Student-Program Alignment and

Teaching to Mastery

Professor Siegfried Engelmann

(University of Oregon)

 

Paper Presented at National Direct Instruction Conference

Eugene, Oregon. 1999   [Italics added by Professor Plum.]


When students are taught to mastery, they become

smarter, acquire information faster, and develop

efficient strategies for learning. Teachers must have an

understanding of what mastery is and how to achieve it

in their students. However, teachers cannot teach to

mastery without referencing the performance of their

students. In addition, teachers cannot teach to mastery

without a program design that supports the approach.

Teaching to mastery is built upon effective

student/program alignment. This paper discusses the

features of a program design that supports mastery,

properties of mastery, criteria and procedures for

measuring mastery, procedures for aligning program

placement with student performance, and the benefits of

mastery.

 

Features of a Program Design that Supports
Mastery

A program design that supports mastery does not

present great amounts of new information and skill

training in each lesson. Rather, work is distributed so

new parts in a lesson account for only 10 - 15 percent of

the total lesson. The rest of the lesson firms and

reviews material and skills presented earlier in the

program. The program assumes that nothing is taught in

one lesson. Instead, new concepts and skills are

presented in two or three consecutive lessons to provide

students with enough exposure to new material that they

are able to use it in applications. So a lesson presents

material that is new today; material that is being firmed,

having been presented in the last two or three lessons;

and material that was presented even earlier in the

sequence and is assumed to be thoroughly mastered.

This material often takes the form of problems or

applications that require earlier-taught knowledge.

 

The amount of new material is relatively small because

most students are not capable of assimilating more. This

design provides for some “overlearning”, but having the

program err in the direction of providing too much

practice is better than providing too little practice. Work

on material presented in the preceding few lessons is

needed to ensure that students are “automatic” with

information or operations that were previously taught.

The review of earlier material assures that students use

and apply what they have learned. Reviews also prompt

students toward an understanding that they expected to

retain and use material learned - not just learn it for the

moment. Basically, most things are taught in the

program so they can be used in applications or problemsolving

settings. Therefore, the program is constructed

so students review and use what they have learned

according to a systematic schedule. Because reviews

are a regular feature of every lesson, the program

design provides daily prompting that material presented

will appear again. Also, applications that involve earlier taught

skills provide the kind of practice that students

need to keep from mixing up different things they are

learning. If students partially learn things, new learning

is easily confused with things that are similar. If

students learn material well, less confusion results.

 

Mastering a step at a time. A program designed with

small amounts of new material in each lesson is

something like a stairway. Like a stairway, it needs

strong support. That support is in the form of the

previously taught skills and knowledge that are logical

underpinnings for what is to come next in the program.

Also, for the stairway to work well, the “steps” in this

series should be about the same size. Certainly, they

can’t be fashioned with the accuracy of a physical

stairway, but they can be designed so they are close to

each other in size.

If we conceive of the program as being like a stairway

that transports students to increasingly complex

performance, we recognise the supreme importance of

mastery, what it is, and how it relates to the curriculum.

The following six points clarify the relationship between

mastery and the stairway.

 

1. The program will function as a stairway if

the student reaches eve ry stair on

schedule. If students are firmly on the fifth

stair (which is analogous to the fifth lesson),

the new learning that students must achieve

to reach the sixth stair is manageable. The

students’ position on the fifth stair represents

a foundation that places the sixth lesson

within stepping distance. Because the

foundation is in place, the sixth lesson does

not overwhelm students with too much new

vocabulary, unfamiliar or unpracticed

operations, too much information, or too many

unknown or unexplained details.

 

2. The steps are levelers of individual

differences. Not all students who stand on

the fifth stair are the same age, learn at

precisely the same rate, have equal

intelligence, or exhibit the same “style” of

learning. However, every student who is

firmly on the fifth step is the same with

respect to the program sequence. Each has

the skill repertoire and knowledge needed to

take the next step and reach that step within

30-45 minutes of instruction. Because

students could not reach the fifth step without

specific skill and knowledge, the stairway

structure of a well-designed program serves

as a leveller. All students with a particular

skill profile are placed on the same stair.

Certainly, the program design does not

guarantee that all students will progress at

exactly the same rate; however, greatest

individual differences occur on the very

beginning levels. On higher levels, after

students have mastered a battery of skills and

knowledge, the difference in rate of ascent

for appropriately placed students is far less

because all students tend to have enough skill

to master the new material at around the

same rate.

 

3. The benefits of the design of the program

are obliterated if a student falls below the

level of a stair. This fact holds for students

who are “smart” as well as those who have a

history of failure. If a student is below the

fifth stair and tried to reach the sixth stair

with one step (which means thoroughly

mastering the sixth lesson in one period), the

student must learn substantially more than

students who are firmly on the fifth stair.

Furthermore, the student must learn this

material during the same amount of time

allotted for students who are firmly on the

fifth stair. Therefore, the student who is

below the fifth stair must learn the material at

a faster rate. The student on the fourth stair

must learn material at twice the rate of

students who are correctly placed. The

student who is on the third stair must learn at

three times the rate. For the typical student,

a step that requires three times the amount of

new learning is too great. Even if the student

is able to perform acceptably on lesson 6

after some repetition, the retention rate of the

student on the subsequent lessons drops

dramatically.

 

4. Just as the design of the program

“guarantees” a successful future for

students who are firmly based on a stair,

the design suggests an unsuccessful

future for a student who is greatly below

that stair. The systematic stairway design

does not provide relief because skills and

knowledge do not go away. Once introduced,

they are used throughout the rest of the

program, either as elements that are used

regularly (such as a word type that is

learned), as details that are embedded in

problems and applications (such as the math

operation of carrying), or as items that are

frequently reviewed (such as identifying the

verb in sentences). Because of this program

design, once a student falls behind, the

student will tend not to catch up. If the

student is initially 3 steps below the lesson,

the student will probably end up a little more

than 3 steps below the next lesson, a little

further below the following lesson, and so

forth until the student is not 3, but 4, steps

below the level of the lesson, then 5 steps

below, and so forth.

This student is not able to benefit from the design of

the problem, because although the program presents

small increments of learning, this student must master

large increments of learning to catch up. For this

student, the program presents a poorly designed

sequence. It requires too much new learning and

does not provide adequate reviews.

 

5. Because the program’s design benefits

are transmitted only to students who are

on the lesson stairs, student performance

must match the level of performance

assumed by each stair. This goal is

achieved if teachers teach to mastery.

Mastery assures that everything that is

supposed to be taught is taught thoroughly

and at the time it is introduced in the program

(not 20 or 30 lessons later).

 

Note, however, that DI programs are designed with

enough redundancy that a student who is absent for

two or three days will not be perfectly lost for the rest

of the year. Also, if students do not master a new

skill on the first day it is introduced, the following

lessons provide at least one - possibly two - reviews

of the introduction so that students will have sufficient

opportunity to learn the skill before it is assumed to be

in their skill repertoire and begins to appear in

applications.

 

The problem occurs when students are not brought to

mastery on skills that will be used later. For instance,

students in Level 1 of Reading Mastery are

supposed to be taught to follow the teacher’s

directions about “touching words” before lesson 30.

The tasks that the teacher presents require students

to follow directions to “Touch the first word ... touch

the next word ... touch the next word ...”

Often students are not brought to mastery when this

series of tasks is introduced. These students have

problems in the lesson range of the 40s because now

they are expected to first “touch the next word...” and

then “sound it out.” If they are not firm on touching the

next word on signal, the activity becomes very sloppy

and students often become confused about what they

are supposed to do. If students are taught on time,

however, they have far less difficulty mastering the

mechanical steps of touching the next word and then

touching the individual letters as they sound it out. The

program design provides for enough practice; however,

that practice must not be mere exposure or practice with

a very low standard of performance. The practice must

lead to mastery.

 

6. Most programs do not require teaching

to mastery. Teaching to mastery is a

foreign practice to many experienced

teachers because most programs do not

require mastery. Instead of providing

continuous skill development, these programs

present topical or thematic units. Students

will work on a particular unit for a few days

and then it will be replaced by another unit

that is not closely related to the first and that

does not require application of the same skills

and knowledge. This design, referred to as a

“spiral curriculum”, is more comfortable for

the program designers, teacher, and students;

however, it is inferior for teaching skills and

knowledge.

 

It is comfortable for the designers because the design

does not have to be careful. The designers do not

have to document that everything that is presented is

“teachable”; the amount of new learning does not

have to be carefully measured. The amount of time

required for a “lesson” does not have to correspond

precisely to a period, because the design assumes that

different teachers will take different amounts of time

to get through a particular “lesson” and “unit”. The

amount of new material is not controlled. The

expectations for student performance is low because

teachers understand that students will not actually

master the material. They will simply be exposed.

The accountability of the teacher is therefore more

“comfortable” because the teacher is not expected to

get through the material in a specified period of time

or bring students to mastery. The spiral curriculum is

more comfortable for students because they are not

required to learn, use, or apply the skills from one unit

to the next unit. They quickly learn that even though

they do not understand the details of a particular unit,

the unit will soon disappear an be replaced by another

that does not require application of skills and

knowledge from the previous unit. The design clearly

reinforces students for not learning or for learning

often vague and inappropriate associations of

vocabulary with a particular topic.

 

If the systematic program is like a stairway, the spiral

curriculum is like a series of random platforms

suspended on different levels. Students are

mysteriously transported from one platform to

another, where they remain for a few days as they

are exposed to information that is not greatly

prioritized. Mastery is impractical with a spiral

curriculum design because many students lack the

background knowledge they need to stand on a

particular “platform”. The poor design relieves the

program designer of assuring that earlier-taught skills

and knowledge are mastered and used. The poor

design also relieves students of the responsibility of

learning to mastery and it relieves the teacher of

teaching to mastery. It therefore promotes poor

teaching and poor learning.

 

In summary, a program that teaches to mastery is like a

stairway. Mastery is the guarantee that students are

able to reach each stair without falling.

 

Properties of Mastery

Clearly, mastery is the handmaiden of a systematic

program. Mastery is effective for a number of reasons.

The most important reason is that mastery permits

teachers to achieve steady reliable progress in student

learning. When teachers teach to mastery, we can

make predictions about student performance. We can

very accurately project where students will be 100

school days from now or 200 school days from now.

Such projections are very powerful, but very foreign to

traditional orientations about learning, which view the

students’ performance as a function of their ability to

learn and motivation to learn. Therefore, to predict that

student X will be accurately reading 30 words per

minute by the end of the kindergarten year would be

something of a contradiction because it assumes that the

teaching somehow controls the student’s learning.

The traditionalist hopes to reach and motivate the

student and hopes that the student does not have some

type of mysterious “learning disability” that interferes

with learning to read. The traditionalist, however, is

unable to predict who will read and who won’t.

Readiness tests are tools that are supposed to predict

performance according to what the student brings to

school. Because they don’t take into account the kind

of reading instruction the student will receive, readiness

tests fail to predict accurately. In fact, the traditional

orientation to reading has a classification for students

who are predicted by readiness tests to succeed but who

fail to learn to read on schedule - specific learning

disabilities. Note that this label holds fast to the

assumption that the student’s failure to learn to read has

to do with a flaw in the student, not a flow in the

instruction. The school or teacher does not have a

“disability”. The student does. In other words, for the

traditionalist, the performance of the student is not

clearly linked to teaching. The more scientific

orientation to teaching that by DI espouses assumes that

the student who meets the entrance requirements for the

program and who is taught appropriately (to mastery and

on schedule) will respond in perfectly lawful ways and

will be reading at a predicted skill level by the end of the

kindergarten year.

 

Individualisation must occur from the beginning.

Projections are keyed to the performance of a student.

Not all children entering kindergarten have the same

projections because not all of them start at the same

place. Those who enter with more skills have a

headstart and are expected to be farther after nine

months of instruction than the child who enters with a

lower skill level. However, even if children begin as low

performers, the prediction is that they will master

beginning reading skills in kindergarten and will be

reading by the end of kindergarten. For the child who

enters with a low skill level, the projected end-of-K-year

performance may be lesson 120. The projection for the

higher performer may be double that number.

The fact that projections are met means that the DI

orientation to teaching and mastery is correct. Students

will learn if the teaching is appropriate. If they fail to

learn, the reason lies not with their inability to learn but

with the delivery system’s inability to teach.

The concept of individualisation is closely related to the

issue of mastery and to projections about students’

performance. The teacher cannot teach to mastery

without referring to the performance of the students

being taught. The teacher bases decisions about what

to do next on samples of each student’s behaviour. This

sample may come from tasks presented to the group,

tasks presented to individual students, or worksheets and

similar work samples. DI is designed so students’

thinking is made overt. The teacher therefore receives

samples of behaviour at a high rate on everything that

is being taught. The teacher uses this information to

judge what rate of presentation is appropriate. If

students have already learned the skill or concept, the

teacher is to move on. If the teacher determines that

some students have not mastered what is being taught,

the teacher corrects the mistakes and possibly repeats

parts of the exercise. If quite a few students missed the

item, the teacher may repeat the entire exercise with the

whole group, which is more efficient than presenting it

to some students individually.

 

In summary, teaching to mastery is possible only if the

teacher keys the amount and type of practice students

receive to the performance of these children.

 

Criteria and Procedures for Measuring Mastery

Teaching to mastery is a difficult procedure for teachers

to learn. They must learn to reference what to do next

according to the students’ performance. They must

learn high, but realistic, expectations for their students.

They must also learn to coordinate mastery with fast

pacing so that the lesson is neither a chore for students

nor busy-work. The teacher uses efficient means of

checking students’ work, of providing additional practice

and firm-ups for students who do not achieve mastery

on skills that were taught, and of providing

reinforcement for trying hard and for succeeding.

 

First-time correct procedures. An important key to

teaching to mastery is the use of first-time correct

procedures. Procedures for inducing mastery require

the teacher to interpret students’ performance. The

primary indicator of mastery is how well students

perform the first time a particular task or exercise is

presented in the lesson. Each time a task is presented,

the group either responds correctly (all students correct)

or incorrectly (some students giving the wrong response

or no response). First-time correct means all students

are correct the first time a task is presented in a lesson.

Also important is how well students perform on the task

or exercise if the teacher presents it more than once. If

the teacher corrects and repeats the task or exercise, it

is important for students to perform correctly the second

time. However, for diagnostic purposes, students’

responses to the first time the task or exercise is

presented provides the most critical information about

where students are positioned on the stairway and

whether they are appropriately placed in the program.

For instance, the first time the teacher asks a question

such as, “Do we multiply or divide to solve this

problem?” or the first time students read a particular

word list, their responses reveal information about the

mastery level the students bring to the lesson.

 

The students’ pattern of correct responses also provides

important mastery information. If they are making too

many mistakes, or if they are not firm on material that

had been taught earlier and that is assumed to be firm,

they are placed too far in the program and should be

moved back. If students give solid indications that they

already know what the lessons is teaching, the students

may not be placed as far in the program as they might

be, and the rate of lesson presentation should increase.

Finally, the “correct-response” patterns of a group

indicate whether all students belong in the group or

whether some should be placed in other groups.

 

Four criteria permit precise interpretation of the correct response

performance for groups and individuals:

 

Criterion1. Students should be at least 70% correct on

anything that is being introduced for the first time.

 

Criterion 2. Students should be at least 90% correct

on the parts of the lesson that deal with skills and

information introduced earlier in the program

sequence.

 

Criterion 3. At the end of the lesson, all students be

virtually 100% firm on all tasks and activities.

 

Criterion 4. The rate of student errors should be low

enough that the teacher is able to complete the lesson

in the allotted time.

 

Again, all the percentages are based on how students

perform the first time a particular task is presented in

the lesson. For material that is assumed to be mastered,

the group should respond perfectly at least 9 out of 10

times.

 

As noted above, students’ first-time performance shows

what they have brought with them to the lesson. That is

the material that is in their memory and skill repertoire.

The performance of students after the teacher repeats

the material indicates only what the students may retain

for possibly less than 10 minutes. That time span does

not measure mastery. When students master a skill they

know it “as well as they know their own name.”

 

All four criteria should be considered in evaluating the

mastery of the group. If students meet the first three

criteria but can’t seem to get through lessons in the

allotted time, something is wrong. The following

sections examine the four criteria in more detail.

 

Criterion 1. Students should be at least 70 percent

correct on anything that is being introduced for the

first time. This percentage is based on the

understanding that even the new skills or procedures

that are being introduced are not composed entirely of

material that is new. Much of it will be familiar.

Therefore, the initial rate of correct responses should not

drop below 70 percent. If students are at mastery on

the preceding lessons, this outcome will occur in almost

all cases.

 

If students perform much below 70 percent, they are not

learning the material. If they are only 50 percent

correct, they may be at a chance level - guessing at the

answers or the steps in the operation. Their responses

are not generated by an overall understanding of what

they are learning. At 70 percent correct, their responses

show that they are much closer to understanding the

new material than they are to taking blind stabs at

responding, and therefore should be able to master the

new material during the lesson.

 

Criterion 2. Students should be at least 90 percent

correct on the parts of the lesson that deal with

skills and information introduced earlier in the

program sequence. Criterion 2 is based on the fact

that students must be completely at mastery on earliertaught

material. When earlier-taught material occurs in

later lessons, no reteaching should be required. If

substantial reteaching is needed, the amount of new

learning that students must achieve to mater the lesson

becomes too great. If students are consistently not at

the 90 percent correct level on material that had been

taught earlier in the program, students need more

extensive firming and more delayed tests. Possibly, the

teacher should use a game format in which she asks

students different questions at the end of the lesson.

Students who respond correctly receive points. When

virtually all students consistently earn points, they have

learned good techniques for learning and retaining

information presented in the lesson.

 

Criterion 3. At the end of the lesson, all students

should be virtually 100 percent firm on all tasks

and activities.

 

Criterion 4. The rate of student errors should be

low enough that the teacher is able to complete the

lesson in the allotted time. Criteria 3 and 4 go

together. When the rate of errors for the overall lesson

is low, the teacher does not need to spend great amounts

of time firming students, and the teacher should be able

to complete the lesson in the allotted time. If students

enter the lesson with skills that permit them to attain 70

percent correct on new material and 90 percent correct

on material taught earlier, students should be able to

achieve virtually 100 percent on all exercises presented

in the lesson. Achieving this performance level may

require a little additional firming, but it should not be

necessary or excessive lesson after lesson. Therefore,

if Criteria 1 and 2 are met, students should easily

achieve Criterion 3 and the teacher should be able to

complete the lesson during the allotted time.

 

Calculating percentages. Several different procedures

are effective for teachers to learn how to “estimate” or

calculate the percentage of first-time-correct responses.

One way is to place sticky tabs in the teacher

presentation book after each task, or affix a sheet of

paper to the page so the teacher can mark whether the

group (or individual) correctly responded to each task.

After the children have responded to ten tasks, the

teacher simply counts the number of tasks that were

correct. If seven were correct, the percentage is 70

percent. (Note: if the teacher repeats a task, she would

not mark the second-time performance the same way

she would mark the first-time performance. She could

circle the second-time performance, note the

performance in a second column, or use another way to

separate the first-time performance from performance

on tasks or exercise that are repeated.)

 

After using a procedure of actually counting the

responses within each exercise, the teacher should try

to make estimates in her head. One way is to “ball

park” patterns in terms of whether students are

performing closer to 50 percent or 100 percent. If they

seem closer to 50 percent (missing a little less than half

of what the teacher presents) their first-time percentage