Education in States and Nations: 1991
(ESN) Indicator 10: Note on secondary education enrollment
Notes on Figure and Tables
Czechoslovakia
Specialized secondary education is included in general secondary education because the students in vocational and technical schools also receive general education.
Denmark
The students in secondary and higher education are generally much older than in Central and Southern European countries and in North America. The enrollment rates for higher education, for example, are higher at the age of 25 than at the age of 20.
France
Participation rates do not include special education, which is classified as ungraded.
Ireland
The data have been influenced by the use of population figures for 16, 17 and 18 year-olds, which were based on estimates supplied to OECD prior to the publication of the final results of the Census of Population for 1991. The revised population totals for these age groups indicate rates of participation in secondary education (males and females) of 88.6, 67.5 and 30.1 percent for 16-, 17- and 18-year-olds, respectively, by comparison with the 85.1, 64.7 and 28.8 percent one finds in the text.
Japan
Students enrolled in upper secondary courses offered by the Special Training Colleges (about 100,000 students in the relevant age group) are not included.
Spain
Participation rates do not take into account either the students enrolled in special education (classified as ungraded), or children whose age is unknown (about 1.3 percent of full-time enrolled students).
Switzerland
Participation rates do not take into account special education at the lower secondary level (4.7 percent of 14-year-olds, 4.0 percent of 15-year-olds and 2.1 percent of 16-year-olds).
Technical Notes
Calculation of full-time equivalent enrollments
See technical notes for Indicator 8.
Contrast between Census- and Current population survey-derived estimates of secondary education enrollment rates
There exist two inconsistencies between the Census Bureau's Current Population Survey (CPS) estimates for secondary education enrollment rates (which provide the U.S. figure in the OECD's Education at a Glance) and the 1990 Census of Population estimates (which provide the basis for the state figures used here). The first derives from the differing administration times of the two surveys - the CPS on or about October 1 and the Census on or about April 1. The second derives from differences in the wording of the pertinent questions that seem to elicit differing responses regarding enrollment status.
The effect of the differing administration times is to raise the April Census enrollment rates above the CPS rates in the older age cohorts (16 and above), when some teens start to drop out of school. This is because many students who started in a secondary grade in the fall semester when they were age x, turn age x+1 before the April Census.
For example, at age 15, enrollment is still virtually universal. Thus, the October CPS shows an enrollment rate for 15-year-olds close to 100 percent. Likewise, the April Census shows enrollment rates in the mid-to-upper nineties for 15-year-olds. But, the enrollment rates drop off for 16-year-olds, to a greater degree in the CPS numbers than in the Census numbers. That's because about half of the 16-year-olds in the Census figures are, in all likelihood, a grade level below about half of the 16-year-olds in the CPS estimate. Students in the same age cohort, then, are, on average, at a lower grade level in the Census than they are in the CPS.
For comparison purposes, the October 1990 CPS secondary education enrollment rates and the standard errors are presented in Tables S13 and S14 below for the 10 largest states.
Table S13 Enrollment in public and private secondary education, by age and state: October 1990
-----------------------------------------------------------------------------
Age
State 14 15 16 17 18 19 20 21
-----------------------------------------------------------------------------
California 100.0 99.0 95.8 79.6 22.6 5.2 2.0 2.3
New York 99.3 97.3 96.5 85.1 34.5 9.1 1.4 0.9
Texas 99.1 98.1 100.0 89.1 37.7 8.3 1.4 2.1
Florida 100.0 96.9 95.9 84.8 23.7 8.1 0.0 3.1
Pennsylvania 98.6 96.6 99.0 90.6 19.3 2.2 1.2 1.4
Illinois 100.0 98.4 96.4 80.8 19.9 9.4 0.0 2.6
Ohio 98.8 97.5 96.9 93.1 19.5 5.3 0.0 0.0
Michigan 100.0 96.1 93.7 83.5 30.6 4.9 3.8 0.9
New Jersey 100.0 97.6 95.8 83.4 27.6 1.9 1.2 0.0
North Carolina 100.0 99.0 94.2 88.0 16.8 5.1 0.0 0.0
UNITED STATES 99.4 98.3 95.3 84.6 23.6 6.0 1.3 1.1
-----------------------------------------------------------------------------
SOURCE: U.S. Department of Commerce, Bureau of the Census, Current Population Survey, October, 1990.
Table S14 Standard errors for estimates in Table S9
-----------------------------------------------------------------------------
Age
State 14 15 16 17 18 19 20 21
-----------------------------------------------------------------------------
California 0.0 0.8 1.7 3.2 3.4 1.8 1.0 1.1
New York 0.8 1.8 1.9 3.9 5.0 2.8 1.2 0.9
Texas 0.9 1.5 0.0 3.4 5.2 2.7 1.1 1.4
Florida 0.0 2.1 2.5 5.1 5.4 3.2 0.0 2.4
Pennsylvania 1.7 2.3 1.3 4.1 4.6 1.7 1.4 1.5
Illinois 0.0 1.8 2.5 4.5 4.9 4.1 0.0 2.0
Ohio 1.4 1.8 2.2 3.4 5.1 2.6 0.0 0.0
Michigan 0.0 3.0 3.6 4.9 6.3 2.9 2.5 1.3
New Jersey 0.0 2.4 3.2 6.2 7.1 2.0 1.7 0.0
North Carolina 0.0 1.7 4.2 4.9 5.9 3.5 0.0 0.0
UNITED STATES 0.0 0.0 0.0 0.0 1.1 0.6 0.3 0.3
------------------------------------------------------------------------------
SOURCE: U.S. Department of Commerce, Bureau of the Census, Current Population Survey, October, 1990.
Adjustments have been made to the 1990 Census-derived enrollment rates at each age in the 50 states and the District of Columbia as follows:
- 1.) A regression equation was designed with the following form:
Y = a + b*X
Y = CPS-derived enrollment rate (October 1990)
X = Census-derived enrollment rate (April 1990)
a = constant (arithmetic adjustment)
b = slope of the relationship (proportional adjustment) - 2.) The regression at each age was run on the data from the 19 states with CPS estimates derived from samples large enough to give reliable estimates. The regression run produced estimates for "a" - constant in the equation and the arithmetic adjustment factor between the Census and CPS estimates - and for "b" - the slope of the relationship and the proportional adjustment factor between the Census and CPS estimates. The eight regression runs produced the output shown in Table S15.
Table S15 Output from regression of October 1990 CPS secondary enrollment rates onto April 1990 Census secondary enrollment rates
---------------------------------------------------------------
Constant Standard Slope Standard
Age (a) error (b) error
---------------------------------------------------------------
14 199.424 0.575 -1.036 0.462
15 19.580 1.407 0.826 0.534
16 111.913 6.359 -0.188 1.120
17 55.280 7.374 0.372 0.415
18 13.477 7.146 0.283 0.369
19 2.768 3.658 0.446 0.766
20 0.038 1.217 0.424 0.487
21 -0.595 1.102 1.211 0.544
---------------------------------------------------------------
The two factors - "a" and "b" - were then used to adjust the Census estimates for enrollments in all 50 states and the District of Columbia. These adjusted enrollments are listed in Table 10a for the United States, in Table 10b for the states, and displayed in Figure 10.
Enrollment in preprimary education
Entry ratio to higher education...