al., 1999). This study focused on the trustworthiness or credibility aspects of these channels, which were a critical determinant of which information sources an individual would rely on, and differentiated beliefs about U.S. media from those about Chinese media. We propose the following hypothesis:

• H8: Greater trust in U.S. media will be related to increased use of U.S. media (a) for information on COVID-19, whereas greater trust in Chinese media will be related to increased use of Chinese media (b) for information on COVID-19.

Finally, the RISP model includes a section linking information seeking and processing with preventive behaviors, which is underexplored in the RISP literature (Griffin et al., 2013). When it comes to health information seeking, the key downstream behaviors would naturally be health behaviors. Based on our research context, we focused on four distinct health behaviors: handwashing, wearing face masks, social distancing, and hoarding food and supplies. While we expected the use of U.S. and Chinese media to be positively associated with handwashing, social distancing, and hoarding food and supplies, we anticipated that the use of U.S. and Chinese media would lead to different outcomes vis-à-vis the wearing of face masks. This was because wearing face masks was a recommended behavior in China since the start of the COVID-19 pandemic, but it was not until much later that the U.S. CDC made the same recommendation, which was also reflected in the media coverage in both countries (Lynteris, 2020; Prasad, 2020). At the time of our study, the recommendation to wear face masks was not salient in U.S. media unlike in Chinese media. Therefore, although we believe that more exposure to media coverage in both countries should lead to more health behaviors, we predict that there will be differences in the wearing of face masks because of the uniqueness of the issue:

• H9: Increased use of U.S. media for information on COVID-19 will be related to increased adoption of handwashing (a), social distancing (b), and hoarding food and supplies (c), and decreased adoption of wearing face masks (d).

• H10: Increased use of Chinese media for information on COVID-19 will be related to increased adoption of handwashing (a), social distancing (b), hoarding food and supplies (c), and wearing face masks (d).

Figure 1 presents our proposed theoretical framework and hypotheses.

Figure 1.

Theoretical framework.

Method

Sample and Procedure

We conducted a two-wave longitudinal survey of a sample of Chinese people living in the United States during the COVID-19 pandemic. The surveys, written in Chinese, were conducted between March 16, 2020, and April 16, 2020. We recruited participants via social media (e.g., WeChat), online discussion boards, and email. The participants initially completed the first-wave survey. One week later, we invited those who provided their emails and consented to further correspondence at Wave 1 to complete the second-wave survey. Each wave offered participants incentives to complete the surveys: 16 random participants from Wave 1 and 10 from Wave 2 received US$100 Amazon gift cards.

At Wave 1, 2,208 participants completed the survey and passed two attention check questions (e.g., “Please select the disagree option”) and at Wave 2, 1,295 participants completed the survey and passed one attention check question (i.e., “Please select the absolutely will option”).¹ We also dropped data from 11 participants because they were not in the United States at Wave 2, rendering a final sample of 1,284 for analyses. Among them, 48.1% identified as female (n = 618), 51.2% as male (n = 658) and the rest as other genders or did not answer the question (n = 8). Participants on average were 26 years old (SD = 4.36) and had a median annual household income level of US$25,000 to US$34,999. The majority had at least a bachelor’s degree (92.6%) and were not U.S. citizens or permanent residents (89.5%, n = 1,149). On average, they had lived in the United States for 4.71 years (SD = 4.02).

Measures

Table 1 presents the descriptive statistics and composite reliabilities of the measures used in this study, as adopted from the research cited in the literature review section. The lead author, who is proficient in both languages, translated the English version of the measures into Chinese, which other researchers on the team validated through back-translation. Wave 1 measures included acculturation (Du & Li, 2015), current knowledge (Griffin et al., 2008), information sufficiency threshold (Lu et al., 2020), informational subjective norms (Griffin et al., 2008), trust in Chinese and U.S. media (Harring, 2013), perceived information gathering capacity (Griffin et al., 2008), perceived distance from China and the United States (Ahadi & Puente-Díaz, 2011), and conspiratorial beliefs (Allsop, 2020). Wave 2 measures included use of Chinese and U.S. media for COVID-19 information, washing hands, wearing face masks,² social distancing, and hoarding food and supplies (see Table 2 for the correlation matrix of the variables used in this study).

Table 1.

Descriptive Statistics and Composite Reliabilities of Survey Measures.

Wave 1 Measures
Latent Variables
Observed Indicators	N	M	SD	Reliability
Acculturation				α = .77
What do you think of your English? (1 = not fluent, 5 = fluent)	1,284	4.15	0.86
Generally speaking, how satisfied are you with your English? (1 = not satisfied, 5 = satisfied)	1,284	3.54	1.10
How many of your friends are from the United States? (1 = none of my friends are from the United States, 5 = all of my friends are from the United States)	1,284	2.5	0.81
Perceived Information Gathering Capacity				α = .78
If I wanted to get more information about coronavirus, I could readily take the time to gather any additional information I might need. (1 = disagree, 5 = agree)	1,284	4.02	0.87
. . . I would know how to separate facts from rumors.	1,284	3.89	0.91
. . . I would know where to go for more information.	1,284	3.97	0.88
Observed Variables		M	SD
Informational Subjective Norms: People who are important to me (e.g., parents, friends, etc.) expect me to know something about the novel coronavirus. (1 = disagree, 5 = agree)	1,284	4.37	0.74
Current Knowledge: How much do you think you currently know about the novel coronavirus? (1 = not at all, 5 = A great deal)	1,284	3.46	0.67
Information Insufficiency: How much more, if at all, do you think you need to know about the novel coronavirus to deal with it adequately? (1 = no need to know more at all, 5 = need to know a lot more)	1,284	3.56	0.97
Trust in Chinese Media: How much trust do you have in Chinese media to give you accurate information about the novel coronavirus? (1 = no trust at all, 5 = complete trust)	1,284	2.91	0.90
Trust in U.S. Media: How much trust do you have in U.S. media to give you accurate information about the novel coronavirus? (1 = no trust at all, 5 = complete trust)	1,284	2.81	0.86
Perceived Distance from China: What do you think of the distance between Chinese society and you? (0 = very close, 100 = very distant)	1,282	39.24	26.21
Perceived Distance from the U.S.: What do you think of the distance between U.S. society and you? (0 = very close, 100 = very distant)	1,283	46.71	25.96
Conspiratorial Belief against China: The novel coronavirus originated from a Chinese bioweapon lab.	1,284	1.59	0.85
Conspiratorial Belief against the United States: The novel coronavirus is in fact a bioweapon made by the U.S. government.	1,284	1.68	0.89
Wave 2 Measures
Observed Variables	N	M	SD
Use of Chinese Media for COVID-19 Information: In the past week, how often did you use Chinese media (e.g., social media, internet, TV, newspaper) to obtain information about the novel coronavirus?	1,284	3.71	1.15
Use of U.S. Media for COVID-19 Information: In the past week, how often did you use U.S. media (e.g., social media, internet, TV, newspaper) to obtain information about the novel coronavirus?	1,284	3.41	1.09
Washing Hands: In the past week, how often did you wash your hands? (1 = not at all, 5 = always)	1,284	4.53	0.61
Wearing Face Masks: In the past week, how often did you wear face masks when going outside? (1 = not at all, 5 = all the time, 6 = did not go outside in the past week)	1,284	4.58	1.29
Social Distancing: In the past week, how often did you avoid close contact with others? (1 = not at all, 5 = completely avoided any close contact with others)	1,284	4.34	0.61
Hoarding Food and Supplies: In the past week, how much, if at all, did you hoard food and supplies? (1 = not at all, 5 = a great deal)	1,284	3.63	0.80

Table 2.

Correlation Matrix.

	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17
1. Acculturation	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
2. Perceived Information Gathering Capacity	.175**	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
3. Informational Subjective Norms	.023	.171**	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
4. Current Knowledge	.142**	.358**	.143**	—	—	—	—	—	—	—	—	—	—	—	—	—	—
5. Information Insufficiency	.068*	−.072**	.098**	−.153**	—	—	—	—	—	—	—	—	—	—	—	—	—
6. Trust in Chinese Media	−.061*	.088**	.100**	.109**	−.057*	—	—	—	—	—	—	—	—	—	—	—	—
7. Trust in U.S. Media	.010	.036	−.042	.009	−.047	.018	—	—	—	—	—	—	—	—	—	—	—
8. Perceived Distance from China	.097**	−.046	−.060*	−.018	−.036	−.211**	.115**	—	—	—	—	—	—	—	—	—	—
9. Perceived Distance from the United States	−.275**	−.083**	−.009	−.087**	−.093**	.085**	−.088**	.053	—	—	—	—	—	—	—	—	—
10. Conspiratorial Belief against China	−.019	−.103**	−.042	−.063*	.058*	−.099**	.072**	.120**	.007	—	—	—	—	—	—	—	—
11. Conspiratorial Belief against the United States	−.073**	−.109**	−.012	−.053	.077**	.223**	−.156**	−.012	.093**	.571**	—	—	—	—	—	—	—
12. Chinese Media Use	−.139**	.046	.118**	.068*	.084**	.234**	−.161**	−.176**	.077**	−.054	.107**	—	—	—	—	—	—
13. U.S. Media Use	.253**	.180**	.088**	.130**	.134**	−.045	.173**	−.015	−.201**	.015	−.079**	.200**	—	—	—	—	—
14. Washing Hands	.106**	.071*	.072**	.037	.157**	.021	−.042	−.004	−.085**	.025	.046	.129**	.102**	—	—	—	—
15. Wearing Face Masks	−.081**	−.007	.028	.024	.061*	.137**	−.139**	−.014	.135**	.048	.153**	.129**	−.037	.060*	—	—	—
16. Social Distancing	.075**	.038	.052	.025	.045	.058*	.021	−.036	−.034	−.042	.002	.034	.082**	.120**	.191**	—	—
17. Hoarding Food and Supplies	.016	−.011	.036	.073**	.075**	.062*	−.095**	−.020	.072*	−.020	.052	.143**	.037	.164**	.225**	.140**	—

^*p < .05. **p < .01.

Analyses

We performed structural equation modeling (SEM) to examine the proposed hypotheses and evaluate the overall model fit. We employed Mplus 8.3 for all SEM analyses. We first constructed a measurement model with good fit and then a structural model with good fit following a two-step modeling procedure (Kline, 2016). We refined the structural model for better model fit based on modification indices and theoretical reasoning simultaneously.

We first screened data for collinearity, normality, and missing values before the SEM analyses. Collinearity was not an issue because all tolerance values were higher than .10 and variance inflation factor (VIF) values were lower than 10 (Kline, 2016). In addition, we used Stata 16 to test for multivariate normality and found that the multivariate normality assumption was likely violated (i.e., skewness >|3| and kurtosis >|20|; Kline, 2016). Therefore, using Mplus 8.3, we adopted a maximum likelihood estimator with robust standard error (MLR), which also dealt with missing values, for its robustness to non-normality.

Results

Model Fitting Results

We first created a measurement model of two latent variables: acculturation and perceived information gathering capacity. The measurement model fit the data well (Kline, 2016): χ²(8) = 11.87, p = .157, root mean square error of approximation (RMSEA) = .019 (90% confidence interval [CI] = [.000, .041]), comparative fit index [CFI] = .998, standardized root mean square residual (SRMR) = .015.

We then constructed the structural model based on the proposed relations in Figure 1. The initial structural model did not fit the data well: χ²(137) = 685.14, p < .001, RMSEA = .056 (90% CI = [.052, .060]), CFI = .842, SRMR = .051. We refined the initial structural model by considering modification indices, theoretical reasoning and empirical evidence. First, we covaried Chinese and U.S. media use to account for common method variance and their shared variance in general media use. Second, we allowed the four health behaviors to covary, due to common method variance and their shared variance in health behaviors. Third, we set the error terms of the two English proficiency items free, because they were highly correlated and because they assessed the same dimension of acculturation. Fourth, we added a path from acculturation to perceived information-gathering capacity, because acculturation into the United States, often indicated by English proficiency levels, is inherently associated with immigrants’ abilities to access and comprehend health information (Oh et al., 2012). Fifth, we added paths from acculturation to handwashing and wearing face masks based on research indicating that acculturation can influence health behaviors (Du & Li, 2015). Finally, we added paths from trust in U.S. media and conspiratorial beliefs against the United States to wearing face masks. As China and the United States treated face masks differently in the early days of the COVID-19 pandemic (Lynteris, 2020; Prasad, 2020), variables indicative of individuals’ general trust in and attitudes toward the two countries could potentially have more direct impacts on the wearing of face masks.

The final structural model fit the data adequately: χ²(124) = 286.85, p < .001,³ RMSEA = .032 (90% CI = [.027, .037]), CFI = .953, SRMR = .032. Overall, the final model explained 16% of the variance in using Chinese media for COVID-19 information, 11% in using U.S. media for COVID-19 information, 4.2% in washing hands, 6.1% in wearing face masks, 0.8% in social distancing, and 2.5% in hoarding food and supplies.

Relation Testing Results

Figure 2 shows unstandardized coefficients with standard errors for all relationships investigated in the final structural model. Starting with acculturation, we observed that greater acculturation into the United States was related to less use of Chinese media, and more use of U.S. media, for information on COVID-19, supporting H1(a) and H2(a). In addition, increased perceived distance from China was associated with less use of Chinese media, but had no significant relationship with U.S. media use, supporting H1(b) and rejecting H2(b). Finally, perceived distance from the United States was associated with neither Chinese nor U.S. media use, rejecting H1(c) and H2(c).

Figure 2.

Unstandardized estimates and standard errors (in brackets) for the structural model.

Note. Significant path coefficient estimates are in bold; CN = effects on or of Chinese media use; US = effects on or of U.S. media use.

*p < .05. **p < .01. ***p < .001.

With respect to the three motivations for information seeking, we found that a greater information sufficiency threshold was related to increased use of both Chinese and U.S. media, supporting H3(a)–(b). In addition, we found that informational subjective norms were unrelated to the use of either U.S. or Chinese media, rejecting H4(a)–(b). Furthermore, a stronger conspiratorial belief against China was associated with more U.S. media and less Chinese media use, supporting H5(a)–(b). Finally, a stronger conspiratorial belief against the United States was related to more Chinese media use, but was not significantly related to U.S. media use, rejecting H6(a) and supporting H6(b).

As for perceived information gathering capacity and trust in media, we found that stronger perceived information gathering capacity was associated with more U.S. and Chinese media use, supporting H7(a)–(b). Greater trust in U.S. media was related to increased U.S. media use, supporting H8(a), and decreased Chinese media use. In comparison, greater trust in Chinese media was related to increased Chinese media use, supporting H8(b), but was not significantly related to U.S. media use.

With regard to the relationship between information seeking and health behaviors, increased use of U.S. media was related to greater adoption of handwashing and social distancing, but was unrelated to hoarding food and supplies or wearing face masks, supporting H9(a)–(b) and rejecting H9(c)–(d). On the contrary, increased use of Chinese media was associated with greater adoption of handwashing, hoarding food and supplies, and wearing face masks, but was not associated with social distancing, supporting H10(a), (c), and (d), but rejecting H10(b).

We also discovered a few unexpected relationships. First, greater acculturation into the United States was related to greater perceived information-gathering capacity and more handwashing, but to less wearing of face masks. Second, greater trust in U.S. media was associated with less wearing of face masks, whereas a stronger conspiratorial belief against the United States was related to more wearing of face masks.

Discussion

As the COVID-19 pandemic continues to inflict damage throughout the world, some minority groups are bearing a disproportionate share of its impacts. We concentrated on one such group, U.S.-dwelling Chinese, who have had to cope with challenges related to acculturation, health, safety, and racism. Recognizing that health information seeking was an essential step in helping maintain and improve health behaviors, we relied on the RISP model to examine the various factors that predicted U.S.-dwelling Chinese’s use of U.S. and Chinese media for information on COVID-19 as well as the consequences of that health information seeking. Overall, we found that acculturation, accuracy and defense motivations, trust in media, and perceived information gathering capacity played a key role in explaining information seeking from an intercultural viewpoint, and that the use of U.S. and Chinese media was associated with different health behaviors. These findings contribute to theory and practice in a variety of ways.

First, we situated the COVID-19 pandemic in an intercultural context, making it relevant to the investigation of acculturation. We employed two types of acculturation measures: a more traditional one based on host country language and friends, and another focused on perceived distance from both host and home countries which might be a better representation of the bidimensional model of acculturation. The traditional acculturation measure supported our predictions, but the perceived distance measure largely rejected them. It is likely that, in the context of health information seeking, perceived distance was not the most pertinent aspect of acculturation. By contrast, the language-based measure was potentially more relevant because utilizing U.S. media required a certain level of English proficiency. In addition, as individuals have limited time for information seeking, time spent on one type of media (e.g., U.S. media) means less time for other media (e.g., Chinese media), which could serve as an additional explanation for why greater acculturation into the United States was associated with decreased use of Chinese media. Relatedly, participants’ general perceived information-gathering capacity, which was not framed specifically for a particular type of media, increased their use of both U.S. and Chinese media. In addition, trust in the media of a specific origin predicted use of those media, further confirming the influence of source credibility in media use.

Second, we examined three types of motivations for information-seeking. Accuracy motivation, as represented by information insufficiency, encouraged the use of both U.S. and Chinese media. This finding was foreseeable because accuracy goals compel individuals to concentrate on obtaining correct information, which is more likely to succeed by exploring a wider variety of media sources. Contrary to our hypotheses, impression motivation, as indicated by informational subjective norms, was irrelevant to individuals’ use of U.S. and Chinese media. This finding is inconsistent with a meta-analysis of RISP studies, which found informational subjective norms to be one of the two most prominent predictors of information seeking (Yang et al., 2014). We suspect that, in the context of a severe global pandemic that is relevant to every individual’s health, the motivation to meet social expectations and impress others is less germane than the goal of maintaining good health. In addition, one key characteristic of this immigrant and sojourner population was that they were probably separated from many of their important others, who remained in their home country. Thus, the influence of impression motivation may have been weakened in this group. Furthermore, as we did not differentiate between types of social expectations (e.g., U.S. or Chinese) our participants might have experienced, it is possible that a specific type of social expectation (e.g., an expectation from one’s Chinese social circle) was associated with the use of a specific type of media (e.g., Chinese media). Future research should examine whether individuals attempt to achieve other goals that might compete against impression motivation when seeking out information relevant to the self, how the influence of social expectations may shift for immigrant and sojourner populations (Lu, 2015), and whether there is a need to differentiate between the various types of social expectations.

The more novel part of our study on motivations considered the two conspiratorial beliefs, which we conceptualized as indicators of defense motivation, the least-explored motivation type in the RISP literature. Our findings support the idea that defense motivation drives individuals. Participants were less likely to use the type of media that they believed might weaken their preexisting beliefs. The only exception was that conspiratorial beliefs against the United States did not reduce the use of U.S. media. This might be because living in the United States somehow forced our participants to get information from U.S. media because of its relevance, regardless of their defense motivation. It should also be noted that media use driven by defense motivation could manifest in numerous ways. For example, some people may simply avoid undesirable information (i.e., selective exposure), whereas others may choose to process that information in a way that reinforces, rather than undermines, their preexisting beliefs (i.e., selective processing; Stroud, 2017). Our investigation only reflected one way in which defense motivation functions, and participants who hold conspiratorial beliefs against the United States might adopt other defensive strategies when dealing with U.S. media. In summary, our study reveals that there is a need to further consider impression and defense motivations as additions to the RISP model as well as to other models derived from it, such as the Planned Risk Information Seeking Model (Kahlor, 2010).

Third, we explored the association between media use and health behaviors, answering the “so what” question of information seeking. Intriguingly, we found that U.S. and Chinese media use was associated with different health behaviors. One might attribute these differences to the behaviors that the media recommended, or made salient, at the time of our survey. For instance, social distancing has been the principal behavior recommended by the United States. In addition, whereas China recommended wearing face masks as an effective prevention measure since the start of the pandemic in China, it was not until much later that the U.S. CDC made the same recommendation (Lynteris, 2020). In terms of the number of health behaviors influenced by media exposure, Chinese media use stimulated more health behaviors than U.S. media use, which might suggest that, at least with regard to prevention measures, our participants relied more on Chinese media. One possible explanation for this finding is that China had contained the pandemic at the time of our survey, while the number of infected cases had begun to rise in the United States (The Center for Systems Science and Engineering at Johns Hopkins University, 2020). Thus, our participants might have relied more on Chinese media for recommendations on which behaviors were proven effective in preventing infection. In addition, among the unexpected findings, three relationships involving wearing face masks are worth noting. All three relationships pointed to a general trend, that a more pro-U.S. orientation (i.e., greater acculturation into the United States, greater trust in U.S. media, and weaker conspiratorial beliefs against the United States) was associated with a lower likelihood of wearing face masks. These findings support the immigrant paradox that, as a person’s level of acculturation increases, the number of both positive and negative health outcomes experienced by that person will increase (Du & Li, 2015).

This study has several limitations. First, as no published directory on Chinese residents in the United States, if any, was accessible to us, we resorted to a convenience sample for this study, which is an unavoidable limitation in the research design. In addition, our sample was biased toward younger, lower-income, and more highly educated individuals. Although we did not ask about occupation, we suspect that a majority of our participants were international students from China, the largest international student group in the United States, who make sizable contributions to the U.S.’s economy while studying or working after graduation in the United States (Timmons & Dwyer, 2020). Like other immigrant and sojourner subgroups, this group has to deal with numerous barriers during their acculturation, including one that is unique to this group, the academic acculturation (Ye, 2005). Investigating their information seeking during the pandemic offers unique insight into the complex population of U.S.-dwelling Chinese. In addition, we recruited our participants through online channels only, which may have precluded people who were not frequent users of those channels from participating in our study. Future research should attempt to employ a more representative sample, or to explore other subgroups within the group of U.S.-dwelling Chinese, including those who do not use online media as their primary means of communication, to assess the generalizability of our findings.

Second, owing to the complexity of our conceptual model, we employed a crude media use measure that did not differentiate between media outlets within each country. We intended to examine participants’ general media use in the two countries, assuming that media use patterns would be more similar for media produced in the same country. It is obvious that individuals have varying levels of trust in different U.S. or Chinese media, and that they may respond defensively only to some media from a given country. Future research should consider more nuanced media use patterns, and then compare their predictors and consequences. In addition, there are other factors proposed in the RISP model that this study did not consider, such as risk perception, efficacy beliefs, and affective responses to risk, that may nevertheless be relevant. Future research that examines similar intercultural issues may wish to investigate these other RISP factors.

Third, compared with many other health information-seeking studies, one key strength of our study was its two-wave longitudinal design, which helped reduce concerns about cross-sectional data. However, one should interpret our findings regarding causality with caution. This is especially true for the relationships between the predictors of information seeking, which we measured at Wave 1, and between health information seeking and health behaviors, which we measured at Wave 2. Future research could employ an experimental design that tests the causality of the proposed relationships (e.g., Lu et al., 2021). In addition, the interval between the two waves was relatively short. Due to the rapidly changing nature of the COVID-19 pandemic, we purposely chose a short interval to ensure the consistency of the social environment (e.g., discrimination, official recommendations for protective behaviors). To examine the time- and context-sensitivity of our findings, they could be compared with findings from other studies that surveyed similar populations at other times during the COVID-19 pandemic.

Fourth, although we examined the consequences of information seeking, the underlying mechanism between information seeking and the adoption of health behaviors remains unclear. In fact, the RISP model proposes that people’s systematic processing of risk information, not just information seeking or heuristic processing, leads them to perform certain behaviors habitually and repeatedly, which are crucial during the COVID-19 pandemic (Griffin et al., 1999). Active information seeking may be accompanied by systematic processing, but this is not always the case (Griffin et al., 1999). To better investigate the linkage between information behaviors (e.g., information processing) and health behaviors through the lens of the RISP model, a panel study design should be employed to examine whether systematic processing of risk information at an earlier time is associated with the stability of participants’ health behaviors across time.

Practically speaking, if the goal is for individuals to become informed citizens who can make sound judgments and decisions regarding their health, then utilizing a wide range of media sources, including those from multiple countries, seems necessary. Our study suggests that exposure to media from a single country is insufficient to encourage people to perform every recommended health behavior. To achieve this goal, efforts should be made to stimulate accuracy motivation, improve the credibility and trustworthiness of media sources, enhance information gathering capacities, and address conspiratorial beliefs and misinformation properly. Regarding the intercultural aspect, attention should be paid to helping immigrant and sojourner populations navigate through media channels from multiple countries, and improve their intercultural media literacy.

Author Biographies

Hang Lu, PhD (Cornell University), is an assistant professor of media psychology in the Department of Communication and Media at the University of Michigan. His primary research focuses on examining the dynamic role that emotions play in influencing environmental, health, science, and risk communication. In particular, he studies emotions as persuasive appeals, dynamic processes, and situational cues, with a recent focus on the experience of multiple emotions.

Haoran Chu, PhD (University at Buffalo), is an assistant professor at the University of Florida. His research examines the influences of psychological distance, mental construal, and social interaction on people’s response to health and environmental risks. He also studies means to foster individual and community resilience against natural or man-made disasters.