Too Much of a Good Thing?

Teacher question:

      As supervisor of reading and language arts K-5, I've stressed the importance of small group instruction during the literacy block as a means to differentiate and to work explicitly with all students. Teachers recognize the importance of flexible grouping but many attempt have divided their class into 5-6 groups and as a result, meet very infrequently with the groups or for only 10-12 minutes at a time. The instructional block for grades 3-5 tends toward whole group instruction with little time for small group. I've suggested that teachers attempt no more than 3 groups so that students receive the appropriate amount of instruction and at the same time, teachers obtain a better understanding of student progress etc... We've also been transitioning away from rigid instructional levels and moving toward scaffolded instructional support using grade level text. 
      Am I being unreasonable? Some teachers insist that we "show them the research." Could you point me toward specific research on small group instruction as well as time spent 'face to face' with the teacher?

Shanahan's Response:

This is a great question. I’ve observed this problem myself over the years: too many groups and too little teaching.

Let’s be extreme about it: We could maximize the amount of teacher time with kids and the amount of explicit instruction that everyone got by doing all school work in a whole group approach. The model classroom would be students at their desks facing the teacher who would stand at the front of the room. In fact, we could probably knock some walls down and have much larger classrooms… or maybe we could get one lesson delivered via television or the Internet to all third graders. We’d certainly get the maximum explicit instruction that way…

Or we could be extreme in the other direction. Grouping is a great idea, it will give kids more chance to respond and the teacher will get a closer look at the students without the distractions of whole class teaching; the kids will even get a chance to do some work on their own (when the teacher is working with other groups). If that is what is important in education, then let’s make every child his or her own small group (no competition for responding in that situation). If there are 30 students in class for 5 hours per day, then each child should get his or her 10 minutes of tutoring. It will be tight working in reading, math, science, social studies, and the arts into those 10 minutes, but when you can work closely with children you can get a lot done, and the assignments will be terrific.  

Teaching is really a balancing act.

Too much whole class instruction and you don’t get a chance to focus on the kids’ learning needs and they don’t get much of a chance to participate. Too much grouping and you reduce the amount of instructional time with a teacher which cuts their learning, too.

The research show us two things: the amount of explicit instruction is very important in student learning and that instruction requires lots of interaction between teachers and students. I want us much explicit instruction as I can get (including teacher explanation, modeling, questioning, guided practice, feedback)… but it is critical that students have opportunities to interact with the teacher and with each other, too (to use their language, to try to answer questions or respond when the teacher will actually be able to pay attention, etc.).

Kulik and Kulik (1992) have shown that within-class grouping can be beneficial, thought the benefits are small and uneven (with a bigger payoff for the better students—presumably because they get more out of working on their own, and with smaller payoffs for smaller classrooms; small groups tend to do better than more individualized plans, including with computer work). Similarly, we have lots of studies showing the importance of amounts of instruction, time on task, and the value of explicit or direct teaching (see Hattie, 2009 for partial summaries of this work). However, this is a very complicated issue and one that you can’t just point to particular research studies. For example, small group work can be beneficial… and yet, it usually requires students to engage in lots of independent seatwork, which is related to lower achievement, especially for the lower students. 

There is no study that coordinates all of these features and factors in one big study to provide us with any kind of picture of what may be optimum in terms of balancing size and number of groups.

When I see whole class instruction, I’m always asking myself, how could the teacher make this more interactive or involving? How could she monitor student success better? Things like think-pair-share and multiple-response cards can help a lot. Moving around the classroom, watching students carefully helps, too; as does getting a lot of written response.

But when I see a bunch of small groups, I ask:  Are these groups really that different? Could this teacher get away with fewer groups?  Often I find that the reading-level differences that teachers are grouping for are just too small and unreliable to make any learning difference at all. Too often I see several repetitive lessons with slightly different text levels… no one learns much in those kinds of classrooms, since there isn’t much productive work for the students. You’d be better off collapsing to fewer groups with the teacher providing more teaching.  

Of course, the worst situation is lots of small groups, small amounts of teaching, and minimum interaction, responsiveness, and participation… the worst of both worlds. That’s why small group work tends to be more effective when the teacher uses cooperative learning practices, or includes lots of explicit instruction. I don’t have a hard and fast rule about numbers of groups, but like you, when the number creeps up over 3, I start to get suspicious. Of course, in a very large classroom with a very heterogeneous population, you might need more small groups. This is especially true if the classes are skewed towards having lots of very low achieving students. In such cases, the reason for more groups isn’t in response to more levels, but it is an effort to ensure that the low kids get sufficient teacher attention. If the classes are smaller (say 25 or less) and they are average in their distributions or are skewed towards higher achievement, then you can usually get away with fewer groups.

Finally, make sure the scheme being used to determine group assignment is a sound one. The teachers asked for research on fewer small groups, What evidence do that have that the way they are grouping provides a learning advantage? (With common core raising levels in grades 2-12, teachers are going to need to rethink how they are matching students to books anyway.)