LinkedIn conducted social experiments on 20 million users over five years

LinkedIn has experimented with more than 20 million users over five years, which, while intended to improve the way the platform works for members, could have had an impact on some people’s livelihoods, according to a new study.

In experiments conducted around the world from 2015 to 2019, Linkedin randomly changed the proportion of weak and strong contacts suggested by the People You May Know algorithm — the company’s automated system for recommending new contacts to its users. Researchers at LinkedIn, MIT, Stanford, and Harvard Business School later analyzed data collected from the tests in a study It was published this month in the journal Science.

LinkedIn’s algorithm experiments may come as a surprise to millions of people because the company hasn’t told users that tests are in progress.

Tech giants like LinkedIn, the world’s largest professional network, routinely run large-scale experiments where they try different versions of app features, web designs, and algorithms on different people. The lengthy practice, called A/B testing, aims to improve consumer experiences and keep them engaged, helping companies make money through premium membership fees or advertising. Users often have no idea that companies are running tests on them. (New York times Uses such tests To evaluate headline drafting and make decisions about the products and features the company releases.)

But changes made by LinkedIn suggest that such tweaks to widely used algorithms could become social engineering experiments with potentially life-changing consequences for many people. Experts who study the societal impacts of computing have said that long, large-scale experiments on people can affect their job prospects, in ways invisible to them, raising questions about industry transparency and oversight of research.

“The results indicate that some users have better access to job opportunities or a significant difference in access to job opportunities,” he said. Michael ZimmerD., associate professor of computer science and director of the Center for Data, Ethics and Society at Marquette University. “Those are the long-term consequences to think about when we think about the ethics of engaging in this type of big data research.”

Studying in Science tested an influential theory in sociology called The strength of weak relationships‘, which asserts that people are more likely to get work and other opportunities through acquaintances with arms than through close friends.

The researchers analyzed how algorithmic changes in LinkedIn affected users’ job mobility. that they I found that social ties are relatively weak LinkedIn has proven to be twice as effective in securing job opportunities as stronger social relationships.

In a statement, LinkedIn said that during the study it “acted consistently” with the company’s user agreement, privacy policy and member settings. The privacy policy Indicates that LinkedIn uses members’ personal data for research purposes. The statement added that the company used the latest “non-invasive” social science technology to answer important research questions “without any experiments on organs.”

LinkedIn, which is owned by Microsoft, did not directly respond to a question about how the company considered the potential long-term consequences of its experiments on the employment and economic situation of users. But the company said the research did not disproportionately benefit some users.

He said the aim of the research is to “help people on a large scale” Karthik Rajkumar, an applied research scientist at LinkedIn and was one of the study’s co-authors. “No one has been put at a disadvantage to find a job.”

Sinan Aral, a professor of management and data science at MIT who was the study’s lead author, said the LinkedIn experiments were an attempt to ensure users had equal employment opportunities.

“To run an experiment on 20 million people and then come up with a better job prospects algorithm for everyone as a result of the knowledge you learn from this is what they’re trying to do,” Mr. Aral He said, “Instead of scanning some people with social mobility and some not.” (Professor Aral conducted data analysis for The New York Times, and received an award Research Fellowship Scholarship from Microsoft in 2010.)

Experiences on users by major internet companies have a checkered history. eight years ago, Facebook Study A description has been published of how the social network has quietly handled posts that appeared in users’ news feeds in order to analyze the spread of negative and positive sentiment on its platform. The week-long trial on 689,003 users, quickly generated a backlash.

The Facebook study, whose authors included a company researcher and Cornell University professor, confirmed that people tacitly agreed to experience emotional manipulation when they signed up for Facebook. “All users consent before creating a Facebook account, and constitute informed consent for this research,” the study said.

Critics disagreed with some, with some attacking Facebook because it invaded people’s privacy, exploited their mood and caused them emotional distress. Others emphasized that the project used an academic co-author to lend credibility to problematic companies’ research practices.

Cornell later said that his internal ethics board had not been asked to review the project because Facebook conducted the study independently and that the professor, who helped design the research, had Do not share directly In experiments on humans.

LinkedIn professional networking experiences have been different in intent, scope, and scale. It was designed by Linkedin as part of the company’s ongoing effort to improve the relevance of its “People You May Know” algorithm, which suggests new member connections.

The algorithm analyzes data such as members’ employment history, job titles, and relationships with other users. It then attempts to measure the probability that a LinkedIn member will send a friend invitation to a proposed new connection as well as the probability that this new connection will accept the invitation.

For the experiments, LinkedIn modified an algorithm to randomly change the spread of strong and weak links recommended by the system. The study stated that the first wave of tests, conducted in 2015, “involved more than four million experimental subjects”. The second wave of tests, conducted in 2019, involved more than 16 million people.

During the tests, people who clicked on the People You May Know tool and looked at recommendations were assigned different algorithm paths. Some of these “therapeutic variables,” as the study called them, caused LinkedIn users to form more relationships with people with whom they had poor social ties. Other modifications caused fewer relationships to form with weak ones.

Whether most LinkedIn members understand that they may undergo experiences that may affect their employment prospects is unknown.

LinkedIn privacy policy He says the company may “use personal data made available to us” to research “workplace trends, such as job availability and skills for those jobs.” that it External Researchers Policy Seeking to analyze company data clearly indicates that these researchers will not be able to “experiment or perform tests on our members”.

However, neither policy explicitly informs consumers that LinkedIn itself may experiment or conduct tests on its members.

In a statement, LinkedIn said, “We are transparent with our members through our search section of our user agreement.”

In an editorial statement, Science said: “We understand, and the reviewers understand, that the experiments conducted by LinkedIn are operating under the guidelines of their user agreements.”

After the first wave of computational tests, researchers at LinkedIn and MIT came up with the idea of ​​analyzing the results of those experiments to test the theory of the strength of weak links. Although the decades-old theory has become a cornerstone of the social sciences, it has not been rigorously proven in a large-scale prospective experiment that randomly assigned people to social bonds of different strengths.

External researchers analyzed the data collected from LinkedIn. The study reported that people who received more recommendations about relatively weak contacts generally applied for and accepted more jobs — results consistent with the weak link theory.

In fact, relatively weak contacts — that is, people with whom LinkedIn members shared only 10 mutual connections — proved much more productive in job search than stronger contacts with users sharing more than 20 mutual connections, the study found.

After a year of contact on LinkedIn, people who received more recommendations for fairly weak contacts were twice as likely to get jobs at the companies where those contacts worked compared to other users who received more recommendations for strong connections.

“We found that these somewhat weak links are a better option to help people find new jobs and much more than stronger relationships,” said Mr. Rajkumar, a researcher at LinkedIn.

The study reported that 20 million users who participated in LinkedIn experiences created more than 2 billion new social connections and completed more than 70 million job applications that led to 600,000 new jobs. The study said that weak links proved most beneficial for job seekers in digital fields such as artificial intelligence, while strong links proved more beneficial for employment in industries that rely less on software.

LinkedIn said it applied the findings about poor relationships with several features including a new tool Notify members When a first or second order connection is assigned. But the company hasn’t made study-related changes to the People You May Know feature.

Professor Aral of the Massachusetts Institute of Technology said the study’s deeper significance is that it demonstrated the importance of powerful social networking algorithms – not only in amplifying problems such as misinformation but also as key indicators of economic conditions such as employment and unemployment.

Catherine Flick, a senior researcher in computing and social responsibility at De Montfort University in Leicester, England, described the study as more of an institutional marketing exercise.

“The study has an inherent bias,” Dr. Flick said. “It shows that if you want to get more jobs, you have to be on LinkedIn more.”