Our hypotheses in the following study were:

The researchers initiated a focused group discussion among the researchers, teachers, and counselors to discuss the methods and procedures required. The members approved purposive sampling, online data collection, and intervention owing to the COVID-19 and convenience factors. Although members suggested different questionnaires, the unanimous consensus was to use the perceived stress scale, Rosenberg’s self-esteem scale, and the Kuppuswamy socio-economic scale. The researchers decided on questionnaires after carefully validating the objectives and suitability of the questionnaires to the Indian culture. The discussions led to checking the reliability and validity of the questionnaires to ensure the quality of the research. The members also suggested a pilot study to confirm the proposed intervention plan.

The focused group members are all from the Amrita School of Engineering, Coimbatore and devised the research strategy from first-hand experience with the potential participants. Thus immediately after the focused group discussion, researchers framed the research outline and sought ethical clearance.

This research gained approval and ethical clearance (AMRITA/SOE/ADMN/DOE/01/2021/01) from the Ethics Committee, Amrita School of Engineering, Amrita Vishwa Vidhyapeetham, Coimbatore, India. Written informed consent was obtained from the participants on the Google form. Participants were also given the necessary details about the intervention outcomes and assured that they had full rights to withdraw at any point of the research. The data collection and analysis were done based on the ethics committee, Amrita Vishwa Vidhyapeetham, India, and the personal identification details of the participants were removed. The tools (Perceived stress scale, Rosenberg self-esteem scale & Kuppuswamy socio-economic scale) ^{46, 47} were added to a Google Form, and a total of 100 students completed the form. Then every student’s stress and self-esteem were calculated. Although the Google form had data from 100 students, not all had high stress and low self-esteem. Hence, a threshold value was chosen for both the variables, where the value for stress was 22 and self-esteem was 31. A total of eighty students fell into this category, upon which they were sent emails regarding their scores and the option to attend the mindfulness-based intervention. Out of the 80 students, 75 responded positively and attended the intervention. Figure 1 indicates the process flow of the study.

The stress and self-esteem scores before the intervention were taken from the Google form, and after the intervention, the students were asked to fill out a new Google form containing the same questionnaires. A follow-up was conducted using the same set of procedures one month after the intervention. The results were analyzed using SPSS 23.0 (IBM SPSS Statistics: RRID: SCR_016479) ⁴⁸ .

Purposive sampling was used to select college students involved in the placement process at Amrita Vishwa Vidyapeetham University in Coimbatore, India. The inclusion criteria were: (1) Students with high stress and low self-esteem or either of the above. (2) Students who were in their final year of college, preparing for their placements. (3) Students with a grade point average of more than 6.5 out of 10 (University placement eligibility policy). The exclusion criteria were: (1) Students who were preparing to pursue higher studies inside or outside India. (2) Students who were planning start-ups. One hundred students who fit the criteria mentioned above, completed a demographic survey shared with them through Google Forms.

In this research, self-esteem is the perception of the sense of worth of the college student involved in placement possess. This self-esteem may have varied due to the adverse situations produced by placements during COVID-19. Stress is the perception of burden and negative feeling due to a problem, person or thing caused by the self-evaluation that they lack the specific skills required to handle that situation. This stress may lead to physical disturbances also. However, physical disturbances were not included in the scope of the study. Henceforth, stress was defined as an overtaxing, and heavy feeling and thinking in college students caused by the placements during COVID-19.

A demographic survey was used to collect participants’ data such as name, age, class, gender, and consent to participate in the intervention. The students were informed about confidentiality and that their details would not be revealed to anyone under any circumstances. At the end of the intervention, the students were asked to provide feedback (if any). The following tools were used to collect data:

The perceived stress scale (PSS) was designed by Cohen et al. ^{49, 50} to measure the perceived stress in individuals. It is a five-point Likert scale ranging from zero to four with the options of never to not very often. It is a 10 item instrument where the total score is summed, and the higher the score, the higher the perceived stress. Some of the items are, "In the last month, how often have you been upset because of something that happened unexpectedly?", "In the last month, how often have you felt that you were on top of things?". This scale’s reliability was assessed by pre-testing it to one hundred and six college students participating in campus placements. The Cronbach’s alpha value of 10-item PSS indicates moderate internal consistency of 0.79, and test-retest reliability for a week showed, Intra-class correlation (ICC) of 0.83. During focused group discussion, the expert team authorized the face validity and content validity of the same.

The Rosenberg self-esteem scale is a 10 item scale for estimating the value of an individual’s self-esteem, it has a four-point Likert scale ranging from strongly agree to disagree strongly ^{50, 51} . Some of the statements are: "I certainly feel useless at times" and "I take a positive attitude towards myself". While conducting the focused group discussion, face validity and content validity were discussed: a pre-testing of this 10-item scale was carried out on 106 college students involved in campus placements. Results showed that Cronbach’s value for internal consistency was 0.85 and ICC value for test-retest reliability was 0.86.

The Kuppuswamy’s socio-economic scale ^{50, 52} is mainly used to measure the status of a person in society by considering parameters such as education, occupation, and salary earned by the head of the family. The parameters, as mentioned earlier, have further subgroups where each has individual scores, the total of which ranges from 3–29. Based on the total score, the students were grouped into the following five classes: upper class, upper-middle class, lower-middle class, upper-lower class, and lower class.

The mindfulness-based intervention used in this research was inspired by Kabat-Zinn ^{17, 50} . The intervention used in this experiment combined five mindfulness meditation techniques and mindfulness-based activities. The outline of the interventions used is tabulated in Table 1. The sessions were conducted online using Google meet, and each session was for two hours. Thus, the students attended one session weekly and, overall, eight sessions. In addition, the college students were asked to practice at least 30 minutes of mindfulness meditation on their own and report to the researchers. The college students were eager to oblige, and they completed the practice for eight weeks and reported their progress weekly once a month.

The quantitative measures were analyzed using SPSS 23.0, segregating the research process into pre, post, and follow-up phases. The demographic variables, namely age, gender and socio-economic status, were analyzed using percentage analysis. Researchers calculated the mean and standard deviation to understand the changes in the variables during pre, post and follow-up phases. One-way ANOVA portrayed the significant difference caused in the variables because of mindfulness-based intervention. Post hoc analysis was employed to identify the significant changes during the pre, post and follow-up phases of mindfulness-based interventions.

The researchers conducted a pilot study to ensure the feasibility of the design and outcome. Out of fifty involved in the placements, thirty had high stress and low self-esteem and among them, a few dropped out due to personal and academic difficulties. The pilot study started by pre-testing the twenty-one college students with the Perceived stress scale and the Rosenberg self-esteem scale followed by a mindfulness-based intervention: eating meditation, breathing meditation and body scan meditation for two weeks (two sessions, total four hours). It was found that the pre-phase mean value of stress was 28.05, and the post was 24.33. Similarly, the pre-phase self-esteem value was 23.80, and the post was 25.85. A one-way ANOVA test showed that there was a significant difference in the stress scores during the pre- and post phases of the mindfulness-based intervention (F (1,40)) = 11.69, p = 0.00). However, for self-esteem the one-way ANOVA indicated that there was no statistically significant difference between the pre- and post phases (F (1,40)) = 2.47, p = 0.12). Although there was a rise in the mean score, it was not statistically significant. In the focused group, discussion members suggested the following inclusions and those were carried out in the study. Expanding the intervention time from two weeks to eight weeks and including the meditation methods such as mountain meditation and loving-kindness meditation A session was devoted solely to connecting with meditation, self-reflection on its benefits and committing to understand and practice MBI.

Although 80 students needed the mindfulness-based intervention, five members opted out due to conflict with college practical lab sessions and personal reasons. Thus, the mindfulness-based intervention commenced with 75 members, and all of them attended all three phases: pre, post and follow-up.

Table 2 shows the demographic variables of the college students. The college students involved in the study were from 19 to 21 years of age, of which 48% of the population were 20-years-old, while 12% were 21-years-old. Twentynine percent of the students were female, and 71% were male. The absence of the socio-economic classes upper-lower and lower among the college students was noted. The majority of the college students were from upper-middle class (50.67%) and upper class (46.67%) backgrounds, while a few were lower-middle class (0.02%).

Table 3, shows the pre-phase mean value of stress was 22.57, which is higher than the post-phase mean value 18.25, indicating that the intervention has reduced the stress levels in the participants. Similarly, the pre-phase mean value of self-esteem was 25.92, which is lower than the post-phase mean value 28.90, indicating that the intervention increased self-esteem among the participants. During the follow-up phase, the mean values for stress and self-esteem were 18.04 and 29.09, respectively, and these scores show that the impact of the intervention was sustained.

Table 4 shows the approximate F calculation of the pre-, post-, and follow-up phases for both the variables (self-esteem and stress). The output of analysis shows that for self-esteem, there was a statistically significant difference between pre-, post-, and follow-up phases as determined by one-way ANOVA, (F (2,222)) = 23.21, p = 0.00). There was also a statistically significant difference between the pre-, post- and follow-up phases for stress as determined by one-way ANOVA (F (2,222)) = 18.15, p = 0.00). The results indicated that mindfulness-based interventions helped college students during the placements.

Figure 2 and Figure 3 show the changes in the mean stress and self-esteem scores across the pre-, post-, and follow-up phases of the intervention. It can be observed from Figure 2 that there is a significant difference in stress during pre- and post- phases of the intervention. However, it can also be observed that the difference tends to decrease between the post- and follow-up phases of the intervention. A similar pattern can also be observed with self-esteem in Figure 3. More specifically, the mean value of stress and self-esteem improved throughout the study.

Table 5 shows the Post hoc analysis for the pre-, post-, and follow-up phases of intervention for stress and self-esteem. The mean stress scores of the college students during pre- and post-phases changed with a significant difference, 4.32 ± 0.75, p = 0.00. Pre- and follow-up phases also significantly differed, 4.53 ± 0.75, p = 0.00. There was no statistically significant difference between the follow-up and post-phases, 0.21 ± 0.75, p = 1.00. This analysis showed a significant decline in the stress scores of the pre- and post-phases because of the intervention. Self-esteem mean score pre- and post-phases (2.97 ± 0.75, p = 0.00.) and pre- and follow-up phases (3.17 ± 0.75, p = 0.00.) were statistically significant. Though there was an increase in the mean score of the post- and follow-up phases, the post hoc analysis revealed that (0.18 ± 0.75, p = 0.00.) were not statistically significant. Thus, we can conclude that the mindfulness intervention raised the self-esteem scores of the college students involved in placements.

College students need to take up new roles and make tough decisions in life. In addition, they have to meet the demands, pressure, and expectations from society and the after effects of the pandemic. Thus, mindfulness-based interventions will aid their focus and set a path for a bright future. Frontline workers like psychologists, psychiatric nurses, and psychiatric social workers can teach mindfulness to college students. The awareness and impact of mindfulness-based interventions should be made mandatory to all colleges and universities. In addition, the government could allocate funds and grants to researchers and trainers who conduct research and training programs on mindfulness.

Purposive sampling was used to conduct this research in a single university. Future research should cover a diverse population, including more colleges in different districts/counties. The same experiment could also be conducted for alumni who are yet to find employment. Focusing on a diverse population and including parameters covering different areas of psychological health will open up new scope for research. Although the post hoc analysis revealed statistically significant results, the mean scores of stress and self-esteem did not continue in the follow-up phase. This could be rectified by assisting the participants to engage more in mindfulness meditation interim sessions. In addition, mindfulness meditation could be introduced as a life skill by creating an online presence on platforms like websites, apps, and courses. This will encourage students to habituate mindfulness meditation as a lifelong learning process.

Figshare: Underlying data for ’Does online mindfulness-based intervention help college students succeed in their job search during the COVID-19 pandemic?’ https://doi.org/10.6084/m9.figshare.19360955.v4 ⁵⁰

This project contains the following underlying data:

Figshare: Extended data for ’Does online mindfulness-based intervention help college students succeed in their job search during the COVID-19 pandemic?’ https://doi.org/10.6084/m9.figshare.19360955.v4 ⁵⁰

This project contains the following extended data:

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).