More Internet Sociology on Dropout Rates

by James A. Bacon

In a blog post yesterday, I observed that, while blacks are more prone to drop out from high school than whites statewide, in some 30 rural Virginia localities, blacks showed a lower dropout rate. In Nottoway County, the drop-out gap between blacks and whites was 14 percentage points for the class of 2012 — in favor of blacks. In next-door Lunenburg, the gap was almost 20 points. I speculated that perhaps the breakdown of white, working class culture highlighted by Charles Murray in “Coming Apart” might have played a role.

An interesting conversation ensued in the comments section. Don Rippert hypothesized that perhaps the difference could be attributed to different levels of religiosity between blacks and whites. If blacks attended church more frequently, perhaps they were more likely to hew to traditional values and, consequently, suffer lower rates of social dysfunction such as divorce, teen pregnancy and other problems that would induce students to drop out of school.

Among the data points he presented in support of his conjecture is the fact that there are 59 churches serving the 5,000 households in Lunenberg County, an average of 84 households per church. That compares to, say, suburban Henrico County, where 140 churches serve 108,000 households, an average of 772 households per church. Furthermore, he cited Carsey Institute data showing the religiosity of America’s “black belt,” where blacks and whites are roughly equal in numbers, are significantly higher than for Appalachia, which is overwhelmingly white.

Rippert’s data is suggestive but it’s too impressionistic to persuade anyone who doesn’t want to be convinced that Nottoway/Lunenburg blacks are more religious than their white neighbors. What we can do, however, is test the idea that dropout rates are correlated with metrics of socially dysfunctional behaviors — out-of-wedlock births and teen pregnancies — that might be influenced by religious views.

Statewide, 28% of all white births in 2009 were non-marital while 67% of all black births were. Here is the breakdown by locality. And here are the numbers for Nottoway and Lunenburg:

The rates were elevated for both races, but the rate for blacks was two to three times higher than for whites. If out-of-wedlock births has any bearing out dropout rates in Lunenburg and Nottoway, I can’t see it.

How about teenage births then? It’s teen births that disrupt a girl’s prospects for completing high school, not a non-marital birth later in life. Statewide, the rate of live births is 11.8 per 1,000 females among whites and 22.4% among blacks. Here are the numbers for Nottoway and Lunenburg:

Nope, that’s not the answer either. Blacks have a higher rate of teen pregnancy.

Whatever accounts for dropouts in Lunenburg and Nottoway, it’s not affecting girls as much as boys. In Lunenburg, 15.8% of girls dropped out compared to 25.4% of boys. The gap is 3.1 percentage points in Nottoway. Perhaps we need to be comparing juvenile male arrest rates, but I can’t find that data.

On the other hand, maybe there are non-sociological factors at work. Maybe there are  dynamic principals and school administrators who inspire black high school students to achieve. Maybe the two counties have dropout prevention programs aimed at black students that I’m not aware of.

Or maybe I’m just chasing a will ‘o the wisp. Perhaps, because the graduating classes of both school systems are small, we’re dealing with small numbers, which make for high rates of variability from year to year. Perhaps the drop-out rates for the classes of 2012 are a fluke and I need to find better things to do with my time.

Share this article


(comments below)


(comments below)


  1. DJRippert Avatar


    Excellent data. One quibble – to understand the dropout rate today (among high school aged people) you’d have to look at the non-marital live births from 14 to 18 years ago rather than 2009.

    Teen pregnancies is a better statistic although pregnancy for an 18 or 19 year old is a much different situation that pregnancy for a 14 or 15 year old.

    Another idea running through my mind is sports. Almost every county requires a minimum GPA for students who participate in sports. Lunenburg County graduated a young man named Anthony Davis who was a very talented football player at Virginia Tech and, briefly, in the NFL. I have looked at some news reports and sports are a big deal in that area. African-Americans are observably over-represented in professional sports like football and basketball. Perhaps the same is true at the high school level in that area.

    However, I am starting to suspect that a small sample size is leading us to statistically insignificant conclusions.

    I believe the only high school in Lunenburg County is Lunenburg Central High School. It has 482 students with approximately 100 in 12th grade.

    41% of the students are African – American and 55% are white.

    It appears that only 59% of the students who enter the Lunenburg Public School system end up graduating from high school (if I understand the statistics from the web site). However, I would assume that most of the students who make it into 12th grade go on to graduate.

    Just as an example, let’s assume a 10% drop out rate among 12th graders.

    That means 90 students currently enrolled in 12th grade will go on to graduate.

    We would expect 41% of 90 = 37 African American graduates and 55% of 90 = 50 white students to graduate.

    A shift of seven students over the course of the four years of high school would give us the 20% graduating class differential (if I have done the back of the envelope math right).

    That could easily be a teacher or two, coach, or minister making the difference. Or, it could just be a statistical anomaly.

    The problem with these statistics is that the sample group is small.

    Of course, as an alternative, we could just call the principal of Lunenburg Central High School and ask why he / she thinks African – American students at that high school are more likely to graduate than white students (at least for one year).

    1. Agreed: to find out, sometimes, you just gotta pick up the phone.

  2. you’re dealing with small numbers and tip toeing into a data swamp if you expand the sample.

    there are folks who spend their entire careers doing this – right.


    In Va, blacks were denied access to a decent education up until the end of Massive Resistance.

    Almost ALL of the parents of black children from that point on – did not even have a high school education and yet there is some kind of a premise here about church attendance and lifestyle?

    It’s a good thing ya’ll don’t “investigate” explosives…. or other things that require specific education and skill in conducting said investigations…..


  3. DJRippert Avatar


    You seem very confused. First, you seem confused by Jim Bacon’s point. Jim has identified that 30 localities in Virginia have situations where the white high school dropout rate exceeds the African – American dropout rate. Like you, Jim understands the history of racism, Massive Resistance, etc in Virginia. He thinks there is probably some lingering prejudice in Virginia so he wonders how it could be that 30 localities are seeing a higher percentage of African – American children graduate from high school than white children.

    Jim believes white society may be falling apart and causing ever more white students to fail to graduate.

    I believe that black society in rural Virginia is more self-reliant and organized than most people think.

    Your points about Massive Resistance demonstrate something of a misunderstanding. As is too often the case, you refer to events “in Virginia”. However, Virginia never operates as a single entity. Between 1956 and 1960 Massive Resistance was widely followed in almost all of Virginia. However, after that disgraceful four year period things started to change. On Jan 19, 1959 (Robert E Lee’s birthday) the Virginia Supreme Court overturned the Clown Show’s Massive Resistance laws. On Feb 2, 1959 Arlington county began to integrate its school system.

    I wasn’t even born when Arlington County started integrating its schools.

    So, when you say that, “In Va, blacks were denied access to a decent education up until the end of Massive Resistance.”. Well, in Arlington, I guess that was 53 years ago. So, a person who was starting first grade in 1959 would be 60 years old by now.

    Richmond didn’t settle its final school segregation suit with the federal government until 1986. However, the capitulation of the Prince Edward County Board of Supervisors to court mandated integration is generally seen as the end of massive resistance in Virginia. So, even in Prince Edward County, a person born in 1958 would have attended an integrated first grade by 1964. That person would be 54 years old today.

    So, when you write:

    “Almost ALL of the parents of black children from that point on – did not even have a high school education and yet there is some kind of a premise here about church attendance and lifestyle?”.

    From what point on?

    The end of massive resistance in Arlington (1959) or the end of massive resistance in Prince Edward County (1964)?

    See, I am 53 years old and I attended integrated public schools in Virginia throughout my life. All of my peers at Groveton High School have high school educations – blacks, white, whatever. And most of us have seen our children graduate from high school, college and go out to the working world.

    You need to:

    1. Understand Jim’s point and …
    2. Draw yourself a timeline of Virginia history and the parents and grandparents of today’s high school seniors

  4. I …

    1. – understand Jim’s point

    2. – agree that it’s been a while since Massive Resistance and that today’s black kids parents probably did have an opportunity at a HS education but the grand parents did not.

    but I need to see bigger populations of blacks across a wider range of Va counties before I’m going to give my credence here.

    we have more and more tendency these days to have lay people with no background … cherry-picking .. what is mostly anecdotal data and trying to use it to draw broad conclusions.

    When/if I see some scholarly articles from folks who have the background and have done the depth and scope of work to support the premise – and it’s backed up by other credible folks.. I’ll put my stock in it.

    I do not mean to offend Jim nor you but let’s see some wider/deeper data here before I’m on board.

    what explains the reverse situation in our urbanized areas?

  5. look at page 6 of this DOE Lunenburg report:

    I cannot access the xls files you reference but this report for Lunenburg on page 6
    shows a white grad rate of 71% and a black grad rate of 59%.

    perhaps you can show your tables that you extracted.

    1. Wow, I’m not sure how to reconcile that report with the state data I quoted. Here’s a PDF of the data I used (with a lot of extraneous info stripped out):

      Here’s where that digested data came from:

  6. DJRippert Avatar

    “I do not mean to offend Jim nor you but let’s see some wider/deeper data here before I’m on board.”.

    I am not offended. I think you are right. So does Jim (per his comments).

  7. the DOE site is one of the ones I have in mind when I say “data swamp” by the way.

    I’m not convinced that I am right or you are wrong but the numbers I found seem to be what I have normally come to expect at many schools.

    perhaps a clue here is in the Caroline report:

    white – 79
    black – 76

    keep in mind also that the STATE average is:

    white – 86
    black – 73


    page 4 explains the difference and in my view a glimpse into the “data swamp”.

  9. reed fawell Avatar
    reed fawell

    Interesting conversation.

  10. look at this for a comparison between the Federal standard for Graduation and Virginia’s standard.

    Now the interesting thing here is how the numbers with respect to blacks CHANGE when applying the Virginia standard but NOT with the Federal Standard.

    you read the attached and tell me what is going on with Virginia’s approach such that it would pervert the reality of the black achievement gap.

    as I said before.. you have to be careful with data … and my experience with the data that VOE plays with is that one can be easily misled if you’re not pretty thorough in understanding the things that VOE does with data.

    Makes one wonder just what we’d actually know about Virginia schools achievement and graduation data if there was no Federal standard , eh?

    1. Larry, I think you’re on to something. Apparently, there is no uniformity in defining dropout and graduation rates. That makes it easier for administrators to game the system. Clearly, Virginia counts a lot of things as “graduating” that the federal definition does not.

      The next trick (which may not be possible through Internet research) is to chronicle how Virginia has expanded the definition of “graduation” over the years, and to analyze how much of the improvement in graduation rates can be attributed to redefining the word than actual changes in behavior.

  11. HillCityJim Avatar

    Possibly the answer to these questions can be found in Dr. Fryer’s report as he footnotes many recent studies. And no, I don’t have the answers.

  12. well… it’s not only no uniformity – it’s purposeful in my view and it was partly responsible for your own premise of what the data might mean.

    The Feds strive for simple, common-sense metrics and Va gravitates towards data that is obtuse and counter-intuitive and if it is picked up and misunderstood – “oh well”.

    Virginia’s approach to education metrics is, in a word, not forthright.

    they seem to delight in getting into the weeds and playing inside baseball games.

Leave a Reply