According to widely reported survey data, women’s substantial gains in education, labor force participation, and earnings haven’t made them better off during retirement.

But that outcome makes little sense, and it is in fact based on incomplete data.

According to data from the Current Population Survey’s Annual Social and Economic Supplement, older women are no more likely to receive retirement income from a pension, 401(k), or IRA than they were a quarter-century ago.

In addition, the survey reports that those who do have retirement incomes have experienced only 21 percent growth in those incomes, and that women’s incomes decline after retirement.

This survey data yielded these results because it fails to adequately capture the substantial gains that women have made in retirement income from sources such as 401(k)s and IRAs. Part of that failure is because of the way the survey worded certain questions, leading respondents to underreport retirement incomes.

More accurate administrative data, such as data based on tax returns from the IRS and earnings records from the Social Security Administration, paint a different picture.

According to an economic study that linked survey data to administrative records, administrative records revealed substantial gains: Women are twice as likely to have retirement income as they were 25 years ago, their retirement incomes have increased by 58 percent, and their incomes remain remarkably steady before and after retirement.

These differences are significant. Administrative data show that median household income for women is 45 percent higher at $60,939, compared to $41,977 reported in the survey data.

Moreover, the administrative data show more equal income growth across women of varying earnings levels.

This new link between survey and administrative data helps solve the puzzle of why women’s significant economic gains haven’t translated into retirement income gains. It only makes sense that more work, education, and earnings by women would raise their retirement incomes.

After all, women’s labor force participation increased from 33 percent in 1950 to 57 percent today; their college graduation rate increased from 12 percent for women born in the 1920s to 27 percent for those born in the 1940s, and to 45 percent for women born in the 1980s; and those working full time experienced a 23 percent increase in real earnings over the past three decades (while men’s real incomes rose only 1 percent).

All these gains provide more opportunities to save and more incomes with which to save. That is why older women are doing better today than in the past.

Data helps drive informed decisions, but bad data drives misinformed and potentially damaging policies.

When it comes to women’s retirement security and economic gains, the widely reported survey data are wrong. In reality, women’s incomes do not decline once they retire, and women have achieved significant growth in retirement incomes.

This revelation should lead policymakers to focus on policies that help enable personal savings and financial security for all Americans, as opposed to crafting policies aimed exclusively at and for women.