One Bad Apple

by Robert Pondiscio
May 12th, 2009

bad-apple Children from troubled families perform “considerably worse” on standardized reading and mathematics tests and are much more likely to commit disciplinary infractions and be suspended than other students, according to a new study.  Writing in Education Next, Scott Carrell of UC-Davis and the University of Pittsburgh’s Mark Hoekstra offer evidence that  “a single disruptive student can indeed influence the academic progress made by an entire classroom of students.”

Carrell and Hoekstra, who are both economists, examined confidential student data from Florida’s Alachua County school district, consisting of observations of students in grades 3 through 5 over an eight-year period. The pair also had access to disciplinary records for every student in their sample, which they cross-referenced to domestic violence data from public records.  What emerged was a compelling set of data that indicates children exposed to domestic violence have more disciplinary problems at school, underperform academically and have a negative effect on peers–resulting in lower test scores and increased disciplinary problems in others.  In essence,  a ”one bad apple” syndrome.  Carrell and Hoekstra title their piece “Domino Effect.”

“A majority of parents and school officials believe that children who are troubled, whatever the cause, not only demonstrate poor academic performance and inappropriate behavior in school, but also adversely affect the learning opportunities for other children in the classroom,”  observe Carrell and Hoekstra.  The pair cite a Public Agenda survey which found that 85 percent of teachers and 73 percent of parents agreed that the “school experience of most students suffers at the expense of a few chronic offenders.”  The study largely validates those concerns. 

Our findings have important implications for both education and social policy. First, they provide strong evidence of the validity of the “bad apple” peer effects model, which hypothesizes that a single disruptive student can negatively affect the outcomes for all other students in the classroom. Second, our results suggest that policies that change a child’s exposure to classmates from troubled families will have important consequences for his educational outcomes. Finally, our results provide a more complete accounting of the social cost of family conflict. Any policies or interventions that help improve the family environment of the most troubled students may have larger benefits than previously anticipated.

Poll teachers in struggling schools, and I will wager a substantial amount that classroom disruption is identified consistently as the primary barrier to student achievement.  Yet it is consistently glossed over or dismissed, typically attributed to a teacher’s lack of classroom management skills.  I have long believed that the time on-task lost to disruption and behavior problems is almost certainly one of the under-discussed root causes of the achievement gap.  This study does a great service by confirming what many teachers and parents have intuited for years: disruption matters and has a negative effect on all students.

School and classroom tone matter enormously–perhaps more than any other factor.  Get it right and everything seems to work.  Get it wrong and nothing does.  This study holds out the promise of sparking a very important discussion about the rights of the individual in the classroom versus the rights of the community.  It’s long overdue. 

(Image via Digital Eargasm)


  1. It makes sense to me that children from troubled homes would bring behavior problems to school. What perplexes me is the children who have seemingly loving, concerned, stable parents; pleasant, functional homes; and average or above-average intelligence, yet still commit serious behavior infractions for every teacher he or she has. If it was just one teacher, you could chalk it up to poor classroom management, personality clash, whatever. But I’m thinking of a former student of mine who had very nice parents and a lovely home and all the advantages thereof…and he was seriously, repeatedly, deeply disruptive in every single class during his long tenure at my school. I have never been able to find a teacher who said, “Oh, he wasn’t so bad.”

    What to do with those children? They don’t require “services” or “interventions” (unless they are mentally ill). So what then?

    Comment by Miss Eyre — May 12, 2009 @ 10:46 pm

  2. Robert writes, “I have long believed that the time on-task lost to disruption and behavior problems is almost certainly one of the under-discussed root causes of the achievement gap.” Me too. In fact I find it uncanny why this is so rarely talked about. I don’t get it. Is it because we’ve all swallowed the party line that teachers who provide engaging lessons and develop good rapport with students don’t have misbehavior in their classrooms, and so complaining about misbehavior implies that we’re bad teachers?

    My own suburban middle school is plagued with typical middle school unruliness (less severe, I’m sure, than what occurs in Bronx, but damaging nonetheless). Yet no one talks about cracking down. There’s a stigma to calling for “sticks” like in-school suspension (which this study implies could be very beneficial for overall student achievement). Teachers should prevent behavior problems by being “in relationship with” (that’s the icky term used in our PD worksheets) students; being a stern disciplinarian thwarts the building of these relationships. Listen, I enjoy good rapport with kids as much as the next warm-blooded human, but I ain’t going to feel much genuine warmth for a kid who constantly talks while I’m talking, or who gives me attitude on a routine basis. Root out such ugly behaviors first with solid discipline, THEN let the (cautious) love-fest begin. My colleagues think love-fest will breed good behavior. They’ve got it backwards.

    One more point: my own experience corroborates the study’s finding that a few “bad apples” can bring down a whole class. This is actually somewhat encouraging, because it suggests that handling a small number of students effectively can yield a major improvement of the learning environment.

    Comment by Ben F — May 13, 2009 @ 12:49 am

  3. I’m speaking from my personal experience as a science enrichment teacher in elementary schools. You cannot rule out teacher quality with regards to “classroom management”. I can tell whether a class respects their teacher within 2 minutes of entering the room. I teach in all classrooms in 4 different schools each year. The teacher is the number one predictor of behavior regardless of demographics. For example, in one school the K classes are mixed race with predominately FARMS children. Three of the classes can be difficult to work with and the teachers are clearly clueless about setting limits effectively. The 4th class is filled with angels; why? Because the teacher expects such behavior. The expectation is evident from the first second I see them. In one of the wealthy schools I work in I see the same exact results. Some classes constantly ignore my instructions and we spend half the time waiting for them to pay attention and there are zero consequences for bad behavior. In other classes the expectations are clear and the children know the teacher means it because she does not hesitate to quickly remove any trouble makers. I want to kiss those teachers! Yes, trouble makers do decrease learning for others – many times my programs are cut short because of misbehavior, but differences in teachers is definitely the number one predictor in my experience. Rather than allowing disrupters to continue to cause problems they are dealt with quickly and effectively. Note that these are suburban schools outside of Wash DC; demographics vary from lots of FARMS students to students of very wealthy neighborhoods.

    Comment by Gina — May 13, 2009 @ 8:17 am

  4. It’s a plausible hypothesis. The practical question is about options: in-school suspension is one, but when and how? (Just saying “go to ISS”?? No.) Behavioral contracts have solid research support, but that’s for relatively mild problems. One relatively unstudied idea (that has probably never been rigorously evaluated) is to make both the student and an adult relative come in on the weekend for a Saturday detention. Lots of plausible ideas, but relatively little research.

    Comment by Sherman Dorn — May 13, 2009 @ 8:32 am

  5. I don’t disagree with you at all, Gina. Teachers matter a great deal. (So do building-wide behavior expectations and policies, as opposed to every teacher going his or her own way, but that’s a subject for another day). But there is a clear tendency to dismiss behavior issues and disruption as completely within a teacher’s control, which is clearly not correct.

    As a related issue, I can’t help but think of the struggling schools banking on earnest rookie teachers, TFA, et al to turnaround. It’s pretty clear that it takes a while to find your voice as a teacher, earn that respect you describe, Gina, and maximize learning time. My hunch is that the schools that extend classroom time and add Saturday sessions are merely bringing the effective on-task time up to where it is in schools with more experienced teachers and less time lost to disruption.

    Comment by Robert Pondiscio — May 13, 2009 @ 8:34 am

  6. I agree with the idea that disruptions, from a number of causes, are a huge elephant in the room. It’s a political problem because there are special-ed, race, SES and gender issues involved, along with ideological biases.

    I also agree that some teachers cope better than others, but any solution that depends on “superstar” teachers is bound to fail because there will never be enough superstars (in teaching or any other field). Helping teachers to improve class management skills is fine, but I think the solution has to include better admin support and the removal of some kids from the classroom, either temporarily or permanently. Some kids are frankly dangerous and some really aren’t suitable for mainstreaming. I can’t escape the thought that homogeneous grouping would also help. Kids who are either bored or totally lost are likely to be disruptive.

    Comment by momof4 — May 13, 2009 @ 6:06 pm

  7. I agree momof4. Far too many school improvement schemes are predicated on the idea that teachers are (or at least ought to be) superstars. I don’t think we’re going to make much progress until we create conditions that allow the average teacher to succeed, and wrestle with the fact that a not-insignificant number of students are ill-suited to traditional classrooms. These problems are most acutely felt in our most challenging schools. And while I don’t mean to sound defeatist, all of the TFA corps members and merit pay in the world does nothing to address that.

    Comment by Robert Pondiscio — May 13, 2009 @ 6:59 pm

  8. Amen Robert and Momof4.

    Comment by Ben F — May 14, 2009 @ 1:26 am

  9. I don’t think you need a study to tell you that the interaction between “bad apples” and “non-deviant” peers has a negative effect on the “good” kids and on the learning environment in general. That just seems obvious.

    As someone who works with students in the Juvenile Justice system, the more important question (at least to me) is how do you prevent this behavior from happening? And when it does happen – as it inevitably will – how do you intervene? For the most at risk kids (the bad apples) I think classroom exclusions will only exacerbate the problem. The kids I wind up working with usually have a long history of in school and out of school suspensions – to say nothing of unmet special education needs. When you look at the amount of money we end up paying on the “bad apple” kids throughout their life, spending a bit more to provide evidence based interventions and supports in and outside of school to prevent further deterioration and promote more positive, constructive behavior, seems like a worthwhile idea.

    Comment by Nick Sheehan — May 14, 2009 @ 10:05 am

  10. The “unmet special education needs” strikes me as the right place to start. At the risk of sounding unsophisticated or worse, hard-hearted, as a former classroom teacher I worry that our concern about exacerbating the problem for the at-risk child has made the situation worse, not better. I don’t think it’s out of line to suggest that providing a safe, productive, warm and nurturing classroom has to be treated as a sacrosanct obligation, especially for our most disadvantaged children. Without that, we’re doomed to spin our wheels.

    Comment by Robert Pondiscio — May 14, 2009 @ 10:23 am

  11. I can’t help wondering if the prevailing ed-school mindset is part of the problem. It seems to me that the most disadvantaged kids need explicit instruction, both in behaviors/attitudes and in reading, math and other subject areas. Waiting for the magic moment of developmental readiness hasn’t worked very well. Why not group all kids homogeneously, from school entry, and teach them the necessary skills and habits? Waiting until they are years behind to offer special help is just wrong. I think it’s time to try the old-fashioned way; maybe fewer kids will have unmet special-ed needs. I remember reading the suggestion that a lot of special ed kids really don’t have learning disabilities; they just haven’t been explictly taught (phonics) to read (in particular).

    Comment by momof4 — May 14, 2009 @ 11:04 am

  12. Robert -
    “I don’t think it’s out of line to suggest that providing a safe, productive, warm and nurturing classroom has to be treated as a sacrosanct obligation, especially for our most disadvantaged children.”

    I couldn’t agree more. But how do we create that classroom? Do we exclude the children who exhibit poor behavior? Or is it possible to include them in the warm nurturing classroom. I think we can, and I don’t think it takes “superstar” teachers.

    Momof4 –
    I agree with you too, at least in terms of teaching behavior. There is a lot of research going on around the framework of positive behavior supports. I’m not an expert on it, but one of the basic premises of this framework is that behavior is learned and therefore can be taught. I’d encourage you to check out and read a bit about it…see if it reflects you’re thinking on the subject.

    Comment by Anonymous — May 14, 2009 @ 1:12 pm

  13. Anonymous, that’s an interesting website. I’ve heard about Positive Behavior Support over the years, but was unclear on what it amounted to (except when done for/with individual students with IEP’s). Clearly, if schools are buying into it, there must be some benefit to those schools.

    However, as I read through the descriptions of how PBS is implemented, I could not help but think, “what parents that had already taken the time and effort to train and nurture their children to respect others, control their impulses, and so forth, wouldn’t be very upset about the huge investment of teacher, student, and administrator time that this system requires? and on another level, how can children who are already adequately socialized not feel that their school experience is awfully mickey mouse if other children are getting a preschool curriculum in proper behavior? if you have an entire school full of children who would benefit from this approach, fine (although sad); but if you implement this system in order to alter the behavior of only 10% or 15%, there’s an equity issue going on.

    Comment by Jane — May 15, 2009 @ 10:14 am

  14. Hey Jane,

    That’s an interesting reaction to PBIS. (BTW i was anonymous, I just forgot to write my name in the post) I don’t agree that there would be an equity issue. If the 10-15% of kids are as a negative a force on the classroom and their peers as this study suggests they are, then wouldn’t other parents want those students’ needs addressed? Then the issue is, how should they be addressed? suspensions? exclusions? or something else?

    To me, the universal interventions do not seem especially intrusive. They seem like the things good schools do inherently: Be clear with students about expected behavior; be clear with students about consequences for inappropriate behavior; praise good behavior regularly; use instances of bad behavior as chance to teach not just punish. Its hard for me to believe any parent could have a problem with a school adopting this kind of mindset. Even if they’ve already taught appropriate behavior, wouldn’t they want it reinforced at school? And when you consider that a positive, proactive approach to behavior has been proven to increase time on task and academic achievement for all students, it seems increasingly unlikely to me that parents would object to PBIS on the grounds that distributes resources unequally.

    It is important to note that in most schools where PBIS is implemented those basic principles only take care of about 80-90% of students. The remaining moderate and high needs students require more focused, individualized attention along the lines of behavioral analysis, behavior intervention plans, and wraparound services.

    Comment by Nick Sheehan — May 15, 2009 @ 1:34 pm

  15. Nick, this may be a question of the inventors/propogators of a particular program over-describing it. The PBIS website makes it sound as if teachers are to spend quite a large amount of time coaching children whose behavior doesn’t fit in the school (yet).

    I have no problem with a moderate amount of such coaching (even though I’d rather see children arrive at school age not needing it). It just seems to me that if 80-90% of students are needing this coaching, something is really wrong that needs to be dealt with earlier in life.

    Comment by Jane — May 15, 2009 @ 5:02 pm

  16. No excuses! Just give those victims of domestic abuse three excellent teachers in a row and they will do just as well as the child from the loving home. (Just kidding!)

    Seriously, if we had the sense to listen to our teachers, or even teachers in places like Canada and Finland, we might solve some of our educational problems.

    Comment by Linda Johnson — May 15, 2009 @ 9:31 pm

  17. “…if you have an entire school full of children who would benefit from this approach, fine (although sad); but if you implement this system in order to alter the behavior of only 10% or 15%, there’s an equity issue going on.”
    I agree with Jane —-although it is much broader than what we have to do for discipline. I think the entire No Child Left Behind issue is because of 10 or 15% of the students. Rather than deal with race and poverty issues they make it the teachers fault for the gap in scores and put huge sanctions on all schools because of a problem with a small number of students that are not taught proper behavior or academics at home.
    The entire school system has become socialized because of a small percentage of students with a gap that has nothing to do with what goes on in schools.

    Comment by M.Beach — May 16, 2009 @ 9:17 am

  18. I don’t think that poverty is the issue; it is the pattern of behaviors that cause chronic poverty. Those who run on impulse and make bad choices stay at the bottom and become the multigenerational underclass; those who make good choices climb up the SES ladder. As I posted on another site, I don’t think this country has ever had such a large percentage of kids who don’t want to be in school and don’t care about education. They are that X% who spoil the barrel for the rest. (their percentage varies widely among schools and communities)

    Jane, there are parents who are very frustrated/unhappy because their well-socialized, well-prepared kids are trapped in a holding pattern because of the disruptive, ill-prepared and uninterested. Many of them remove their families to private schools or home-school. Bad behavior and attitudes are contagious. From my experience, the reluctance of families in the leafy suburbs to accept a large influx of low-SES kids has much more to do with behavior than race or income. I’ve seen the effects first-hand.

    I don’t know much about the Canadian system, but I think I’ve heard that there is a gap between the First Peoples and others; is that accurate? I’m very wary about extrapolating from practices in Finland; it’s not only a much smaller population, it is much more homogeneous. Japan is often held up as an example and I am equally unconvinced of its validity; not only is Japan exceedingly homogeneous, it is ferociously dedicated to education (“education moms”, cram schools, suicides after GATEWAY exam failures).

    Comment by momof4 — May 17, 2009 @ 9:28 am

  19. [...] it’s simply more of what’s not working during the week, well, no thanks.  I’ve banged on this drum endlessly over the years, but the real enemy of achievement in low-performing schools is the time [...]

    Pingback by More vs. More of the Same « The Core Knowledge Blog — March 22, 2010 @ 11:45 am

RSS feed for comments on this post. TrackBack URL

Leave a comment

While the Core Knowledge Foundation wants to hear from readers of this blog, it reserves the right to not post comments online and to edit them for content and appropriateness.