An Evaluation of a College Preparatory and Readiness Program for Deaf Students.

Deaf and hard of hearing students have, on average, lower academic achievement than heating students. The standardized achievement scores of deaf and hard of hearing students, for example, are significantly below those of heating students, particularly in mathematics and English (Allen, 1986). In examining the 1974 norming of the sixth edition of the Stanford Achievement Test and the 1983 norming of the seventh edition, both of which included representative samples of deaf and hard of hearing students, Allen (1986) found that these students lag behind their hearing counterparts in reading and mathematics. The deficit was more pronounced in reading comprehension than in mathematics computation. In addition, studies show that deaf youth graduate with diplomas from high schools at rates considerably lower than their heating peers (Schildroth, Rawlings, & Allen, 1991). According to a 1989 study, of deaf students in transition from school to work conducted by the Center of Assessment and Demographic Studies at Gallaudet University, only about 50 percent of the deaf students exiting high school graduated with a diploma (Allen, Rawlings, & Schildroth, 1989). In contrast, census data for the general population for the same year show a graduation rate of 81 percent (U.S. Bureau of the Census, 1997). In addition, a considerable number of deaf students are still graduating from high schools throughout the nation with a third- to fourth-grade reading level (Bowe, 1991). The Annual Survey of Deaf and Hard of Heating Children and Youth, conducted by the Center for Assessment and Demographic Studies, found that only one-fourth of deaf students who enter postsecondary institutions read at fifth-grade level or above (Allen, 1994). Thus, even those deaf students who pursue higher education do so with very limited skills.

A major obstacle to employment for many deaf and hard of hearing adolescents is their lower level of English literacy and numeracy compared to hearing peers. Poor English skills may well be the most serious obstacle to both initial employment and future advancement in the changing workforce. Although more jobs are opening up today than ever before, these jobs require higher levels of English literacy and numeracy than in the past (Silvestri & Lukasiewicz, 1989).

Managers interviewed about their workers' skills stated that basic academic skills in mathematics and English are needed for the entry-level jobs they seek to fill. Some managers noted that, although academic skills are not needed in their entry-level jobs, these skills are needed for higher jobs in their companies into which entry-level workers can move. Employers noted that upword movement is possible from entry-level jobs only if workers posses adequate basic skills at the time of entry (Rosenbaum & Binder, 1997).

Studies have consistently found relationships between academic achievement and productivity. Cognitive ability has been found to be the strongest predictor of on-the-job performance in many occupations (Hunter & Hunter, 1984). There is also an association between test scores and performance (Barrett & Depinet, 1991; Bishop, 1993; National Research Council, 1989) and between course work (and academic skills) and wages and employment (Cameron & Heckman, 1993; Daymont & Rumberger, 1982; Kang & Bishop, 1986; Murnane, Willet, & Levy, 1995). One study found that, although grades in school did not improve the wages of new high school graduates immediately after graduation, they had a strong payoff for these graduates' earnings 10 years later (Rosenbaum & Roy, 1996). Even more relevant to our study are the consistent findings that educational attainment is positively associated with earnings; earnings increase as one moves up the degree ladder (U.S. Bureau of the Census, 1991). Moreover, even within the same levels of educational attainment, literacy level is positively associated with higher wages and a reduced likelihood of being unemployed (National Center for Education Statistics, 1996). Therefore, deaf students with college education and with higher literacy levels are more likely to succeed in the workforce.

There are many postsecondary educational opportunities available to deaf and hard of hearing students today. A 1994 study conducted by the Center of Assessment and Demographic Studies at Gallaudet University on the transition patterns of young deaf adults found that two thirds of participants indicated they were either currently enrolled in a postsecondary institution or had some postsecondary education experience (Rawlings, 1994). Of the students who attended a postsecondary program, 50 percent attended a 4-year college or university, 25 percent attended a 2-year college, and 17 percent attended a vocational technical school. The availability of programs for deaf and hard of hearing students in colleges, universities, and technical and vocational institutions and the increase in services and financial support from rehabilitation agencies are evidence of the increasing postsecondary educational opportunities now available to students who are deaf or hard of heating (Allen, Rawlings, & Schildroth, 1989).

Unfortunately, the expansion of postsecondary educational opportunities for deaf and hard of hearing students has not been accompanied by the necessary improvements in academic preparation at the elementary and secondary levels. To remedy this situation, education and rehabilitation agencies serving deaf and hard of hearing students need to assist these students in acquiring English and mathematical skills that are important to future occupational and economic success.

To address the lower educational achievement of deaf and hard of heating high school students and to prepare them for the transition into postsecondary careers, which will in turn contribute to their success in the workforce, the Career Development Department at the League for the Hard of Hearing developed the College Preparatory and Readiness Program. The program's goal is to prepare these students to qualify for postsecondary education. The study reported here assessed the effectiveness of this program in improving deaf students' standardized scores in mathematics and English toward application to postsecondary education.

Method

Description of the Program

The College Preparatory and Readiness Program is conducted annually at the League for the Hard of Hearing, providing intensive English and mathematics tutoring to high school students who are deaf or hard of hearing. It also includes workshops designed to facilitate a better understanding of problems commonly encountered in a campus/student residence setting and to help in the college application process. The program, geared toward students who are planning to obtain a postsecondary degree, lasts four weeks. Classes consist of two hours of mathematics and two hours of English language remediation a day. In past years, classes ranged in size from six to seventeen students and have been taught by two peer tutors. The peer tutors are college students with strong educational backgrounds in English and mathematics and knowledge of, or experience with, deaf and hard of hearing students.

The Director of the Career Development Department, who supervises the Program, selects two tutors from a pool of applicants. Tutors then meet with the Director several times prior to the start of the program. During these meetings, the tutors review the purpose and history of the program as well as curricula from past years, including lesson plans and textbooks. The tutors use this information in creating their own lesson plans for the upcoming sessions.

Young college students serve as peer tutors in order to engender the high school students' identification with them because of their similar age. This identification is believed to facilitate the students' motivation to succeed. Previous research on peer tutoring demonstrates that this method of teaching is very successful with various populations (e.g., Schloss, Kobza, & Alper, 1997; Simmons, Fuchs, Fuchs, & Mathes, 1995), including deaf students (Burley, Gutkin, & Naumann, 1994).

Both traditional and cooperative learning methods are used in the class, including handouts, individual and group reviews, blackboard outlines, educational games, homework, and quizzes. Textbooks are reviewed every year before the start of the program and new texts are selected when deemed appropriate.(1) The tutors use their judgment in implementing or modifying the plans for each lesson based on students' performance and compliance with the planned lessons. The Director of the Career Development Department holds planning meetings with the tutors twice daily during the course of the program, once before the start of classes and once at the end of the day. At these meetings, the tutors have the opportunity to discuss their students' progress and to voice questions and concerns with each other and with the Director.

In previous years, most of the tutors have been hearing college students who are not fluent in American Sign Language. In order to meet the needs of all students attending the program, certified sign language interpreters are used for students who communicate via American Sign Language.

To recruit students, a program description with referral information is sent each year to district offices of the New York City Vocational and Educational Services for Individuals with Disabilities (VESID), Board of Education Hearing Education Services, area high schools for the deaf, and area high schools with programs serving students who are deaf and hard of heating. Students' participation is paid for by VESID or privately. To be eligible for the program, students have to be in high school, have a hearing loss, have to be interested in attending college or post high school technical trade school, and have at least a fourth grade reading level. In addition, an audiological evaluation and a psychological evaluation providing reading and mathematics grade equivalents are required.

The purpose of the mathematics program is to strengthen the students' knowledge of basic mathematics and to increase their competence in handling more complicated questions and calculations. The course emphasizes word problems and focuses on computation skills, recognition of operations, and the application of mathematical concepts. It covers basic arithmetic, algebra, and geometry. The English program focuses on strengthening reading comprehension, building vocabulary, and helping the students identify and write structured English. In addition, students practice various test-taking skills, such as multiple choice, true/false, matching, and fill-in responses.

In the sessions evaluated here, the students were given one version of the Stanford Achievement Test at the start of the program. At the conclusion of the program, another form and/or level of the Stanford Achievement Test was administered to see whether significant gains were made in mathematics and English skills over the course of the 4 weeks.

The Evaluation

The first evaluation of the program was conducted in the spring of 1997. This effort included an analysis of the sessions conducted in 1990, 1994, 1995, and 1996, and was based on students' scores on the Stanford Achievement Test.(2) Pre-test and post-test Stanford Achievement Grade Equivalents were obtained on the following subtests: mathematics computation, concepts of numbers (place value, percents, decimals and the interrelationship of fundamental operations), mathematics application (mathematics applied to real life situations), reading comprehension, spelling (1994 only), vocabulary (1996 only), and language (grammar, punctuation, and dictionary skills).

The results of this evaluation of the 1990 through 1996 sessions of the College Preparatory and Readiness Program were communicated to the Director of Career Development and the tutors of the 1997 session. Recommendations were made for greater emphasis in areas in which students needed further improvement.

A second evaluation was conducted at the end of the 1997 session, and included students' scores on the Stanford Achievement Test. Pre-test and post-test Stanford Achievement Test Grade Equivalents were obtained for the following subtests: mathematics: problem solving (measurement, estimation, problem-solving strategies, number relationships, number systems & theories, patterns and functions), mathematics: procedures (computation using symbolic notation, computation in context, rounding), vocabulary, reading comprehension, and language.

An analysis of pre-test grade equivalents and post-test grade equivalents was conducted for each of the sub-tests. Grade equivalent scores represented the average performance of the hearing students on which the test was standardized. To avoid practice effect, different forms and/or levels of the test were administered before and after the program.(3) Although the difficulty level of the Stanford Achievement Test has not changed, editions of the test have. The 7th edition of the test was administered to the students in 1990, 1994 and 1995, and the 8th edition of the test was given in 1996. The publishers of the Stanford Achievement Test state that there are no significant changes between the two editions and, therefore, comparisons could be made across editions (Psychological Corporation, Customer Service, Personal communication, May 21, 1997). In 1997, however, the 9th edition of the Stanford Achievement Test was administered to the students. Although the English subtests--vocabulary, reading comprehension; and language--remained the same in this edition as in previous editions, the mathematics subtests underwent changes (Gallaudet Research Institute, 1996). The mathematics computation, concepts of numbers, and mathematics application subtests that constituted previous editions were replaced by mathematics: problem solving and mathematics: procedures. Since the 9th edition of the test has no mathematical computation subtest, direct comparisons of this edition with any previous editions are not possible in mathematics.

In addition to the objective scores, students' subjective evaluation of the program was assessed using pre- and post-program surveys. The pre-program survey included ten questions concerning the students' expectations of the program and their performance in it. It also included questions about their plans after graduation from high school and about how they had learned about the program. The post-program survey included twelve questions parallel to the pre-program survey in which the students evaluated the program, their tutors, and their own improvement within the program.

Participants

1990-1996 Sessions. Thirty students participated in the program from 1990 to 1996. Two other students dropped out of the program in 1996 and, thus, were excluded from the study. Eleven (36.7%) of the students were male and 19 (63.3%) were female. Three students (10%) were in the 10th grade, 23 (76.7%) were in the 11th grade, and four (13.3%) were in the 12th grade. The students' mean age was 19.08. The majority of the students, 80% (n = 24), attended schools for the deaf and 20% (n = 6) of the students were partially mainstreamed.(4) None of the students was fully mainstreamed.

The students attended an average of 91.9% of the classes. Fifty percent (n = 15) of the participating students preferred to use sign language during their intake interview, 40% (n = 12) preferred total communication, and 10% (n = 3) preferred oral communication. Pure tone average (PTA) in the better ear was recorded for 28 of the 30 students, the majority of whom, 67.9% (n = 19), had a profound hearing loss with PTA's of 90 dB or greater in the better ear. The average PTA for the whole sample was 90.9 dB.

1997 Session. Seventeen students participated in the 1997 session. Ten (58.8%) students were male and seven (41.2%) were female. Three students (17.6%) were in the 10th grade, nine (52.9%) were in the 11th grade, and five (29.4%) were in the 12th grade. The students' mean age was 18.25. The majority of the students, 88.2% (n = 15), attended deaf schools and only 11.8% (n = 2) were fully mainstreamed. None of the students was partially mainstreamed.

The students attended 96.3% of the 19 classes held during the 1997 session. Most (47.1%, n = 8) students preferred to use total communication, fewer used sign language exclusively (41.2%, n = 7), and only two (11.8%) of the students preferred to communicate orally. Pure tone average (PTA) in the better ear was recorded for 13 of the 17 students. Fifty percent of the students had a profound hearing loss with PTA's of 90 dB or greater in the better ear. The average PTA for the full sample was 91.8dB.

Comparing the students in the two studies shows that students did not differ on age, grade, gender, preferred mode of communication, percent of classes attended, and pure tone average in the better ear. Significant differences were found only for school setting, whereby several partially mainstreamed students, but no fully mainstreamed students, attended the 1990-1996 sessions. In contrast, two fully mainstreamed students and no partially mainstreamed students attended the 1997 session, [Chi square] (2, N = 47) = 7.02, p [is less than] .05.

Results Study I: 1990-1996

Assessment of Mathematics and English Skills

To compare the students' English and mathematics achievement, an overall mathematics score was computed by averaging their scores on the subtests mathematics application, mathematics computation, and concepts of numbers, and an overall English score was computed by averaging their scores on the subtests language and reading comprehension.(5)

Using a correlated t-test, we found that the average post-program score in mathematics (M = 9.98) was significantly higher than the pre-program score (M = 8.44), t(29) = 7.75, p [is less than] .001. In contrast, there was no significant difference between pre- and post-program scores in English (Ms = 5.55 and 5.46, respectively). Table 1 shows students' pre- and post-program scores in mathematics and English from 1990 through 1996, including the various subtests and depicting effect sizes.

Table 1
Pre- and Post-Program Stanford Achievement Test Mean Grade
Equivalents

TEST                    n     M       M      t       p      ES    n
                             PRE    Post

MATH                    30   8.44   9.98    7.75   <.001   0.76   17
COMBINED

Math: Computation       30   9.52   10.95   4.43   <.001   0.58
Math: Application       30   7.15   8.57    4.37   <.001   0.49
Math: Concepts          30   8.66   10.40   5.07   <.001   0.47
Math: Problem Solving                                             17
Math: Procedures                                                  17

ENGLISH                 30   5.55   5.46    0.57   n.s.    0.67   17
COMBINED

English:                30   5.95   5.94    0.04   n.s.    0.49   17
Reading Comprehension

English: Language       24   4.99   4.85    1.06   n.s.    0.78   17
(except 1990)

English: Spelling       5    6.64   7.10    0.73   n.s.    0.27
(1994 only)

English: Vocabulary     11   4.49   5.65    3.39   n.s.    0.53   17
(1996 & 1997)

TEST                     M      M      t       p      ES
                        Pre    Post

MATH                    7.19   8.42   4.01   <.001   0.77
COMBINED

Math: Computation
Math: Application
Math: Concepts
Math: Problem Solving   6.12   8.42   4.99   <.001   0.64
Math: Procedures        8.25   8.43   0.28   n.s.    0.28

ENGLISH                 5.73   6.72   4.47   <.001   0.93
COMBINED

English:                5.80   5.75   0.13   n.s.    0.82
Reading Comprehension

English: Language       5.67   7.69   6.86   <.001   0.87
(except 1990)

English: Spelling
(1994 only)

English: Vocabulary     5.44   9.41   8.92   <.001   0.49
(1996 & 1997)

Relationship between Scores and Demographic Variables

Several demographic variables were analyzed to determine their relationship to students' mathematics and English achievement scores (i.e., class attendance, gender, age, grade in school, pure tone average in the better ear, preferred mode of communication, school setting, and program year). These variables were analyzed in relation to the students' pre-test and post-test scores and to the degree to which their scores improved over the course of the program.

Gender. The results of an independent samples t-test found that male students achieved significantly higher scores than females in the mathematics pre-test ([M.sub.male] = 9.54, [M.sub.female] = 7.81), t(28) = 2.51, 12 [is less than] .05 and post-test ([M.sub.male] = 11.00, [M.sub.female] = 9.38), t(28) = 2.05, p [is less than] .05. Males and females did not differ significantly in the degree to which they improved in mathematics. In contrast to the gender differences in mathematics, English pre-test ([M.sub.male] = 5.18, [M.sub.female] = 5.76) and post-test ([M.sub.male] = 5.47, [M.sub.female] = 5.46) scores were not significantly different between male and female students.

Age. A bivariate correlation revealed that age was negatively correlated with pre-test scores in English (r = -.48, p [is less than] .01), yet positively correlated with progress made in the same subject (r =.45, p [is less than] .05). In other words, although older students tended to have lower mean pre-test scores, they also tended to improve more than younger students in this area over the course of the program.

School setting. School setting had an effect on both pre- and post-test mathematics scores. The mean mathematics pre-test score of students in schools for the deaf was higher than of those who were partially mainstreamed ([M.sub.deaf] school = 8.95, [M.sub.partial mainstream] = 6.42), t(28) = 3.23, p [is less than] .01, as was their mathematics post-test score ([M.sub.deaf] school = 10.45, [M.sub.partial mainstream] = 8.08), t(28) = 2.59, p [is less than] .05. Students who attended schools for the deaf scored at least two grade equivalents higher than students who were partially mainstreamed.

Program year. Finally, a one-way analysis of variance was used to test whether program year had an effect on students' test scores and improvement. It was found that program year had a significant effect on post-test English scores, F(3,26) = 4.52, 12 = [is less than] .05, and on improvement scores, F(3,26) = 3.84, p [is less than] .05. Post hoc scheffe tests showed that students' mean post-test English score in 1994 (M = 7.20) was significantly higher than the mean score in 1995 (M = 5.01) and 1996 (M = 4.99). Difference scores reflected the same findings ([M.sub.1994] = .92, [M.sub.1995] = -.19, and [M.sub.1996] = -.38). Thus, while students' mean pre-test score was similar in each program year, the progress that students made in 1994 was exceptional compared to the lack of progress in the other years.

The variables of class attendance, grade, preferred mode of communication, and PTA in the better ear had no statistically significant relationship with students' achievement in the program.

Study II: 1997

Assessment of Mathematics and English Skills

An overall mathematics score was computed by averaging grade equivalents in the mathematics: problem solving and procedures sub tests. An overall English score was computed by averaging grade equivalents in the reading comprehension, and language subtests.(6) Using a correlated t-test, we found that the average post program score in mathematics (M = 8.42) was significantly higher than the pre-program score (M = 7.19), t(16) = 4.01, p [is less than] .001. Similarly, the mean English post-test score (M = 6.72) was a significant improvement over the pre-test score (M = 5.73), t(16) = 4.47, p [is less than] .001. That is, on average, the students improved one grade in their English skills. Table 1 shows students' pre- and post-program scores in mathematics and English, as well as all the subtests, including effect sizes.

Relationship between Scores and Demographic Variables

The effects of class attendance, gender, age, grade in school, school setting, preferred communication mode, and PTA in the better ear on students' performance were examined. Unlike the analysis of these variables in the 1990-1996 sessions, only age and preferred mode of communication were associated with students' performance. The fact that there were fewer associations between demographic variables and test scores in the 1997 analysis compared to the previous sessions may be the result of the smaller sample size.

Age. As in previous sessions, age was negatively correlated to both pre- and post-test English scores, demonstrating that the older students tended to have lower scores both before and after the program ([r.sub.pre] = -.72, p [is less than] .01, [r.sub.post] = -.76, p [is less than] .001). Similar results were obtained in the pre- and post-test mathematics scores, as older students scored significantly lower than younger students ([r.sub.pre] = - .59, p [is less than] .05, [r.sub.post] = -.56, p [is less than] .05). The relationship between age and post-test scores is unlike that of previous sessions, where improvement was positively correlated with age.

Mode of communication. A one-way ANOVA showed that students' preferred mode of communication also had a significant effect on English scores, [F(2,14).sub.pre] = 3.79, 12 [is less than] .05; [F(2,14).sub.post] = 4.27, 12 [is less than] .05. Post hoc scheffe tests showed that students who communicated orally scored higher than students who used sign language on both the pre-test ([M.sub.oral] = 9.50, [M.sub.sign] = 3.77) and post-test ([M.sub.oral] = 9.70, [M.sub.sign] = 4.83). Scores of total communication users were in between these two groups and did not differ significantly from either.

Students' Expectations and Evaluation of the Program

As previously mentioned, most sessions were conducted orally with a sign language interpreter. It appears, however, that this mode of communication is not one to which the students have frequent exposure. Most (n = 14, 80%) students in the 1997 session (the first time this information was obtained) reported that their teachers used total communication in the classroom. Other students (n = 2, 13%) reported that their teachers used sign language, and one student (7%) reported using oral communication without an interpreter. None of the students reported having interpreters with non-signing teachers in the classroom. It is unknown whether using the mode of communication to which the students are accustomed would result in greater gains in their mathematics and English scores.

Students reported in the beginning of the program how much progress they hoped to make, and indicated at the end of the program how much progress they believed they had made. Before the program students expected to improve their mathematics skills on average "somewhat," 2.69 on a scale of 1 ("not at all") to 4 ("a lot"). After the program, students believed they had improved to a similar extent with a mean of 3.13. In English, however, the students' expectations of improvement before the program (M = 2.75) differed significantly from their feelings of how much they had improved at the end of the program (M = 3.81), t(16) = 4.58, p [is less than] .001, such that their assessment of their own improvement exceeded their expectations.

To compare the students' feelings of improvement with their actual improvement, improvement scores were converted to a 4-point scale based on the scores' distribution into quartiles. Using correlated t-tests, students' subjective feelings of how much they improved in mathematics and in English were compared to their actual improvement scores. The students' belief about their level of improvement ([M.sub.math] = 3.12, [M.sub.English] = 3.82 on a 4-point scale) was significantly higher than their actual improvement in both mathematics (M = 2.47), t(16) = 2.52, 12 [is less than] .05, and English (M = 2.59), t(16) = 4.58, 12 [is less than].001, signifying that students had somewhat unrealistically high assessment of their own progress.

Overall, the majority of the students (n = 9, 56%) were "very satisfied" with the program, others (n = 7, 44%) were "somewhat satisfied," and none of the students said he or she was not satisfied with the program.

Discussion

In all years of the 4-week College Preparatory and Readiness Program students gained on average about 1.5 grades in mathematics. Progress in English was less consistent, but after communicating this fact to the program director and tutors, subsequent gains in English of about one grade were evident as well. These results show not only that deaf students can significantly improve their mathematics and English scores by attending a short and intensive summer program, but also that conducting an evaluation of such a program can result in higher program effectiveness. Because structural differences in curriculum or teaching methodology were not quantified, they cannot be credited for the substantial improvement made by the students in 1997 compared to the previous sessions. Future studies may record various aspects of the program (e.g., the extent to which each teaching method is used), which will allow analysis of the effect of classroom methodology and curriculum on students' achievement.

The elevated mathematics and English scores would make these students more likely to be admitted to college, which should then lead to better work opportunities and success, because educational attainment is associated with earnings (U.S. Bureau of the Census, 1991). Even for students who decide not to go to college, higher mathematics and English scores would lead to better work performance. The relationship between numeracy and literacy skills in high schools is consistently associated with work performance, wages, and reduced unemployment (Barrett & Depinet, 1991; Bishop, 1993; National Research Council, 1989; Rosenbaum & Roy, 1996). In a follow-up study, students who attended the program in previous years will be contacted and their educational and occupational success will be assessed. This future study will enable us to determine whether the improved performance resulting from the program had an effect on these students' performance in subsequent schooling and in the workplace.

Differences between Mathematics and English Scores As other studies have shown (Holt, 1994), there was a disparity between the students' mathematics and English skills. In the 1990-1996 sessions, there was an average gap of nearly three grade levels between the students' mean mathematics skills and their mean English skills upon entering the program, which increased by the end of the program, as students made gains in mathematics but not in English. In the 1997 session, the gap between students' mathematics and English skills in both the beginning and the end of the program was about 1.5 grade levels, with higher scores in mathematics than in English. Clearly, when designing remediation programs for students who are deaf, more emphasis needs to be placed on English skills than any other area. As noted before, English skills are crucial for entry level positions as well as for future advancement in most fields. In addition, English literacy determines the ability to perform tasks requiring other skills (e.g., mathematics), that are presented in English.

A close examination of students' mathematics scores seems to suggest that during the 1997 session students' scores were lower than their scores in previous years. This difference, however, reflects the changes introduced in the 9th edition of the test. To reiterate, the 9th edition of the Stanford Achievement Test has no pure computation subtest. The National Council of Teachers of Mathematics states that mathematics literacy must be viewed as encompassing more than simply computational tasks (Daniele, 1993). Addressing this concern, all items in the 9th edition's mathematics: problem solving subtest are word problems, and most of the items in the mathematics: procedures subtest are word problems. Reading comprehension level has been demonstrated to be related to the arithmetic problem-solving level of deaf students (Pau, 1995), such that if the student is unable to understand a problem presented in a text format he or she will not be able to solve it correctly. Mathematics subtests that depend heavily on English skills, therefore, tend to become in part reading tests for deaf and hard of heating students (Holt, 1994). Thus, the new structure of the 9th edition accounts for the lower mathematics scores of students in the 1997 program as compared to students in previous years.

Demographic Variables

Gender. Male students in our first study (1990-1996) had higher mathematics scores than females. In a review of the literature, Kluwin (1994) cited "low level but persistent gender effects" (p. 466) in the education of deaf children. Kluwin found in his own study that boys with hearing loss were more likely than girls to be mainstreamed in a specific class, usually mathematics. Studies about gender differences in mathematics achievement of heating students yield similar results. Most reviews of the literature find that boys' achievement levels in mathematics surpass girls' achievement level at some point in their schooling (Brandon, Newton, & Hammond, 1987). Our results, therefore, reflect gender differences in mathematics found both among hearing and deaf students.

Our results indicated no significant gender differences in English achievement, in accordance with some other studies (Holt & Allen, 1989; Slate & Fawcett, 1996). The results of one study (Allen, 1986), however, indicated that female students with hearing loss had higher reading comprehension scores than male students with hearing loss. Similar results are found in studies of heating students, where girls show a slight advantage of boys in spelling (Allred, 1990) and languages in general (Stumf & Stanley, 1996). Many educational experts believe, however, that even these slight gender differences are narrowing. Our results, therefore, are in line with general findings about gender difference in academic achievement showing either no gender differences in English or very slight differences.

Age. In both evaluations, age was negatively related to preprogram English scores: older students had lower scores than younger ones. Since students with disabilities can stay in school until the age of 21, and many do stay, these results may reflect that students who do poorly academically stay in school longer, hence the negative relationship between age (but not grade) and English achievement. In addition, age was positively associated with level of improvement in English, such that older students showed more improvement than younger students. One possible explanation for this effect is that the older students may have attended the program knowing that they were particularly weak in English and were especially motivated to improve in this area in order to get into postsecondary schools.

School setting. The results of the first evaluation also indicated that students attending schools for the deaf scored significantly higher in mathematics than students who were partially mainstreamed in public high schools. These results appear to counter prior findings that students in local school settings show advantages over students in schools for the deaf (Allen, 1986; Holt, 1994; Wood, Wood, Kingsmill, French, & Howarth, 1984). In some of these studies it is unclear to what degree students in the local schools were mainstreamed (Allen, 1986; Wood et al., 1984), and in all studies no information is given about which subjects the students studied when they were mainstreamed. Furthermore, we do not have information about which academic subjects our own partially mainstreamed participants studied in integrated classrooms. They may or may not have had English and mathematics instruction with hearing students. Therefore, it is not entirely clear what the differences observed between students from different educational environments mean. Future research would need to specify the academic subjects in which students attend classes with deaf or hearing peers in order to draw firm conclusions about the role of mainstreaming in academic achievement.

Mode of communication. The 1997 results find that students who communicate orally score higher in English (although not in mathematics) than students who prefer to use sign language. These results reflect that those students who prefer to communicate orally have a higher level of English skills than those students who prefer to communicate in sign language. Evidently, the use of oral language as a mode of communication aids in the development of English of deaf and hard of hearing students who often develop these skills more slowly than heating students (Allen, 1986; Holt, 1994).

Program year. It is not clear why English scores in 1994 were higher than any other session of the program since neither the curriculum nor the teaching strategies changed significantly from year to year. Because structural differences in curriculum or teaching methodology were not quantified, they cannot be credited for the substantial improvement made by the students in 1994 or 1997. Future studies of the College Preparatory and Readiness Program may include an analysis of the effects of the classroom methodology and curriculum on students' achievement.

Students Surveys

The pre- and post-program surveys show that before the start of the program students had expected to improve their English skills only "somewhat," whereas following the program they felt they had improved "very much." In mathematics students both expected and felt that they improved "somewhat." These results correspond with the students' actual greater improvement in English than in mathematics. In addition, when feelings of improvement were compared to the students' actual improvement scores it was found that the students were overly optimistic about the degree to which they had actually improved. Students' responses to the survey also indicate high level of satisfaction with the program.

Conclusion

This evaluation of the College Preparatory and Readiness Program shows that the educational needs of deaf and hard of heating high school students can be addressed and their achievement markedly improved using a specialized, short, and intensive summer program. The generally low academic achievement of deaf and high school students (Bowe, 1991), therefore, need not be a hindrance in applying for postsecondary education if more rehabilitation agencies conduct similar programs.

Despite the program's success in improving students' scores by two grade levels over a 4-week period, students at the end of the program still perform below grade level. This obviates the need for similar programs earlier in the students' educational careers than junior or senior year in high school. It is apparent that this type of educational intervention is needed at all stages of deaf youngsters' educational experience in order for them to be fully prepared to enter postsecondary education, as well as a competitive workforce.

Endnotes

(1.) A complete list of textbooks is available from the first author.

(2.) The program did not run from 1991 through 1993 due to the coordinator's leave of absence.

(3.) In 1990, 1994, and 1995 students were given the Advanced Battery Level 1-Form E at the beginning of the program, and Advanced Battery Level 1-Form F at the end of the program. In 1996, the Advanced Battery Level 1 test was administered at the beginning of the program, and the Advanced Battery Level 2 test at the end of the program. In 1997, the Intermediate Battery Level 3-Form S was administered at the beginning of the program, and the Advanced Battery Level 1-Form S at the end of the program. The Stanford Achievement Test in the Intermediate Battery Level 3 (all forms) is designed for hearing students in the 6.5 to 7.5 grade levels, the Advanced Battery Level 1 (all forms) is designed for hearing students in the 7.5 to 8.5 grade levels, and the Advanced Battery Level 2 (all forms) is designed for hearing students in the 8.5 to 9.9 grade levels.

(4.) No information is available about what academic subjects students were mainstreamed in.

(5.) These were the only two subtests that were administered consistently over the different years of the program, and were therefore the only ones to be used in computing an overall score of English.

(6.) Although scores from a third subtest, vocabulary, were available for this year, it was omitted from the calculated English score in order to maintain consistency with scores computed for the previous years. Analyses including the vocabulary subtest demonstrate even greater gains in English than without it.

Authors note:

We thank Florence Rosen, the coordinator of the College Preparatory and Readiness Program for her assistance in carrying out the evaluation of the program and for helpful comments on a previous draft of this article. We also thank three anonymous reviewers for comments on a previous version of this article.

References

Allen, T.E. (1986). Patterns of academic achievement among hearing impaired students: 1974 and 1983. In A.N. Schildroth & M.A. Karchmer (Eds.), Deaf children in America (pp. 161-206). San Diego. CA: College-Hill Press.

Allen, T.E. (1994). Who are the deaf and hard of hearing students leaving high school and entering post secondary education? Paper submitted to Pelavin Research Institute. Gallaudet University, Washington, DC. http://www.gallaudet.edu/~cadsweb/whodeaf.html.

Allen, T.E., Rawlings, B., & Schildroth, A. (1989). Deaf students and the school to work transition. Baltimore, MD: Paul H. Brookes.

Allred, R.A. (1990). Gender differences in spelling achievement in grades 1 through 6. Journal of Education Research, 83(4), 187-193.

Barrett, G.V., & Depinet, R.L. (1991). A reconsideration of testing for competence rather than for intelligence. American Psychologist. 46, 1012-1024.

Bishop, J. (1993). Improving job matches in the U.S. labor market. In: M.N. Bailey, (Ed.), Brookings Papers in Economic Activity: Microeconomics. (pp. 335-400). Washington DC: Brookings Institution.

Bowe, F. (1991). Approaching equality: Education of the deaf. Silver Spring, MD: T.J. Publishers.

Brandon, P. R., Newton, B.J., & Hammond, O.W. (1987). Children's mathematics achievement in Hawaii: Sex differences favoring girls. American Educational Research Journal, 24 (3), 437-461.

Burley, S., Gutkin, T., & Naumann, W. (1994). Assessing the efficacy of an academic hearing peer tutor for a profoundly deaf student. American Annals of the Deaf, 139(4), 415-419.

Cameron, S. & Heckman, J. (1993). The nonequivalence of high school equivalents. Journal of Labor Economics, 11, 1-47.

Daniele, V.A. (1993). Quantitative literacy. American Annals of the Deaf, 138 (2), 76-81.

Daymont, TN. & Rumberger, R.W. (1982). Job training in the schools. In: R. Taylor, H. Rosen, & F. Pratzner (Eds.), Job training for youth. Columbus: National Center for Research in Vocational Education, Ohio State University.

Gallaudet Research Institute. (1996, December). Achievement Testing of Deaf and Hard of Hearing Students: The 9th Edition Stanford Achievement Test. Washington, D.C.: Gallaudet Research Institute.

Holt, J.A. (1994). Classroom attributes and achievement test scores for deaf and hard of heating students. American Annals of the Deaf, 139 (4), 430-437.

Holt, J.A., & Allen, T.E. (1989). The effects of schools and their curricula on the reading and mathematics achievement of hearing impaired students. International Journal of Educational Research, 13(5), 547-566.

Hunter, J.E., & Hunter, R.F. (1984). Validity and utility of alternative predictors of job performance. Psychological Bulletin, 96, 72-98.

Kang, S. & Bishop, J. (1986). Effects of curriculum on labor market success immediately after high school. Journal of Industrial Teacher Education, 23(4), 15-29.

Kluwin, T.N. (1994). The interaction of race, gender and social class effects in the education of deaf students. American Annals of the Deaf, 139(5), 465-470.

Murnane, R.J., Willet, J.B., & Levy, F. (1995). The growing importance of cognitive skills in wage determination. Review of Economics and Statistics, 77, 251-266.

National Center for Education Statistics (1996). Labor market outcomes of literacy and education. Indicator of the month. Washington D.C.

National Research Council (1989). Fairness in employment testing. Washington DC.

Pau, C.S. (1995). The deaf child and solving problems of arithmetic: The importance of comprehensive reading. American Annals of the Deaf, 140 (3), 287-290.

Rawlings, B.W. (1994). Postsecondary education: Its impact and outcomes. In: T.E. Allen, K.H. Lam, B.W. Rawlings, D.E. Rose, & A.N. Schildroth (Eds.) Young deaf adults and the transition from high school to postsecondary careers. Gallaudet Research Institute: Occasional Paper 94-1. Gallaudet University, Washington, DC.

Rosenbaum, J.E., & Binder, A. (1997). Do employers really need more educated youth? Sociology of Education, 70, 68-85.

Rosenbaum, J.E., & Roy, K. (1996, August). Long-term effects of high school grades and job placement. Paper presented at the annual meeting of the American Sociological Association, New York.

Schildroth, A.N., Rawlings, B.W., & Allen, T.E. (1991). Deaf students in transition: Education and employment issues for deaf adolescents. The Volta Review, 93(5), 41-53.

Schloss, P.J., Kobza, S.A., & Alper, S. (1997). The use of peer tutoring for the acquisition of functional math skills among students with moderate retardation. Education & Treatment of Children, 20(2), 189-208.

Silvestri, G., & Lukasiewicz, J. (1989). Projections of occupational employment, 1988-2000. Monthly Labor Review, 112(11), 42-65.

Simmons, D.C., Fuchs, L.S., Fuchs, D., & Mathes, P. (1995). Effects of explicit teaching and peer tutoring on the reading achievement of learning-disabled and low-performing students in regular classrooms. Elementary_ School Journal, 95(5), 387-408.

Slate, J.R., & Fawcett, J. (1996). Gender differences in Wechsler Performance Scores of school-age children who are deaf or hard of heating. American Annals of the Deaf, 141 (1), 19-23.

Stumpf, H., & Stanley, J. (1996). Gender-related differences on the College Board's Advanced Placement and Achievement test, 1982-1992. Journal of Educational Psychology, 88(2), 353-364.

U.S. Bureau of the Census (1991). Does education pay off? Washington, D. C.: U.S. Government Printing Office. (Bureau of the Census Statistical Brief SB/91-4)

U.S. Bureau of the Census (1997). School enrollment -- social and economic characteristics of students, October 1995 (Update). Washington, D. C.: U.S. Government Printing Office. (Current Population Reports, Publication NO. PPL55).

Wood, H.A., Wood, D.J., Kingsmill, M.C., French, J.R., & Howarth, S.P. (1984). The mathematical achievement of deaf children from different educational environments. British Journal of Educational Psychology, 54, 254-264.

Yael Bat-Chava, Ph.D., Director of Research, League for the Hard of Hearing, 71 West 23rd Street, New York, NY 10010. Email: [e-mail]3