How have state testing programs affected teaching practices? This question is key to measuring the success of the standards and accountability movement, and teachers are a prime source to provide insight into the answer.
- High stakes for schools or teachers and high stakes for students (High/High).
- High stakes for schools or teachers and moderate stakes for students (High/Moderate).
- High stakes for schools or teachers and low stakes for students (High/Low).
- Moderate stakes for schools or teachers and high stakes for students (Moderate/High).
- Moderate stakes for schools or teachers and low stakes for students (Moderate/Low).
Alignment
Teachers responded to a series of questions about the match between the state standards and tests and their own instruction and assessments. For each item, teachers indicated their degree of agreement on a 4-point Likert scale (strongly agree, agree, disagree, or strongly disagree ). We have collapsed the strongly agree and agree categories here to give a sense of overall agreement.
More than three-fourths of all teachers, regardless of their state's stakes level or their grade level, indicated that their district's curriculum was aligned with the state-mandated testing program. Somewhat fewer teachers (60–65 percent), regardless of stakes level or grade level, agreed that the state-mandated test was compatible with their own daily instruction.
When asked whether district-required instructional texts and materials were compatible with the state-mandated test, 55–65 percent said that they were. Differences existed among groups for this question, however. The highest percentage of teachers agreeing with this statement, 65 percent, was from High/Low states—those with high stakes for schools or teachers and low stakes for students. And fewer elementary teachers (56 percent) than high school teachers (66 percent) agreed with this statement.
About half of the teachers indicated that their own tests had the same content as the state tests. This percentage was higher—close to 60 percent—in High/High states and in High/Low states. A higher percentage of elementary and middle school teachers (58 percent) than high school teachers (50 percent) agreed that their tests had the same content as the state test.
Fewer than half of the teachers said that their tests were in the same format as the state tests. This percentage also varied according to the stakes attached to the tests: 51 percent of teachers in the highest-stakes states and only 29 percent of teachers in the lowest-stakes states agreed with this item. Fewer high school teachers (38 percent) than elementary or middle school teachers (49 percent) agreed.
In summary, although 75 percent of all teachers said that their district's curriculum was aligned with the state testing program, this figure dropped slightly when we asked teachers about the alignment of their own instruction with the state testing program. It dropped again when we asked about the alignment of instructional materials with the state-mandated testing program, and again regarding alignment of their own tests' content and format with those of the state tests. A closer alignment of instructional materials and teacher tests tended to occur, however, in states that placed high stakes on their state assessments.
Narrowing of Instruction
Using a 5-point scale (increased a great deal, increased moderately, stayed the same, decreased moderately, or decreased a great deal ), teachers who had taught for more than one year indicated whether they had changed the amount of time spent on various activities in order to prepare students for the state-mandated testing program. The teachers rated the amount of time spent on core instructional areas that appeared on the state tests; noncore areas, such as fine arts or foreign language; and activities, such as field trips or school assemblies.
The results indicated that many teachers believe that tests have a narrowing effect on what they teach. Especially in states with high stakes for students, teachers indicated that they had increased the amount of time spent on tested areas, a finding similar to that of Shepard (1990). It is not surprising to find that they had also decreased the amount of time spent in the other two areas, especially activities.
For example, 43 percent of teachers in High/High states and about one-third of teachers in states in the other high-stakes categories indicated that they increased a great deal the time spent on instruction in tested areas, compared with only 17 percent of teachers in Moderate/Low states. Conversely, 16 percent of teachers in High/High states, as opposed to 7 percent of teachers in Moderate/Low states, said that they had decreased a great deal the time spent on instruction in fine arts. It is interesting to note that the percentage of teachers in High/High states saying that they decreased the amount of time spent on fine arts is far less than the percentage saying that they had increased time spent on tested areas. Although some teachers take the needed additional time from fine arts, others apparently take the time from foreign language, student performances, field trips, or a combination of non-tested areas.
Differences also existed by grade level. Elementary teachers said that they increased time spent on tested areas and decreased time spent on noncore areas and activities to a greater extent than did high school teachers. Middle school teachers fell somewhere in between, but their responses were closer to those of the elementary teachers.
We also asked teachers to indicate the extent of their agreement with this statement: The state-mandated testing program leads some teachers in my school to teach in ways that contradict their own ideas of good educational practice. Roughly 75 percent of the teachers agreed. Higher percentages of teachers in states with high stakes for either teachers/schools or students agreed (71–77 percent) than did teachers in the Moderate/Low group (63 percent). Larger percentages of elementary teachers (78 percent) agreed with the statement than did high school teachers (67 percent) or middle school teachers (73 percent).
This latter finding is particularly troubling because it indicates that statewide testing programs, especially high-stakes ones, are leading to an outcome that policymakers do not intend—namely, that teachers teach in ways that contradict their own views of good instructional practice. To bring students to the required level, teachers apparently find it necessary to use teaching strategies that they don't believe in. As reported elsewhere (Pedulla et al., 2003), teachers in high-stakes states indicate that they spend a great deal of time preparing students specifically for the state tests. These test preparation activities may be the activities teachers referred to in our survey as running counter to their views of good instructional practice.
Pressure
The survey included a series of questions about the pressure teachers feel as a result of the state-mandated testing program. Teachers were asked to react to the following statements.
Teachers feel pressure from the district superintendent to raise scores on the state test. More than 90 percent of teachers in states with high stakes attached to their tests agreed with this statement; a slightly smaller percentage of teachers (84 percent) in Moderate/ Low states agreed. Larger percentages of elementary and middle school teachers agreed (92 percent) than did high school teachers (85 percent).
Teachers feel pressure from the building principal to raise scores on the test. More than 80 percent of teachers in states with high stakes agreed with this statement, compared with only 68 percent of Moderate/Low teachers. Elementary and middle school teachers (85 percent) were more likely to agree than were high school teachers (76 percent).
There is so much pressure for high scores on the state-mandated test that teachers have little time to teach anything not on the test. Although 80 percent of teachers in High/High states agreed with this statement, a much lower 56 percent of Moderate/Low teachers agreed. Again, higher percentages of elementary and middle school teachers agreed (78 percent) than did high school teachers (61 percent).
Teachers were even more likely to agree with the survey items on pressure than they were to agree with the item about spending more time on tested areas. All of the percentages on the items about pressure are extremely high, indicating that teachers, and especially elementary and middle school teachers, feel a great deal of pressure as a result of teaching in a state with a statewide testing program. Not surprisingly, the pressure is even greater when high stakes are attached.
- High school teachers often teach multiple grade levels, whereas elementary teachers teach just one; therefore, high school teachers would be less likely to desire a transfer.
- Elementary teachers teach several tested subject areas; high school teachers generally teach only one. Thus, elementary teachers may feel more pressure.
- The high school curriculum may already match the framework of the state testing program better than the elementary curriculum does. If so, more instructional change would be necessary at the elementary level, thus creating more stress for elementary teachers at the tested grade levels.
- High school teachers may view the state test as a reasonable measure of what they were teaching before the testing program went into effect, whereas elementary teachers may view the test as more of a departure from what they have been teaching.
These explanations are speculative, but all seem plausible.
Overall Views of State-Mandated Testing
- Overall, the benefits of the state-mandated testing program are worth the investment of time and money.
- The state-mandated test measures high standards of achievement.
Fewer than 30 percent of all teachers agreed with the first statement. This percentage did not differ among the groups of teachers, regardless of stakes or grade levels. Thus, a healthy majority of all teachers surveyed (more than 70 percent) did not see great benefit in the state testing program.
About half of all teachers, regardless of their state's stakes levels, agreed that the state test measured high standards of achievement. These percentages differed markedly by grade level, however. Fifty-six percent of elementary teachers, 48 percent of middle school teachers, and only 35 percent of high school teachers agreed.
Overall results from the survey paint a revealing picture. Most teachers believe that their district's curriculum is aligned with the standards measured by their state tests. The vast majority of teachers also experience a great deal of pressure related to the testing program. Teachers in high-stakes states have spent more time on tested areas and less time on nontested areas since the implementation of their state testing program. Unfortunately, teachers believe that some of this instructional time does not represent good education practice. In keeping with this sentiment, many teachers do not believe that the state-mandated testing program is worth the time and money invested.
Educators must examine these findings in the context of their particular state's desired outcomes. For example, policymakers may have designed their state's accountability system with the desired goal of getting teachers to spend more time on tested areas and less on other areas. If so, the results presented here provide evidence that this occurs. If requiring more time on tested areas was not their intention, policymakers should reexamine the testing program and possibly change it to decrease this effect. These survey results can guide our thinking when examining existing testing programs or designing new ones so that we can achieve the outcomes that we want and avoid creating unintended and undesired outcomes.
