CHARTS AND TABLES

 All Charts and Tables Updated for 2002-03

The data presented on this page shows how various demographic and achievement variables are related to one another.  The charts provide a graphical representation of these relationships. 

The correlation coefficients which accompany each chart are a statistical method for measuring the strength of the relationship between two variables.  Click here for a brief explanation of correlations.

CHART 1. Academic Performance and TQI Rating

CHART 2. Academic Performance and Poverty

CHART 3. Academic Performance and Percent Students Who Are English Language Learners

CHART 4. TQI Rating and Poverty

CHART 5. TQI Rating and English Language Learners

CHART 6 (a and b). TQI Rating and Student Ethnicity   NEW!

CHART 7. TQI Rating and Number of Schools

CHART 8. Average TQI for All Schools in California (by Year)  NEW!

CHART 9. Rate of change required to meet NCLB "highly qualified" teacher requirement  NEW!

TABLE 1. Correlations Among Variables

TABLE 2. Distribution of Underqualified Teachers in California (by API Rating)

TABLE 3. Distribution of Underqualified Teachers in California (by Poverty)

TABLE 4. Distribution of Underqualified Teachers in California (by Language Proficiency)

TABLE 5. Number of Districts by "Spread" of Underqualified Teachers

CHART 1. Academic Performance and TQI Rating

Note:  API Rankings are from 2002-03.

TOP of Page

CHART 2. Academic Performance and Poverty

Note:  API Rankings are from 2002-03.

TOP of Page

CHART 3. Academic Performance and Percentage of English Language Learners

Note:  API Rankings are from 2002-03.

TOP of Page

CHART 4:  TQI Rating and Poverty

TOP of Page

CHART 5:  TQI Rating and English Language Learners

TOP of Page

CHART 6:  TQI Rating and Student Ethnicity (a)

• In schools where 60% or more of the student population is WHITE, the average TQI rating is 8.6
• In schools where 60% or more of the student population is ASIAN, the average TQI rating is 7.7
• In schools where 60% or more of the student population is HISPANIC, the average TQI rating is 5.8
• In schools where 60% or more of the student population is BLACK, the average TQI rating is 4.9

CHART 6: TQI Rating and Student Ethnicity (b)

• In schools where 90% or more of the student population is WHITE, the average TQI rating is 9.0
• In schools where 90% or more of the student population is NON-WHITE, the average TQI rating is 5.3

TOP of Page

CHART 7. TQI Rating and Number of Schools
NOTE 1: While the API is "norm referenced," the TQI is "criterion referenced."  This means that a TQI rating is based on a set of criteria (e.g., the percentage of underqualified teachers at the school) and not on how one school's numbers compare with another's.  It is possible under this method for all schools to have a TQI of 10 if the percentage of underqualified teachers was sufficiently low.  As indicated in the graph below, nearly half of the schools in California already have TQI ratings of 9 - 10.  It is, of course, the other half of the state's schools (and especially those with very low TQI ratings) that we should be most concerned about with respect to the qualifications of their teachers.  With the norm-referenced API, there will always be the same number of schools with 10's, 9's, 8's, etc. regardless of how well or poorly schools perform on the achievement tests on which the API is based.  

Click here to see how the TQI is calculated.

TOP of Page

CHART 8: Average TQI for All Schools in California (by Year)

CHART 9: Rate of change required to meet NCLB "highly qualified" teacher requirement

From 1999-00 to 2002-03, the percentage of underqualified teachers statewide fell at an average annual rate of 9%. If we assume that 20% of California's teachers were not "highly qualified" in 2002-03 (see note below), and the percentage of underqualified teachers continues to drop at the same rate, 15% of our teachers, as shown in the chart below, will be underqualified in 2005-06, the year when NCLB says schools may no longer employ any teacher who is not "highly qualified."  If California were to double the rate of reduction with an average annual rate of 18%, then 11% of our teachers statewide would be underqualified.  In order to comply with NCLB by 2005-06 California would need to reduce the percentage of underqualified teachers by an annual rate of 60%!

Enter your own figures into the Highly Qualified Teacher Projection Calculator and see where California will stand in 2005-06.

Notes:
1. Why should we assume that 20% of our California's teachers would not be considered "highly qualified" when the percentage reported here and elsewhere is closer to 11%?  What has been reported on this website and by organizations such as the Center for the Future of Teaching and Learning is the percentage of teachers who are not fully certified.  To be considered "highly qualified" by NCLB, one must not only be fully certified but also have demonstrated appropriate subject matter knowledge (in ways that are very different from what California currently requires).  The federal definition boosts the number of underqualified teachers because it has raised--or, some would argue, changed--the qualification bar.  In its Consolidated Application sent to the Federal Department of Education in August 2003, California's State Board of Education estimated that 52% of all teachers statewide currently are not "highly qualified" in at least one subject.  In high-poverty schools, it estimates that 65% of our teachers fall into this category. These estimates are likely to drop significantly if the federal Department of Education accepts California's plan for HOUSSE (High Objective Uniform State Standard of Evaluation) which would essentially allow California to apply its current methods of evaluation to most existing teachers.  But even if the HOUSSE plan is accepted, all new teachers will be required to comply with the federal definition.  An estimate of 20% is an educated guess but, if anything, it is probably low. 

2. NCLB's requirements apply only to schools receiving federal Title I funds.  Since California is still permitted to employ teachers who are not "highly qualified" in schools that are not eligible for Title I funds, can't we simply make sure that our underqualified teachers teach only in these schools?  Unfortunately there are good reasons to believe this will not happen. Schools become eligible for Title I funds by having a large student population that is poor.  Non-Title I schools--those with wealthier students--have relatively little problem attracting and keeping "highly qualified" teachers. Few people would be happy if these teachers were suddenly replaced with ones who were not qualified enough to teach in Title I schools.  If, however, school districts were to employ strategies to attract "highly qualified" teachers to their Title I schools (see success stories), there is a good chance real progress could be made.  But isn't this a zero-sum game?  Aren't we just shifting the staffing problem from one place to another? No.  Most middle-class schools have access to large pools of qualified teaching candidates.  If teachers in these schools were persuaded (and not forced) to transfer to hard-to-staff schools--by creating conditions that enable them to succeed--most districts would not find it difficult to replace them.

TOP of Page 

TABLE 1. Correlations Among Variables

TABLE 2. Distribution of Underqualified Teachers in California (By API Rating)

Year: 2002-03

HIGH API schools have an API rank of 9 or 10. 
LOW API schools have an API rank of 1 or 2.

TABLE 3. Distribution of Underqualified Teachers in California (by Poverty)

Year: 2002-03

In HIGH STUDENT POVERTY schools 90% or more of the students qualify for the free or reduced lunch program.
In LOW STUDENT POVERTY schools 10% or fewer of the students qualify for the free or reduced lunch program.

TABLE 4. Distribution of Underqualified Teachers in California (by Language Proficiency)

Year: 2002-03

In HIGH ELL schools 75% or more of the students are classified as English Language Learners.
In LOW ELL schools 25% or fewer of the students are classified as English Language Learners.

TABLE 5. Number of Districts by "Spread" of Underqualified Teachers

* Spread is calculated only for districts with 5 or more schools

TOP of Page

AN EXPLANATION OF CORRELATIONS
A strong positive correlation occurs when increases in Variable A are strongly associated with increases in Variable B.  Smoking of cigarettes is, for example, positively correlated with lung cancer.  A strong negative correlation occurs when increases in Variable A are often associated with decreases in Variable B.   Increases in gasoline prices are negatively correlated with the quantities of gasoline purchased by consumers.  Correlation coefficients range from +1 (strong correlation) to -1 (negative correlation). A correlation coefficient close to 0 indicates that no correlation exists between the two variables. 

To say that two variables are strongly correlated does not mean necessarily that changes in one causes change in the other.  It could be that people who smoke a lot also live in places with lots of air pollution.  Unless one accounts for the influence of this third variable, air pollution, and any others that might cause lung cancer, one cannot say with confidence that cigarette smoking causes cancer.  Researchers have, of course, accounted for the other variables which is why the surgeon general can say with confidence that cigarettes are known to cause cancer. 

What about the correlations among the variables under consideration here?  Most importantly, is there a causal link between a school's TQI rating and its API rank, or is the .3 correlation merely a coincidence?  Several studies that examine the relationship between teacher qualifications and student academic performance provide strong support for a causal link.  David Berliner's recent study is especially noteworthy.  In his conclusion he writes:

This study addressed one of these factors—the effectiveness of certification on student achievement.  We found what might be expected of those who choose to do complex work, namely, that those who trained longer and harder to do that work do it better.  Common sense and empirical data agree. Despite our lack of understanding of how it is accomplished, and despite the extreme variability in the programs of instruction (surely masking both excellent and dreadful programs), the present research study supports the assertion that university-prepared teachers are of higher quality than those prepared without an approved program of preparation (see also Evertson, 1984; Darling-Hammond, 1997a). 

In this study regularly certified teachers significantly outperformed under-certified teachers with children who are most at risk of school failure and school dropout.  These already low-achieving children, when assigned to the classrooms of under-certified teachers, made gains that were approximately 2 months less per school year on three different subtests of the SAT 9. This is about 20% less academic growth than they would have made had they been assigned to a teacher with regular state certification. (David Berliner, "The Effectiveness of 'Teach for America' and Other Under-certified Teachers on Student Academic Achievement: A Case of Harmful Public Policy.")

TOP of Page