Introduction

While it is generally believed that the anti-bilingual movement in California's schools is a recent phenomenon, it has actually been a long-standing issue. Crawford (1992, p. 66) found that California legislation mandating English-only instruction began in 1855 and reached a peak in the 1878-79 constitutional revisions, in violation of the Treaty of Guadalupe Hidalgo of 1848. Although the English-Only amendment was dissolved in 1966, an organization called Official English, headed by former U.S. Senator Hayakawa, was successful in passing an English-Only initiative, Proposition 63, in 1986, by a 2-to-1 popular vote (Crawford, 1992 p. 61). From the outset, Macedo (Zeyrep, 2000, p. 21) finds, the history of the anti-bilingual movement has not been about methodologies, but "these assaults are fundamentally political." Studies of bilingualism, (Baker, 1996, pp. 352-353) include more than linguistics, sociology, and psychology, but also power structures and political systems. While minority languages themselves are seldom the cause of conflict, they are viewed as a political problem to be solved by assimilation. Casanova (1991, p. 170), in her review of the history of bilingual education, cites several instances of political intervention, including the use of the AIR study of 1974 and the Baker/de Kanter report of 1981, to attempt the defeat of federal bilingual education legislation. Both have since been found to contain significant errors in methodology, which arrived at conclusions unfavorable to the use of the primary language for instruction.

As the Proposition 227 campaign against bilingual education raged in 1998, the same kind of political tactics were used. The campaign coincidentally followed on the heels of Proposition 187, the anti-immigrant initiative. As had been predicted by Krashen (1996, pp. 43-50), bilingual education was once again under direct attack through misinformation to the public. Krashen's research verifies that, when the public, including parents, is properly informed about bilingual education, the overwhelming majority supports it. Faltis and Hudelson (1998, pp. 15-16) have found that misinformation about bilingual education has been used effectively to reduce student enrollment in bilingual programs. Research on the "mother tongue," as the name implies (Dicker, 1996, pp. 25), has well established its value in the development of personal identity. Yet, the English-Only movement insists that languages other than English be eradicated in this country. In his haste to convince the public that the schools were the ones who misinformed parents, Ron Unz, the initiative's principal author, included a provision in Proposition 227 that permitted parents or guardians to request a waiver and select the bilingual education option. Since then, many parents have chosen the waiver.

Since the June 1998 passage of Proposition 227 (Now EC 300-340) by public initiative in California, there have been mixed messages regarding its success. While some reports claim that English language learners are learning English and achieving higher through English language immersion, according to state test results, others show that the proposition has failed miserably (Mora, 2000, p. 20). However, it is important to note that not all ELLs are enrolled in English immersion programs. Table 1 lists the various programs in which California ELL students are enrolled.

These programs options are described as follows:

Program options 1-3 are classified by the State Department of Education as SEI, as mandated by EC 300-340, although No. 1 does not provide for SEI content instruction. Thus, 71.1% of all ELL students are considered to be placed as required by the law.

Because it is generally believed that Proposition 227 destroyed bilingual education, any improvements in student test scores would tend to be attributed to SEI programs. This is, of course, a simplification, as can be seen in Table 1. Bilingual education, although serving only 11.4% of ELL students, is still one of several options available in many school districts.

The continuation of bilingual education programs is due to the parent/guardian exception waivers (Section 310). Indeed, it is one of the three primary requirements of the new law, the other two being the SEI instructional provision (Section 305) and the adult education provision (Section 315). While the latter two appear to be interpreted in several ways by school districts, it is the parent/guardian exception waiver provision which clearly separates school districts that continue offering bilingual education programs and those that dismantled them almost immediately. While there are some who accuse districts that received a high number of waivers of manipulating the process, the opposite may be true. No one will deny that Unz stated throughout his campaign that parents should ultimately decide the kind of program by which their ELL children should be served. Indeed, in its publication of emergency regulations in reference to parent/ guardian exception waivers, the California State Department of Education (1998) ordered school districts to:

Table 1
Distribution of ELL Students (K-12) by Program

Inform parents/guardians of a child's placement in the structured English immersion program and notify them of their waiver opportunities. Specifically:

Clearly, from the first legally required day for the implementation of Proposition 227, every school district in the state was notified and should have known the interpretation by the SDE (State Department of Education) on the implementation of its parent/guardian exception waiver provision. While this provision should have received as much attention as the instructional provision, it did not. Indeed, according to Quezada (2000, pp. 17-18), while many school districts complied with this state requirement, there were also many that either failed to properly inform parents in a timely manner or found ways to avoid implementing the wishes of parents. As a result, parents are filing complaints with the State Department of Education, and at least two school districts, Pittsburg Unified and Oceanside School District, have been found in violation of this provision of EC 300-340 and other provisions on as many as 20 counts. In the meantime, their ELL students have been denied their educational rights under the law.

The Problem

This study identified three school districts, in three distinct geographic locations, which complied with the parent/guardian waiver provision from the outset, resulting in the continuation of bilingual education programs as alternatives to the structured English immersion (SEI) programs and mainstream programs. One of these school districts is located in the northern portion of the state, another in the central coast portion, and the third in the Los Angeles area. Besides their geographic location, these three school districts were selected for the study because of their compliance with Proposition 227's parent/guardian exception waiver provisions and their offering of SEI and alternative instructional programs.

Statement of the Problem

It was the purpose of the study to determine whether significant differences existed in opinions about language acquisition/development and the school's three options under Proposition 227, between parents who requested a waiver and those who did not, in an environment where all parents had a complete opportunity to be made aware of these factors.

In this study, "bilingual parents" refers to parents/guardians who requested an exception waiver for their ELL (Spanish-speaking) children so that they could enter or remain in a bilingual education program, and "SEI parents" refers to parents/guardians who did not request a waiver for their ELL (Spanish-speaking) children from the requirement for structured English immersion. The following hypotheses apply to this study:

Background of the Problem

In June 1998, immediately after the passage of Proposition 227, each of the three school districts in the study initiated district and school level action to prepare official documents and programs in accordance with the new law, to gain approval at all levels, and to disseminate the school district decisions to all personnel and parents/guardians. Thus, support, communication, and agreement on policies and practices were maximized. This provided the opportunity for all parents/guardians in each of the school districts to be fully informed of each area of the new law and to select the instructional option of their choice.

School district documents for each of the school districts illustrate the thoroughness of their approach. Presumably in anticipation of the passage of Proposition 227, school district administrators prepared parent/guardian exception waiver request information for approval by their school boards. Once approved, all school principals were expeditiously informed and provided with bilingually prepared forms for distribution to parents. Each was charged with orienting school personnel. Every conceivable effort was made to inform parents of the new law, its effect on their children, and the instructional options available to them, without biased statements. Before parents were asked to decide on an option, handouts which clarified each option, and even definitions, were provided to each parent/guardian. Two of the school districts even provided parents with bilingual glossaries of terminology used in the options. School level meetings for parents and guardians followed district-approved agendas, some with "talking points," covering every facet and instructional factor of the law. In order to assure consistency, all school and district personnel who made contact with parents were fully informed of district policy, in compliance with the law. Each was able to discuss options with parents, as long as they followed district guidelines. Parents and guardians who did not attend meetings were contacted by telephone so that they also had a complete opportunity to be fully informed. Finally, they were able to make their choice, in some cases on separate forms so that the choice was quite clear and exhibited its reasons, as required by law.

Design for the Study

In order to maximize the outcomes of the proposed study, the investigator sought to conduct it in these school districts after it became apparent that they provided an excellent environment for this purpose. Having complied completely with the new law's parent/guardian exception waiver provision, and having developed a potentially well-informed parent population, these districts were outstanding choices for this study.

Population and Sample

Once each of the three school districts had agreed to participate in the study, the investigator met with district administrators and three school principals per district to discuss details of the study and determine their interest in participating. Each of these districts has a very high percentage of Latino students. At the time of the meetings, they were:

Within the Latino population of each district, at least 50% were ELL students. Procedures for identifying and gaining approval of parents to be interviewed were determined. At the outset, it was anticipated that three schools in each district would provide the study with 20 bilingual parents and 20 SEI parents each, thus 60 in each group, or 120 per school district. Later, as the need for additional participating parents became evident, more schools were added, but with the same ELL student population characteristics. In the Northern and Central Coast school districts, it was agreed that parents/guardians would be interviewed in their homes. In the Southern district, it was preferred by the school district that they be interviewed at school sites.

Each participating school agreed to solicit parent/guardian participation, with equal attention to each of the sample groups. In the Northern and Central Coast districts, this was initially conducted by letter, but needed to be reinforced by telephone. In the Southern district, all parent contacts were made orally, either personally or by telephone. Every effort was made to develop a total sample of 360, with 180 in each group. By the time of the planned interviews, the sample size had been attained.

Instrument

The instrumentation for the study consisted of an interview schedule containing 24 items, to be personally administered by the investigator and research assistants in the language of choice of each interviewee. Items 1-5, 7-11, 13-17, and 19-23 were numerical statements, for response on a five-point Likert Scale, from "Strongly Agree" to "Strongly Disagree." The intervening questions (6, 12, 18, and 24) were open ended, with responses to be categorized upon collection and percentages determined for each category.

The first set of five statements was about language acquisition and development. The first open-ended question was on the most effective form of school-based language acquisition. The second set of five statements was about SEI programs. The second open-ended question was on how the interviewee was informed about the available educational options. The third set of five statements was about mainstream programs. The third open-ended question was on how a Spanish-speaking child would do in a mainstream classroom. The fourth set of five statements was about bilingual programs.

The fourth open-ended question was on how the interviewee arrived at the decision to choose a program option for his/her child. All of the statements in the numerical area of the interview were drawn from information disseminated by school districts and/or research.

Procedures for the Study

Fifteen research assistants were selected and trained to assist in data collection. Six were included in the Northern district and six were included in the Central Coast district, since interviews were to be conducted in homes. In the Southern district, only three assistants were included, with the investigator also conducting interviews, all on school sites. Once each assistant was selected by means of an application and telephone interview, a training site, date, and time were identified. At the time of the training all details of the study and lists of interviewees were provided; however, the group identification of the interviewees was not divulged.

In all 360 cases, appointments were arranged by letter or telephone. In the Northern and Central Coast districts, these were made by research assistants. In the Southern district, they were made by school office personnel. Research assistants in the Northern and Central Coast districts began, in turn, to arrange for interviews, while the Southern district assistants awaited the three consecutive dates for interviews.

As was expected, there was an attrition rate by the time interviews were completed. The final samples were: (a) Northern, 96 total (51 Bilingual, 45 SEI); (b) Central Coast, 94 (46 Bilingual, 48 SEI); and (c) Southern, 106 (53 Bilingual, 53 SEI), for a total of 296 (150 Bilingual, 146 SEI). While the home interview data collection in the Northern and Central Coast district occurred over a period of two months, the school-interview data in the Southern district was completed in a three-week period, with three consecutive days of interviewing, followed by a two-week hiatus and a fourth, final day of interviews.

Each interview was conducted individually by one interviewer. The interviewee was asked to choose a language for the interview. Then, without explanation or additional information, each item was presented and recorded in turn. After the interview, additional information was provided, if asked for. As soon as possible, the interview schedules were collected and grouped by the investigator. Group identification was not recorded on each interview schedule until the interview was completed. Responses to open-ended questions were recorded verbatim.

Analysis of the Data

Two different statistical procedures were used to analyze the data in the study. The mean responses to numerical items, based on a Likert Scale, were analyzed for statistical significance using t-tests for unpaired data. The responses to the open-ended questions were placed into categories to determine percentages. The differences between percentages within each group responding to each category were tested for statistical significance using a Chi Square goodness-of-fit test. For the purpose of determining significance, a p level of .10 was required.

Results

The study yielded useful numerical data regarding the differences of opinion on language acquisition/development, the structured English immersion program option, the mainstream program option, and the bilingual education program option. It also provided useful data, through open-ended responses, on differences of opinion on language acquisition, sources of information, language submersion, and parent reasons for choosing an educational program for their children.

Hypothesis One

The results show a statistically significant difference between the two groups in the area of language acquisition/development, with the Bilingual Parent group in greater agreement. Hypothesis One is rejected when measured by numerical data in Items 1-5. Specific results for each item are illustrated in Table 2.

These results demonstrate that the two groups are in agreement on statements about the development of basic interpersonal communication skills and that cognitive academic language requires more time to develop. The highest mean was 2.05, suggesting that parents in both groups, on average, tended to agree with all of the statements. In the three areas of significant difference, the bilingual parent group responses were in greater agreement. They differ on the use of the use of the primary language for instruction most significantly. The second most significant difference is in the area of transferability of concepts from L1 to L2. Thirdly, they differ on the relative importance of learning subject matter, with the SEI parent group agreeing less.

Hypothesis Two

The results show statistically significant differences between the two groups on language acquisition. Hypothesis Two is rejected, when measured by Item 6. This was an open-ended question: "What do you believe is the best way for Spanish-speaking children to acquire English in school?" Responses were placed into six categories and percentages were used to determine significant differences, as seen in Table 3.

These findings resulted in statistically significant differences. The individual categories record slight differences for (a) and (c), with greater agreement by the SEI parent group. Significant differences were found for the other categories, with greater agreement by the SEI parent group on (b), (d), and (f), and by the bilingual parent group on (e), as could be expected, given the results of the numerical study.

Table 2
Results of Numerical Responses on Language Development

ns = No significant differences

Hypothesis Three

The results show no statistically significant difference between the two groups in the area of structured English immersion. However, as Table 4 illustrates, there are significant differences on items 10 and 11. Hypothesis Two cannot be rejected, as measured by Items 7-11.

Table 3
Results of Open-Ended Responses on Language Acquisition

ns = No significant differences

Table 4
Results of Numerical Responses on Structured English Immersion Program ns = No significant differences

ns = No significant difference

Both groups seemed equally unsure about items 7, 8, and 9. The significant differences on items 10 and 11 are based on greater agreement by the bilingual parent group. The high mean of 2.54 on item 11 in the SEI parent group indicates a degree of unsureness.

Hypothesis Four

The results show no statistically significant difference between the two groups in the area of sources of information, as illustrated in Table 5. Hypothesis Four cannot be rejected, as measured by item 12. This was an open-ended question: "How were you informed about the educational options for your child(ren)?" Responses were placed into six categories and percentages were used to determine significant differences.

Table 5
Results of Open-Ended Responses on Means of Information

ns = No significant differences

There were no statistically significant differences in any of these categories. However, it should be noted that 62.7% of the bilingual parent group but only 50.3% of the SEI parent group were informed at meetings. On the contrary, 33.3% of the SEI parent group versus only 28% of the bilingual parent group relied on written messages and the media. Also, 10.9% of the SEI parent group, as opposed to only 4.7% of the bilingual parent group, were informed by other parents and relatives.

Hypothesis Five

The results show no statistically significant difference between the two groups on the mainstream option. Hypothesis Five cannot be rejected. As Table 6 shows, there are only significant differences in items 13 and 17.

The lowest mean in this area was 2.00 and the highest was 2.51, indicating less agreement by both groups. On the question of for whom the mainstream classes are designed (item 13), there is a significant difference, with the bilingual parent group agreeing more with the statement. The results for item 17 indicate significantly greater agreement by the bilingual parent group.

Hypothesis Six

The results show statistically significant differences in the area of the effects of submersion as illustrated in Table 7. Hypothesis Four is rejected, as measured by item 18. This was an open-ended question: "How would a Spanish-speaking child do in a mainstream classroom?" Responses were placed into six categories and percentages were used to determine significant differences. Table 7 illustrates the findings.

Table 6
Results of Numerical Responses on Mainstream Program

ns = No significant differences

Table 7
Results of Open-Ended Responses on English Language Submersion

ns = No significant differences

No significant differences were found in (b) and (e). However, more than twice as many bilingual parent group members as SEI parent group members agreed with (a). The opposite was true for (c), where SEI parents outnumbered bilingual parents 2.5 to 1, and (d), where the ratio was almost 4 to 1.

Hypothesis Seven

The results show statistically significant differences between the two groups on the bilingual education option. Hypothesis Seven is rejected. Table 8 illustrates the results, which found each item to yield significant differences, with the bilingual parent group in greater agreement.

Table 8
Results of Numerical Responses on Bilingual Education Program

ns = No significant differences

These results clearly illustrate the area of greatest disagreement in the study. The statistical differences are greatest overall, and the significance of differences on each individual item are also greatest.

Hypothesis Eight

The results show no overall statistically significant differences between the two groups on reasons for choosing a program. Hypothesis Eight cannot be rejected, as measured by item 24. This was an open-ended question: "How did you arrive at your decision to choose a program for your child?" However, as Table 9 illustrates, there are significant differences in (d) and (e).

These results demonstrate that, while there were clear differences in numbers, only the differences in (d) are significant to this study, since the reasons vary in (e). This demonstrates that there were significantly more SEI parents who did not select a program, so it was selected for them by the school, in accordance with the law. Although not statistically significant, it should be noted that, while 14 more bilingual parents than SEI parents made a personal choice, twice as many SEI parents as bilingual parents were influenced by their partner or a relative.

Table 9
Results of Open-Ended Responses on Reasons for Choosing a Program

ns = No significant differences

Summary

The results of this study show statistically significant differences between the opinions of the bilingual parent group and the SEI parent group on language development, language acquisition, the effects of submersion in a mainstream classroom, and the bilingual education option. It shows no such difference in opinion on structured English immersion, sources of information, the mainstream option, and reasons for choosing a program. However, even in areas where no overall significance was found, there were specific numerical items or open-ended categories which yielded significant results. On the structured English immersion option, the two groups differed significantly on the primary goal of SEI and the one-year provision for mainstreaming students, with the bilingual parent group in greatest agreement. While none of the categories in sources of information yielded significant differences, the data indicates that the SEI parent group relied primarily on written messages, the media, other parents and relatives for their information. By contrast, the bilingual parent group gained more of its information at meetings. On the mainstream option, the two groups differed significantly on two items, with the bilingual parent group in greater agreement that: (a) Mainstream classrooms are for students with native English language fluency, and (b) former ELL students are grouped with native speakers. It is also important to note that a significantly larger number of SEI parents state that they did not themselves select a program for their children.

Conclusions

This study was conducted in school districts with well-informed parents, which provided an ideal laboratory environment for the comparison of parent opinion. While the school districts are distinct geographically and in size, they conducted themselves in similar fashion with regard to Proposition 227. In anticipation of the initiative's passage, they prepared themselves to meet each of its provisions. An examination of each district's documents verifies that, in addition to organizing the instructional programs to meet the new law's requirements, the parent/guardian waiver was also given full attention. In so doing, these districts met both the spirit and the letter of the law. As this study was conducted, then, it was apparent that nearly all of the parents/guardians were in close contact with the school district and had been given every opportunity to comprehend the effects of the new law on the lives of their children. The data demonstrates that neither group of parents was in disagreement with any of the statements in the interview. Indeed, the highest mean on any of the numerical questions was 2.95.

Yet, given these ideal conditions, significant differences were found. These differences exhibit some interesting results. In the area of language development, the parent groups differed in terms of the role of the primary language. SEI parents rejected the concept of a common underlying proficiency, preferring to view the two languages as separate, without the existence of transferability. These parents, therefore, perceive that English language proficiency is far more important than conceptual development. These same parents, then, consistently favor immersing ELL students in English as quickly as possible. They wholeheartedly reject bilingual instruction for their children, while at the same time, they significantly disagree with school district descriptions of these programs. While SEI parents realize that ELL children who are "submerged" into mainstream classes may have great difficulty, they do not believe they will necessarily fail. Bilingual parents, on the other hand, agreed more strongly on the role of the primary language and bilingual instruction as a means of acquiring English, believe that the submersion of ELL students results in failure, and were in very strong agreement with the school district's description of bilingual instruction.

As the results show, even in areas where there was no overall significant difference, these two groups differed significantly on specific items and categories of open-ended questions. Bilingual parents exhibited a greater understanding of English learning as the primary goal of the SEI option, as well as the new law's provision for mainstreaming ELL students after one year, as described by the school districts. SEI parents agreed with school district descriptions to a significantly lesser degree on the role of mainstream classrooms and their placement of former ELL students. The SEI parent group had significantly more people who became informed by indirect means and were less involved in making the decision on the placement of their child(ren).

The role of parents in determining the academic success of their children cannot be overstated. Reese et al. (Macías and García Ramos, 1995, pp. 205-230) found that Latino parents' academic expectations for their children are linked to the children's classroom performance. Parents who understand the educational options for their children and expect success from their children are likely to see them experience success. The parents of ELL students must be involved in school discussions about the value of the native language and allowed to affirm the legitimacy of their children's discourse (Banks and Banks, 1999, p. 394). The schools should not peddle the archaic notion to parents that their children's use of Spanish is a problem, even as they simultaneously acquire English at school, in compliance with state law. Instead, states Nieto (1996, p. 188), the schools should recognize that the way in which teachers and schools view (the language of their students) may be even more crucial to student achievement.

Recommendations

The implications of this study are many, with the first being that every school district should be as thorough as those in this study in complying with EC 300-340. At the very heart of this new law is the right of every parent or guardian to choose the program option which they believe is best for their children. This, of course, cannot be accomplished fairly unless they are fully informed, without bias on the part of school and district personnel. Parents need to understand the options as well as be informed about their child's progress within his/her existing program.

Since this study makes it clear that less well-informed parents relied on written information and the media, school districts should make every effort to meet directly with parents and provide accurate information. However, school district board policy should be very clear and implemented consistently in each school, without regard to the personal philosophies of school administration or staff. All school personnel should be encouraged to accurately and objectively discuss educational program options.

It is highly likely that more highly informed parents in these three school districts chose bilingual education because their bilingual programs had been successful in the past. While this is true for many other school districts in California, it has not been true for others in which parents may not have been given this opportunity, even if the school districts had successful bilingual programs. In all fairness to the ELL students of our state, the successful records of bilingual programs, including those of the three school districts in this study and others, should be shared throughout the state. Until parents are provided these options for their children, school districts will not have complied with the new law.

It is also highly recommended that the State Department of Education increase its compliance reviews to assure that all school districts make sincere, concerted efforts to educate ELL students. While programs which assure that each ELL acquires English language skills as efficiently and effectively as possible, school districts must also be required to assure access to the core curriculum. As this study has made clear, fully informed parents will not settle for less.

References

Baker, C. (1996). Foundations of bilingual education and bilingualism (2^nd ed.). Avon, England: Multilingual Matters.

Banks, J., & Banks, M. (Eds.). (1999). Multicultural education: Issues and perspectives (4^th ed.). New York: Wiley and Sons.

Casanova, U. (1991). Bilingual education. Amsterdam: John Benjamins.

CDE, California State Department of Education. (1998). Emergency procedures for Proposition 227. Sacramento: State of California.

California State Department of Education. (1999). Language Census Report. Sacramento: State of California.

Crawford, J. (1992). Hold your tongue: Bilingualism and the politics of "English Only." Massachusetts: Addison-Wesley.

Dicker, S. J. (1996). Languages in America. Clevedon, England: Multilingual Matters.

Faltis, C. J., & Hudelson, S. J. (1998). Bilingual Education in elementary and secondary school communities. Massachusetts: Allyn and Bacon.

Kerper, Mora (2000). Jumping to conclusions. The Multilingual Educator, 1 (1), 20.

Krashen, S. D. (1996). Under attack: The case against bilingual education. Culver City, CA: Language Education Associates.

Macedo, D. (2000). Decolonizing English only: The democratic power of bilingualism. In Z. F. Beykont (Eds.), Lifting every voice: Pedagogy and politics of bilingualism. Cambridge, Massachusetts: Harvard Education Publishing Group.

Nieto, S. (1996). Affirming Diversity: The sociopolitical context of multicultural education (2^nd ed.). New York: Longman Publishers.

Quezada, M. (2000). And the beat goes on. The Multilingual Educator, 1 (1), 17-18.

Reese, L., Gallimore, R., Goldenberg, C., and Balzano S. (1995). Immigrant Latino parents' future orientations for their children. In R. F. Macías, & R. García (Eds.), Changing schools for changing students: An anthology of research on language minorities, schools and society.

Acknowledgements

The author is grateful to School District Superintendents: Dr. Alfonso Anaya, Richard Duarte, and Dr. Magdalena Carrillo Mejia, Bilingual Coordinators: Jeanne Herrick, Stephanie Purdy, and Joanne Slater, and Principals: Alex Bennett, Tom Donfrio, Ileene Gershon, Jerry Jepsen, Dennis Johnson, Todd Macy, John Olague, Ruben Pulido, Dr. Roxanne Regules, and Tony Zubia, as well as their School Secretaries and other Administrators. I would also like to thank Project Secretary, Tawnya Conradi, to whom I owe a great deal for her work on my study. Without their permission and gracious assistance, this study would not have been possible.