In most states, employment interview questions related to marital status, family planning, age, ethnicity, religion, or sexual preferences are either barred by law or may be used as evidence to support discrimination claims.1–5 The term potentially illegal refers to such questions, which are intended to reveal, or have the effect of revealing, a job applicant’s membership in a class protected by federal and state civil rights laws. Residents are recognized by the courts as “employees” who are protected by these civil rights statutes that bar discrimination on the basis of race, sex, national origin, religion, age, and disability.1–5 Some data suggest, however, that significant violations to such laws occur in graduate medical education. In 1990, more than 35% of respondents to the Association of American Medical Colleges (AAMC) Graduation Questionnaire reported that they were asked questions about family planning or marital status during residency interviews.6 A 1990 study from the AAMC revealed that 38% of female applicants were asked questions about their pregnancy intentions during residency interviews. In 1996, 45% of all respondents to the AAMC Graduation Questionnaire reported being asked such questions.7 Over half of respondents to the 1993 American Medical Student Association survey reported being asked questions that they felt were “inappropriate, uncomfortable, or possibly discriminatory” during residency interviews.8 Similarly, 90% of respondents to a 2010 study of a single medical school graduating class indicated that they had been asked at least one potentially illegal or discriminatory question.9
To our knowledge, no one has conducted an analysis by specialty of the prevalence of potentially illegal questions during residency interviews. In 2000, 230 respondents to a study of urology residency applicants noted that they had been asked potentially illegal questions; 100% of those female respondents reported being asked about marital status, and 60% reported being asked about family planning.10 In 2001, 86% of respondents to a survey of emergency medicine residency applicants also identified a number of similar potential violations.11 In 2005 and 2006, emergency medicine residency applicants indicated that they were asked potentially illegal questions approximately 30% of the time.12 The findings of these limited studies suggest that significant violations to equal employment regulations are occurring. In addition to negatively biasing applicants against a program or specialty, these questions are potentially illegal under both federal and state regulations. In this study, our intent was to quantify the number of potentially illegal residency interview questions being asked to applicants in five specialties.
All 2006–2007 National Residency Matching Program applicants in five specialties (internal medicine, general surgery, orthopedic surgery, obstetrics–gynecology [OB/GYN], and emergency medicine) were eligible to participate in our study. We developed our survey instrument based on a review of the literature and a study group consensus. In addition, a survey methodologist edited the instrument. In February 2007, the Electronic Residency Application Service (ERAS) staff sent invitations via e-mail to all eligible applicants. Included in the e-mail was a link to our survey, hosted on a public Web site (www.surveymonkey.com), and the confirmation that respondents’ information would remain confidential and that only the research team would have access to individual responses. The ERAS staff sent a second e-mail invitation one week later. They sent these e-mail invitations so that respondents’ identities would not be revealed to the research team, and they generated a custom ID for each respondent so that all data would remain anonymous and securely encrypted.
Respondents were told that the purpose of the study “concern[ed] the interview experiences of medical students applying to residency programs in the United States” and that their answers would “help improve the overall interview process for future classes.” The survey first asked respondents to recall if, during any of their residency interviews, they were asked questions relating to the following categories: gender, age, marital status, couples matching, current children, intent to have children, ethnicity, religion, or sexual orientation. For example, the survey asked, “Were you asked any questions about your gender in relation to your specialty choice or training?” The survey also asked what effect such questions had on respondents’ decisions to rank programs as they did. The survey clearly stated that responses should be based on questions initiated by residency interviewers regarding information that the applicant had not previously disclosed to the interviewer either in his or her file or in the interview prior to the question being asked. Respondents then were asked to quantify how many programs asked such questions and to provide free-text examples. Finally, the survey asked for demographic information, such as specialty, age, gender, the number of programs to which the applicant applied, the number that offered interviews, and the number of programs with which the applicant interviewed. See Supplemental Digital Appendix 1, http://links.lww.com/ACADMED/A134, for a copy of the complete survey instrument.
We imported responses into a Microsoft Excel spreadsheet, then into STATA (Version 10.1, STATA Corporation, College Station, Texas), wherein we calculated descriptive statistics, including comparisons of percentages between subpopulations.
The Alameda County Medical Center institutional review board approved our survey instrument and protocol prior to distribution.
Prevalence of potentially illegal questions
The ERAS staff sent invitations to 11,983 applicants. Of those, 7,028 completed surveys for a response rate of 58.6%. Of respondents, 4,557 (64.8%) reported that they were asked at least one potentially illegal question. Of the potentially illegal questions asked, questions related to marital status and whether the applicants currently had children were most common. Of respondents, 3,816 (54.3%) reported that they were asked about their marital status, and 1,923 (27.4%) reported that they were asked whether they currently had children. The next most common question involved applicants’ plans for childrearing or childbearing. Of respondents, 1,073 (15.3%) reported that they were asked such a question. See Table 1 for complete data on the potentially illegal questions respondents reported being asked and List 1 for free-text response examples of those questions and respondents’ comments.
Among those respondents who reported being asked a potentially illegal question, questions about marital status had the highest mean, with an average of 5.0 (standard deviation [SD] 3.6) programs per applicant asking such a question. Questions about having children (mean 4.0, SD 3.1), ethnicity (mean 3.6, SD 3.3), gender (mean 3.5, SD 3.0), age (mean 3.3, SD 2.9), family planning (mean 3.2, SD 2.6), religion (mean 2.6, SD 2.5), and sexual orientation (mean 2.5, SD 2.1) followed.
Prevalence by gender.
Regardless of specialty, women were more likely than men to receive questions about their gender, marital status, and family planning (see Table 2). More women than men received questions about their gender (579/3,279 [17.7%] versus 130/3,448 [3.8%], P < .001), marital status (1,887/3,251 [58.0%] versus 1,794/3,413 [52.6%], P < .001), and family planning (707/3,227 [21.9%] versus 337/3,408 [9.9%], P < .001). We found no difference between the sexes in the number of questions asked about age, children, ethnicity, or sexual orientation (see Table 2 for complete data). Fewer women than men, however, received questions about their religion (245/3,167 [7.7%] versus 392/3,366 [11.6%], P<.001).
Free-Text Response Examples of the Potentially Illegal Questions Asked During Residency Interviews and Survey Respondents’ Comments, 2006–2007
* I had one interviewer ask me point bank totally out of context was I married, did I have children.
* I was asked about my marital status at almost every interview. On the whole, the residency interview process was the most unprofessional interview process I have ever experienced.
* I was asked about my marital status but not how it related to my specialty choice.
* I found it unusual that every program asked me whether or not I was married. The question was often followed by a remark like: “Oh, I’m not supposed to ask that—it’s illegal.” But, it didn’t stop them. Besides, they know I would be a fool to follow up with any legal recourse. I need them desperately. And, they only kind-of need me. So, I answered. Wouldn’t you?
* EVERYONE asked about being married! It’s not an option in the Midwest.
* I was asked about having children during residency and lied and said we had no plans of having children.
* I had a number of questions as to whether I would be having children in the near future/during residency. I am hesitant to say that the field of orthopedic surgery has entered the 21st century with regards to attitudes/concerns regarding employing women.
* There was one program where I volunteered that I was currently pregnant and was told by the residency director after my statement: “Well being a mother will be your most important role—you’ve made your choice.” And I was dropped from their rank list.
* Someone asked me if I was pregnant.
* I was asked repeatedly about my religion, mainly due to me being from a state with a particularly predominant religion. This bothered me to no end and was incredibly unprofessional in all instances. I believe this is one of many incredibly poor reflections on our profession.
* I went to a school in Israel. They asked me if I was Jewish.
* Lots of very offensive “How do you expect an old fart like you to handle a residency?”
* I had two interviews in which the question of whether my age would be a problem during my residency was asked.
* Before I even had a chance to shake one of my interviewer’s hands, he told me that I was too young to be starting residency and that I should probably take a year off.
* “Do you have a girlfriend?”… Answer: “No.”… “Do you like girls?”; “You do like girls.”… “What do you think of her?” pointing to the secretarial [assistant].
* I have found it to be very difficult to be a gay medical student at my medical school and interviewing for residency was no easier. In fact, a couple programs made me feel that if I were to let them know my sexual orientation I would not be considered.
Prevalence by specialty.
Among those respondents who indicated their specialty, those in OB/GYN and general surgery reported the highest prevalence of potentially illegal questions about gender (OB/GYN: 162/756 [21.4%]; general surgery: 214/876 [24.4%]). Respondents who indicated a surgical specialty also were most likely to receive questions about their marital status (general surgery: 593/868 [68.3%]). Respondents who indicated an OB/GYN specialty also were most likely to receive questions about family planning (200/750 [26.7%]). Questions about age, children, ethnicity, religion, and sexual orientation were less common, but we still found significant differences between specialties (see Table 1 for complete data).
Effects of potentially illegal questions on overall rank list
Female respondents were far more likely than male respondents to report that receiving questions about gender, marital status, or family planning caused them to rank a program lower on their list. This effect was most pronounced in general surgery and orthopedics (see Table 3 for complete data).
Among orthopedics respondents who were asked a question about gender, 45.1% of women (23/51) reported that it made them unlikely or very unlikely to rank that program, whereas only 14.3% of men (1/7) did. This difference also was pronounced among emergency medicine respondents: 39.1% of women (18/46) and 0% of men (0/2) stated that they would not rank a program because of such a question.
Overall, female respondents who received a question about marital status were unlikely or very unlikely to rank a program more often than male respondents. This effect was most pronounced in orthopedics (women: 13/54 [24.1%] versus men: 17/259 [6.6%]) and general surgery (women: 57/252 [22.6%] versus men: 27/338 [8.0%]).
Female respondents in all five specialties were more likely to not rank a program if they received family planning questions. In orthopedics, 60.7% of female respondents (17/28) who received such a question reported being less likely to rank that program compared with 8% of men (4/50). In general surgery, 45.8% of female applicants (66/144) who received such a question reported being less likely to rank that program compared with 9.9% of men (7/71).
The results of our survey suggest that residency applicants routinely are asked potentially illegal questions—almost two-thirds of respondents reported being asked at least one such question. These findings are supported by published studies6–12 that underscore that medical students continue to face this problem during residency interviews.
The courts consider residency interviews to be “employment practices.”2–5 Although interviewers then may pose informal questions trying to “get to know” the applicant, they are prohibited from asking questions reflecting concerns about residency coverage issues or the applicant’s family support.1 This is a common misperception among interviewers. Frequently, interviewers have the best intentions and are not trying to figure out “secrets” or “privileged” information but are merely asking informal questions, trying to learn more about an applicant. Unless the questions are asked to all applicants, though, they can be perceived as discriminatory and, in some states, are illegal.
No federal statute makes it illegal, per se, to ask about a protected classification. However, if asked, the question can be used as evidence of discrimination if the applicant is not hired. Individual states, however, have different statutes on employment discrimination.13–20 For example, California’s Fair Employment and Housing Act makes it illegal for an interviewer to ask any question that suggests a “limitation, specification, or discrimination as to race, religious creed, color, national origin, ancestry, disability, marital status, sex, age, or sexual orientation.”5 Asking such a question, then, is a violation of California law and potentially a source of damages, even if the applicant is ultimately hired.
Regardless of the legality of such questions, we conducted our survey to study the prevalence of potentially illegal questions in residency interviews and to identify the impact of such questions on applicants’ decisions. We found that merely by being asked a question about family planning, for instance, about half of female respondents reported that they would rank that program lower. Knowing that asking potentially illegal questions to applicants could change how successful a program is at recruiting women should prompt program directors to stop these practices from occurring. Doing so would improve the program’s ability to recruit and retain highly qualified applicants.
To achieve this goal, we recommend that program directors educate residency interviewers on the federal and state laws governing equal employment hiring practices. Interviewers’ practice of asking these questions is at best inappropriate and may be illegal if they are asking such questions to one group more often or if applicants’ answers influence how a program ranks them. Program directors may not be aware that some interviewers even are asking these questions, and they may not understand the harm that could come from interviewers doing so. However, a lack of knowledge may not be the only problem, as many applicants reported that interviewers often prefaced potentially illegal questions with the statement “I know I’m not supposed to ask you this but….” This often-repeated comment suggests that interviewers know the parameters of residency interviews but choose to seek out privileged and protected information from applicants regardless. Program directors likely need additional education themselves as well as more oversight of their interviewers’ practices.
Applicants may think that refusing to answer a potentially illegal question would remove them from consideration by the program. Thus, we recommend that medical students learn how to identify and respond to such questions. For instance, when asked about childbearing, applicants could discuss how they deal with multiple demands on their time or how supportive their family and community are of their career goals. As program directors, we recommend these answers over others in which applicants lie or provide a response that they think the interviewer is expecting to hear.
Our study shares limitations with other survey-based research projects. First, our results were dependent on participants’ memory, as they completed their surveys months after they participated in residency interviews.
Second, respondents who were asked potentially illegal questions during their residency interviews may have been more likely to complete our survey, affecting our findings. Through a sensitivity analysis of responses, however, we determined that the percentage of respondents who likely were asked a potentially illegal question should fall between 38.0% (if no nonrespondent was asked such questions) and 79.4% (if all nonrespondents were asked such questions). Assuming a conservative position that all nonrespondents were not asked a potentially illegal question, our findings confirmed that more than a third of the entire applicant pool were asked potentially illegal questions.
Third, survey responses were based on respondents’ interpretation of the interview questions rather than on the actual questions asked. Although there is no way to monitor all residency interviews for potentially illegal questions, asking applicants immediately after the interview season rather than months later could be a more effective alternative that allows applicants to reflect on all of their interviews even if they still may be unable to recall specific questions.
Fourth, we did not use the words “illegal,” “inappropriate,” or “discrimination” in the introduction of the survey to limit bias. See Supplemental Digital Appendix 1, http://links.lww.com/ACADMED/A134, for a copy of the complete survey instrument. Still, the wording of the questions may have prompted respondents to report a higher prevalence of potentially illegal questions if they had been offended by the questions they were asked.
Finally, applicants may have recalled that an interviewer asked a potentially illegal question when instead, the applicant was the one to bring up the controversial issue and the interviewer only continued the line of questioning. Our survey specifically asked respondents to consider topics “raised by residency interviewers about which [they] had NOT previously disclosed information.” Although such recall bias is unavoidable if the interview is not recorded, researchers can reduce it by surveying applicants immediately after the interview rather than at the end of the season.
In this study, we found that a large number of residency applicants were asked potentially illegal questions. Female applicants were more likely than male applicants to report being asked a question about gender, marital status, or family planning and were more likely to not rank a program that asked such questions. Program directors must recognize that asking such questions has a negative impact on residency applicants, which should prompt them to change their interview practices. Leaders in graduate medical education should consider educational outreach targeting both applicants and programs regarding acceptable interview procedures and guidelines. Developing a formal interview code of conduct may be necessary to address the potential flaws in our process for selecting future practitioners and medical educators.
Acknowledgments: The authors wish to thank Ms. Moira Edwards and Ms. Renee Overton of the Electronic Residency Application Service for their invaluable help and Michael Ward, JD, for his employment law opinions.