Do you know the story about your conception? I do. According to my parents, I was the “best mistake” they ever made. I wouldn’t suggest asking your parents, because you might hear something that you aren’t happy about. My parents’ answer included a contraceptive failure and a washing machine.

My personal story is fairly common – the unintended birth, not the washing machine. In the mid to late 2000s, an estimated 37 percent of births were unintended. Twenty-three percent were mistimed, like me: My parents wanted to have a child at some point, but not then. The rest were unwanted.

However, unintended births are hard to count. It’s very difficult to measure whether a conception was intended. But having the data is important: Accurate measurements of unintended fertility allow researchers to assess population growth, women’s reproductive autonomy and the impacts of unintended births. It can also show how policy changes affect unintended pregnancy. For example, if Roe v. Wade were to be overturned, researchers would want to understand the impact on fertility.

Researchers who want to know what couples are thinking at conception have two major strategies: asking women years before or sometime after pregnancy. But both strategies have substantial flaws. This highlights how difficult it is to measure theoretical concepts – especially such emotional and complex ones as the decision to have a baby.

Differing measurements

Figuring out what people are thinking right before they got pregnant is difficult without surveyors constantly knocking on bedroom doors.

Because it’s invasive, costly and just plain weird and impossible to pursue women at the moment they become pregnant, many demographers ask women after they’ve become pregnant or had a child. For example, one government survey asks, “Thinking back to just before you got pregnant with your new baby, how did you feel about becoming pregnant?” Intended births include mothers who reported they wanted to become pregnant then or sooner. Unintended births include both mistimed (i.e., “I wanted to be pregnant later”) and unwanted births (i.e., “I didn’t want to be pregnant then or at any time in the future”).

But there are some problems with retrospective questions. Researchers suspect that women may rationalize their past behavior by providing a socially desirable response or one that reflects their current situation or worldview. After women become emotionally attached to their baby, they may not want to say their baby was mistimed or unwanted. It’s easy to imagine a woman who is happy in her role as a mother – or believes she should be – saying that she wanted a child at conception even if she actually didn’t. The opposite could also occur. My colleague and I found evidence for both types of rationalization, though the former was more common.

Some demographers suggest asking before conception. This means asking people about their intent to have children in the future, then looking at what happens later. If someone said she didn’t want another baby but had one later, then that birth would be considered unwanted.

In this prospective strategy, women have fewer reasons to misreport, so this might provide a better estimate of intent at conception. Indeed, this method suggests that the percent of unwanted births is much higher: 26 percent, as opposed to the 9 percent estimated through retrospective measures.

But the prospective strategy is also flawed. Women could have changed their minds between their answer and conception. In the data used in my research, women were asked the prospective question every two years. A lot can change in two years. While retrospective measures underestimate unwanted births, prospective measures overestimate unwanted births.
An issue of both strategies is that women may not have certain unambiguous intentions or preferences at conception. People could have both strong positive and strong negative feelings about having a child or feel uncertain at conception. Intentions might not be discrete categories, instead existing on a continuum. And there might be different aspects of intentions, such as how much planning someone does for a baby and how happy someone is about a pregnancy.

Perhaps the most vexing assumptions of both strategies is that women have conscious intentions or preferences at the time of conception. Certainly, for some, childbearing is a conscious decision. But others may have motivations that are not connected to a conscious intent for childbearing. For example, not using condoms can signify trust between partners. Some couples who have unconscious desires for a child may not use birth control consistently or effectively, but others may be guided by different motivations. People start to think consciously about childbearing over time, in tandem with developments in other domains such as relationships and work, but some people get pregnant before they’ve made fertility plans.

A better way?

Should demographers keep on trying to measure unintended fertility? Some may say the best way to push forward is to abandon this concept because it’s constraining researchers’ thinking about fertility decision-making.

However, retrospective questions have been helpful in a lot of ways. They highlight socioeconomic and racial or ethnic disparities, as well as a link between unintended pregnancy and poor health. Even if retrospective measures don’t quite capture what demographers want to know, they can help pinpoint women who might need some extra help coping after birth.

There may be yet another, better way to measure intended fertility. It would require a careful balance. If the measure isn’t complex enough, we might miss how people really make decisions about fertility. If it’s too complex, it can become incomprehensible.

More broadly, researchers need to decide how much error they’re willing to tolerate. Many theoretical concepts are measured in ways that might not completely reflect the underlying concept. Social scientists need to interrogate important concepts like unintended fertility, but also need to use clear measures that adequately capture reality – even if they aren’t perfect.

Heather M. Rackin does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.