Determinants of Student Achievement: New Evidence from San Diego Web Appendix C Additional Information about Student Demographics, School Resources and Test Scores in San Diego and Other Districts This appendix supplements the data on student demographics and school resources presented in Chapter 2. In addition, the appendix adds to the analysis of test scores in Chapter 2. There, we summarized patterns by taking simple averages of test scores across grades. The tables presented in this appendix show grade-by-grade achievement in the five largest districts of California as well as for the state as a whole. Student Demographics Table C.1 shows the ethnic mix of the five largest districts and the entire state in 1997- 1998 and 1999-2000. No district exactly matches the ethnic and racial mix of students in the state as a whole. San Diego has significantly greater percentages of African American and Filipino students than the state, a slightly smaller percentage of Latino students, and in 1999- 2000 roughly 9 percent fewer white students. Three of the other four largest districts have a greater share of Hispanic students and a smaller percentage of African American students, and all four have a smaller percentage of white students. In spite of these differences, the five largest districts have one thing in common: a far smaller share of white students than is found in the state as a whole. Recent trends in ethnic mix also merit attention. In all districts but San Francisco Unified, the percentage share of Hispanics increased more than that of any other group between 1997-1998 and 1999-2000. All districts experienced similar trends over the two-year span shown in the table. Statewide and in each of the five districts, the share of whites has declined; in two districts—Fresno and Long Beach Unified—the share of Asians has also declined 1
24
Embed
Web Appendix C Additional Information about Student ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Determinants of Student Achievement: New Evidence from San Diego
Web Appendix C Additional Information about Student Demographics, School Resources and
Test Scores in San Diego and Other Districts
This appendix supplements the data on student demographics and school resources
presented in Chapter 2. In addition, the appendix adds to the analysis of test scores in Chapter 2.
There, we summarized patterns by taking simple averages of test scores across grades. The
tables presented in this appendix show grade-by-grade achievement in the five largest districts of
California as well as for the state as a whole.
Student Demographics
Table C.1 shows the ethnic mix of the five largest districts and the entire state in 1997-
1998 and 1999-2000. No district exactly matches the ethnic and racial mix of students in the
state as a whole. San Diego has significantly greater percentages of African American and
Filipino students than the state, a slightly smaller percentage of Latino students, and in 1999-
2000 roughly 9 percent fewer white students. Three of the other four largest districts have a
greater share of Hispanic students and a smaller percentage of African American students, and
all four have a smaller percentage of white students. In spite of these differences, the five largest
districts have one thing in common: a far smaller share of white students than is found in the
state as a whole.
Recent trends in ethnic mix also merit attention. In all districts but San Francisco
Unified, the percentage share of Hispanics increased more than that of any other group between
1997-1998 and 1999-2000. All districts experienced similar trends over the two-year span
shown in the table. Statewide and in each of the five districts, the share of whites has declined;
in two districts—Fresno and Long Beach Unified—the share of Asians has also declined
1
Determinants of Student Achievement: New Evidence from San Diego
noticeably. Overall, trends in SDUSD seem to match state trends more closely than is the case in
the other districts.
Table C.2 examines another measure of diversity: the percentage of students by English
Learner (EL) status—formerly known as Limited English Proficient or LEP—in SDUSD and the
state as a whole. The table shows that in 1999-2000, SDUSD enrolled a larger share of EL
students than the state as a whole.
A related category of students, Fluent English Proficient (FEP), consists of students who
were formerly EL students but who have been redesignated by their district as fluent in English.
Table C.2 shows the percentage of students who are EL, the percentage who are FEP, and the
percentage of EL students who are redesignated as FEP each year for the five largest districts and
California as a whole. All of the five largest districts have significantly greater shares of
students who are EL and FEP than occur statewide. Among the districts, SDUSD’s share of
students who are EL matches the state average most closely.
Some analysts have used the proportion of EL students who are redesignated as FEP
students as a measure of the success of schools’ programs to teach English to English Learners.
In both years, SDUSD’s redesignation lags behind the state average and the rate in the other four
largest districts except Fresno. However, it would be a mistake to judge the districts narrowly on
this measure, given that districts vary in the conditions under which they allow students to be
redesignated Fluent English Proficient.1
Table C.2 also shows the percentage of students eligible for free or reduced-price meals.
This percentage is a widely used indicator of poverty in school populations. The figures show
1 For instance, a report released in spring 2002 summarizes the results of the state’s new California English Language Development Test (CELDT), now given to English Learners throughout the state. Results indicated that many students were fluent in English on the basis of CELDT results but were still designated EL by their districts, with some fairly sharp variations across districts. See Tully, Tapia, and Sacchetti (2002).
2
Determinants of Student Achievement: New Evidence from San Diego
that in these districts a large percentage of students live in disadvantaged circumstances. With
the notable exception of San Francisco Unified, the districts have much higher proportions of
students eligible for assistance than occurs in the state as a whole.
School Resources
Our next set of tables illustrates variations in key school resources among the five largest
districts and the state as a whole. Table C.3 shows the pupil-teacher ratio. Between 1997-1998
and 1999-2000 all districts have matched patterns at the state level fairly closely. The main
exception is San Francisco Unified School District (SFUSD), which in both years had a
significantly lower pupil-teacher ratio than elsewhere. SFUSD is also the only district among the
five to buck the trend toward lower pupil-teacher ratios over the period. SDUSD mimics state
patterns quite closely but in both years has a slightly lower pupil-teacher ratio than observed
statewide.
The next few tables highlight various measures of teacher characteristics. Table C.4
reveals that the five districts differ considerably from each other and from the state as a whole in
the share of teachers holding a master’s degree or higher. SDUSD has a considerably higher
percentage of teachers holding a master’s degree than the state or the other four largest districts.
However, the large variations among districts may not be as large as meets the eye. As noted in
the table, some of the other districts have large proportions of teachers with some education
beyond a bachelor’s degree but less than a master’s degree. Many of these teachers may be just
a course or two away from completing their master’s degree. Using “more than a bachelor’s
degree” as the measure of whether a teacher is highly educated makes SDUSD’s ranking fall
from first to fourth. Overall, then, it seems that teachers in San Diego Unified have somewhat
3
Determinants of Student Achievement: New Evidence from San Diego
but not markedly higher levels of education than seen in other districts and the state, with more
teachers holding master’s but fewer holding “at least some postgraduate education”.
Table C.5 shows various measures of teacher experience. Each of the largest five
districts employs teachers whose experience matches the state average reasonably closely.
SDUSD had the least experienced teachers among the five large districts. For instance in 1999-
2000, average years of teaching experience in SDUSD was 11.1, compared to 12.7 statewide.
Moreover, 25 percent of San Diego’s teachers were in their first or second years of teaching in
1999-2000, compared to only 16percent in California as a whole. Los Angeles Unified School
District (LAUSD) most closely resembles SDUSD in this regard, although even there only 20
percent of teachers were in their first or second year of teaching in 1999-2000.
Table C.6 shows the percentage of teachers holding a full credential. Comparisons across
districts based on this table should be viewed with caution: Our estimates of teacher credentials
for SDUSD come from our own compilations using district data because in the CBEDS data San
Diego is missing teacher credentials for a large number of teachers in 1997-1998. At any rate,
the overall patterns seem clear: SDUSD has a far larger percentage of teachers holding
credentials than is observed in the state as a whole. Fresno and San Francisco Unified School
Districts have roughly comparable percentages of credentialed teachers.
Test Scores
Beginning in spring 1998, California initiated a new state test, the Stanford 9, which has
been given annually to all students in grades 2 to 11 since 1998. The Stanford 9 is a standardized
test that has been normed using a national sample of students. This provides a national
performance yardstick against which California’s students can be compared. Throughout the
report we focus on math and reading scores on the Stanford 9. Our reason is simple: Although
4
Determinants of Student Achievement: New Evidence from San Diego
the Stanford 9 includes additional subject areas in certain grades, the math and reading tests
stand out as representing the very core of educational achievement.
Table 2.1 showed that San Diego students’ scores have matched the scores of students
statewide quite closely. But this table did not disaggregate overall scores into the scores of EL
and non-EL students. Much of the gap in reading performance between SDUSD students and the
national norming sample reflects the fact that many students in the district are English Learners.
Tables C.7 and C.8 show the performance of the five largest districts and California as a whole
for English Learners and non-English Learners, respectively. Table C.7 shows that in all
districts, EL students lag considerably behind national norms (based on a national sample in
which only 2 percent of test-takers were EL). In both reading and math, SDUSD’s EL students’
initial performance in 1997-1998 closely reflected that of EL students statewide. Over the two-
year period, gains in performance of SDUSD’s EL students in reading slightly exceeded the
statewide gains, while gains in math slightly lagged the statewide gains. Gains relative to those
for EL students in the other districts were in the middle of the distribution.
Table C.8 shows corresponding data for students who were not English Learners. This
group includes both native English speakers as well as students whose native tongue is not
English but who have been identified by districts as Fluent English Proficient. Just as Betts,
Rueben, and Danenberg (2000) found statewide, when we examine this group, the performance
of California students approximates national norms. In San Diego, non-EL students
outperformed national norms in reading and math even in the inaugural year of the test. (That is,
more than 25, 50, and 75 percent of students met or exceeded the 75th, 50th, and 25th percentiles
of national norms in 1997-1998, respectively.) By 1999-2000, this better-than-the-national-norm
performance still held true in SDUSD and had also become true statewide, as well as in some but
5
Determinants of Student Achievement: New Evidence from San Diego
not all of the other districts. Over the two-year period, gains in reading in SDUSD exceeded
those in the state and the other districts, while gains in math almost exactly matched those
statewide.
The analysis we have presented here and in Chapter 2 summarizes test scores by district
by taking an average of outcomes across grades 2 through 11 in each district. Some readers will
be interested in a more detailed analysis by individual grades. We provide analogues to Tables
2.1, C.7, and C.8 by grade level in Tables C.9 to C.14. Among the many interesting patterns
shown in these latter tables, SDUSD improvements in reading achievement have generally
outpaced statewide improvements, but have done so most noticeably in the lower grades. For
instance, Table C.9 shows that the percentage of SDUSD students in grade 2 scoring at or above
the 25th percentile rose from 64 to 78 percent between 1997-1998 and 1999-2000, compared to
an increase from 61 to 71 percent statewide. A second interesting finding from Tables C.9 and
C.12 is that by 1999-2000, SDUSD students in the lower grades performed the best against
national norms in reading, but for math it was the lower and higher grades where SDUSD
student performance appeared the strongest against national norms.
Conclusion: The Overall Degree of Similarity between SDUSD, Other Large Districts, and the State’s School System
Overall, this appendix has revealed both dissimilarities and similarities between the five
largest districts and between these districts and the state educational system as a whole.
SDUSD’s student population matches the statewide racial mix quite well but has more African
American and Filipino students and fewer white students. Similarly, SDUSD students are more
likely to be eligible for meal assistance and to be English Learners than the average student
statewide. We consider that these differences from state averages, if anything, enhance the
policy relevance of our research: Betts, Rueben, and Danenberg (2000) and others have
6
Determinants of Student Achievement: New Evidence from San Diego
convincingly shown that in California, low student achievement is very much an issue of
language, poverty, and, to some extent, race. SDUSD provides an almost ideal setting in which
to explore the reasons why disadvantaged students and those who are English Learners lag
behind, and whether certain types of school resources are particularly conducive to increasing
achievement among such students.
Turning to school resources, SDUSD appears to have roughly the same mix of resources
seen in other districts statewide. Perhaps the most striking difference is that teachers in SDUSD
are much more likely to hold a master’s degree than teachers in the other four largest districts or
those in the state as a whole. On the other hand, other districts have far larger percentages than
SDUSD of teachers who hold some post-baccalaureate education, so it is important not to
overstate the gap. In a similar vein, SDUSD’s teachers are more likely to hold a full credential
than in the comparison districts. Finally, counterbalancing the relatively high numbers of
teachers with master’s and full credentials in SDUSD, SDUSD’s teachers have less teaching
experience than do teachers elsewhere. Overall, the pattern of teacher qualifications in the
district is reasonably close to state averages.
Finally, our comparisons of student achievement confirmed that on the whole SDUSD
provides a representative microcosm of California as a whole. Indeed, the distribution of test
scores in SDUSD matches the California distribution more closely than does that of any other of
the five largest districts. In all of the districts we examined, student performance initially lagged
behind student performance observed nationally but caught up partly or fully over the period of
study. Further, much of the achievement gap between students in SDUSD and the United States
reflects the disproportionate number of English Learners in SDUSD. This pattern mimics the
California/United States comparison very closely. As for gains in achievement over time,
7
Determinants of Student Achievement: New Evidence from San Diego
SDUSD’s students appear to have gained in reading achievement somewhat more quickly than
students statewide, while math gains mirror statewide gains very closely.
Overall, San Diego appears to provide a district that is quite representative of patterns
and trends statewide. Perhaps the biggest difference between SDUSD and the state’s school
system is that SDUSD has relatively more students who are English Learners and more students
who are economically disadvantaged. As we have argued, this difference makes San Diego all
the more important to examine.
8
Determinants of Student Achievement: New Evidence from San Diego
Table C.1
Student Demographics: Total Enrollment and Breakdown by Race/Ethnicity
District 1997-1998 1999-2000 Change Between 1997-
1998 and 1999-2000
STATE TOTAL
Enrollment 5,727,303 5,951,612 224,309
African American not Hispanic 501,303 8.8% 509,637 8.6% 8,334 -0.2
American Indian or Alaska Native 49328 0.9% 50750 0.9% 1422 0.0
Asian 466,399 8.1% 479,073 8.0% 12,674 -0.1
Filipino 137,126 2.4% 141,045 2.4% 3,919 0.0
Hispanic or Latino 2,319,072 40.5% 2,513,453 42.2% 194,381 1.7
The percent of students who were redesignated FEP is calculated by dividing the number of redesignated students by the prior year's count of English Learners. The percents reported for English Learners and FEP Students use the current year's total enrollment as the denominator.
** The data for this measure are based on an "unofficial" enrollment count that includes public and
nonpublic school data. For further information about the free/reduced meal data, see http://data1.cde.ca.gov/dataquest/gls_calworks.htm.
11
Determinants of Student Achievement: New Evidence from San Diego
Table C.3
Pupil-Teacher Ratios
District 1997-1998 1999-2000
Change Between 1997-1998 and
1999-2000 STATE TOTAL 21.6 20.9 -0.7 SAN DIEGO UNIFIED 20.5 19.1 -1.4 FRESNO UNIFIED 21.1 20.5 -0.6 LONG BEACH UNIFIED 23.9 22.3 -1.6 LOS ANGELES UNIFIED 22.0 20.9 -1.1 SAN FRANCISCO UNIFIED 17.2 18.4 1.2
12
Determinants of Student Achievement: New Evidence from San Diego
Table C.4
Percentage Distribution of Teachers by Education Level and Total Number of Teachers
1997-1998 District Master's
Degree or Higher
Less than a Master's Degree
No Education Level Reported
Total Number of Teachers
STATE TOTAL 31.2 68.3 0.5 272,459 SAN DIEGO UNIFIED 57.3 42.7 0.0 6,836 FRESNO UNIFIED 10.3 89.7 0.0 3,776 LONG BEACH UNIFIED 31.3 68.7 0.0 3,622 LOS ANGELES UNIFIED 26.1 71.1 2.9 31,513 SAN FRANCISCO UNIFIED 23.0 77.0 0.0 3,620 1999-2000 District Master's
Degree or Higher
Less than a Master's Degree
No Education Level Reported
Total Number of Teachers
STATE TOTAL 30.4 69.3 0.4 292,012 SAN DIEGO CITY UNIFIED 53.2 46.8 0.0 7,395 FRESNO UNIFIED 10.7 89.3 0.0 3,929 LONG BEACH UNIFIED 29.3 70.7 0.0 4,128 LOS ANGELES UNIFIED 25.0 73.6 1.4 34,652 SAN FRANCISCO UNIFIED 24.4 64.0 11.6 3,340 Continued on next page
13
Determinants of Student Achievement: New Evidence from San Diego
Table C.4 (continued)
Difference District Master's
Degree or Higher
Less than a Master's Degree
No Education Level Reported
Total Number of Teachers
STATE TOTAL -0.8 0.9 -0.1 19,553 7.2% SAN DIEGO UNIFIED -4.1 4.1 0.0 559 8.2% FRESNO UNIFIED 0.4 -0.4 0.0 153 4.1% LONG BEACH UNIFIED -2.1 2.1 0.0 506 14.0% LOS ANGELES UNIFIED -1.1 2.5 -1.5 3,139 10.0% SAN FRANCISCO UNIFIED 1.5 -13.0 11.6 -280 -7.7% Note: The "Master's Degree or Higher" category was created by aggregating Doctorate, Master's Degree +30, and Master's Degree categories, while the "Less than a Master's Degree" category was created by aggregating Bachelor's Degree+30, Bachelor's Degree, and Less Than Bachelor's Degree categories. The aggregation categories reported here were chosen to minimize the problems of apparently inconsistent data reporting of education categories over time and across districts. For example, if a group of teachers were reported as Master's Degree +30 in 1997-1998 and then reported as Master's Degree in 1999-2000, these teachers appear in the same aggregate "Master's Degree or Higher" category in both years. It should be noted, however, that if the aggregation is done by separating into "Higher than Bachelor's" and "Bachelor's or Less" then the results are very different. If this alternate aggregation pattern were used, Fresno would have the highest percent of teachers in the higher education category for both years.
14
Determinants of Student Achievement: New Evidence from San Diego
Table C.5
Teacher Experience
District 1997-1998 1999-2000 Change between 1997-1998 and 1999-2000
STATE TOTAL Average Years Teaching 13.2 12.7 -0.5 Average Years in the District 10.7 10.3 -0.4 % of Teachers in their First or Second Year
17.2% 15.7% -1.5
SAN DIEGO UNIFIED Average Years Teaching 11.1 11.1 0.0 Average Years in the District 11.0 10.8 -0.2 % of Teachers in their First or
Second Year 20.1% 25.3% 5.2
FRESNO UNIFIED Average Years Teaching 13.4 13.9 0.5 Average Years in the District 11.3 11.8 0.5 % of Teachers in their First or
Second Year 14.2% 9.6% -4.6
LONG BEACH UNIFIED Average Years Teaching 12.0 11.5 -0.5 Average Years in the District 9.7 9.2 -0.5 % of Teachers in their First or
Second Year 20.2% 18.7% -1.5
LOS ANGELES UNIFIED Average Years Teaching 12.3 11.6 -0.7 Average Years in the District 11.0 10.5 -0.5 % of Teachers in their First or
Second Year 16.9% 20.4% 3.5
SAN FRANCISCO UNIFIED Average Years Teaching 14.8 12.0 -2.8 Average Years in the District 13.7 11.9 -1.8 % of Teachers in their First or
Second Year 2.5% 18.4% 15.9
15
Determinants of Student Achievement: New Evidence from San Diego
Table C.6
Percent of Teachers with Full Credentials
District 1997-1998 1999-2000
Change Between 1997-1998 and
1999-2000 STATE TOTAL 87.3 86.1 -1.2 SAN DIEGO UNIFIED 94.9 94.3 -0.6 FRESNO UNIFIED 89.9 93.9 4.0 LONG BEACH UNIFIED 78.0 72.5 -5.5 LOS ANGELES UNIFIED 74.9 71.5 -3.4 SAN FRANCISCO UNIFIED 94.0 82.6 -11.4
* Note: The 1997-1998 CBEDS data for San Diego is missing credential data for 28% of the sample of teachers. As a result, CBEDS reported that only 71.1% of San Diego teachers were fully credentialed that year. (For the other four districts reported here, as well as for the state total, CBEDS has valid data on credentials for at least 95% of teachers.) Using data from the San Diego Unified School District, we have calculated that 97.2% of teachers were fully credentialed in 1997-1998, and 96.6% were fully credentialed in 1999-2000. For 1999-2000 we report the CBEDS figure. For 1997-2000 we report a figure calculated from our own data, which has been normalized by subtracting the difference between our Figure and the CBEDS figure for 1999-2000.
16
Determinants of Student Achievement: New Evidence from San Diego
Table C.7
SAT9 Test Score Distribution: EL Students
Reading Math Unweighted Average Across Grades Unweighted Average Across Grades
Change between Change between 1997-1998 1999-2000 97-98 and 99-00 1997-1998 1999-2000 97-98 and 99-00
Determinants of Student Achievement: New Evidence from San Diego
Table C.8
SAT9 Test Score Distribution: Non-EL Students
Reading Math Unweighted Average Across Grades Unweighted Average Across Grades Change between Change between 1997-1998 1999-2000 97-98 and 99-00 1997-1998 1999-2000 97-98 and 99-00