It’s a tedious exercise reviewing the ways in which the pretense of ostensibly neutral journalists to judge truth via PolitiFact investigations is tilted severely toward their political leanings. Indeed, if it weren’t for an explicit meter and the assertion that they are dealing only in facts, it wouldn’t be worth the effort.
But in order that I may exorcise today’s demon, I have to point out the ignorance and bias on display in Eugene Emery’s finding that it is “mostly true” that “only 14 percent of Catholics agree with the Vatican’s position that abortion should be illegal.”
An objective assessment must acknowledge that there are two parts to the question, with a third qualifier necessary for an understanding of the results:

1. What is the Church’s position on the matter?
2. Do Catholics agree with that position?
3. In what sense are the respondents “Catholic”?

In order to answer the first question, Emery did not call up our very accessible bishop, Thomas Tobin, for an explanation. Rather, he thumbed through the Catechism of the Roman Catholic Church and quoted it as follows:

The “Respect for Human Life” section of the Catechism of the Catholic Church says, “Since the first century the Church has affirmed the moral evil of every procured abortion. This teaching has not changed and remains unchangeable. Direct abortion, that is to say, abortion willed either as an end or a means, is gravely contrary to the moral law.”

The theological/philosophical language “willed either as an end or a means” ought to have provided Emery a clue that he should tread carefully. As with theological notions such as double effect and just war, the difficult matter is where the line is between willing an action and accepting it as a consequence of an action undertaken for a morally positive end.
Removing an unborn child from his or her mother’s womb prematurely may be necessary to prevent both of their deaths, and it may thereafter be impossible to preserve the child’s life. As a moral matter, such an action is substantially different from “willing” an abortion “as an end or a means.” That is especially true “as an end,” and inasmuch as the death of the child, of itself, will never save the life of the mother, “the means” are a step removed, as well.
When it comes to determining the law of the land — what should be illegal — we must acknowledge that the judgment of professionals and those who find themselves in horrendous positions is unavoidable. So, this moral teaching of the Church would translate as “except to save the life of the mother.”
As with all translations of religious teaching into law, semantics are a problem. The procedure of “abortion” implies the intent of killing the child. That is, in the public debate and in the law, there is no separate term for “fatal early delivery,” or some other construct that might be more accurate to describe a circumstance in which a doctor does something to save the mother’s life that unavoidably ends the child’s life. In public discourse, we simply call all such exigencies “abortions.”
Therefore, given the only four options available in the surveys (legal always, legal most of the time, illegal most of the time, illegal always), it isn’t really the case that the Catholic Church requires the last. A far more useful survey would draw a line in the third category to distinguish between “life of the mother” and other exceptions.
Of the three surveys that Emery reviewed, only the one conducted and promoted by the pro-choice activist group gives any such indication. They found that 51% of the 923 Catholics supposedly representing the 70 million or so U.S. Catholics believe that abortion should be legal “in just a few” cases or “never.” The percentage climbs to 75% among those who “attend mass frequently” (44% and 31%).*
The survey goes a bit farther and gives some evidence of what “just a few” cases might be, although it does so in terms of what health insurance (whether public or private) should cover. In that case, the 14% who are against all abortions increases minimally to 16% who don’t think any should be covered by insurance. The next marker, though, is 24% who don’t believe it should be covered in any case except those that “threaten the life of the woman.” Of the options, that is most closely the official Catholic view.
Adding in the qualification of Church attendance, however, the percentage moves up considerably. And the qualification is certainly relevant, I’d argue, because it doesn’t tell us much to know that people who don’t assent to the bishops instruction to go to Church regularly also don’t assent to much more emotionally difficult teachings.
In summary, according to a poll commissioned by a group that actively advocates for access to abortion while calling itself “Catholic,” 30% of Mass-goers don’t think insurance should cover abortion in any case, and 42% don’t think it should cover nothing or only life saving procedures. Adjust that a bit, if you like, to account for those who think the procedure should be legal but not covered even by private insurance, but it’d likely be a small tweak.
From a devout Catholic’s point of view, assent to the Church’s teaching is still too limited, but it’s significantly different from 14%, and one needn’t accept even the highest arguable number to think “mostly true” is more than a little bit of a stretch.
* For some perspective on a 923 sample size for a 70 million person population, consider that the Bureau of Labor Statistics surveys 1,200 people in Rhode Island alone to determine unemployment. That means that bias in the methodology could have a huge effect.
For example, the survey document does not define what the survey sample bases ” proper proportions” on, the total population or the Catholic population. If the former, it could skew toward wealthy Northeastern whites who attend Church only occasionally. That may be a representative sample of people who happen to be Catholic in the general population, but not of people who are Catholic.