Introduction

There is considerable heterogeneity in diet behaviors among Asian American (AA) ethnic subgroups. For example, saturated fat intake is lower and sodium intake is higher among Chinese, Korean, and Japanese adults, relative to non-Hispanic White adults in the U.S.; whereas, intake of saturated fats and refined carbohydrates is higher among Asian Indian adults 1]. Prior research also suggests that acculturation is an important influence on diet among AA adults, such as higher intake of nutrient-poor, energy-dense foods and lower consumption of fruits and vegetables consumption with longer duration in the U.S. [2]. However, the extent to which food purchasing behaviors differ between AA ethnic subgroups or those of different acculturation levels remains unclear. This research is critical for addressing issues related to food purchasing behaviors via intervention (e.g., offering healthy, culturally-specific grocery items in online retail settings).

The COVID-19 pandemic and other recent events have fueled an increase in incidents of discrimination among AA adults [3], which may contribute to anxieties about leaving home to shop for groceries and spur AA consumers to shop online. Previous surveys suggest that internet use is higher among English-speaking AA adults relative to other race/ethnic groups [4], but there is a dearth of disaggregated data about online food purchasing behaviors and attitudes from AA adults. In our previous work, we explored changes in diet and food shopping behaviors among AA adults due to COVID-19 [5], but we did not explore online shopping practices in general or differences by acculturation score. For example, online retailers may be an important source for culturally-specific (e.g., produce indigenous to Asia) or culturally-preferred foods (e.g., specific brands of seasonings) among AA adults, but this may differ across the spectrum of acculturation. To address these gaps, we sought to examine differences in online grocery shopping behaviors among AA adults using data disaggregated by AA ethnic subgroup and levels of acculturation.

Methods

We administered an online survey to a nationally-derived nonprobability sample of 3,084 AA adults from June 9–15, 2020 using Dynata, an online surveying company that recruits volunteer research participants [6]. The sample was recruited to approximately match the distribution of gender and age of Asian adults residing in the U.S. [7] Potential participants completed an online consent, followed by a brief pre-screening questionnaire. Eligibility criteria included identifying as Asian, being age 18 years and older, and being able to read and speak English.

Open REDCap, an online survey platform, was used to create and distribute the survey. The survey was designed to assess sociodemographics, health status, diet behaviors, and food shopping behaviors, using questions described elsewhere.[5] A 10-item version of questions adapted from the Marin Short Acculturation Scale was also included in the survey, which yields a total acculturation score ranging from 10 to 50 [8]. All procedures were approved as exempt by the Institutional Review Board. Duplicate responses (n = 37) and implausible skip patterns (n = 29) and participants who identified as more than one of the three ethnic subgroups (defined by country of origin) (n = 124) were excluded. The final sample included 1,737 East Asian, 570 South Asian, and 587 Southeast Asian (n = 2,894) adults. The median completion time for the survey was 17.4 minutes (IQR: 12.1, 25.3).

We used logistic regression to compare differences in online grocery shopping (yes/no) by Asian ethnic subgroup (East Asian as referent) and acculturation score (dichotomized using median score). We also examined responses to questions regarding participants’ primary grocery store (type, reasons); shopping for Asian grocery items (yes/no, store type, reasons); shopping for online groceries (yes/no, frequency, location, type of groceries, reasons, intentions), including Asian grocery items; and whether there were other ways they were getting food at home that they were not using before COVID-19 (none, online orders, neighbors, family, food delivery business, food from my child’s school to eat at home). We controlled for age, gender, household income (<$20,000 vs. ≥$20,000), educational attainment (high school or less vs. more than high school), and household size (1, 2, 3, ≥ 4). We used a two-sided alpha of 0.05 as the threshold for statistical significance. Stata version 15.1 (StataCorp LP, College Station, TX) was used for all analyses.

Results

In our sample, the average age was 43.2 (SD = 16.5) years, 60% reported having at least one foreign-born parent, 39.8% reported being born outside of the U.S., 87.3% reported annual household income above $20,000, and 78.9% of respondents reported having a post-secondary education degree (Table 1). Approximately 29.2% of participants reported shopping online for groceries in a typical month, and 36.4% of those adults shop online for groceries at least once a week (Table 2). About 6% of participants reported shopping online for groceries at iFresh or another type of Asian online retailer (Table 3). A higher percentage of South Asian adults (45.4%) reported shopping online for groceries in a typical month compared to Southeast (30.3%) and East Asian adults (23.4%). Southeast (vs. East) Asian adults had lower odds of buying fresh fruits and vegetables (OR = 0.7; 95% CI: 0.5, 0.97) and higher odds of buying sugar-sweetened beverages (OR = 1.5; 95% CI: 1.02, 2.3) online.

Table 1 Sociodemographic characteristics of a nationally-derived nonprobability sample of 2,894 Asian American adults, overall and by Asian American subgroup (n = 2,984)
Full size table
Table 2 Grocery shopping behaviors and attitudes (%) of a nationally-derived nonprobability sample of 2,894 Asian American adults, overall and by Asian American subgroup (n = 2,984)a
Full size table
Table 3 Grocery shopping behaviors and attitudes (%) of a nationally-derived nonprobability sample of 2,894 Asian American adults, overall and by Asian American subgroup
Full size table

Among participants who did not shop online for groceries, South Asian adults had lower odds of reporting high prices (OR = 0.7; 95% CI: 0.6, 0.9) and extra delivery fees (OR = 0.6; 95% CI: 0.4, 0.7) as preventing them from starting; and lower odds of reporting free shipping (OR = 0.5; 95% CI: 0.4, 0.6) and lower prices (OR = 0.5; 95% CI: 0.4, 0.6) as what would motivate them to start shopping online for groceries (Table 3). The pattern was opposite for the variety and quality of goods online, with a higher prevalence of South Asian adults reporting them as potential motivating factors. We also observed higher odds of reporting shopping online for groceries among those who reported getting food via online orders (OR = 4.8; 95% CI: 3.0, 7.8) and food delivery businesses (OR = 3.0; 95% CI: 1.7, 5.2) for the first time since COVID-19.

The acculturation score of the overall sample was moderate [36.0 (SD = 8.0)]. Compared to East Asian adults [37.0 (SD = 7.8)], we observed a lower acculturation score among Southeast [35.6 (SD = 7.8)] and South [33.2 (SD = 8.1)] Asian adults (Table 1). Participants with a high (vs. low) acculturation score had higher odds of reporting shopping online for groceries due to low prices (OR = 1.6; 95% CI: 1.1, 2.3) and lower odds of reporting shopping online for groceries due to the variety of special foods, such as Asian grocery items (OR = 0.7; 95% CI: 0.5, 0.98) (Table 3). Participants with a high acculturation score also had lower odds of reporting poor quality foods as preventing them from shopping online for groceries (OR = 0.6; 95% CI: 0.4, 0.7). Participants with a high acculturation score were not less likely to buy groceries on the internet, but they had lower odds of reporting shopping online for Asian groceries specifically (OR = 0.7; 95% CI: 0.5, 0.9).

Discussion

We found that almost 30% of AA adults shop online for groceries in a typical month, which is slightly lower than the 39% of participants in the Nielsen National Consumer Panel who reported ever shopping online for groceries in July 2020 [9]. Our results also suggest that less-acculturated AA adults are more likely to shop for Asian grocery items online, potentially due to the availability of special foods. South Asian adults were the least acculturated subgroup in our sample, for example, and they were also more likely than other subgroups to purchase Asian grocery items online. It is possible that less-acculturated South Asian adults may be able to find culturally-preferred foods (e.g., spices) in popular online retail websites, whereas, less-acculturated East and Southeast Asian adults may be able to find preferred items in their primary brick-and-mortar stores (e.g., soy sauce, fish sauce).

The variety and quality of goods were more likely to motivate South Asian adults to start shopping online for groceries; whereas, Southeast Asian adults, who reported lower income levels, were more likely to shop online due to low prices. Southeast Asian adults were also more likely to report buying sugar-sweetened beverages online, and less likely to report buying fresh fruits and vegetables online. Taken as a whole, these findings suggest that initiatives and changes designed to promote healthy food purchases among AA adults should be tailored to the food shopping considerations of ethnic subgroups. These findings echo our recent report examining changes in diet and food shopping behaviors among AA adults due to COVID-19, wherein a higher percentage of Southeast Asian adults reported not having sufficient financial resources to safely acquire an adequate supply of food compared to other AA adults [5].

Our work has important limitations. Because we recruited participants from an online panel, responses may be more generalizable to adults who are proficient at using the internet. Respondents also had higher levels of income and education than the total AA population in the U.S., so results are less generalizable to those with lower food budgets. However, we observed wide variation in generational status and acculturation score in our sample, and we had a large sample size overall. We also collected data disaggregated by country of origin, which allowed us to compare responses by ethnic subgroup.

Given the capacity for online grocery shopping to address racial/ethnic inequities in health, including mitigating disparities in food access and protecting individuals during emergencies [10], our work highlights which issues related to online shopping could be addressed by interventions and for whom these changes would have the greatest impact. For example, offering high-quality and culturally-specific grocery items may increase the use of online grocery shopping among less-acculturated AA adults, similar to other culturally-tailored programs. Our findings also strengthen the argument that disaggregating responses by racial/ethnic subgroup is an essential next step in survey research. Future researchers may want to explore differences in the availability of culturally-preferred food items and brands in online versus brick-and-mortar stores, and how the availability of Asian-specific online retailers influences online food purchasing behaviors.