Are Fake Resumes Ethical for Academic Research?

(Photo: Beat Bieler via

“Audit studies” have been popular in labor economics research for 10 years.  The researcher sends resumés of artificial job applicants in response to job openings. Typically there is a crucial difference in some characteristic of the person that indicates a particular racial/gender/ethnic or other group to which one person within a pair of resumés belongs while the other does not.  The differential response of employers to the difference in the characteristic implied by the resumés is taken as a measure of discrimination in hiring.

Is this ethical?  Employers are deliberately lied to; and unbeknownst to them they spend time in an activity that cannot lead to a hire.  If they look at 2000 resumés, spending only 30 seconds on each, and spend 2 minutes each calling back 10 percent of them (typical in such studies), employers are being asked to spend about 23 hours of their time in this activity.  Valuing their time at $50/hour, this part of the time cost of the research,  essentially a tax borne by participants who must be kept ignorant for the research to succeed, is over $1000. 

I find this line of research to be unethical—it certainly violates the “informed consent” that is always required in biomedical studies.  Nor is it even necessary:  For example, we learned a huge amount about the impact of racial characteristics, in their case, names, from the excellent study by our own Steve Levitt and his coauthor Roland Fryer.


Mr. Hammermesh,

Didn't Levitt and Fryer do precisely what you called unethical to preform their research?

While I agree it should be limited, I also feel occasional studies of the sort are important.


Neil (SM)

No, they used publicly available data.


I disagree. Informed consent applies to individual persons, not corporations. But it doesn't stop here. Job candidates as well as consumers are also constantly subjected to uninformed and non-consensual prodding, prying, cross-checks and intrusive analyses by employers and companies (internet cookies, bots, RFIDs,..). And what do we make of bogus job ads that serve the only purpose of promoting companies to the public for the fraction of the cost of a full-blown marketing campaign? Companies aren't especially known to show much consideration either for the hundreds of candidates that will waste 10 hours or so on fine-tuning their response. Of course this should NOT be part of the equation, right? After all the company doesn't pay for it. They transfer the costs, so we can simply close our eyes on it? And the same could be said of real job ads. So many applicants, only one winner. How much time is wasted on this? Doesn't it ever count at all?
A more interesting question would be, how much asymmetry in the balance of power between employers and employees is enough asymmetry in favour of employers? The "divide-and-conquer", artificially maintained asymmetry that exists as of today is certainly enough to allow employers to use Human persons (with needs for food, shelter, health care etc..) pretty much as they please, at a much better bargain than even slavery used to, meaning they don't even have to take into consideration the basic needs of the people they employ. This may be legal, but it doesn't quite look ethical to me. So in the grand scheme of things, your concern about companies' "informed consent" on research that primarily concerns anti-social behaviour in a profoundly corrupt system is, in the best of cases, heavily misplaced.


mango punch

I think the "Well Loved" and "Hot Debate" highlights should be based on proporiton of likes to dislikes, not on difference... for instance this is 31 likes to 24 dislikes = 1.29 and "Well Liked" whereas a few posts above is 8 likes to 5 dislikes = 1.6 and "Hot Debate"d.


It's absolutely unethical. As a hiring manager of technical specialists (at a small division of a large company), I spend considerably longer than 30 seconds reviewing resumes that I receive for positions that I NEED to fill. Any bogus resume not only wastes my time and my company's resources, but it potentially harms legitimate candidates. I often have only a short time to make the hire, with a very limited number of available interview slots, and if an acceptable candidate is bumped from the first cut by a ghost, the day or two that it takes to give up on reaching the fraud may mean that it's too late to bring in the real person for an interview.


This is a cost associated to doing business. The public needs this data to understand the level of discrimination taking place in the workplace. Consider it from the applicants perspective and how much time belonging to the qualified applicant the discriminatory employer is wasting. Furthermore what that employer is doing is illegal however there is no law that states you cannot submit a fictitious resume for the purpose of doing research.

This is not a biomedical study which seeks to find test subjects to evaluate human behavior. Such a study serves the primary purpose of facilitating the work of its researchers. This is on par with a state, local or federal agency visiting an employer to check health and safety standards. Such an effort serves the primary purpose of protecting the greater good of the public.

Discriminatory hiring practices is regulated under law. How can you evaluate whether or not this law is effective without the use of a scientific study? What if all employers decided to opt out of such research? Would not this amount to a loss of key knowledge that would allow the public to know how effective this law is in practice? Its great to have lots of environmental protection laws but if there are never any EPA inspectors to actually go onsite to companies and measure pollution what good are such laws?



You make some good points. But as a hiring manager (not an HR screener) who has to spend time going through resumes I have mixed feelings. All resumes are promotional in that they highlight the best abilities of the person applying. I do my best to get the best people for the job. Typically I only choose three to five for followup interviews or phone conversations depending whether they are local or not. My positions are highly technical and require people skills. A bogus application not only deprives me of a choice but wastes my time. This is an example of an external cost of the research being done.


If there wasn't the suggestion that recruiters acted unethically themselves there wouldn't be a need for such studies. If there were hundreds of studies that showed that recruiters acted fairly then it's a waste of time and money to do any more, but that's not the case.

Publicizing the results of these studies can only help in making the recruiter think twice before ruling out a candidate based on a prejudice, and while it might be annoying for the good recruiters it is a necessary evil, similarly to dope-testing for clean athletes, nobody likes doing it but you accept it so that in the end you end up with a clean game.

Eric M. Jones.

Why worry? The higher up the corporate ladder you hire for, the more faked resumes you get. Apparently HR feels pressure when a GM, CEO or VP is hired. Checking on THEM might cost HR their jobs.

At the last place I worked they hired for GM, a CRACKHEAD with a resume 50% invented. This took me ten minutes on Google. But whom dare I tell?

The next job I get, I'll list my astronaut wings, Medal of Honor, Nobel Prize, and CIA affiliations.


Somebody could maintain a register of companies prepared to accept these fake resumes. Each company would nominate a price. If a researcher sends them the fake resume, then they also have to send the price after the experiment is complete.

The problem would be getting enough companies to sign up.


If the research doesn't have a control to determine if the individuals being discriminated against still wouldn't have been hired without the R/G/E "markers", it's tough to take it too seriously. Seems like that would be tough to do, but who knows.

Slightly off topic re the linked paper: I'm curious if the study looked at migration over time to places like that where the study was done. Specifically, are there areas of the country where the use of distinctively "black" names is more prevalent, and has there been net migration from those areas? I can't recall ever coming across a girl named Shanice when I lived in California, but neither had I ever come across a girl named Taylor. Until I moved to the south. Have white Californians named Preston or Hunter increased at the same rate Shanice and DeShawn have? It had always been my impression that the names cited in the study were, for the most part, the southern black variant of the southern white practice of giving children waspy surnames as their first name - basically different sides of the same coin.



How many other studies involved uninformed participants? Some studies are predicated upon the participants not knowing they are participating in a study.

Joe Dokes

Social Science researchers and society as a whole needs to know what level discrimination exists in hiring. This appears to be the only way to collect this data. The data serves two purposes, it highlights the disparity in hiring and keeps HR departments aware of their own potential bias.

My question: If you disagree with this type of data collection, what other method would you use to collect the same data, that you find ethical? How else would you enforce anti-discrimination laws?

Having qualms about the current method of data collection is fine, but simply not collecting the data is not a viable solution.


Joe Dokes

Joe J

"How else would you enforce anti-discrimination laws?"
As far as I know these are not used to enforce any kind of law. There is also the question of people or companies having free choice, choice is discrimination.

Actually not collecting the data is a viable choice too.


One standard of ethical as far as research is concerned - is whether or not anyone is hurt. I was sent a book to review recently. For all I know, the whole thing might have been a fake. I was not hurt- learned in the process of trying to review it something about myself.


How about if we enlarge the study to include ageism as well? A few months ago I started seeing a requirement to include high school graduation dates on some on-line HR systems. That seems to be a great screening tool to exclude older applicants since you can't provide bogus dates (which would be grounds for not hiring or termination) and you still get to side step the "how old are you" question. The other sleazy thing companies do is post a position on-line which is used as a justification for hiring an internal candidate. So..who's being dishonest i9n this case? Also, other than (arguably) wasting time and resources reviewing bogus resume's, what is the downside for discriminating companies? Probably nothing. As a reminder, ethics is a two way street. Not a lot of ethical companies these days.


I wonder if some resumes aren't being rejected these days because they come up as questionable during a pre-screening process of checking against public information online. In my line of work, professionals with no online presence are suspect.

Mark Troy

I don't get it. Where's the rest of the article?


I think your rhetoric is misleading. If hundreds of firms share in the $1000 cost, then I don't see the giant ethical problem with the experimental design, as the cost to any individual firm is extremely low.

jobu babin

While I think it is a practice with serious ethical questions, it still seems like compensation after the fact would be reasonable. Or are our research budgets stretched too thin to do the right thing?