A nationwide survey conducted in Japan showed that while men are comfortable with gender equality on the job, they tend to be less so when it comes to interpersonal relationships. In an article originally published in the Japan Times, Kiho Muroga and Charles Crabtree show that notions about women’s roles seem to be more conservative outside the workplace.

*     *     *

In the last few years, the Japanese government has been trying to increase the share of women in the workforce, in general, and in leadership positions, more specifically. This focus underscores the vital importance of women to the future of both the labor force and economy. Concurrent to this push for greater workforce equality, women organizations across Japan have also pushed for reforming gender norms and expectations both at home and in the workplace.

Despite the enormous attention that has been paid to these issues, there is little data about how Japanese residents think about them. In February, we fielded a survey with a national sample of 2,389 Japanese residents to address this problem. In one part of the survey, we asked men and women to tell us how much they supported female equality in the workplace, the #MeToo movement, and feminism. Specifically, we asked them to rate their support on a 0–100 point scale, where higher values indicate more support.

The data provide a set of rich findings about Japanese views on gender issues. To begin with, there are no obvious differences in the extent to which men and women support female equality in the workplace. On average, men rated their level of support at 76.69/100. Perhaps surprisingly, though, this is not substantively or statistically different from women’s level of support (i.e., 75.89/100). If anything, men are more supportive of female equality in the workplace. At least in this one area of gender relations, Japanese men and women initially seem to see basically eye-to-eye.

When you elicit information about public attitudes in other areas, though, women and men in Japan seem to have very different views. For example, when we asked about the extent to which individuals support the #MeToo movement, we see that the patterns of replies differ across gender groups. Japanese men on average indicated 63.47/100. Women, though, rated their support at 67.43/100, or about 4 points higher. This difference is statistically significant.

It’s also substantively meaningful, hinting that male support for gender equality in Japan might not extend past certain aspects of corporate life. While men might support women colleagues being treated equally in terms of hiring and promotions, for instance, they seem opposed to the idea that this equality would extend to interpersonal relationships within organizations.

We also found a similar disconnect between men and women’s views on feminism; women express more support (70.65/100) than men (66.46/100). The difference is important, indicating that women support feminism about 1/5 of a standard deviation more than men. This provides additional evidence that Japanese men might accept or even support equality in the workplace but are less enthusiastic about that equality in other spheres of life.

There are two ways of looking at the findings. On the one hand, the gap between men and women might not be as large as one would think looking at the various national-level measures of gender equality in Japan. On the other hand, men continue to lag behind women in their support for movements aimed at increasing equality.

One implication of these findings is that public beliefs about women’s roles outside the workplace remain more conservative than norms about their positions in the workplace. The government has spent considerable resources trying to increase gender equality on the job. Our work suggests that similar programs aimed at improving equality outside the office are also necessary. For instance, developing school-based efforts to broaden children’s views about men and women—and their roles at home, in society, and in the labor force—might be a good place to start. Much more work needs to be done to assess how the views we document are created and how they can be changed.