The LinkedIn experiment changed job prospects for millions — and it raises a red flag: privacy experts

Academics suggest Canada needs stricter laws on consent and how companies use people's data

A five-year study by LinkedIn of its nearly 20 million users raises ethical red flags because some unknowing participants in the social experiment may have limited job opportunities, advise experts in data privacy and human resources.

Online networks and social media platforms randomly vary the number of strong and weak acquaintances in user suggestions "People You May Know" to test an old theory: that people are more likely to land a new job through distant acquaintances than their close contacts.

The results of the study, published in Science Magazine on September 15, by LinkedIn, MIT, Stanford and Harvard researchers, confirm the idea: users who indicated contacts with whom they had only 10 mutual friends, doubled their chances of landing a new job, compared to shown. people with 20 mutual friends.

But it also means LinkedIn users whose algorithms are flooded with "close contacts" — those with 20 or more mutual friends — connecting with fewer opportunities through the networking site.

Given the possible consequences, it's unlikely that many people will knowingly consent to their networks, and their livelihoods, being manipulated as they are for this study, said Jonathon Penney, a law professor whose research focuses on internet, society, and data policy at Osgoode Hall in York. University. Law school.

'There's no way they'll agree'

That's "a large number of people who could be negatively affected in terms of job prospects just because of this study," Penney said of 20 million subjects. More than five million participants are said to have come from North America in the 2019 research phase.

"Most users, if you ask them, will say there's no way they would approve of this kind of study... I'm very concerned about ethics."

While academia holds strict ethical and disclosure standards, it is not uncommon for marketing or media companies to use algorithms to measure the success of a new product or service. This is a process known as A/B testing, in which the user has access to various online tools or experiences to analyze how one engages with them.

In an email to CBC News, a LinkedIn spokesperson said the company hopes to use the data to customize its service.

"Through these observations, we were able to determine that you are more likely to get the job from an acquaintance than your best friend," LinkedIn said in an email. "We can't wait to see how this research helps employers, recruiters and job seekers change the way we think about the labor market."

Comprehensive privacy policy

While the company never notifies its users of the experiment while it is in progress, its privacy policy states that LinkedIn may use member profiles to conduct research.

But online privacy experts who spoke to CBC News suggested that the standard privacy policies that people click on when signing up for online platforms give companies too much freedom in how they use people's information.

In fact, the purpose-limiting principle in Canada's Personal Information and Electronic Documents Protection Act (PIPEDA) states that user data may only be used for the purposes stated at the time of collection — but companies often push the envelope, says Ignacio Cofone, Canadian Research Chair in intelligence law. creation and data governance at McGill University.

"The problem ... is that companies very rarely know the purpose of using the data in the future," Cofone said in an interview. Thus, "the legal way in which business has evolved has allowed for a very broad purpose [the use of user profiles]."

The LinkedIn study "is a perfect illustration of how empty consent means in our online interactions for companies," Cofone continued. For example, it takes a person 250 hours to read the average number of privacy policies they approve in a year, he says—and those policies often change unilaterally.

Penney said he recognized the purpose of the LinkedIn study: a practical look at big data and human behavior. And the research has been the subject of an institutional review board for human subjects research, unlike Facebook's 2014 stealth psychological experiment, which sparked an investigation from UK data protection authorities.

Nonetheless, Penney says accepting a lengthy and intentionally vague privacy policy at the time of registration does not equate to the "informed consent" required of a typical human subject study - especially one that may have real-life consequences.

There are often significant hurdles that university-level studies need to overcome to conduct research on human subjects, Penney said. "You have to be very [precise] about the research and its purpose. If there's fraud, there's often additional safeguards to be put in place."

He also shared concerns that LinkedIn might use their study to test new avenues for profit.

“You can easily imagine that the kind of affordability of design LinkedIn is testing could be used for intention bias, where the best jobs [and] hiring opportunities are funneled to wealthier users,” Penney said.

Liking richer users

The platform has made important changes to offering benefits to paid users in the last five years, said Neil Wiseman, a senior consultant for Pivotal recruitment and HR services in Mississauga who uses LinkedIn in his work.

LinkedIn's premium subscription, starting at $30 per month, allows users to directly contact anyone on the platform. Those with free accounts, meanwhile, can only contact the people they are connected to.

"When people reach out [via LinkedIn Premium], I try to give them something of value. They take their time, and they pay to contact me," says Wiseman. And he notes that those who directly reach companies or hiring managers typically see more success in the job market.

Relying on algorithms

Refer HR, a recruitment firm that has served 42 corporate clients since opening in Vancouver in 2019, is also scouring LinkedIn for potential hiring candidates, general manager Kobe Tang said. Recommendations made by LinkedIn's algorithm play an important role in search and eventual hiring, he said.

The website is also an important space for Canadian tech workers after notable layoffs by Shopify, Wealthsimple, Hootsuite and Unbounce in 2022, said Refer HR marketing manager Rob Gido.

"Adding so-called weaker [connections] definitely increases your chances of finding new opportunities and jobs," says Gido.

The Office of the Privacy Commissioner (OPC) said in an email that it had not received a complaint regarding the study, but if it did, it could prompt an investigation.

But Cofone and Penney say Canada's leniency in privacy laws around consent is one sign of how the law is less stringent than its counterparts around the world. The EU's general data privacy policy has been updated twice since a Canadian law came into effect 22 years ago, while the country's personal protection laws saw no major changes in that time.

Penney said he would like to see legislative changes that give the federal privacy commissioner more powers for investigation and enforcement — and that limits how the company's privacy policies can be used when it comes to personal data, Penney said.

The act should be updated to reflect underlying user rights – and instead place the onus on the companies that step on it, Cofone said. Whether jobs will be harmed by companies' use of user profiles, for example, "we shouldn't absolve them of responsibility just because they have the illusion of consent," he said.

"If Canadians are not happy to be the guinea pigs in platform studies like this, they should vote with their feet for the party proposing stronger data protection and privacy laws," Penney said.

"Politicians should pay attention to this issue  this kind of platform practice can take root in social and economic inequalities."

Post a Comment

Previous Post Next Post