Sexual Harassment: What’s the Best Way to Handle a Sexual Harassment Investigation?

In our various operations, we’ve been dealing with several sexual harassment complaints. We have three questions:

1. When we get into the he said/she said situation and there are no other witnesses and no other evidence, what do we do?
2. How can we interview witnesses without laying out the whole story?
3. How do we decide how many witnesses to interview and whom to interview?
   — Harassed HR Manager in Modesto

400+ pages of state-specific, easy-read reference materials at your fingertips—fully updated! Check out the Guide to Employment Law for California Employers and get up to speed on everything you need to know.

For this challenging three-part query, we sought the advice of Rebecca A. Speer of Speer Associates/Workplace Counsel.

1. He said/she said. First off, the key to “he said/she said” situations is to do everything reasonably in your power to uncover ‘corroborating evidence’— information that would support the complainant’s or the accused’s version of events. In my experience, even the most stark-seeming he said/she said situations can offer ample opportunity for corroboration. And it’s one of your main duties as a diligent investigator to seek it out.

    

   Corroboration takes many forms and is not limited to the accounts of direct witness-observers to an alleged incident. For instance, if a female employee contends that, over a period of time, her manager engaged in offensive behavior toward her in private, outside the presence or earshot of anyone, I would want to know the following:

   • Did she complain to anyone about this behavior around the time it supposedly occurred? Did she engage in any conversations with anyone (e.g., close co-workers or a supervisor) that could offer insight into whether certain events occurred and, if so, precisely what occurred?
   • Similarly, did the accused engage in any conversations or make statements during the relevant time period(s) that offer insight into whether he engaged in the alleged offensive behavior?
   • Did anyone notice a change in rapport between the complainant and the accused around the relevant time period(s), signaling possible tensions?
   • Do any e-mail communications or other documents between the complainant and accused, or between either of them and others, provide information and insight into alleged events?
   • Do any circumstances occurring during the relevant time period(s) (such as an extended absence or precipitous drop in performance by the complainant) suggest that the alleged events occurred?
   • Does any other information provide some support for the complainant’s allegations or the accused’s version of events? For example, have there been similar complaints about the accused in the past or, alternatively, does the accused have a history of exemplary behavior?
   • Does any information exist indicating a motivation the complainant might have to lie or exaggerate about the matters in question? The accused?

   Ultimately, it’s important to do everything in your power not to stop an investigation in its tracks when you encounter the ‘he said/she said’ dilemma. When you cannot identify any direct witnesses to the alleged incidents, don’t give up. Instead, focus on the ‘ripple effect’—the events or circumstances that you would expect to see if the alleged incidents occurred (or, alternatively, did not occur). Doing so will bring you closer to determining whether sufficient evidence exists to support the complainant’s or the accused’s version of events.

2. Interviewing witnesses effectively. “As for your second question, interviewing witnesses is both an art and a science. As the question suggests, one dilemma an investigator often encounters is how to interrogate a witness without revealing too much—either about the details of a complaint or about information gathered from other witnesses during the investigation. Ultimately, the interview needs to be effective; therefore, you must probe witnesses for information central to the issues at hand. But you must also be discreet: you must guard to the extent possible the details of a complaint and information gathered from other witnesses.

    

   One general strategy that helps to accomplish this goal is to adopt the tried-and-true ‘funnel’ technique to questioning a witness—begin with general questions and progress to the more specific. For instance, let’s assume you are attempting to discover whether the accused inappropriately placed his hands on the complainant’s buttocks during an office holiday party and then pulled her closely toward him. Avoid asking the witness at the outset of the interview, ‘Did you see Bob grab Nancy at the holiday party last year?’

   Instead, probe around the issue, moving from general to specific questions. For instance, ask whether the witness has ever observed behavior by the accused—at any time, toward anyone—that he or she considered inappropriate or unprofessional. Next, move to whether the witness has observed inappropriate or unprofessional behavior at an office social event. Then, at the holiday party in question. On a different line of questioning, ask whether the witness has observed the accused engage in inappropriate behavior toward anyone. Then, toward female subordinates. Then, the complainant. Very likely, you will get the answers you need or cover necessary factual ground well before you get to more specific questions that tend to reveal the details of a complaint.

   Of course, questioning techniques—and the level of detail you must achieve—will differ in the interview with the accused. But the above technique works well with other witnesses.

3. Deciding whom to interview. “And, regarding your third question, deciding whom to interview and how many people to interview involves a careful balancing of two competing objectives: the need to be thorough and the need to protect confidentiality. How you strike that balance will depend on numerous factors, such as the nature, extent, and complexity of the factual questions at issue; the number of witnesses who are believed to have information relevant to those issues; and the fruitfulness of interviews you have conducted as you proceed through the investigation.

    

   You need to continue your interviews until you have satisfied yourself that you have adequately explored the factual questions at issue. Do important gaps exist in information you have gathered? Have you reconciled competing witness accounts? Have you adequately sought corroboration for the complainant’s and accused’s versions of events?

   Ironically, investigations that don’t quickly uncover information to support a complainant’s allegations tend to be the longest and broadest. Before you reach the conclusion that insufficient evidence exists to support a complaint—that the complaint is ‘without merit’—you must be able to show that you looked under enough rocks. In other words, your investigation must have been diligent enough to have uncovered possible supportive evidence.

   Investigations typically will be faulted for being too narrow—for not involving enough interviews—rather than for being too broad. So, if a doubt exists in your mind regarding whether you should speak with additional witnesses, opt for a broader investigation, within reasonable limits. This is especially necessary if you are attempting to close factual gaps or satisfy yourself that you have looked extensively enough for information.

   In choosing whom to interview, consider the following:

   • Don’t come up with a definitive witness list at the outset of the investigation. Instead, decide whom exactly you will interview as the investigation progresses. Oftentimes, interviews with certain prospective witnesses become unnecessary as you uncover information at earlier points in the investigation.
   • In deciding on the most appropriate witnesses, ask: who has been identified as having knowledge regarding a particular issue? Or, who is best positioned to possess that knowledge? Be sure that you can articulate a clear rationale for selecting someone as a witness.
   • You will need to ask the complainant and the accused, and other witnesses as well, whom they suggest that you interview. Be sure to ask the reasons they have suggested someone as a witness. Take that information into consideration as you decide whether an interview with a particular person is necessary.

   Ultimately, no hard-and-fast rules exist for deciding the scope of an investigation. In the end, it all comes down to the exercise of good judgment, thoughtfulness, and caution.

   Rebecca A. Speer is founder and principal of Speer Associates/Workplace Counsel, an employment law and employee relations consulting firm in San Francisco.