Chapter 5: Education
5.1 Third Grade Academic Proficiency
Definition
Third grade academic
proficiency is measured by the scaled scores of third grade students in
language arts and math on the New Hampshire Education Improvement and
Assessment program (NHEIAP).[1]
The scaled scores published by the Department of Education are "the mean
scaled scores of all the students for the school/district at that grade
level." Scores between 200-239 are ranked as novice achievement, scores
between 240-259 are ranked as a basic achievement, scores between 260-279 are
deemed proficient, and scores between 280-300 reflect advanced achievement.
Findings
In 1999, statewide results for
third graders reflected small gains over 1998. The state average scaled score
in language arts improved from 248 to 251 (both scores in the basic range);
math scores improved from 256 to 257 (also in the basic range).
The scores of some schools
increased and others decreased. More schools improved their scaled score in
language arts than improved in mathematics. Improvement was not particularly
related to how well a school's students did in 1998. The lower performing
schools were just as likely to increase or decrease their scores as the higher
performing schools.[2]
In contrast to the lack of
correlation between improvement in scores and past performance, a strong
correlation does exist between average scaled scores and community
economics. Figure 5-1 shows the scaled
scores of schools grouped by percentage of students in each school receiving
free or reduced-price school lunches.[3] Schools serving a higher percentage of
lower-income families tend to have lower average scores.
Figure
5-1[4]
Of the 60 elementary schools with fewer than 10% of their students receiving free or reduced price meals,
only 10 had language arts scaled scores below the state average and only 11 had
below average scores in math. Of the 35 schools with more than 40% of their students receiving free or reduced price meals,
only 7 had a language arts score that was above average, and only 8 had above
average math scores.[5]
State averages do not, however, tell the whole story.
Figures 5-2 and 5-3 are scatter charts that display each of 245 elementary
schools for which both 1999 NHEIAP third grade scaled scores and 1997 school
lunch data were available. These
charts show the large variation of scaled scores among schools serving similar
percentages of low-income households. There are some schools with more than 40%
of their student populations eligible for free and reduced-price meals that
have higher average scores than other schools with fewer than 10% of their
students so eligible.[6]
Figure
5-2
Figure
5-3
Significance
Community economics are closely tied to, but not the sole
predictor of, academic proficiency scores.
Assessment of schools should take into account a number of issues,
including:
v how
an individual school compares to the state average;
v how
results in an individual school change over time;
v how
a school compares to a sub-group of schools serving reasonably comparable
populations.
Responsibility for such analysis and reporting lies with the
New Hampshire Department of Education. Recent legislation requires the
Department to produce a "report on the condition of education statewide
and on a district-by-district and school-by-school basis." Analysis such
as that begun here should be an important component of such a report.
Additionally, the reasons for different results among schools serving similar
populations should be investigated. Program, staffing, and administrative
differences should be recorded and investigated to determine which, if any,
school-based practices lead to higher student achievement.[7] The lessons
learned from the higher performing schools may well shed light on the resources
needed to provide an adequate education.
5.2 Reading Ability of Fourth Graders
Definition
Reading ability of
fourth graders measures the percentage of fourth grade
students able to read and understand passages at defined proficiency levels on
the National Assessment of Educational Progress (NAEP) standardized tests.[8]
Findings
As part of the NAEP, a sample of approximately 2,000 New
Hampshire fourth graders took tests of their reading ability in 1998, 1994, and
1992.[9]
Students are scored in a manner similar to that of New Hampshire's own NHEIAP
testing program; each student is designated as being in the Below Basic,[10]
Basic,[11]
Proficient,[12] and
Advanced[13]
category. A single scaled score is also calculated for each student and
averages are derived from the scaled scores.
In comparison to the other states, New Hampshire's fourth
grade readers do very well. Scores from the 1998 fourth grade test (NAEP), rank
New Hampshire second in the nation, behind Connecticut, and in line with
neighbors Maine and Massachusetts (Vermont schools did not participate).
Thirty-eight percent of the sample of New Hampshire fourth graders scored at or
above the proficient level in 1998, 1994, and 1992. The New Hampshire results are displayed in figure 5-4 and
compared to the national average.
Figure
5-4
The NAEP test provides data on the impact of family, school,
and peers on school performance.
With respect to
families, the NAEP data reveals that (1) income, (2) education, (3)
mobility, (4) frequency of family discussions about school studies, and (5) the
literary climate of the home all affect school performance.
Family income is clearly related to average student reading
performance. Students from families with income low enough to qualify them for
free and reduced-price school meals have lower average scaled scores than the
students from higher-income families as shown in Figure 5-5. Of note is that, in contrast to our strong
national and regional showing, New Hampshire students with higher incomes
scored lower than similarly situated students in the Northeast.
Figure
5-5
The level of education of the parent(s) of fourth graders is
also strongly related to their average score.
As shown in figure 5-6, the more educated the parents, on average, the better
the students are able to read. In all categories, New Hampshire students do
better than their national counterparts. This difference was most pronounced,
however, at the higher education levels.
Figure
5-6
Family mobility is another factor that impacts school
performance as measured by the NAEP. Students who have changed schools more
often have lower average reading scores than those who have had a more stable
school environment. Changes in schools from second through fourth grade are
almost entirely attributable to families who move. Students in the New
Hampshire sample who did not change schools had an average scaled reading score
that was 28 points higher than those who changed schools three or more times,
as seen in Figure 5-7.
Figure
5-7
Students who discuss their
schoolwork at home more often are more likely to have better reading skills
than those who do so less frequently. Nationally, 18% of fourth graders say
they never discuss their studies at home. In New Hampshire, only 12% are in
this situation.
Finally, the literacy climate of the home plays a role in
how well students perform on the test.
The students in the New Hampshire sample who have an encyclopedia at
home had an average score of 228, while those who do not had an average score
of 215. Similarly, students who
receive magazines regularly at their homes scored an average 230, whereas
students who do not scored 212.
Nationally and statewide, students who choose to read for fun outside of
school perform better than those who do not, with those New Hampshire students
who do so daily scoring 19 points higher than those who never do so.
Figure
5-8
School factors
that impact NAEP performance include the amount of reading assigned and the
length of writing assignments.
Figure 5-9 shows that those students who read ten or fewer
pages a day in school and for homework do not fare as well as those who read
eleven or more. It also indicates that there is little difference once students
reach the eleven-page level. New Hampshire students do better than the national
average at all levels of reading. In any given demographic group, however,
students who are assigned little reading do not do as well as those who are
required to read more than ten pages a day in school and for homework.
Figure 5-9
The regularity with which students are asked to write long
answers to questions about their reading is also related to reading
achievement, as seen in Figure 5-10.
Figure 5-10
There are other school-based factors that do not appear to be consistently related to
differences in reading performance. Among these are: how often students do
group projects related to their reading, whether reading is aloud or silent,
and whether the library is used for doing school assignments, for borrowing
books, for using computers, or as a place of quiet study.
Finally, there are non
school-based and non-family based factors that relate to reading
achievement, such as the attitude of peers.
The attitude toward school by friends is strongly related to the
student’s reading ability. Those fourth graders who report that their friends
make fun of those who try to do well in school have much lower average scaled
reading scores than those whose friends are more positive in offering support
and reinforcement. In the New Hampshire sample, 17% of the students agree or
strongly agree with the statement, "My friends make fun of people who try
to do well in school."
Figure 5-11
Significance
This detailed analysis of the reading ability of New
Hampshire fourth graders shows that:
§
Schools can make a difference in fourth graders'
average reading abilities by providing adequate opportunities to read (eleven
or more pages a day) and writing at length about their reading (once or twice a
month).
§
Factors external to the school are also critically
important. The home environment (parents' education, family income, access to
magazines and books, regularity of discussing schoolwork at home, consistency
of residence) is directly related to fourth graders' reading abilities.
§
The positive or negative attitudes of close friends and
peers are also strongly associated with reading ability.
Raising the level of reading competency is achieved by
designing programs that reflect the interrelationships between school, home,
and peers.
5.3
Tenth Grade Academic
Proficiency
Definition
Tenth grade academic
proficiency is measured by the scaled scores of tenth grade students on the
New Hampshire Education Improvement and Assessment program (NHEIAP).[14] NHEIAP tests for tenth graders assess
language arts, mathematics, science, and social studies.[15] The scaled scores published by the
Department of Education are "the mean scaled scores of all the students
for the school/district at that grade level." Scores between 200-239 are
ranked as novice achievement, scores between 240-259 as a basic achievement,
scores between 260-279 as proficient, and scores between 280-300 as advanced
achievement.
Findings
Tenth grade results showed
small gains in 1999 over 1998. The state average scaled score in language arts,
mathematics, and social studies improved by 2 points each, while science
remained the same. The scores of some schools increased and others decreased.
Figure 5-12 is a scatter plot
that displays each high school's language arts scores as a single dot. Dots
above the diagonal line are schools with higher scaled scores in 1999 than
1998. Dots below the line are schools where the scaled score declined. A few
schools with large increases or decreases are labeled.
Figure
5-12
The range of scaled scores among schools is quite
small. For example, the range in
language arts scores from lowest scoring to highest scoring high school was
only 18 points in 1998 and 15 points in 1999.
There is, however, a large difference among the numbers of
students achieving different levels of proficiency among the high schools.
Figure 5-13 illustrates the percentage of tenth grade students in each school
in 1999 who reached each of the proficiency levels in language arts. Students
who were not tested or who were only in the "novice" category fall to
the left of the baseline, while students scoring, basic, proficient, and
advanced fall sequentially to the right of the baseline. The schools are
ordered from high to low on the basis of the percentage of students scoring
basic and above.
Figure 5-13
There appears to be some
relationship between high schools and economic cluster and test results.[16] Figure 5-14 charts that relationship.[17]
Figure 5-14
Significance
That there is some connection between community economics
and high school clusters and the NHEIAP results is demonstrated by the fact
that students attending high schools in the "wealthier" clusters are
slightly more likely to fall in the basic and proficient categories than in the
novice category when compared to the students in the "poorer"
clusters. And seven of the ten schools
that had the lowest percentage of students scoring in basic and above, were in
the poorest economic cluster.
However, some high schools serving
some our poorest communities scored in the top 20.[18] This raises the question of what programs
and approaches those schools have that may be contributing to the academic
competency of their students. As with
the third grade scores, answers to these questions would help the state
determine the resources needed to provide an adequate education.
5.4 High School Dropout Rate
Definition
High school dropout
rate is the percentage of students registered in grades 9-12 in a given
school who withdraw during a given school year.[19]
Findings
In the 1997-98 school year, 2,676 of the 56,878 students
enrolled in 76 of New Hampshire’s 78 public high schools withdrew without
completing their high school education.[20]
This is an average dropout rate of 4.7%, slightly less than the 4.9% of two
years before. The rate was highest in Newport, where more than 10% of the
students dropped out during the year, and lowest in Wilton-Lyndeboro, Durham,
and Hanover, where less than 1% of the students dropped out during the year.[21]
The percentage of youth that choose to leave high school
early is related to the economic context of the communities in which they live.
Figure 5-15 displays the single-year dropout rate for each of the five clusters
of high schools. Youth in the two wealthier clusters have greater success in
completing their high school education than those in the middle and poorer
clusters. The average dropout rate of schools in the poorest cluster is twice
that of communities in the wealthiest cluster.
Figure
5-15
Within each cluster there are wide differences among
schools. Figure 5-16 illustrates this, each dot representing a high school’s
dropout rate within its respective economic cluster. The schools with the
highest and lowest dropout rates in each cluster are labeled.
Figure
5-16
As shown in Figure 5-17, there is not a strong or consistent
difference in dropouts among schools of different size. The dropout rates vary
little, with slightly higher than average rates in the smallest (under 750
registered students) and largest (1500 or more) schools.
Figure
5-17
Significance
Without a high school degree, individuals are at a great
disadvantage in our technology-driven economy, which requires a highly skilled
labor force. According to the U.S.
Department of Education, high school dropouts are more likely than high school
graduates to earn less money in low-end, dead-end jobs, receive public
assistance and, particularly if female, become single parents in their
teens. High school dropouts also
represent a disproportionate number of inmates in correctional facilities.[22]
There is a clear relationship between community economics
and high school dropout rates. Such a
relationship threatens to perpetuate the cycle of poverty -- as youth from
poorer communities who dropout wind up in low paying jobs. But there are signs of hope -- instances
where the dropout rate is not as clearly linked to community economics. We need to explore why dropout rates in
some of the wealthier schools are high and why dropout rates in some of the
poorer schools are low. Are there
programs or approaches in particular schools that are making the difference for
teens? The answers are critical to
ensuring that our schools continually increase the percentage of teens that
complete high school and therefore enter the adult world with a more promising
future.
Bullets
·
Teens from poorer communities are more likely to
dropout of school.
·
Early warnings signals for school dropouts include:
poor grades, inability to read at grade level, poor attendance, and behavioral
problems.[23]
·
Effective dropout prevention programs link the student
to broad community supports.[24]
5.5
Post-High School Plans
Definition
Post-high school plans
count and students who have graduated from high school and have indicated their
plans for post-secondary education or employment.
Findings
According to statistics published by the NH Department of
Education, 52% of New Hampshire's 1998 public high school graduates planned to
attend a four-year college. An
additional 16% planned to attend some other post-secondary education, including
two- and three- year programs. Thus, just over two-thirds of all recent public
school graduates planned to further their education in some way.[25]
Between the two periods, 1991-1995 and 1996-1998, there has
been an increase in the percentage of students planning to attend four-year
colleges after graduation. The state average increased 4%, from 48% to 52%. The
state average in students planning to attend some form of post-secondary
education also increased 4%, from 64% to 68%.
Some schools stand out as having greatly increased numbers of students
now planning to attend four-year colleges, up by 10% or more in some cases.[26] Other high schools have an increase by more
than 10%
in the number of graduates pursuing some form of
post-secondary education. [27]
The economics of the communities that a school serves are related
to how many students plan to go on to a four-year college, as shown in Figure
5-18. A nearly equal percentage of students from each economic cluster planned
to attend other post-secondary programs in 1996-1998. With respect to college plans, however, the percentages vary by
cluster, from an average of 58% planning to go on to a four-year college
program in the wealthiest cluster to an average of 43% and 45% in the two
poorest clusters.[28]
Figure
5-18
There are, however, differences between communities of
similar profiles. Figure 5-19 displays a dot for each school in each cluster
and labels the highest and lowest school for each cluster. It shows quite a few
high schools in the wealthiest cluster that have lower percentages of graduates
planning to obtain further education than some schools in the poorest cluster.
The range within each cluster is about 35%.
Figure
5-19
Significance
There is an encouraging, albeit small, increase in students
from all economic clusters planning to attend four-year colleges between
1991-1995 and 1996-1998. As with
national trends, wealthier communities in New Hampshire had a larger percentage
of students planning to attend a four-year college than poorer communities.
It is important to look at individual schools where the
percentage of graduates with plans for higher education has jumped
considerably. What is happening in the
communities and schools with higher ranks in each cluster that may not be
happening in those that rank low? While it would not be fair to compare the
situation at Stratford High School to that at Bow High School, it does seem
reasonable to make a comparison between Stratford and Lin-Wood High School
given their comparable socioeconomic characteristics.
Finally, we must continue to support programs that expand
access to post-secondary education and increase student aspirations for
continuing education. This, of course,
includes programs to help with the cost of college education, which is well
beyond the means of many low-income families.
5.6 What We
Would Like to Know About Our Children and Education
This
chapter on education provides more information on student performance than past
editions of KIDS COUNT New Hampshire. Given the current focus on defining an
adequate education and the grounding of that discussion in measurable outputs
such as test scores and dropout rates, the data are of heightened
importance. What follows are
suggestions for improvement -- suggestions that will necessitate additional
resources to ensure that improved data collection does not come at a cost to
other core functions of the Department of Education.
First, with respect to testing, the NHEIAP testing process
should collect more information, similar to that collected by the NEAP
national tests, so that characteristics of students who do well and schools
that do well could be closely analyzed by the Department of Education. The
Department of Education has informed us that, beginning in December of 2000,
school district profiles will compare NHEIAP results among schools serving
similar communities. This practice will
allow for a meaningful way for the public to compare statistical results and
will provide tools to help improve our public schools. With respect to the NAEP, New Hampshire
should participate in the national testing process and the Department of
Education should analyze and make available the results relative to factors
that are related to higher achievement among New Hampshire students in all
tested subjects at all grade levels.
Finally, with respect to the SAT, the Department of Education should
compile information on the numbers and percentages of students at each public
high school who take the SAT exams in anticipation of applying for college.
Such information, which the department indicates will be included in the school
profiles available in December of this year, should come from the national
database of the College Board.
Second, with respect to dropout
data, the Department is in the process of changing the way in which it collects
information on dropouts and graduates.
Such change is welcome, because the existing data are not accurate or
comparable from district to district. Information on each school dropout should
be submitted to the Department of Education, which should then publish annual
statistics and profiles of the dropout population. Good data here should lead
to better dropout prevention programs.
Third, with respect to special
education data, the department's SPEDIS database of special education
students is capable of producing much more useful information than we were able
to obtain. The Department should prepare regular quarterly and annual public
reports that provide statistics on counts and cost of special placements of
special education students. These should include counts by town as well as
district, counts of newly added students by grade level, and counts of cases by
cost category and by disability code.
Fourth, with respect to the cost per pupil for education, the data on school district budgets
currently collected on an annual basis by the Department of Education should be
broken down on a per pupil basis so that each district can see its expenditure
pattern in light of those of comparable districts. At present, significant
types of expenditures are excluded from the published per pupil expenditure
figures. The department should revise and expand its calculations comparing
school expenditures so that such questions as how much is being spent per pupil
for textbooks and building maintenance in each district can be answered. The
full costs of transportation, out-of-district tuition, and debt service should
be built into per pupil figures used to compare districts.
Finally, the Department of Education should explore
compiling information on local commitment
to education. Such information might include: teacher turnover rate, salary
scale indicators, volunteer time in schools, student participation in special
academic events, among others. Much of
this information requires data not actively collected by the Department of Education
from local school districts. The Department, together with major stakeholders
in public education, should develop such a measure as part of its "School
Report Card," required under RSA 193E.
Endnotes
[1] There were
16,946 third grade students in public schools in May, 1999. Of these, 96% took
the language arts test and 98% took the mathematics test.
[2] Tables
displaying the change in each elementary school 's scores from 1998 to 1999 are
displayed on our website.
[3] Free and
reduced price school lunches are available to students at public schools.
Family income determines a child's eligibility and parents sign their children
up for the program. Each school has some percentage of students who apply and
are determined to be eligible to participate.
The percentage of students who receive free and reduced-price lunches at
a school reflects the economic condition of the families that the school
serves. Figures for each school are
taken from "New Hampshire Department of Education Free and Reduced Lunch
Percentages" dated December 3, 1997. There are a few exceptions to schools
that link free and reduced price lunch to ability to pay. For example, the Monroe School District
offers free lunches to all students regardless of income. As parents do not apply, the school reports
none eligible for free and reduced price meals.
[4] A table of
data from which this graph is created is available on our website.
[5] Tables with
the 1999 NHEIAP language arts and math average scores of all schools, with the
percentage of students receiving free and reduced price meals at each school
listed, are provided on our website.
[6] For example,
in language arts, the following schools with 40% and above of students
receiving free and reduced school lunches between 247-250 (below the state
average and yet still above a significant number of schools from wealthier
communities): Bluff School, Woodland
Heights Elementary, Pleasant Street School, Conway Elementary School, Hilltop
School, School Street School and Warren Village School. The following schools with less than 10% of
the students receiving free and reduced price meals had scores below 247 on
language arts and thus below the scores of schools from the poorest
communities: Salisbury Elementary
School, Webster Central School, Weston School, Walpole Middle School, and Smyth
Road School.
In math, the following schools with 40% and above of
students receiving free and reduced school lunches scored between 255-253
(below the state average and yet still above a significant number of schools
from wealthier communities): Woodman
Park School, Richards School, Woodland Heights Elementary, Warren Village
School, Walpole Middle School, Marlborough Elementary School, and Ossipee
Central School. The following schools with less than 10% of the students
receiving free and reduced price meals had math scores below 253 and thus below
the scores of schools from the poorest communities: Daniel J Bakie School, Maple Avenue School, Salisbury Elementary
School, Weston School, Smyth Road School and Webster Central School.
[7] The Department of
Education should be provided sufficient resources to conduct this kind of
action research as soon as possible.
[8] The National
Assessment of Educational Progress (NAEP) is a federal program that tests samples
of students in many states to measure their level of achievement. New Hampshire
has participated in this program for some years. The results of the NAEP are
valid at the state level and allow comparisons to other states within a
specified range of error. The sampling process, however, does not allow for
comparison between school districts.
[10] Not able to
read at the Basic level.
[11]
"Should demonstrate an understanding of the overall meaning of what they
read. When reading text appropriate for fourth graders, they should be able to
make relatively obvious connections between the text and their own experiences
and extend the ideas in the text by making simple inferences."
[12]
"Should be able to demonstrate an overall understanding of the text,
providing inferential as well as literal information. When reading text
appropriate to fourth grade, they should be able to extend the ideas in the
text by making inferences, drawing conclusions, and making connections to their
own experiences. The connection between the text and what the student infers
should be clear."
[13]
"Should be able to generalize about topics in the reading selection and
demonstrate an awareness of how authors compose and use literary devices. When
reading text appropriate to fourth grade, they should be able to judge text
critically and, in general, give thorough answers that indicate careful
thought."
[14] Observers
of the tenth grade testing, comments made by some tenth graders who have
completed the tests, and some analysis of results all indicate that some
students do not take the test seriously because they know it will have no
effect on their school grades. While this weakness of the NHEIAP test regime
does not occur in the lower grades, it is also true that the weakness should
affect all results from all schools about equally. Therefore, results of the
tenth grade NHEIAP are still useful for comparative purposes.
[15] There were
14,210 tenth grade students in public schools in May, 1999. For each subject
94-96% of the students took the test.
[16] For an
explanation of economic clusters as they related to high schools, please see
Appendix C.
[17] Scores on
mathematics, science, and social studies are similar to the language arts scores.
[18] These
schools are listed in tables on our website.
[19] The
drop-out data presented in this section has limitations. The Department of Education report collects
only the number of students (by grade, gender and race) who dropout during the
school year. The report does not include students who finished the school year
but did not return in September.
[20] Data are
taken from a New Hampshire Department of Education data file on school dropouts
that did not contain data for Pittsburg High School or Kearsarge Regional High
School. Further, dropout data for the three Manchester high schools are
provided individually, but are aggregated by the Department. We therefore
imputed the average Manchester rate to each of the three schools. The
Department is now engaged in a study to revise the definitions and methods of
collecting dropout information from New Hampshire's public schools.
[21] Analysis by
school would improve if dropout rates were averaged over four or five years.
[22] Kaufman,
Phillip, et al., Dropout Rates in the
United State 1998 Statistical Analysis Report, National Center for
Education Statistics, U.S. Department
of Education, Washington, DC (November 1999).
Website: http://nces.ed.gov/pubs2000/2000022.pdf.
[23]1995 Kids Count Data Book: State Profiles of
Child Well Being, The Annie E. Casey Foundation, Baltimore, MD (1995).
[24] Success in School: Education Ideas that
Count, supplement to KIDS COUNT Data Book 1997: State Profiles of Child Well
Being, the Annie E. Casey Foundation. Baltimore, MD (1997).
[25] Data for
this section was obtained from 4 issues in the series An Informational Study: New Hampshire Public High School Graduates,
published by the New Hampshire Department of Education from July 1996 through
June 1999.
[26] Lin-Wood
(40% to 75%), Portsmouth (46% to 62%), Pelham (41% to 58%) Wilton-Lyndeborough
(43% to 53%), Lisbon (42% to 52%), Coe-Brown-Northwood (35% to 51%), Woodsville
(37% to 47%), Colebrook (32% to 43%), Hillsboro-Deering (29% to 39%), Pittsburg
(22% to 38%), Raymond (22% to 38%), and Stratford (18% to 32%).
[27] Alton (from
65% to 76%), Franklin (51% to 67%), Groveton (55% to 66%), Stevens High School
(53% to 65%), Merrimack Valley (48% to 65%), Nute High School (47% to 61%),
Somersworth (43% to 55%), and Raymond (40% to 52%).
[28] A table listing
the post high school education plans of graduating seniors for all high
schools, listed by economic cluster, is available on our website.
