Non-immigrant STEM students in the US. Is the US still a magnet for attracting STEM talent from foreign countries?

Final Projects, Visualization


One of America’s Most vital Exports, according to a New York Times Magazine article, is the export of education, although it never goes abroad. An excerpt from this article states that at the University of Illinois at Urbana-Champaign, 1/5th of tuition revenue relies solely on Chinese students. Now imagine how much more money there would be if we accounted for all non-immigrant students who do not have citizenship in America. On the other hand, for many non-immigrant students, America has been a beacon of quality education. Higher education in the US can often lead to career opportunities within the US and is a path that is highly sought after by non-immigrant students who choose to stay in the US after graduation. There are several visa options for non-immigrant students, but the typical route goes somewhat like this:

Firstly, the non-immigrant student gets enrolled in a program and finishes the program. After completing a program, the student applies for Optional Practical Training (OPT) and works for one year, and extends to STEM OPT for two years if eligible. Finally, the student switches their visa to an H-1B visa, the Specialty Occupations visa, during their OPT period or after their OPT period. STEM OPT extension incentivizes foreign talent to pursue a STEM-related degree as it gives more opportunities for career growth in the US.

Typical flow of how F-1 students change their visa status to H-1B

For context, the F-1 visa is the most popular type of student visa for non-immigrant students. This would mostly entail enrolling in a full-time academic program and would lead to an Optional Practical Training or OPT extension that allows students to have work experience in their field of study for a duration of one year. The STEM OPT, however, is an additional option for students in any STEM field to pursue two more years of Optional Practical Training. 

As a group of two non-immigrant students, we want to investigate the overall circumstances of non-immigrant students in the United States. In this project, we aim to observe and analyze trends pertaining to the academic and professional career path of non-immigrant students who study in the US, looking specifically at the STEM field/industry. The questions that we are curious to learn about are as follows:

    1. What are the trends among non-immigrant students in the US? What are the Countries of citizenship and the number of students from their respective countries? And how many students are participating in STEM OPT?

    2. How many non-immigrant students continue to work after graduation & how many are in STEM? Looking specifically at F-1 SEVIS Record vs. STEM OPT extensions, is there a drop or an increase?

    3. What are the top industries that sponsor H-1B visas? Where are the states and cities?

    Materials & Methods – Datasets and Tools

    To start, we decided to look at datasets provided by the United States Customs and Immigration Service (USCIS) and US Immigration and Customs Enforcement (ICE). From there, we selected around four datasets, three distinct datasets on non-immigrant students and one on company, industry, and H-1B visa holders.

    The datasets on non-immigrant students were obtained from the US Immigration and Customs Enforcement (ICE) website. We looked specifically into the data in 2021 on the total number of active SEVIS records by country of citizenship and the total number of SEVIS records with authorization to participate in STEM OPT by country of citizenship. We focused on this data because we wanted to analyze and visualize the data on where most non-immigrant students come from and how many of them are studying STEM subjects. Next, we looked into the data on employment authorization document (EAD) issuances, where we found information on how many non-immigrant students applied for employment authorization from 2007 – 2021. The data includes how many students are employed and unemployed with an EAD and categories for students under OPT and STEM OPT.  For our last dataset, we looked into USCIS to find the data on companies that sponsor H-1B visas for foreign workers in 2021. The data contains the employer’s name, NAICS, Tax ID, approval and denial, and location.   

    After obtaining our datasets, we moved to Microsoft Excel to highlight relevant information we wanted to visualize. We joined the unemployed EAD holders and employed EAD holders tables to make it easier to visualize the difference in Tableau.

    Using a nested IF function, we transform the NAICS code of companies that sponsor H-1B visas to their respective industry categories.

    Following that, on ArcGIS, we created a map visualizing the country of citizenship of non-immigrant students. Because we didn’t have access to the desktop version of ArcGIS, we had to use a third-party website called Geoapify to convert the name of the countries in our data to geographic coordinates.

    Image from Geoapify
    Image from ArcGIS

    Charts and other visualizations were made by Tableau to support the narrative we built. Some joining of tables within Tableau was done when there were different datasets with joinable variables, namely ‘year’. We also put a filter on the count of H-1B to visually strengthen the message we are trying to deliver and ensure that the charts do not look overwhelming. 

    Image of filter for the count of H-1B in Tableau

    We intentionally use high-contrast color combinations for our visualization to make them easy on the eyes. For example, the use of blue and yellow, green and purple, and shades of blue.


    General Demographic of non-immigrant students in the US

    Answering our first question, which is to find out the general demographic of students coming to the US to study, we’ve selected a 2021 dataset that contains two variables. The first one [blue/teal] is the total number of SEVIS IDs– a unique ID given to every non-immigrant student coming to the US to study– and the second one [yellow] is the number of SEVIS ID Records of students registered to participate in STEM OPT.

    Layering both variables on top of each other not only allows for a look at where students hold citizenship, as well as the number of students from a particular country, but it also communicates a projection of a trend. A country such as Iran has a larger yellow bubble. It shows that a larger number of students retaining their student status in the US, working in a STEM-related field or industry, while the number of students from that country in 2021 is lower. This means there is a retention of STEM talent in the US specific to Iranian citizens.

    View the interactive ArcGIS Map using this link

    Non-immigrant students who choose to join the workforce by participating in OPT

    Looking at the trends of non-immigrant talent in the US on a general scale, the number of those who choose to participate in OPT has increased over time. From 2008 to 2022, we can assume that there has been a major increase in non-immigrant students studying in the US. From 2014 to the end of 2017, the number of students participating in OPT doubled. The purple layer, which displays the number of non-immigrant students who get hired while they are participating in OPT, showed an increase in overtime as well. We can see that the ratio of employed vs. unemployed stays on a plateau of ~50% after 2016. From this, we see that the prospects of talent retention saw an increase even with the existence of political and global factors, such as the presidential election of Donald Trump, as well as the COVID-19 pandemic, which posed a barrage of barriers for foreign individuals in the US.

    Bar chart showing the number of non-immigrant students who choose to participate in OPT by joining the workforce

    Scrutinizing non-immigrant students who are in the STEM field, we decided to use the same data from 2017-2021 but overlay employed individuals regardless of the field they are in [purple], with data of employed individuals who are specifically in a STEM field or position [blue]. We decided to highlight the year from 2016 to 2019. From 2016-2017 both general and STEM fields saw an increase. However, in 2017-2019, while there was a drop in employment from 2017-2019 for general employment, there was an increase in the STEM field instead.

    To make sense of the drop in 2020 onwards, we’ve decided to take into account the COVID-19 pandemic and quarantine period. According to resources from the DHS, maintaining lawful status for non-immigrant students means that they have been enrolled in a program and be considered “full-time” by their respective institutions. At a point back in 2020, pre-virtual classes meant that students had to show up for classes in person to maintain their lawful status, at least until the summer of 2020 when many schools, including Harvard, sued the US Immigration and Customs Enforcement (ICE) to withdraw a directive that would have prevented any non-immigrant students from maintaining their status and taking classes online.

    It is also important to note that while there is a drop in general employment in 2020, looking at the visualization, we can see that individuals employed in a STEM field did not see as steep of a drop.

    Employed individuals in general [purple] in contrast with employed individuals in STEM [blue], from 2007-2021.

    Employment for H-1B, switching status from student to professional

    One final piece we’ve used to investigate the topic of research pertains to how non-immigrant students transition from a student status to a professional status in the US. We decided to overlay the number of H-1B sponsors by industry, state, and city. From the dashboard, we can see that most professionals who get sponsored to stay in the US to work are in larger cities in the US. New York City, given its history, has maintained its status as an international city, having the largest amount of foreign talent sponsored to stay and work in the US. This is followed by San Francisco and other cities like Chicago, Houston, and Boston. By state, however, it seems like California has the highest amount of foreign talent, followed by New York, New Jersey, Texas, Massachusetts, and Illinois.

    By industry, at first glance, we can see that “Professional, Scientific, and Technical Services” contain the largest amount of foreign talent. While it is hard to directly categorize an industry by whether it is STEM-related or not, we can still see that industries such as “Manufacturing”, “Health Care and Social Assistance”, and “Information”, industries that are by names associated with the STEM field are among the top industries that have a large number of foreign talents.

    Dashboard of the number of H-1B sponsors by industry, state, and city.


    To summarize the trends we are seeing, we can generalize that a huge portion of talents who are interested in studying STEM subjects come from Asia. We also see that the number of international students who pursued post-graduation employment increased exponentially from 2008 to 2022, which indicates that the number of non-immigrant students also increased. Although there is a decline in the number of general employment from 2017 to 2021, we note that the number of employment in the STEM field/ industry increased, which indicate two possibilities. The first possibility is that the STEM industry hires more than the other industries. Second, more foreign talents are interested in working in the STEM industry compared to the previous years. Lastly, regarding how many companies sponsor H-1B, companies in the STEM industry and located in major metropolitan areas have a higher tendency to sponsor foreign talents for H-1B visas.


    This project is the result of a semester-long course on information visualization. Our team of two worked together on all fronts to figure out a topic of interest, find datasets that support our research topic, and finally produce a narrative that we would like to educate others on. We were able to utilize all of the materials that we have learned closely from other projects throughout the semester. We created visualizations and stories that are not only compelling but also answer the research questions that we have proposed. We also believe that this project can be used to build another layer of research regarding how H-1B visa holders transition to permanent residents.