This study used a concurrently embedded mixed-method design combining quantitative and qualitative research (Fig. 1). This was conducted as a pilot study; quantitative research was carried out using a nonequivalent pre-post-test control group, and qualitative research was done through focus group interviews.
Participants were recruited through invitations on flyers, posters, and internet bulletins of social welfare institutes for migrant workers. The inclusion criteria were migrant workers, who were legally employed adults under 40 years of age, working full-time, with intermediate-level Korean or Level 3 or higher qualification in the Test of Proficiency in Korean (TOPIK), a smart cellular phone, and the ability to use mobile applications. TOPIK Level 3 is equivalent to intermediate-level Korean language proficiency, which makes it possible to perform basic language functions necessary for using various public facilities and maintaining social relations without difficulty. Illegal migrants and those who had participated in other health promotion programs in the preceding three months were excluded. We contacted 66 migrant workers who showed interest in this study but 19 did not meet the inclusion criteria. A total of 47 migrant workers from 9 countries (i.e., Cambodia, Mongolia, Vietnam, Myanmar, etc.), 25 in the intervention group and 22 in the control group agreed to participate in 12 intervention weeks. During the intervention, five dropped out and the final analysis included data from 42 workers, 21 in each of the two groups.
During the 12-week intervention period, a total of six sessions, consisting of four group activities (healthy physical activity, healthy dietary habits, acculturation, and stress management) in the form of workshops involving discussion and practice, and two cultural activities (university tour and dental check-up, marathon participation in a community sports event) in the form of outdoor activities using community resources were organized. All sessions were based on living lab principles, which encourage participants to solve their own health problems themselves. The first session on “Healthy physical activities” consisted of stretching and muscle exercises in accordance with the 2014 Musculoskeletal Disease Prevention Manual of the Korea Occupational Safety and Health Agency. Participants learned stretching and muscle exercise movements followed by a discussion about their benefits and specific ways to implement them in real life. Second, the session on “Healthy dietary habits” involved the dissemination of information on intake of a low-sodium diet to help cultivate healthy dietary habits. Participants assessed their taste for salt intake with sample foods and were invited to discuss their usual dietary behaviors and healthy dietary practices. In addition, participants were asked to consult the Ministry of Food and Drug Safety website to find restaurants that served low sodium meals in their neighborhood. The third group activity session stimulated participants' interest in Korean cultural events and enhance access to community resources by introducing 80 community programs to enable early adaptation and assimilation of migrant workers into Korean society. Lastly, participants were introduced to stress management techniques based on a better understanding of their personalities and that of others using DISC, a test tool for understanding propensity and useful for classifying behavior into four personality types (i.e., Dominant, Influencer, Steady, and Conscientious) so as to improve communication methods with peers of different personality types . At the end, participants were able to identify and understand their own personality type and that of the others; how migrant workers can communicate better in order to establish good interpersonal relationships was also discussed.
The contents and location of the cultural activities were decided based on participants’ preferences. First, migrant workers were offered a tour of a private university campus, from where the researchers hailed, and informed about its history and culture. By participating in a free dental care program at the university hospital, they were instructed in the importance of regular dental care. Second, a marathon along the Han River was organized; by participating in a 5 km marathon together, participants and researchers were able to appreciate the benefits of community resources used to organize physical activities and cultural events.
Health promotion lifestyle
The health promotion lifestyle profile (HPLP) was developed by Walker et al.  and originally comprised 26 items but was later shortened by Jeon et al. . The HPLP uses a four-point Likert scale (1 = never, and 4 = routinely), with higher total scores representing better lifestyle conditions for improving health. Cronbach’s α was .70 in this study.
The health literacy scale was developed by An and Yang  to measure the ability of migrant married women to read, write, understand, and use health information. The questionnaire comprises 10 items rated on a five-point Likert scale (0 = strongly disagree, and 4 = strongly agree). The mean level of health literacy in migrant workers and that of migrant married women (original study cohort) with intermediate-level Korean were found to be similar. Cronbach’s α was .71 in this study.
Self-care agency was measured using a questionnaire developed by Evers et al.  and translated by Kim.  It comprised a total of 15 items with responses ranging on a five-point Likert scale (1 = strongly disagree, and 5 = strongly agree). Scores for self-care agency were directly proportional to the total score. Cronbach’s α was .70 in this study.
Sense of Community
Sense of community was measured using the Korean version of the Brief Sense of Community Scale (BSCS), which was based on Peterson and colleague’s  scale and modified by Oh et al. . The scale includes eight items, rated on a five- point Likert scale (1 = strongly disagree, and 5 = strongly agree). Some of the BSCS items were changed for the purposes of this study: for e.g., “peer/friend” to “neighborhood” and “peer group” to “neighbor.” The sense of community score was directly proportional to the total score. Cronbach’s α was .87 in this study.
Social support was measured using a scale developed by Zimet et al.  and reconstructed by Shin and Lee . This scale consists of 12 items: there were 4 questions each on support from the family, support from meaningful others, and support from friends. Each response was measured on a five-point Likert scale (1 = strongly disagree, and 5 = strongly agree). Scores for social support were directly proportional to the total score. Cronbach’s α was .88 in this study.
Acculturation was measured using a questionnaire based on Barry’s East Asian Acculturation Measurement and modified by Jang . This scale has 10 items measured on a five-point Likert scale (1 = strongly disagree, and 5 = strongly agree). The acculturation level was directly proportional to the total score. Cronbach’s α was .72 in this study.
Focus group interviews
Focus group interviews were conducted to freely discuss individual experiences, perceptions, and knowledge related to the intervention through active interaction with participants. It included an opening question, introductory question, transitional question, key questions, and a final question. Key questions were: “What are the useful conjectures found in today's health community activity?” “What are the barriers to engaging in health behavior in your daily life?” and “How can we overcome barriers and engage in health behaviors?”
This study was conducted in accordance with the guidelines set out in the Declaration of Helsinki, and all procedures involving study participants were approved by the ethics committee of the University of Yonsei. This study was conducted from June to September 2019. Researchers trained in ethics explained the study purpose, method, and process to participants before data collection; written informed consent was provided by all the participants.
All participants participated in two surveys: pre-test and post-test, and quantitative data were collected using self-reported questionnaires. Participants who did not understand the questionnaire were provided explanations in person by trained research assistants. Qualitative data were collected on aspects not measured by the survey to supplement the results of the questionnaire. Interviews were conducted in three 40-minute rounds by doctoral students trained in qualitative research methodology. Since migrant workers have slower speech delivery in Korean, interview data were not recorded, but transcribed on the spot.
Quantitative data were analyzed using SPSS Statistics 25.0 (IBM Corp, Armonk, NY, USA). All continuous variables were described as mean ± standard deviation, and categorical variables as frequency. Group differences were analyzed using the independent t-test, Mann-Whitney’s U-test and the chi-squared test or Fisher’s exact test. The pre- and post- intervention scores were compared and analyzed using Wilcoxon’s signed rank test.
Qualitative data were analyzed using qualitative content analysis. The two researchers (Y.K & J.K) read and verified interview transcripts to study migrants’ health engagement experience and their willingness to participate in the health behavior program. Meaningful data from the transcripts were extracted, compared, contrasted, and classified, and the main concepts and themes derived. Additional themes were drawn up to check the validity of the analytical methods used and results were further reviewed by one nursing professor with experience in qualitative research. The themes that emerged from the analysis were agreed upon after discussion between the researchers.