Ok. I’ve seen way to many articles and blog posts lately where people utterly fail at interpreting data accurately. Apparently, the recipe for interpreting good data poorly isn’t much of a secret. If you want to jumpstart your efforts in doing so, simply 1) assume you already know what the data will tell you, or 2) ask the wrong question(s), or both.
The observed correlation between race and poverty is a good example. Let’s ask two questions.
1) What race of people are at the highest risk of being in poverty? Blacks and Hispanics.
2) What race is a typical person in poverty? White.
Most people tend to ask only one of these questions, usually depending on their general political/ideological point of view, and the answer to one or the other is likely to surprise most people. How can both be true at once? This is possible because there are significantly more White people in the U.S. than people of any other race. So if you pick a random person living in poverty they are more likely to be White than any other race. But Blacks and Latinos are at a higher risk of being in poverty statistically. So if you randomly pick any Black or Latino in the U.S., they are more likely to live in poverty than any randomly selected White or Asian.
So asking only one of the above questions would result in an incomplete picture. However, it’s important to keep in mind why we’re asking these questions in the first place. Are we looking to place blame? Prove superiority/inferiority of a group of people? I would submit that such activities are not worthy of the attention of thinking people. Nor do I think that statistics would be an effective vehicle for analyzing such things anyway. Assuming that anything at all should be done by government or by communities to combat poverty, our goal should be to identify the kinds programs and policies that might be the most effective in doing so. This means understanding risk (i.e. rates) more than flat numbers, because a policy that can be aimed at a smaller, higher risk population is more likely to be implemented properly and to have significant impact than a policy aimed at a larger population, a low risk population, or both. Understanding risk is not about placing blame, but rather about targeting finite attention and resources to those who need it most.
Speaking of poverty and stats, today I’ll award the honor of detailed critique to Erika Eichelberger of Mother Jones, for grouping together so many problematic/poorly supported opinions in one post supposedly denouncing myths about poverty. Some of her assertions may be true, but there are serious issues with at least a few of them.
Let’s look at some of Erika Eichelberg’s “busted myths.” (I’ve kept her original numbering to make cross-referencing easy.)
6. “Go to college, get out of poverty [is a myth]”.
First, claiming this is a myth is ridiculous, and Eichelberger backs it up with the weakest kind of stat- a flat number that gives no clue as to the rate of college educated people in poverty, and provides no ability to compare this to the rate of those without a college degree in poverty.
Second, this contradicts general trends regarding educational attainment and earning levels: https://www.census.gov/prod/2002pubs/p23-210.pdf.
Third, not all degrees are equal. Someone that goes to “I smoke pot all day University” and gets a degree in Communications without doing any work does not have the same market value as someone that goes to a good school, studies hard, and graduates with a degree in a Science, Technology, Engineering, or Mathematical field. The US Census captures data on work-life earnings, and it is quite clear that not all degrees are created equal from the standpoint of earning potential. http://www.census.gov/prod/2012pubs/acsbr11-04.pdf.
1. “Single moms are the problem [is a myth]”.
First, saying single moms “are the problem” is indeed a mistake – it places the blame on the single mothers themselves, which does not necessarily follow from the data and is unhelpful. However, it IS true that female householders with no husband present and children under 18 experience poverty at 4 time the rate of married couples with kids under 18, and 2 times the rate of male householders with no wife present and kids under 18. And when comparing among female householders with no husband present and kids under 18, Hispanics and Blacks are again at a higher risk of poverty than Whites and Asians. This makes Eichelberger’s “myth” busting inaccurate. http://www.census.gov/hhes/www/poverty/data/historical/hstpov4.xls
Second, lacking this larger context, Eichelberger fails to understand what her own quoted stats actually mean. She says “Only 9 percent of low-income, urban moms have been single throughout their child’s first five years. Thirty-five percent were married to, or in a relationship with, the child’s father for that entire time.” But if 35% of urban moms were married to, or in a relationship with the child’s father the entire time, that means that 65% weren’t! That’s hardly insignificant.
What’s more, Eichelberger seems to think these stats make the situation better than we might otherwise think for single mothers. But it’s more likely that the problem of poverty is worse for single mothers than we think, not better. The Census data I point to above looks specifically at marital status, i.e. “female householder without husband present and with children under 18.” This would completely miss the presence of unmarried fathers, who would most likely be invisibly lumped into the “female householder with no husband present” category. If the Census data instead captured “female householders without husband or unmarried male cohabitant present, and with children under 18 present”, truly measuring single mothers, the measured poverty rate for this group is likely to be worse than the data captured by Census that ignoring live-in boyfriends – not better!
2. “Absent dads are the problem [is a myth]”/3. “Black fathers are the problem [is a myth]”.
First, again, calling people “the problem” is pejorative and unnecessary, so Eichelberger appears to take opposing arguments in their worst light – something that should generally be avoided when attempting a critique. A better question is whether the absence of fathers tends to contribute to the prevalence of poverty in the lives of their kids and their children’s mother. Additionally, recognizing whether there is a trend of fathers living apart from their children at a higher rate among some demographics compared to others, that could help focus efforts on communities of people that are at higher risk. This is the context under which race might be relevant.
Strictly speaking, whether a father lives apart from his kids and/or is less involved in his kid’s daily lives is only indirectly related to whether his kids live in poverty in the present. Whether he provides monetary support is the primary question, and nothing Eichelberger raises addresses that question. His level of involvement now may affect the poverty level of his children in adulthood, but that’s not addressed here, either. It’s likely that most perpetuations of this “myth” fail to address this point directly, to, but looking solely at Eichelberger’s article we can’t tell that. But she does nothing to contradict a narrative that absent fathers support their children less, which makes this “myth” still un-busted.
Second, if unmarried mothers tend to live in poverty at higher rates than any other kind of household (despite the fact that some of them have live-in boyfriends), does it not follow that absent dads likely contribute to the problem? This certainly seems common sense, although it would take more specific data to establish this. See below.
Third, the CDC study Eichelberger cites elsewhere appears to contradict her assertions on how involved dads that live apart from their kids really are. She references work by Dr. Laura Tach at Cornell University to support saying that 60% of dads that live apart from their kids see their kids daily, but this apparently doesn’t translate into involvement in any of the following activities, as the CDC study’s numbers for daily involvement in each activity studied are much lower.
Chart 1.) Level of paternal involvement by activity and race for fathers that live apart from their kids
So her position that absent fathers stay involved significantly in their children’s lives doesn’t really hold water, at least as far as the activities the CDC measured. This is particularly true when comparing how involved fathers are when they live with their kids vs when they don’t (compare Charts 2 and 3 below.)
Chart 2.) Level of paternal involvement by activity and race, for kids under 5, when the father lives with his kids
Chart 3.) Level of paternal involvement by activity and race, for kids under 5, when the father lives elsewhere
Fourth, while Eichelberger is correct in asserting that Black fathers in the same living situation do tend to be more involved in their kid’s lives than White fathers- both Eichelberger and the CDC authors miss the point when they stop there. Black fathers live apart from their kids almost 3 times as often as White fathers, and statistically, whether or not a father lives with his kids is the most significant predictor of his level of involvement with his children: http://www.census.gov/hhes/families/files/cps2013/tabH3.xls.
Chart 4.) Percentage of kids whose fathers live elsewhere, by race
Below (Chart 5) is the result if we take both living arrangements and involvement levels for each activity given each living arrangement into account. Comparing these results with those in charts 2 and 3 gives a pretty stark picture of the impact that a father living apart from his kids is likely to have on his involvement in their lives.
Chart 5.) Total level of paternal involvement by activity and race for kids under 5
Looking at this data we have no visibility into the level of significant correlation between a father’s living situation and his involvement in monetarily supporting his kids. But we can see that to whatever extent a lack of paternal involvement adversely affects children, Black children are likely to be affected the most, not the least. This runs counter to Eichelberger’s viewpoint based on a narrower view of the stats. It also runs counter to the conclusions of the CDC study’s authors and to the rather one-dimensional interpretation of the same CDC study attempted by Tara Culp-Ressler at Think Progress.
In the case of Black men that have heard the stereotype about Black men being poor fathers, the CDC study offers a hopeful counter to this, showing that Black fathers tend to be more involved than others when compared within the same living situation. But the bottom line for fathers is that the single biggest thing they can do to stay involved in their kid’s lives is to stay living with them. And this should be especially take to heart within the Black community as the prevalence of fathers living apart from their children is significantly higher among Blacks than among Whites (or even Latinos.)
Eichelberger demonstrates enough of carelessness and lack of relevant knowledge in addressing these 4 “myths” that taking the rest of her post with a huge grain of salt (or simply ignoring it) would be more prudent than taking it at face value. A well-tuned B.S. meter is a must in our world of 24 hour news and highly productive “news” channels, sites, and blogs that spout opinion more plentifully than facts. Working on this kind of critique certainly helps me calibrate my B.S. meter. I hope it does the same for my readers.