To report on the background of news industry interns, we first created a survey on a Google Form, circulated it on social media, asking current and recent news interns at top major news publications to fill out a form to find out more about whether they got paid, their financial situation during the internship, their race, and their educational background.
We decided to focus on seven major news organizations that ran internship programs during the summer of 2018 — The New York Times, Washington Post, Wall Street Journal, NPR, Los Angeles Times, Chicago Tribune and Politico.
The New York Times and Washington Post’s list of interns for those years were already online. The Los Angeles Times and Politico gave us their list of interns when we asked for them. The Wall Street Journal, NPR and Chicago Tribune declined to provide their intern lists.
For those organizations, we were able to use a combination of internal memos, interviews, LinkedIn searches, survey responses and social media to identify the interns. We also emailed interns we identified and asked them to fill out the survey.
We created a survey to systematically request the interns’ race, if they got paid, their financial situation during the internship, their previous internships,to confirm their schools and if they wanted to be interviewed by us.
The number of interns at each publication varied from six or seven (Politico and the Chicago Tribune) to 33 (NPR). Since the sample sizes for each organization was small, the differences in percentage of interns from intensively selective between organizations should be treated with caution. To give an example, an increase of two interns from intensively selective schools in an intern class could increase the percentage of interns from intensively selective schools by 7%.
After identifying interns, we decided to exclude interns in non-editorial departments, such as the business, marketing and product departments. We included graphic design and social media interns.
To establish the interns’ education background, we tried to send sent surveys directly to more than 150 students through email, Twitter direct messages and other methods. We used LinkedIn searches where we were lacking information on their higher education.
We assembled a list of about 150 interns.
About 70 former interns responded, and we completed follow-up interviews with 16 of them.
After collecting a list of the interns and their school, we needed to systematically rank the selectivity of a particular school. We chose a system already used by a Harvard economics professor, Raj Chetty, which is based off an admissions competitiveness index created by the publication Barron’s, which publishes a popular guide about colleges. The Barron’s index describes how competitive admission is to a certain school. The index takes GPA, SAT scores, acceptance rates and class rank of the school’s incoming students into account.
Chetty’s dataset of colleges and universities was generated originally from the Department of Education’s IPEDS database and College Scorecards database. From this dataset, we excluded two-year or less institutions and for-profit four-year institutions. We found 1,357 schools in the U.S. after excluding these schools.
The same selectivity dataset by Chetty was also used in an Upshot column in The New York Times titled, “Some Colleges Have More Students From the Top 1 Percent Than the Bottom 60. Find Yours.”
Chetty divides school’s selectivity in this way:
“Ivy Plus,” (combining the eight Ivy League schools with Stanford, MIT, Duke and the University of Chicago), which represent the top 1% of the nation’s four-year colleges; schools in this category were rated with Barron’s top selectivity group.
“Elite” universities such as Northwestern and NYU. There are 65 in this category, which represents 5% of the nation’s colleges; These schools also received Barron’s top selectivity rating.
“Highly selective” schools such as UC Berkeley and University of Florida. There are 99 in this category, which represents 7% of the nation’s colleges. This included Barron’s 2nd selectivity group.
“Selective,” schools represent 74% of the nation’s more than 1,300 four-year not-for-profit schools; this included Barron’s 3rd, 4th and 5th selectivity group.
“Nonselective or other,” which represents the remaining 13%. They include newly established schools and colleges that went to too little information. This included Barron’s lowest tiers of selectivity. Other also included schools without enough information to be categorized.
For our analysis, we consider all universities in these three tiers — the top 13% most selective four-year colleges nationally — as “intensely selective.”
In our dataset, at least 20 interns had a journalism master’s degree. However, we still only counted their undergraduate degree when determining their school’s selectivity since there was no dataset that described the selectivity of graduate schools, which have a separate admissions process than undergraduate schools.
We used different data to conduct the analysis of the educational background of Voices students. The dataset we had of Voices students is made of a document that generally identifies the school that they obtained their most recent degree, which sometimes were graduate degrees. We used that information for the data analysis, the best available before the publication of this story.
Analysis and graphs were done in the programming language R primarily with the tidyverse package.