How to Score Modesty in Japanese Candidates

When a Japanese candidate says "I still have a long way to go," the sentence tells you about manners, and sometimes about a sincere belief. It tells you very little about performance. Most interviewers score it as performance anyway, because they assume people who did well can say so. A 2007 meta-analysis by Steven Heine and Takeshi Hamamura found that Western samples inflate their self-assessments by 0.87 standard deviations on average. East Asian samples inflate by minus 0.01, which is to say they don't.

How big is the gap in self-assessment?

Close to a full standard deviation, and it comes from one side. Westerners rate themselves up. East Asians, on average, don't.

Heine and Hamamura's paper, in Personality and Social Psychology Review, pooled published studies of self-enhancement, such as how much better than "the average person" people think they are. Across 91 direct East-West comparisons, the average difference was 0.84. Within cultures, self-enhancement was 0.87 for Western samples, minus 0.01 for East Asian samples, and 0.52 for Asian Americans, who sat in between.

There's a detail here that is easy to read backwards. The East Asian figure is zero. It means no inflation. It doesn't mean deflation. So "Japanese candidates understate themselves" is true relative to the interviewer's ruler. If the ruler assumes everyone rounds up a little, the person who reports straight reads as one grade lower.

Most of the samples were students, and none of this was measured in interviews. What it shows is narrower: self-assessments from different cultures aren't on the same scale.

Is it politeness, or do they mean it?

Both, and from the interviewer's chair you can't tell which.

One layer is convention. Japanese has a stock of modest phrases. Mada mada desu: I'm not there yet. Tamatama desu: it just happened to work out. Mawari ni megumaremashita: I was lucky with the people around me. In a Japanese setting these carry almost no information. The speaker is following a rule of conversation, much as "How are you?" in English isn't a question about health.

The other layer is real. In 2000, Heine, Toshitake Takata and Darrin Lehman published an experiment in Personality and Social Psychology Bulletin designed to rule out the "they're just being polite" explanation. They skipped questionnaires and covertly recorded what participants did in a lab. Canadians were reluctant to conclude that they had done worse than their average classmate. Japanese were reluctant to conclude that they had done better. The authors' conclusion was that the difference goes beyond self-presentation.

So telling a Japanese candidate "no need to be modest, just be direct" may get you the same version, delivered directly.

A more recent number points the same way. Japan's Children and Families Agency surveyed people aged 13 to 29 in five countries in November and December 2023, online, about a thousand per country. Asked whether they were satisfied with themselves, 57.4% in Japan agreed or somewhat agreed. The figure was 73.2% in the United States, 73.9% in Germany, 75.6% in France and 72.3% in Sweden. Counting only those who plainly agreed, Japan was at 16.9% and the United States at 36.4%.

That is a youth attitude survey. It isn't data on job candidates, and it is mixed up with the Japanese habit of avoiding the ends of a scale, which I covered in In Japan, "neither agree nor disagree" is an answer. But when you ask a candidate "how do you think you did?", you are handing out the same questionnaire in the room.

What happens when modesty is scored literally?

The score moves toward people who praise themselves. There is an experiment on this.

In 2013, Delroy Paulhus and colleagues published a two-part study in the Journal of Applied Social Psychology. First, 72 participants played job applicants in videotaped interviews. Then 222 raters evaluated applicant videos that varied by narcissism (high or low) and by heritage (European or East Asian). The best evaluations went to the European-heritage narcissists. When the researchers broke the behavior down, the narcissism advantage came mainly from frequent self-praise.

The study looked at a Western interview setting, and its East Asian-heritage participants weren't job seekers in Japan, so it shouldn't be carried over whole. I think the mechanism carries over, though. The raters paid for self-praise, and whether a person praises themselves depends more on culture than on ability.

In a Japan subsidiary this can happen through an interpreter. Mada mada desu gets translated, goes into the interview notes, and can become "candidate acknowledges a gap in ability." Nobody was unfair. A sentence of manners got filed as a sentence of evaluation.

One underrated candidate is an accident. The systematic part worries me more. If the mechanism holds, the Japanese candidates in a mixed pool sit lower as a group, and the ones who get through are the ones whose way of talking is closest to what the interviewer is used to. The process ends up selecting for speaking style, with ability coming along as a passenger.

The interviewer does the calibrating

The fix sits on the interviewer's side. Stop asking for self-assessments and ask for facts and scope.

"How would you rate your performance?" "Give yourself a score out of ten." "What are your strengths?" None of these produces evidence from anyone, and with a Japanese candidate they come with an extra discount. Ask which part of the project the candidate decided. Ask what the number was at the start and at the end. Ask who signed off and what that person said. Answers to those questions leave no room for adjectives, so they leave no room for modesty.

Another phrase Japanese candidates use a lot is minna de yarimashita: we all did it together. It points the opposite way from the candidate in "It was all me": finding what a candidate really did, and it comes apart the same way. Ask who did what. Ask which call was theirs.

Using one ruler means giving nothing away at either end. Don't add a culture bonus for modest candidates. Some modest people really didn't deliver. And don't give points for a confident candidate's adjectives. If every line on the scorecard needs an event behind it, and the rule applies to both kinds of candidate equally, the calibration is already done. You don't need a second standard for Japanese candidates.

The misreading runs the other way too. A Japanese interviewer who hears a foreign candidate list achievements tends to write "too self-assertive." It is the same mistake. A habit of speech has been read as character.

It doesn't end with the interview. Plenty of companies have a self-rating box on the performance review, and the self-rating feeds into the final grade. When locally hired Japanese staff and expatriates fill in the same form, the side that doesn't round up loses a little every year. My guess is that many companies have fixed the interview questions and left that box exactly as it was.