Expanding the Knowledge Base on Teacher Learning and
Collaboration: A Focus on Asian American English Language
Learners

Final Report: Project 2.2

Principal Investigator:

Ji-Mei Chang
jmchang@email.sjsu.edu
San Jose State University
Department of Special Education

Co-author: Ward Shimizu
San Jose State University
Department of Special Education

Project Period: Fall, 1998-Spring, 2001

References

Appendix A: Project Related Publications and Dissemination

English Publications Introduced CREDE Standards and Project 2.2

Conference Presentations Introduced CREDE standards and Project 2.2 in English

International Project: Replicating Project 2.2's Research Model in Taiwan

Publications and Presentations

Chinese Publications that Introduced CREDE Standards and Project 2.2

Introduction

This school-based research was conducted as an effort to address educational issues confronting English language learners of Asian descent who were not stereotypical model minority students and were placed in the sheltered program in the present study, hereafter, the target students. In general, the target students were expected to succeed in school despite having limited English language proficiency (LEP), varying levels of teacher and peer support in largely sink-or-swim classroom environments, challenging economic conditions, and/or having school identified mild learning disabilities. These educational conditions do not bode well for our target students' futures, particularly given the push for higher standards for graduation and the usage of standardized test scores to determine high school graduation eligibility. At best, the target students will enter the job market without vocational training because without a high school diploma, they do not qualify for vocational school. Therefore, this research project endeavored to provide the target students with a glimmer of hope and the possibility of building a better life within and beyond school.

This final report is organized in five sections: background and purpose of the study, research goals and objectives, research design, results and discussion, and conclusion and implications.

Background and Purpose of the Study

In 1991, the Principal investigator (PI) conducted a field-based study to explore a home-school-community based conceptualization of Chinese American students with school-identified learning disabilities (LD) (Chang, 1993; 1995a). The study was conducted in urban and inner city schools in Northern California. The findings revealed that when elementary-aged Chinese American students were LEP, LD, poor, and enrolled in an inner city or urban school, they were very likely to experience missed learning opportunities on any given school day. Furthermore, this study revealed that a majority of these students remained LEP and weak in reading comprehension throughout their middle school education (Chang, 1995a).

In the 1995-96 school year, the PI received funding from the U.S. Office of Special Education Programs to conduct a collaborative action research project with six LD resource specialists at six inner-city elementary schools in the district where the PI conducted the 1991 study (Chang, 1996). With the participating school district's administrative support, we provided ongoing district-based professional development support to implement Collaborative Strategic Reading (Klingner & Vaughn, 1998), a extension of the reciprocal teaching model (Palincsar & Brown, 1984). In addition to the six LD resource specialists, over 40 students with LD participated. The students came from diverse ethnic and language backgrounds, such as Asian, African, and Hispanic. We also attempted to forge teacher-parent partnerships to help participating parents/siblings learn the strategies and use them while reading with their children in order to generate more home practice (Chang, Shimizu, & Liu, 1997; Chang, Hernandez, & Lai, 1997). Over the course of the study, we learned the following key findings:

• Teaching students to apply a set of reading comprehension strategies may result in the procedural application of the strategies, but not necessarily comprehension of the text. Based on audio and video transcripts of the students, we observed a high percentage of instructional time being devoted to procedures, such as whose turn to speak, which cue card to use, or how to use the cue cards. The actual discussion of the contents, specific clarification of vocabulary, or concepts was not a prominent part of the dialogues.
• We observed that the reading comprehension strategies demonstrated a heavy reliance on verbal exchanges and did not provide meaningful alternatives (e.g., graphic organizers, real objects, role-play, or drawings, etc.) to jointly construct meaning from text.
• District- or school-based support for ongoing professional development activities was critical when participating teachers systematically learned and integrated a research-based reading comprehension intervention program designed for target students. In this study, the participating district assigned a specialist to coordinate the intervention team across six elementary schools.
• The reading intervention study conducted within a resource program by a team of resource specialists and his/her instructional associate lacked an avenue to help participating students systematically transfer the reading comprehension strategies acquired in a pull-out based special education resource program to their homeroom during other content area reading activities. To help target students transfer the reading comprehension strategies and apply them during their typical day within and beyond school, it was very important to have planned collaboration between special education and general education teachers as well as between special education teachers and parents/siblings of the target students.
• Training workshops for parents/sibling need to be offered more than once, so they have time to practice at home and return for more feedback. By engaging participating parents in hands-on cooperative group training activities using the exact steps that their children used in the classroom further enhanced their ability to read with their children at home.

The key findings obtained in the 1995 study were important features in the present study. The present study engaged participating teachers in a collaborative effort to develop a responsive language-literacy intervention model and forge partnerships with their students' parents and/or family members. The specific design considerations were addressed in the section on Specific Research Considerations.

The importance of the present study can be summarized briefly in the following areas. First, the process of forming a research-based professional development model that fostered teacher collaboration and constructed home-school partnerships took place in an authentic school setting, i.e., the context of a Title I middle school. The middle school had a high enrollment of English learners and resembled the student and teacher demographics of numerous schools in urban areas across the country. Second, the design and research objectives explored further ways to provide meaningful language-literacy intervention initiated by classroom teachers. Third, through participant observations, the researcher documented the processes involved in bridging theory and practice to further explore how the Five Effective Standards for Pedagogy proposed by the Center for Research on Education, Diversity, and Excellence (CREDE) were applied beyond classroom into teacher education and home practices.

Research Goals and Objectives

This project had two major interrelated goals. The first goal was to facilitate teacher collaboration and learning to enhance the target students' language-literacy development. The participating teachers field-tested and advanced a responsive language-literacy intervention program within an inclusive environment designed for the target students' Title I middle school. Specific research objectives included (1) assisting teachers in providing a responsive language-literacy program for the target students; and (2) providing school-based professional development opportunities guided by CREDE standards. We believed that by immersing participating teachers in context where they receive assisted performance within their zone of proximal development (ZPD) (Vygotsky, 1978; Tharp & Gallimore, 1988), they might be more likely to adopt the same type of instructional approaches, or adopt CREDE standards, when working with their students.

The second goal was to examine how CREDE's Five Standards for Effective Pedagogy, which are based on a sociocultural theory of education, were applied across three interrelated contexts: professional development, classroom intervention, and school/home/community partnership. The major findings were organized by infusing research question-findings within the three interrelated contexts.

Research Questions

1. To what extent did the team of teacher-leaders facilitate the language and literacy development of the target students in the following areas:

• concept development,
• participation in constructing knowledge from print,
• perception of themselves as readers, and
• forging partnerships with their parents?

1. What were the features of collaborative intervention methods as demonstrated by the team of teacher-leaders?
2. What were the collaboration processes and contexts that fostered target students' language/literacy development and content area learning?
3. What were the contexts that enhanced the participating district's ability to institutionalize a research-based professional development model that focused on educating Asian American and other English learners who were at-risk and/or learning disabled?
4. What were the mediation processes and contexts that forged partnership with parents?

Research Design

Since the participating teachers were dedicated field practitioners and promoted teaching transformation in their classrooms, their active involvement in all aspects of research processes was deemed important. Therefore, we implemented collaborative action research (Sagor, 1992) with the participating teachers. Sagor (1992, p. 9) defined collaborative action research "as any effort toward disciplined inquiry." Our team defined the present research project as our systematic and collaborative way of learning how to improve the education of particularly challenged students in the classrooms. In addition to being a systematic and collaborative approach, collaborative action research also provided the flexibility for adopting qualitative and quantitative methodologies to address our research questions.

At the outset of this three-year collaborative action research process, we agreed to assume the roles of learner, researcher, and peer supporter. As learners, we remained open and flexible to continuous learning. As researchers, we gathered data through various measures, triangulated multiple data sets, and strived for accuracy. As peer supporters, we listened with understanding and empathy and thought interdependently; hence the project team exercised and applied productive habits of mind (Marzano, 1992) for both personal and professional development. Ultimately, these three roles enabled us to improve teaching, learning, and supporting each other within and beyond the project through joint productive activities and assisted performance within our ZPD.

We employed a mixed-method design. We used quantitative methodology to measure student achievement to address the research questions. In order to document the effectiveness of the classroom language-literacy intervention program, the team tested the statistical significance among groups of target students from whom we collected two sets of standardized achievement tests scores (before and after they participated in the study). The school district routinely collected these achievement tests scores; for schools, raising such test scores was their most demanding task.

We also adopted a qualitative methodology, including participant observation, to study how CREDE's standards may be applied across the three contexts: professional development, classroom intervention and home practices. The PI assumed the role of an active participant observer (Spradley, 1980), initially serving as an observer, and then co-teaching when it became feasible and meaningful. Assuming such a role, allowed the PI to achieve the following research objectives: (1) gain the acceptance of participating teachers and students as well as build team morale, (2) co-construct or modify the research activities based on observations and mutual understanding, (3) learn more about the middle school's and classroom's sociocultural rules of behaviors and expectations, (4) monitor the levels of fidelity of treatment, and (5) document how the proposed intervention program evolved over time. In Year III, we were only able to conduct a follow-up study to interview participants in the Year II study. The planned activities for classroom intervention in Year III were hampered by administrative changes at the school and the sheltered classroom was dismantled.

Design Components for Three Contexts

The present study explored how CREDE standards were used in three interrelated contexts to maximize the effect for professional development, language-literacy development of target students, and forging partnership with parents, sibling or friends of the target students. Within each research context, the following components formed an integral part of the research activities in specific context.

Professional Development Context

CREDE standards. The training activities were aligned with CREDE's Five Standards for Effective Pedagogy (Dalton, 1998; Rueda, 1998). One goal of this project was to study how CREDE's Five Standards might be applied to assist performance in the participants' ZPD as we engaged in the research activities. A decision was made early on for the team to explore and model how such standards would be implemented in the research team's professional development activities throughout the research phases. In order for participating teachers to provide teaching-learning activities guided by these standards, the research team's ongoing planning and training activities must naturally incorporate such practices, so we would all see and experience what it looks like to use the standards this project was advocating.

Planes of Analysis. Rogoff and her associates (Rogoff, 1995; Rogoff, Baker-Sennett, Lacasa, & Goldsmith, 1995) proposed that development is through participation in sociocultural activities and also a collaborative process. Hence her theory provided the team another perspective to closely examine our professional development context, activities and process. Furthermore, Rogoff's (1995) planes of analysis were a means with which to analyze factors observed from three planes: personal, interpersonal, and community. A phenomenon is closely examined by foregrounding the plane in which the phenomenon occurred and scrutinizing the phenomenon's relationship to other planes in the background. Hence, the planes of analysis helped us uncover various factors within the data sets that affected the district and school administrators' ability (1) to sustain the participating teachers' collaborative effort to support the target students and (2) to institutionalize our project's professional development and classroom intervention model for the target students.

There are three planes in Rogoff's framework: community, interpersonal, and personal. Factors observed in one plane may not be understood fully without analyzing the relationship with other two planes. The community plane: Since the school is subject to district and state mandates, in this study, this plane signifies the district and school contexts, including at least, the shared rules, values, priorities, dominant teaching methods, and/or administrative priorities and practices in the school. The interpersonal plane: This plane includes faculty support for ongoing collaboration and built-in scheduling for team-preparation, an administrative structure that supports on-going communication with parents and a home-school partnership, and/or a shared decision-making process over school-related issues, etc. Collaborations occurring in this plane were observed between teacher-teacher, teacher-instructional associates, teacher-administrator, teacher-family members, and/or teacher-parent-community members. The community and personal planes, of course, influence activities observed in this plane.

Classroom Intervention Context

CREDE Standards. CREDE's Five Standards were adopted as a guideline to strengthen student and teachers' ability to focus on making meaning from print or declarative knowledge, rather than focusing on procedural knowledge as we had observed in the previous study conducted by Chang (1996) in which participating special education resource specialists focused more on the procedural aspects of a collaborative strategic reading pedagogy, such as students' turn taking, specific role of each member in a cooperative reading group, as well as the use of exact scripted speech.

Teacher contribution. The classroom intervention model was co-constructed with the participating teachers to generate an intervention model that facilitated content learning, language, and literacy development in a sheltered classroom environment for target students in a Title I middle school. Many participating teachers were experienced mentor teachers in the district, and they know what worked well for target students. Teachers were encouraged to systematically infuse what worked in promoting target students' language-literacy development.

Theory of Multiple Intellilgences (MI). The overall intervention model incorporated current findings on ways to use the theory of MI as tools to assist both teachers and students to tap fully into everyone's multiple abilities beyond linguistic means in our attempt to promote language-literacy development (Chang, 1999; 2001b). Although MI was never intended to shape curriculum, instruction and assessment in the field of school education, Kornhaber and her associates (Kornhaber, 1994; 1999; Kornhaber & Krechevsky, 1995; Kornhaber, Fierros, & Veenema, 1998) at Harvard University's Project Zero systematically examined how MI worked as a reform movement over the years. They studied 41 schools using the theory of MI and found that "MI theory works as a grassroots reform by initially resonating with and validating educators' existing philosophies, beliefs, and practices. At the same time, it provides educators with a useful tool to organize and develop their own practice." (Kornhaber, 1999, p. 184). Such practices included, but are not limited to, the teachers' emphasis on student strengths, project-based learning or curriculum integration, use of diverse entry points as attention getter (Gardner, 1999), and arts-infused curriculum, etc.

Curriculum Integration. To maximize students' ability to practice, apply and transfer specific strategies taught in the project, the intervention was implemented within an integrated context, that is to say in both language arts and history blocks.

Reciprocal Teaching (RT) Strategies. The research team first discussed adopting the Collaborative Strategic Reading (CSR) model (Klingner, 1998; Chang, Shimizu, & Liu, 1997), patterned after a previous study (Chang, 1996). Two elementary resource specialists, who participated in the 1996 study, gave a CSR training workshop. However, shortly after workshop, the present research team decided to continue using the reciprocal teaching methodologies for two reasons. First, their district had already adopted and provided RT training for use in classrooms. Second, two of the participating teachers were the school designated RT trainers and mentor teachers. Based on the RT design, the teachers taught their students to use a set of four reading comprehension strategies within small groups, while instruction was carried out initially using a scripted dialogue approach.

School-Home/Community Partnership Context

CREDE Standards. The planning and implementation process were guided by CREDE standards as a means to yield positive and meaningful partnership. The participating teachers generated school-home/community partnerships aimed at supporting the target students' transfer of effective language-literacy development strategies from the classroom to home practice while reading with parents/guardians, siblings or friends. Using CREDE's standards as guidelines, the participating teachers could further model and explain CREDE's standards to participating parents/sibling/friends and its effect on student learning.

Theory of MI. MI as tools was used to demonstrate areas in which parents and individuals could value a child's abilities within perceived at-risk factors for poor school performance or school-identified learning disabilities, explore career options, adopt multiple pathways for home teaching or practices for school skills.

Reciprocal Teaching Strategies. A handbook was designed to provide home reading practices. The handbook included a step-by-step guide for each of the four reading comprehension strategies as well as a guide for coach-partner reading in the cycle of reading for home reading support.

Research Phases

In the Year I and Year II studies, the research activities were carried out over two phases. In Phase 1, participating teachers engaged in training activities as well as in managing logistics, such as obtaining parental permission. In Phase 2 the classroom intervention was implemented. During Phase 2, the PI participated in the classroom one to three times a week. In Year II, we added Phase 3 -- Team-Facilitated Family Literacy Nights. The research phases were patterned after the PI's 1996 study described earlier in this report, in which the basic design was to gradually shift the use of proper reciprocal teaching model's reading comprehension strategies from the teacher to the participating students. The goal for this project was to help students incorporate such strategies when reading course material in school. Through participant observation, the PI documented the process and identified how each of the major classroom intervention elements identified in research design were integrated and implemented to co-produce a responsive teaching-learning environment for students and teachers over the course of the studies project.

Research Setting

The research site was a Title I middle school located in Northern California with a high enrollment of students from diverse ethnic and language backgrounds. The PI contacted the district for assistance in locating just such a school. District administrators recommended the participating school in April 1998. Upon the PI's first visit to the school, the ELD coordinator/teacher and special education resource specialist (RS) acknowledged the need for an intervention program to enhance target students' language and literacy development. These two teachers were instrumental in supporting and recruiting colleagues and students throughout the project.

The ELD teacher and RSP indicated that they had seen many target students failing in their middle school. For this very reason, the ELD teacher and three of her colleagues initiated a sheltered program for incoming sixth graders at this middle school specially designed to meet the needs of English learners who performed below all other incoming fifth graders based on a set of selection criteria. These teachers were not surprised when the qualifying students for this very first program were primarily Asian or of Asian American descent.

Research Participants

• Participating Teachers

There were two groups of teachers in the present study:

1) Year I teachers included one English language development teacher (ELD Teacher A), three special education resource specialists (including SE A), and one first-year general education teacher (GE A).

2) Year II teachers included ELD Teacher A, one ELD Math teacher, one ELD Science teacher, two special education resource specialists (including SE A from Year I and a new specialist) and two new general education teachers (including GE A). Three Year I teachers continued through Year II of the project.

Each of the participating teachers received three-units of credit from San Jose State University for each semester they participated. Their tuition was supported from another professional development grant directed by the PI.

• Participating Students

The second group consisted the following student participants:

1) one group of English learners enrolled in the sixth grade Sheltered program in Year I (n=26)

2) one group of English learners in the sixth grade Sheltered program in Year II (n=20)

3) one group students with special needs who participated in Family Literacy Nights and other classroom activities (n=10)

We only received parental permission for students in the sheltered program to participate in the full range of research activities. Since we had limited parental permission for students enrolled in other classrooms we did not include their data in the final analysis. In general, all students were permitted to participate in language-literacy intervention activities implemented by his or her own teacher as a part of classroom routine work.

• Participating Parents/Guardians/Siblings/Friends

The third group of participants was parents and family members who attended the project-sponsored Family Literacy Nights in Year II. In the first Family Literacy night, we had over 60 participants consisting of parents and family members. Only one student and her family did not attend the first Family Literacy Night.

Instrumentation: Data Collection and Analysis

We adopted a multi-method approach to address the five research questions guiding the studies. Data collection and analysis related to instrumentation were organized in three sections: (1) student measures, (2) teacher measures, and (3) measures for the Family Literacy Nights.

Student Measures

Reading Interviews: Design and Analysis. Interview questions were listed on a form, and students wrote responses to eight questions. Two of the questions had two parts. This measure was adopted from Goodman, Watson and Burke (1987) to address Research Question 1 regarding students' concept development, their perception of themselves as readers, and forging partnerships with their parents. We collected Year I students' responses to the same reading interview three times: at the beginning and at the end of the intervention in Year I, and a follow-up interview in Year II. For Year II students, we conducted one reading interview in Year II and one in year III.

The interview forms were divided into four stacks according to the date they were conducted, and then duplicates were made of all the forms. Each stack of forms then was cut into strips so that each strip contained a question and the student's response to that question. The strips were sorted by question, then themed by the PI, the research associate, and a research assistant. The theming activity was based upon the one presented in Vaughn, Schumm, and Sinagub (1996). The research assistant was given training in theming student responses prior to this activity. After completing this task for question one, they compared their results with regards to the different themes appearing in students' responses. When there were discrepancies between theme categories, they discussed and gave reasons for their categories, and reached a consensus for either keeping the category or subsuming the category into another themed category. This process was repeated until all four stacks were completed.

Stanford Achievement Tests Ninth Edition (SAT9): Design & Analysis. To address "participation in constructing knowledge from print" and "concept development" we adopted the district and state test, the SAT9. These tests were routinely administered to each grade in the spring semester. The participating students in the sheltered program had two sets of scores, one collected in spring of 1998 while the target students were in fifth grade, and one collected in April 1999, after our designated intervention in the Year I study. We conducted a multivariate analysis of variance and the follow-up univariate repeated measures analysis of variance using SPSS v.9 for Year I and SPSS v. 10 for Year II scores.

Work Samples: Design & Analysis. Writing samples of various types were collected over the years as evidence to support student participation. A set of 24 Roman Magazines, produced by Year II students was collected as evidence of their concept development and ability to utilize instructional language for school learning. The Roman Magazine was a major project-based learning activity that integrated concepts introduced in language arts, history, science, and math. The contents and artistic presentation of students' Roman Magazines were analyzed as evidence of student learning and concept development over time as they were immersed within an integrated learning environment across language arts and history as well as science and math activities when appropriate. ELD Teacher A conducted the analysis according to a set of scoring rubrics.

Teacher Measures

Structured Interview. The interview obtained baseline information regarding participating teachers' interpretation of CREDE standards and MI as tools. It also allowed us to gather information related to how they perceived their classroom practices were aligned with CREDE standards. Teacher responses were collected during Phase 1 of the study as part of the course work for the three-units of university credit. Their responses were transcribed, and two research assistants did the content coding (Miles and Huberman, 1994) and the researcher analyzed for the data for any patterns. This baseline information was compared to their instructional activities and materials collected at the end of the study.

Unstructured Interview. Unstructured interviews were conducted over time during monthly research meetings, e-mail correspondences, classroom observations, and hallway discussions concerning their observations, insights, or other changes regarding the studies. The data were collected regularly to help the team make changes in a timely manner and solve problems.

Videotaped Interview. Semi-structured interviews were videotaped. The purposes of collecting these responses was to document (1) the participating teachers' perception of how they implemented the three components of the intervention program, and (2) evidence as they implemented the intervention program in their classroom. The interviews were conducted in their classrooms at the end of Year I and Year II. The responses were transcribed in summer 1999 and 2000. The constant comparative method (Glaser & Strauss, 1967) was used to examine the sets of transcripts in order to reduce and categorize data into themes related to predetermined categories. These categories were related to the three components of the language-literacy intervention program: RT, CREDE standards, and MI applications. The researcher grounded the codes in participating teachers' terms to document their perception of the use of these components that formed the basis of the language-literacy intervention program as it evolved

After the responses were categorized, the PI had discussions with the participating teachers about their perception of their use of the three components and verified though videotapes. This process helped the team reach a shared understanding about the integration of three components to form the language-literacy intervention program for the target students. It also helped teachers voice their concerns and perceived obstacles in carrying out instructional conversations (IC) with small groups within their classroom on a regular basis.

Videotaped Classroom Activities. We collected on-going evidence of the classroom implementation of the intervention program. The video tapings also monitored levels of administering the major components of the proposed classroom-based language-literacy intervention program, e.g., CREDE standards, MI as tools, etc. on days when the PI was unable to participate in the classroom. The participating ELD teachers routinely videotaped and/or took digital photos of classroom events as well as documenting student progress of their use of the English language. In general, the participating students were comfortable with being photographed and videotaped; and teachers received parental permission to video tape and photograph ongoing student performance.

Participant Observer's Field Notes. The PI kept field notes to document entire class periods in sequence as she participated in classroom activities. The PI kept two columns of notes. On the left side, Note Taking, she recorded the events as observed on a minute-by-minute sequence. On the right side, Note Making reflected the PI's thoughts and questions that were later clarified with the teachers. Some dialogue and exact quotes were recorded when feasible. Field notes were taken one to three times per week in Year I, or a total of 30 field notes. Field notes were taken on 25 different occasions during the Year II study. The classroom observations generally covered an entire period of language arts or history, an average of 45 minutes in length. In addition, some class periods were tape recorded and transcribed shortly afterward.

The units of analysis for PI's Field Notes centered on how the teachers implemented each reading-language intervention component either in small groups or the entire class, such as any time the class used RT strategies, MI as tools, and/or other instructional activities led by an ELD teacher to enhance reading, writing and/or language development during the observed instructional blocks. The PI conducted a grounded analysis of field notes and categorized the units into three pre-determined categories: RT strategies, MI as tools, and each of five CREDE standards as an ongoing process. For example, activities introducing a questioning tree to teach students generating how to generate on-the-surface and under-the-surface questions were categorized as RT strategies; role plays before or after reading a passage were categorized as MI as tools; scanning unknown words before reading were categorized as RT strategies; paired readings were categorized as RT strategies and CREDE Standard 1; teaching vocabularies through visual cues were categorized as MI as tools and CREDE Standard 2; and connecting events occurred in reading passages to students' daily life were categorized as CREDE's Standard 3.

The patterns then were verified periodically with the teacher to clarify the intent and coding categories. This ongoing process of data collection and analysis also helped the team and impacted the research processes. For example, early in Phase 2 of the Year I study, noticeable patterns surfaced in the field notes prompting the team to move from using RT strategies in a small group setting to paired or partnered reading, with a coach and a partner. Given the diverse language abilities, the paired approach seemed to be more appropriate for a middle school instructional block. Consequently, the team promoted a cycle of reading processes, rather than a more linear process of having pre-, during, and after-reading stages. From the ongoing data analysis process, it became clear that there was a need for an in-class IC demonstration lesson conducted among a small group of students. A specialist from CREDE Central conducted this lesson and helped clarify an integration of CREDE standards, MI, and RTM for the research team.

SPC. SPC is a rubric for assessing classroom enactments of CREDE's Standards for effective teaching (Hilberg, Doherty, Tharp, & Epaloose, in press). The levels of enactments per each of the five standards (Dalton, 1998) were ranked in the following categories:

1) Not observed–The standard was not observed within a teacher-designed activity.

2) Emerging–one or more elements of the standard were enacted within an instructional activity.

3) Developing–this was a partial enactment of the standard embedded within teacher-designed activities. For example, Students collaborate on a joint product (Standard 1: JPA.).

4) Enacting–This was a complete enactment of the standards within teacher-designed instructional activities. For example, teacher and students collaborated on a joint product either within small-group or fully inclusive whole-class activities.

5) Integrating–This was the highest level reflecting a teacher demonstrating skillful integration within a single activity to enact two or more CREDE standards simultaneously.

In Year II, we videotaped at least two segments of each teacher in their classroom while working with students. The PI reviewed the tapes with each teacher privately. The PI and teacher then evaluated the teacher's performance using CREDE's SPC. In a qualitative assessment of their own teaching practices, the participating teachers compared their teaching performances to each CREDE standard. They thought these standards were commonsensical to good teaching, and they also reported how they were implementing the standards in their classrooms.

Measures for Family Literacy Nights

• Video and Video Transcripts. Each of the four Family Literacy Nights was video taped by two hired eighth graders who attended the participating school. On the first night when the activities were conducted in three adjoining classrooms, each room was recorded by one video camera. When the remaining three Family Literacy Nights were conducted in the school library, there were two video cameras per night placed on either side of the library to record the entire event. Each of the tapes was transcribed according to a pattern introduced at the Each Teach Fall Conference sponsored by CREDE (Rutherfold, Scirota & Majalca, 1999) to document the major activities that took place each night. The unit of analysis focused on how teachers enacted JPA in each of the activities provided at the Family Literacy Nights using Standard Performance Continuum (SPC).

• SPC. This tool was used by the team to analyze to what extent we have enacted the Standard 1: Joint productive analysis as the key foundation for the four Family Literacy Nights.

• Participating Family member's Evaluation Form. This form collected feedback and suggestions from participating adults at the end of each Family Literacy Nights. Each response was typed and tallied. Two research team members coded the responses and categorized them into patterns.

• Participating Student's Evaluation Form. This form collected feedback and suggestions from participating students at the end of each Family Literacy Night. Each response was typed and tallied. Two research team members coded the responses and analyzed the data using the same theming method as when we analyzed feedback from parents or family members.

• Participating Parents' Feedback on Parents' Handbook. A form was designed and placed in the handbook to collect feedback and suggestions from participating adults at the end of second Family Literacy Nights. Each response was typed and tallied. Two research team members code the responses and analyze the data using the same theming method as when we analyzed the students' responses.

• Observers' Feedback Regarding CREDE's Five Standards. This device was designed to achieve three purposes (1) a shared understanding about each CREDE standard used in four Family Literacy Nights, (2) preparing the research team to plan, implement, and evaluate the jointly produced family literacy night activities, and (3) generating feedback for the first team's presentation and interaction with participants.

• Student Behavior Log. This device was adopted from Lazear (1999) as a means for family members to rank their child's typical behaviors observed outside of the classroom. The Behavior Log yielded a profile that may have described their child's multiple intelligences. The data were not collected and analyzed for research purposes; data were primarily used to help family members connect with two concepts we introduced in the second Family Literacy Night: (1) valuing their child's multiple abilities beyond school diagnosed levels of English language proficiency and/or learning disabilities, and (2) adopting multiple pathways to assist their child's learning of new vocabulary and concepts introduced in language arts, history, science or math.

• My Multiple Intelligences. This booklet was designed in the Year I study when we explored ways to adopt MI as tools as a part of the language-literacy intervention program within the sheltered program. The booklet contained a ten-item checklist per the eight intelligences identified by Gardner, and was co-constructed among team members and field-tested with participating students in Year I. It was used specifically as an introductory activity only. Since the checklist and language were student-oriented, we used it with the student group on the second Family Literacy Night when we broke into two adult and student group activities. The data in this booklet was not collected or analyzed for research purposes; it was used primarily as a conversational focus with their parents or other family members regarding their children's perceived multiple abilities.

Results and Discussion

Professional Development

Professional development activities were central to the project. Within school, teachers focused on classroom intervention; beyond school, they forged partnerships with participating students' families and friends. In this section, information is organized to address the following research questions and preliminary findings:

• What were the collaboration processes and contexts that fostered target students' language/literacy development and content area learning?

• What are were the contexts that enhanced the participating district's ability to institutionalize a research-based professional development model that focused on educating Asian American and other English language learners who were at-risk and/or learning disabled?

The findings related to both research questions were affected by diverse factors within the school and/or the district. To address these two questions, we found the research tool, planes of analysis (Rogoff, 1995) most helpful because a foregrounded action or factor observed in one plane may have its origins in another plane.

Contexts That Facilitated Teacher Collaboration & Student Learning

The collaborative process and contexts that facilitated the target students' language and literacy development can be presented from both the personal and interpersonal planes. Over the course of two years, we observed that all participating teachers were professional, respectful of one another, and committed to supporting their students in a Title I middle school. The participating school was recommended to PI through District Administrators based on two selection criteria: (1) High enrollment of Asian American English learners and (2) feasibility for teacher collaboration. By the time the study began, the middle school had already established a culture for teacher collaboration. One important support for collaboration activities was the team-planning blocks regularly scheduled for first period on Wednesdays.

The General Context. In Year I at this Title I school, we recruited five teachers–ELD Teacher A, SE A, GE A, and two special education resource specialists. ELD Teacher A and SE A recruited participating teachers in the Year I and Year II studies. In both Year I and Year II, we carried out training activities in Phase 1: Preparation of each study. The training activities focused on modeling, discussion, and modification of major elements in the proposed language-literacy intervention program. In Year II, ELD Teacher A recruited ELD Math Teacher and ELD Science Teacher from the same sheltered program to participate in the study; one of the special education specialists recruited a new general education teacher to whom she was assigned as a mentor. Outside of the sheltered program, we have two pairs of mentor-mentee teacher partners including SE A and GE A. Together, we explored the language-literacy intervention program and co-sponsor four evening Family Literacy Nights.

The Research Context. The initiation of the sheltered program was co-constructed by a group of four ELD teachers, including Teacher A, in the participating middle school in the school year of 1998-1999. Among the four teachers who initiated the sheltered program, two were experienced mentor teachers (ELD Teacher A and ELD Teacher B) and two were relatively new (ELD Math and ELD Science). Through personal commitment and collaboration, these teachers set out to provide early intervention for the lowest performing incoming sixth graders. ELD Teacher A and ELD Teacher B also served as co-coordinators for school's ELD program. The major curriculum and instructional activities in the sheltered program were co-designed by the four ELD teachers on a regular basis.

The four ELD teachers screened and recruited roughly 60 incoming sixth grade students prior to the fall 1998 semester and divided them into two ability groups based on test scores and overall English proficiency. ELD Teacher B who did not participate in the project taught the more advanced students and ELD Teacher A taught the group with the lowest scores. Over two consecutive school years between 1998 and 2000, the lowest performing incoming sixth grade English learners were all of Asian descent. Some of them also were referred and identified in school as having learning disabilities (LD), either by their elementary or subsequently by middle school teachers. ELD Teacher A taught the same cohort of students with ELD Math and ELD Science teachers. The focal point of the present research project was ELD Teacher A and her two classrooms, language arts and history classes, in which the target students were enrolled.

Context Required for Institutionalizing The Research-based Program

To address why the district was unable to institutionalize the project's professional development model and classroom intervention for target students, we analyzed the phenomenon in the community plane. Institutional support from the district office and school administrators was very strong in Year I, but changes in administrators at both levels brought shifts in priorities and agendas. In Year II, while planning and sponsoring the four Family Literacy Nights, we witnessed a parade of three substitute school principles. In Year III, the key sheltered program that was designed for incoming sixth grade English learners with the lowest performance profile, mostly the target students–was dismantled without consulting parents or teachers. By now, only ELD Teacher A, citing our research evidence showing positive student achievement, pushed to have the sheltered program reinstated for low performing English learners in future years. Sadly, all of the special education resource teachers have now left the district, citing a lack of administrative support.

The PI's observation revealed that not all administrators' priorities included supporting the target students. They also did not seem to have high expectations of target students' school performance. For example, one of the substituting principals heaped praise on a teacher when his regular education class raised their SAT9 scores even though the class had an average score in the upper 80^th percentile on their previous SAT9 tests.

However, no such praise or words of encouragement was given to either the ELD teachers or the sheltered English program when the target students made significant gain in both Year I and Year II studies. Target students scored below the 50^th percentile in SAT9; this might have contributed to the administrators' lack of enthusiasm. Much of the focus was on the scores themselves rather than the realistic and significant gain on student outcome contributed by the collaborative effort among the team of participating ELD teachers and target students over the entire school year.

Classroom Intervention

The ultimate goal of teachers' collaboration, learning, and professional development was to enhance student achievement through the designated classroom intervention. This section describes the features of the language and literacy intervention program and reports the positive results on student achievement. It addresses two related research questions:

• What were the features of the collaborative intervention methods as demonstrated by the team of teacher-leaders?

• To what extent were the teacher-leaders able to facilitate the language and literacy development of the target students in the following areas:

1. concept development,

2. participation in constructing knowledge from print,

3. perception of themselves as readers, and

4. establishment of effective partnerships with their parents?

Classroom Intervention Program

In the current educational climate–heavy state and district emphasis on standardized test scores–ELD Teacher A had to cover the many literacy skills beyond reading comprehension that are measured by SAT9 tests. Because the participating students had numerous and severe needs in overall language and literacy development, participating Year I teachers decided at the outset to explore and identify strategies, activities, and/or pedagogies that promoted language and literacy development among the low performing target students. The target students usually showed insufficient prior knowledge, limited written vocabulary, weak writing skills, poor self-concept, and selective attention spans for classroom tasks. Using a collaborative action research model, participating teachers worked together to design a classroom intervention program aimed at enhancing participating students' language and literacy development.

The following three sections address the first research question listed above–that is, they describe the collaborative intervention methods utilized by the ELD teachers in the study. These methods emphasized integrating literacy development into a wide range of activities; accordingly, observational data were organized to reflect how such integration occurred in students' typical classroom activities. Specifically, data is reported in the following areas: Curriculum Integration; integrating RT strategies & CREDE Standards; integrating RT Strategies, MI and CREDE Standards.

Curriculum Integration. ELD Teacher A regularly planned instructional units with an ELD Math teacher and ELD Science teacher. Taking themes from the sixth grade history/social studies curriculum (e.g., ancient Greece and Rome), these teachers supported students' reading, writing, listening, speaking, history, mathematics and science learning activities by carefully integrating them across classes. For example, when the history class covered early Rome in the district-adopted textbook, the language arts class read Mystery of the Roman Ransom (Winterfeld, 1971) and other relevant books and articles. In addition to covering the math and science content designated by district standards, teachers integrated themes, concepts, and vocabularies from the language arts and history units. For example, the science teacher helped students cook food as it was prepared in ancient Rome. Fortuitously, the three classrooms were adjacent and interconnected, so that even the students' physical environment reflected the integrated thematic instruction.

Integrating RT Strategies & CREDE Standards. Through field-testing, the RT strategies which were widely adopted in the district were integrated to form a part of the language-literacy intervention program. Specifically, four RT strategies were used to boost reading comprehension–predicting, clarifying, summarizing and questioning. Using RT strategies had several advantages: 1) RT was adapted to address CREDE's standards, 2) the teachers were familiar with RT methodology, 3) each subject area required some reading, 4) RT did not require drastic changes in any of the teachers' existing lessons, and 5) the RT strategy generating questions at the end of reading cycle provided a meaningful link between reading and writing activities. For example, Students were required to respond in writing for both text-implicit and text-explicit questions.

In Year I, the research team established a partner reading procedure built on RT strategies to help the target students engage in joint productive activities to make sense from print. Each pair of students was provided with a step-by-step guide identifying key activities for each stage of the reading process–before (preview/review, predict, scan for new words), during (clarify and summarize), and after (questioning and review). Occasionally, target students used the paired reading guide as a reference for prompting their memory of the next activity for each reading stage; hence this guide was amenable for target students to engage in paired reading. For our project, having a step-by-step guide also made it easier to explain the reading steps to parents or siblings for home practices.

At the beginning of the study, we were aware that classroom conversations associated with CREDE Standard 5, Teaching through Instructional Conversations (IC), differ fundamentally from scripted lesson talk (The National Center for Research on Cultural, Diversity, and Second Language Learning, 1995; Dalton, 1998). Knowing the differences between these two instructional methods, the team did not ask students to follow or recite the scripted statements printed in the partner reading procedure. Rather, the scripts were used as a guide only. IC was perceived as enlightening because it reminded teachers to solicit students' sharing of their prior experiences or knowledge about a topic being studied. ELD Teacher A then attempted to weave students' prior knowledge with current events or other related examples within the students' experiences.

For a variety of reasons we anticipated that carrying out authentic text-based IC as illustrated in the literature and videotape might be challenging in the sheltered classroom even though the classroom was arranged to facilitate small group activities. In both Year I and Year II, as judged by participant observations, ICs were not consistently enacted within small groups in language arts as a means to help individual students construct meaning from print. Typically ELD Teacher A set up paired reading, choral reading, or large group interactions to clarify words, summarize text or generating questions. She also sat by individual learners, listened to their reading, and provided occasional support.

In Year I, we invited a specialist from CREDE Central to demonstrate instructional conversation in ELD Teacher A's language arts classroom. ELD Teacher A thereafter tried this approach but eventually abandoned it. Instead, she set up several activity centers in her classroom and to facilitate small group instruction in reading. In addition, ELD Teacher A consistently worked in small groups to support target students while they worked on various projects during history blocks. The integration of language arts and history did allow ELD Teacher A to converse with students either individually or in small groups in order to help them link old and new knowledge or clarify misconceptions. Although ELD Teacher A did not strictly adhere to the guidelines of IC during reading activities in small group settings, helpful conversations with the target students occurred naturally whenever she worked with them. Her students enjoyed multiple opportunities to clarify concepts throughout a typical school day, in part because the ELD Teacher A also provided extra help before and after school hours.

In addition, ELD Teacher A consistently adopted techniques from the Dimensions of Learning, such as helping target students analyze perspectives. Hence, her integrated lessons provided her students opportunities to engage in many challenging activities through joint productive activities (CREDE Standards 1 and 4). Her students were skilled in carrying out the project-based learning activities not just for integrated language arts-history lessons but also for integrated language arts-science lessons. Such transfer of skills and information obtained from one subject to another was observed as one major benefit of curriculum integration in the sheltered program.

Integrating RT strategies, MI & CREDE Standards. Teaching vocabulary and language skills were two of ELD Teacher A's major instructional activities. In addition, she modeled proper syntax, semantics, and pragmatics. Over time, she consistently applied CREDE's Standards 2 and 3. For example, in the Before-Reading stage, she asked students to scan the text and write down unknown words. She then modeled various word attack skills listed in the Partner-Reading Guide. Since the target students were primarily English learners, she helped students clarify new vocabulary words through various means, such as relating the words to students' prior experiences or to current events. She also regularly demonstrated or modeled new vocabulary words and concepts whether these appeared in the literature book used in language arts or in the subject areas of history, math and science. Because the sheltered program was theme-based, and these students were in a cohort attending all subject matters together in math, science, language arts and history, some of them started to remind each other how some words showed up in other classrooms.

In the Before-Reading stage, all ELD teachers also adopted MI-oriented diverse and vivid entry points to start a unit or introduce abstract concepts. For example, to generate student interest in the subject matter, they showed relevant videotapes, slides, photos, and other artifacts, and they presented a dramatic narrative on legends about Early Rome. Within reading activities, ELD Teacher A and students explored various MI-oriented pathways to learn new words, such as drawing, role play, use of real objects, as well as locating images from textbook or internet. Drawing was used extensively across the before, during and after reading activities. For example, drawing served as a quick tool for teachers to assess learners' prior knowledge and identify any misconceptions. Drawing also helped learners to self-monitor their progress and validate their own learning.

In the After-Reading stage, students created word maps and posters to reinforce instruction in main ideas and vocabulary while taking advantage of students' affinity for designing icons. To reach the instructional goals for reading and writing for understanding, a large poster of a "questioning tree" was displayed in all three classrooms to help students generate questions and check understanding after they finished reading. The questioning tree had six branches, labeled who, when, where, what, why, and how. Writing assignments drawing on one or more of these questions addressed both "on-the-surface" (text explicit) and "under-the-surface" (text implicit) comprehension. The ELD teacher regularly videotaped student performances and role play to analyze and document their oral language development.

In the sheltered program, role-play also was used, both as an entry point and as an After-Reading activity. For example, after listening to and reading various legends about how Rome was established, students chose partners and wrote the stories according to their own account. Each team then prepared its own skit and reenacted the birth of Rome. The ELD Teacher A regularly videotaped such student performances. At the end of the role-play activity, each student wrote a final piece on how Rome was born. A comparison of students' stories written before and after the role-play reflected student growth in both vocabulary and use of detail.

Student Achievement.

The unit of analysis in this section consisted of the target students in Year I and Year II. Each group of learners in the sheltered program received language arts and history instruction from ELD Teacher A. In addition, each group of learners received math instruction from teacher ELD Math and science instruction from teacher ELD Science. At the beginning of Year I, we focused on building a language-literacy intervention program based upon findings from the PI's previous research. Later in Year I, we implemented the program. In Year II we continued to field-test language-literacy intervention strategies and centered on forging teacher-parent partnerships to broaden practices of reading strategies beyond school.

Because one goal of the participating district and school was to enhance target students' language-literacy development as measured by statewide results, SAT9 data were collected pre- and post-intervention. Using this data, we measured students' concept development and participation in constructing knowledge from print before and after they participated in the study. In addition, we collected four sets of qualitative data to enhance our understanding of target students' language-literacy development and to address the sub-questions on student achievement. These data sets included students' responses to repeated reading interviews; work samples (including quizzes, writing assignments and products of project-based learning, such as posters, Roman Magazines, etc.); videotaped transcripts of students' participation in the Family Literacy Nights; and teacher interviews regarding the target students' performances.

Based on classroom observations as well as the four sets of qualitative data, both the Year I and Year II programs improved target students' overall classroom performance in areas such as participation, use of receptive and expressive language demonstrated through oral presentations, role-play, and written expression. For example, in regard to the students' written expression, we compared their drafts to their final products for an integrated language-arts history project. Their drafts did not observe the conventions of writing and were unorganized. In their final products, the students used proper writing conventions and organized their ideas more effectively. The observations and data indicated that the target students grew in other areas as well. Towards the end of the school year, the target students demonstrated improved self-confidence as readers and as students; all felt ready to enter the seventh grade. Furthermore, all of them were comfortable with conducting web search activities to complete their classroom projects.

Concept Development & Participation in constructing meaning from print. The team defined concept development as "students' ability to understand word meaning when they encountered unknown words in reading." To study target students' abilities to understand word meaning, we triangulated data from three sources: PI's field notes from participant observations, students' structured reading interviews, and SAT9 scores. The field notes were used to corroborate student responses in the structured interviews. Student interviews included questions relevant to making meaning from print. Interview questions included: (1) What do you do when encountering unknown words in reading? and (2) If you knew someone who was having trouble reading, how would you help that person?

Although the students' reading instruction introduced a wide range of word attack skills and clarifying strategies (structural analysis, looking at the picture when available, etc.), when asked what they did when they encountered an unfamiliar word, the students generally reported using only one or two strategies: sounding it out or looking it up in a dictionary. Students also identified these same two strategies as the ones used most frequently to help others. Students responded that they were less likely to consider looking at a picture in the book, read before or after the sentence that has the unknown word, or using the context clues without prompting. Given their responses to the interview questions, the target students in general did not seem resourceful when came to clarifying unknown words when responded to structured reading interviews; however, based on classroom observations, many of them applied different strategies when "cued" by their teammate or ELD Teacher A.

When it comes to reading for meaning, one year of sheltered instruction, though valuable, did not seem to be enough. The infrequency with which students spontaneously used many of the reading skills to which they had been introduced suggests the need for reinforcement of these skills beyond the sheltered program itself. It would not be surprising that middle school students have not yet developed the self-regulation strategies to be resourceful readers; nor should we assume that parents would automatically nurture such skills. Indeed, parents' comments during the Family Literacy Nights indicate that within many Asian and Asian American families, learning by rote might still be the dominant approach. Our school/home/community partnerships were designed to assist teachers in helping students transfer research-based classroom intervention strategies to home practices. In such a collaborative environment, teaming with parents, siblings, or friends offers a feasible way for students to practice reading at home using the RT strategies presented in classrooms. Practicing language-literacy skills beyond school is vitally important for our target students, who are still grappling with ways to construct meaning from print and who typically demonstrate limited resourcefulness in reading performance.

SAT9 Scores. In this section, we report findings based on Year I and Year II target students' SAT9 test scores. Multivariate analysis of variance and the follow-up univariate repeated measures analysis of variance statistical procedures were applied to Year I data using SPSS v.10. The same procedures were applied to Year II data using SPSS v.9. A summary of the results is presented in Tables 1 and 2. Results from both years were compiled in the same tables so that changes across years could be investigated.

In general, the Year II program (year 1999-2000) demonstrated very similar effects on reading, mathematics and writing as the Year I program (1998-1999). Both programs demonstrated a strong impact on reading. The Year II program, however, demonstrated a stronger impact on language while its Year I counterpart showed a stronger effect on math. (See Table 1.)

The overall program impact on students was somewhat different between the Year I study and Year II study. In Year I, the effect size (Eta coefficient) of the combined score (multivariate) indicates that the program had a strong significant impact on math, reading and approaching significant effect on students' language skills. Year II study, on the other hand, revealed that the program had significant impact on all three academic skills. A follow-up analysis is presented in Table 2.

SAT9 Reading: The results of the univariate repeated measures ANOVA of the Year II study aligned with those of the Year I study. For reading, only vocabulary showed a significant pre-post score difference. However, the effect size was larger for study II (.67) than for study I (.58), indicating that the Year II program was more effective in improving vocabulary.

SAT9 Language: The results for the three language skills are mixed. The Year II program produced significant progress on promoting students' language mechanics with an effect size jump from .001 to .58. However, the effects on expression and spelling decreased from an effect size of .20 and .19 to .08 and .09.

SAT9 Mathematics: Even though both Year I and Year II programs have shown significant effects on mathematical problem solving and mathematical procedure skills, the effect size of problem solving and procedure skills were .70 and .43, respectively and decreased to .36 and .21 in Year II respectively. The decrease in effect may be contributed by target students' classroom behaviors; ELD math teacher reported that Year II group was not good at keeping on task.

Both Year I and Year II's language-literacy intervention program had its unique strengths as reflected from SAT9 test scores. Multivariate statistics indicated that Year I was successful (Effect size >=0.30) in reading and mathematics but not in language, while the Year II program was successful in promoting all three academic skills for participating students. The univariate statistics revealed that target students in Year I program did extremely well in vocabulary (reading), problem solving and procedure (mathematics). Target students in Year II improved in vocabulary (reading), problem solving (mathematics) and mechanics (language).

It is important to remember that all target students scored below the 36^th percentile when they took the exam in the fifth grade. The fact that most students in both Year I and Year II scored between the 36^th and 50^th percentile in spring of the sixth grade means that the sheltered program's language-literacy intervention program was effective as reflected from the effect size presented in Tables 1 and 2. But of course when compared with state and national norms, these students scored in the bottom half. The finding suggests that when the target students lag well behind their English-speaking peers upon entering the sixth grade, their academic needs cannot be completely met with just one year of sheltered instruction experiences.

Based on school observations and ELD teachers' experiences, the current emphasis on testing as the primary measure of student achievement hurt both targets students and their teachers. For example, by the time of the spring tests for sixth-graders, many felt defeated by standardized tests; in fact, two or three students in Year I and Year II studies refused to take the tests, and many more participated reluctantly. These students demonstrated good oral language skills and improved work samples and self-confidence for school learning; as confident as they were as learners, however, they lacked confidence as standardized test takers. Moreover, since many of the target students refused to take the SAT9 or scored lower than the 50^th percentile, a majority of seventh-grade teachers was unwilling to accept them into their classes, for fear that the target students would lower their overall class score.

School-Home/Community Partnership

Forging a school-home partnership is critical for target students to achieve school success. However, such a partnership may not be an easy task in a middle school. In this section, information was organized to reveal the research findings to the following question:

• What were the mediation processes and contexts that forged partnerships with parents?

The research team entertained various ideas in Year I to forge partnership with parents or family members, but was unable to carry through even after we produced videotapes for parents regarding certain classroom intervention strategies. In Year II, the research team reached a consensus early on to initiate a school/home/community partnership by sponsoring four Family Literacy Nights, or more precisely, two sets of two Family Literacy Nights. One group consisted of the target students from ELD Teacher A's sixth grade sheltered classroom. The other group was students placed in a sixth- through eighth-grade special education resource program, many of whom were not target students. Each two-hour evening event was co-planned and co-presented by participating teachers and resource specialists.

In this section, we examine two events provided for the target students and their family members/friends. The mediation process identified in the research question was defined as the processes involved in the joint productive activities (JPA) among all participants (Dalton, 1998; Hilberg, Doherty, Tharp, & Estrada, 1999). More specifically, the mediation process involving a) all research collaborators before, during, and after the Family Literacy Nights, and b) the JPAs between the participating students and their family members.

JPA was defined as "any collaborative interaction that led to a joint product. Collaboration can take many forms: shared ownership, authorship, use, or responsibility for a product. It [JPA] can also include division of labor for a product." (Hilberg, et al., 1999, p. 8). In the context of this project, we promoted both tangible and intangible products. Examples of tangible products included: a drawing of a "questioning tree" used by parents and children as a guide to generate different questions after they finished reading to check for comprehension, a child's MI profile constructed independently by the child and a parent, and a completed learning log. Examples of intangible products included: participants' abilities to articulate strategies to identify word meanings from a passage, generating text-implicit questions, a shared understanding between teachers and participating family members about the importance of parent-teacher partnerships for reading and language development, a change in attitude towards identifying their child's multiple abilities beyond school-valued high test scores, etc.

Our research team implemented JPAs as the basis for integrating CREDE's other four effective standards and MI as tools in this project. Enacting JPAs encouraged teachers, parents, and students to participate in dialogues while working together to produce a common product, or in the case of this research project, achieving each Family Literacy Night's specific objectives, such as valuing the child's multiple abilities and transferring classroom strategies for home practices. Findings related to this particular research question were organized in two parts. The first part focused on findings related to the JPAs observed around the Family Literacy Night events as well as related feedback obtained from all participants each of the evening. At each Family Literacy Night event, there were pre-event, during-event, and after-event activities. The second part synthesized findings across events sponsored for the two different groups of participants. We also entertained some implications for field practices and future research.

JPA for Family Literacy Nights

We adopted the features provided in SPC as the basis upon which to evaluate mediation processes used to forge partnerships with parents. Using SPC, we analyzed the extent to which we had enacted JPAs in various activity settings. Such information further advanced our shared understanding about JPAs and facilitated future activities. Briefly, to enact the JPA, there is a continuum, as designated in SPC. Here are the key features for a fully implemented JPA:

Feature 1 - an activity that engaged participants to collaborate on a joint product.

Feature 2 - a small group based activity.

Feature 3 - active participation in JPA.

Feature 4 - sustained participation by a teacher or an adult.

Based on these features, we thought by predicating the activities on JPAs when working with families, we would achieve the goal of having all participants fully engage in sustained dialogues or provide meaningful modeling when needed. In short, it would help us provide assisted performance within the ZPD of an individual participant or family. Through JPAs, as delineated in the SPC flowchart, we achieved the following purposes of these Family Literacy Nights: (1) introducing four RT reading comprehension strategies for home practices, (2) adopting the theory of MI as tools for family members to value their child's multiple ability as well as to apply multiple paths, beyond rote learning, to motivate practices beyond school, and (3) extending CREDE's five standards for effective pedagogy for home practices. In the following sections, we described how JPAs were observed through the entire process as it occurred in pre-, during-, and after-event activities designed within Phase 2 of our Year II study.

Pre-Event JPA

Throughout these scheduled planning meetings, we employed CREDE's standards to guide, plan, and design activities implemented in the Family Literacy Nights. All of the research collaborators participating in the Year II study attended an initial 2 and 1/2-hour planning meeting held at a restaurant. There were three ELD teachers, two regular education teachers, and two special education resource specialists. ELD Teacher A, ELD Math, and ELD Science were the leaders for Team One and co-presented at the Family Literacy Night on 10/7/99 and 11/2/99. The two regular education and two RSPs formed Team Two and co-presented at the Family Literacy Night on 10/19/99 and 12/2/99. In addition, the PI and research associate invited a high school SDC teacher who expressed interest in co-sponsoring an event with her colleagues at a nearby high school. All in all, there were eight planning meetings held, two meetings for each event.

During the initial meeting, the first item on the agenda was to schedule the four Family Literacy Nights for the following groups: a) Group One-students enrolled in the sixth grade sheltered program, and b) Group Two-students placed in a sixth-through eighth-grade special education resource program. Teachers in Team One taught the students in Group One, hence Team One co-planned the events for Group One. Team Two teachers taught Group Two students, hence they co-planned the events for Group Two. The next item on the initial meeting agenda was the conversion of CREDE's standards to an Observer's Feedback Form used at the very first Family Literacy Night involving Group One. All of the research collaborators studied and discussed the items listed in the observer's form as a way to connect the activities with CREDE's standards. We knew at the outset that it would be challenging for each observer to provide detailed observations using this observation form; however, the observers were motivated to adopt and use the form since they were conducting their event with Group Two students and their family members two weeks after Group One's first event.

The enactment of JPAs and production of products stemming from the JPAs during the pre-event stage was consistently high. All of the planning members were engaged in brainstorming ideas and modifying activities and materials based on group feedback. For example, after observing the high participation rate during the first event, the entire team realized the importance of calling parents to attend. Therefore, all of the teachers and specialists called their students' parents to personally invite and encourage them to participate in each event. The tangible product for this JPA was teachers' enthusiastic and personable phone calls placed by all research collaborators to each of the prospective participating families. The intangible products included teachers' levels of motivation and confidence to provide meaningful and successful events.

During- and After-Event JPA

Data presented in this section were collected and triangulated from four sources: (1) videotapes and transcripts obtained from the videotapes of all four events, (2) feedback from participating parents and students, (3) feedback from presenters and observers, and (4) PI's field notes. We then synthesized findings and drew implications from the two events for suggestions for field practices and future research. Since all of the activities were videotaped for future analysis, no external observers were scheduled after the first event conducted by Team One. These events primarily were videotaped by two eighth-graders from the same school. When the events were conducted in the school library, we used two cameras to videotape the activities.

At the closing of the first event for each group, we provided each participating student with a certificate of participation. After parents expressed interest in receiving one, we created a family certificate and presented it to them at the end of the last event for each group. In addition, during each of the four Family Literacy Nights, bottled water and snacks were provided for the attendees.

First Family Literacy Night (held on 10/7/99).

There was active participation among the students and their family members. Twenty-three out of 24 students came with either one or two parents. Some brought their entire family. On average, three members per family attended the event, including older or younger siblings and/or both parents. The entire group convened in one classroom for the opening remarks, then divided into three smaller groups. Each small group attended a science, math, or history/language arts session, and then rotated to another classroom for a different subject-area session. Each session lasted approximately 20 minutes. After the groups had attended all three subject- area sessions, they reconvened in the history-language arts classroom for a debriefing and were given an opportunity to provide feedback.

Opening Session. ELD Teacher A opened the event by saying; "We want to set up a partnership with you. This is why you're here and what we can do together to help make your child successful in school, and increase their reading comprehension and decoding skills…" After being introduced to the project, each participating family was presented with the Year I deliverable, the Parents' Handbook Together, We Can Help Your Child Read Better Through Multiple Paths (Chang & Shimizu, 1999). ELD Teacher A used the Parents' Handbook to highlight why reading is so important by saying, "If you can't read, you're really missing out and so much of school is reading–even in PE. They need reading in all of their areas, plus in their daily life." The Parents' Handbook presented the RT strategies as they were used in their children's history and language classroom.

The opening setting, a history-language arts classroom, barely accommodated the large number of attendees, but was too small to accommodate a whole group activity. Many families brought their younger children; so additional space and activities were necessary to keep them happily occupied without interrupting the dialogues or activities. Though it was intended for parents to experience what it was like for their child to move from room to room within the sheltered program, our experience with the undersized classrooms helped us rethink the JPAs in our second event; consequently, we decided to relocate future events in the school library. Given the limited space and according to the levels and features of JPA delineated by SPC, we only partially enacted JPAs between teachers and family members. However, the JPAs within family members were high.

Small Group Sessions. Following the opening session, each ELD teacher explained and modeled how they reinforced target students' use of the four RT strategies. The intent was to engage family members in this first event by providing them with opportunities to experience how the RT strategies were used in four content areas: math, science, history, and language arts. In addition, parents learned about the school's homework hotline from the ELD Science teacher as well as the Student Agenda, a calendar designed by the ELD Math teacher.

The Parent's Handbook provided diagrams illustrating the four RT strategies and provided suggestions for parents who wanted to help their children at home using multiple pathways, such as role-play, drawing, and using the questioning tree. We focused was on ways for parents to engage in joint productive activities with their children at home as a means of supporting their children's language and literacy development.

During the first night, fewer fully enacted JPAs occurred between teachers and participating family members due to a lack of sustained participation by the teachers. All of the ELD teachers circled around the room to support different family teams. However, a higher degree of JPAs occurred within each child-family team when they participated in Paired Reading. While in this activity setting, each team followed the RT guidelines provided in the Parent's Handbook and participated in a JPA. For example, one mother, father, and daughter team used the strategies modeled by the teacher. They read, used the RT strategies, and completed the learning log worksheet together. Using the SPC's flowchart, this team enacted the JPA by (1)