English Language Proficiency and Content Assessment Performance: A Comparison of English Learners and Native English Speakers Achievement

As a result of the accountability requirements established in Title III of the Elementary and Secondary Educational Act (ESEA) legislation, English Learners (ELs) are expected to make progress in both content area academic achievement and English Language Proficiency (ELP). In Tennessee ELs progress is measured by administering WIDA-Access to assess English language proficiency, and Tennessee Comprehensive Assessment Program (TCAP) standardized assessments to measure content academic achievement. The purpose of this study was to compare and analyze the performance levels of ELs who achieved the exit criteria on WIDA-Access state mandated English proficiency assessment and their subsequent performance on English Language Arts and Math TCAP assessments. Specifically, a comparison of EL’s achievement on TCAP was compared to the achievement on TCAP of non-ELs. Independent samples t-tests were performed on data from 302 elementary and middle school ELs and non-ELs that participated in WIDA-Access and TCAP assessments in 2015. Data analyses concluded that English Language Arts and Math TCAP scale scores were significantly different between ELs and non-ELs. Achievement levels in both English Language Arts TCAP and Math TCAP for ELs, who achieved the exit criteria on WIDA-Access, were lower than the achievement levels of non-ELs. Discussions of the findings in this study along with implications of using these assessments to measure ELs growth is provided in relation to the increased demands on measuring both the academic achievement and English language progress for ELs.


Introduction
The student population of English learners is increasing in the United States, and as a result the demands of educational agencies to meet their needs have also increased.Cellante and Donne (2013) noted that presently 9.4% of the student population, or 4.6 million students, are identified as English learners (ELs).As noted by Horsford and Sampson (2013), English learners (ELs) have become the fastest growing population of students in K-12 public education.Title III of the Elementary and Secondary Educational Act (ESEA) established criteria and accountability measures that state and local educational agencies must meet when serving the educational needs of ELs.Under No Child Left Behind (NCLB) requirements, as reported by Estrada and Wang (2013), ELs are held to the same level of accountability on state mandated content assessments as their English speaking peers (TDOE, 2013).As a result of the re-authorization of ESEA, now referred to as Every Student Succeeds Act (ESSA), the same English Learner accountability measures set forth in NCLB remains (Burnette, 2016).
Additionally, with this legislation, states are required to annually administer an English Language Proficiency (ELP) assessment to ELs to measure the four language domains of reading, writing, listening, and speaking as described by Wolf, Everson, Lopez, Hauck, Pooler, and Wang, (2014).According to Hopkins, Thompson, Linquanti, Hakuta, and August (2013), state educational agencies and local education agencies must meet the established accountability requirements via these proficiency assessments.Although local school districts have the discretion to choose the program model that is utilized in servicing the needs of ELs, the Tennessee Department of Education adopts the English language proficiency assessment that is administered to ELs across the state (TDOE, 2013).
World Instructional Design Assessment-Access is the ELP assessment that is administered to ELs in the state of Tennessee.Tennessee is a consortium member along with 36 other states who adopted World-Class Instructional Design and Assessment Access (WIDA-Access) to measure English Language Proficiency of ELs as noted by Huebeler and Lenard (2013).As WIDA-Access is for the third year learners in Tennessee, research is needed to assess the performance level of ELs when compared along with their subsequent proficiency on the state mandated content assessment Tennessee Comprehensive Assessment Program (TCAP).
Currently, the state of Tennessee utilizes the Tennessee Comprehensive Assessment Program (TCAP) to measure student achievement in language arts, math, science, and social studies in grade levels 3-8 (TDOE, 2015).Concurrently, while participating in an English as a Second Language (ESL) program, ELs participate in the TCAP along with non-English learners.According to Siegel (2014), there is a direct correlation between effective and equitable assessments and English Learner achievement.Analyzing and comparing the performance of ELs versus non-English learners on academic content assessments will address the educating goals of ELs (Hopkins et al., 2013).Ultimately, to truly assess English Learner proficiency and ensure student success, a consistent, ongoing, and academic enriched assessment system working simultaneously with an effective ESL program is required.

ELP Assessments and Academic Content Assessments
Title III of the Elementary and Secondary Educational Act (ESEA) established criteria and accountability measures that state and local educational agencies must meet when serving the educational needs of English learners (ELs), as noted by August, Artzi, Kuchle, and Halloran (2015).States are required to annually administer an English Language proficiency assessment to ELs via local educational agencies in order to meet accountability requirements that are determined by the Elementary and Secondary Education Act (ESEA).These guidelines are directly correlated to No Child Left Behind Act (NCLB) policies according to Hopkins et al. (2013), and continued in ESSA (Burnette, 2016).
To meet national guidelines, the Tennessee Department of Education adopted the WIDA-Access English language proficiency assessment as the measurement that is utilized to determine English proficiency for English Learners (TDOE, 2013).Tennessee also requires all students, including EL students, to take the Tennessee Comprehensive Assessment Program (TCAP), which measures student achievement in language arts, math, science, and social studies in grade levels 3-8 (TDOE, 2015).As ELs participate in the TCAP along with non-English learners, understanding the correlation between the English Learner performance on ELP assessments and academic content assessments is imperative to EL success as addressed by Hauck, Wolf, and Mislevy (2013).
As ELs are held to the same accountability level on content assessments as non-English language students, it is imperative that students obtain academic English language proficiency in order to perform successfully on these mandated assessments (Cook, Linquanti, Chinen, & Jung 2012).Based on this fact, this study addressed the issue of ELs' performance on content assessments by analyzing their performance on ELP assessments and their subsequent achievement on academic content assessments.Specifically, by comparing EL's achievement on TCAP as compared to the achievment of non-ELs, the potential gap between ELs' communicative language proficiency and academic language acquisition could be explored.In order to provide effective services to ELs in schools, the validity of the adopted proficiency scores of the WIDA-Access which can be used as a predictor in determining ELs' academic content achievement is vital.Cook et al. (2012) reported that Title III of the Elementary and Secondary Educational Act (ESEA) established criteria and accountability measures that state and local educational agencies must meet when serving the educational needs of English learners.August et al. (2015) cited that this further established the changes in academic needs, assessments, and accountability of educational agencies as it pertained to English learners (ELs).As per the current system of English learner assessments, the success of ELs varies in districts and states.Bailey and Carrol (2015) reviewed ELs requirements and examined language and content assessments, concluding that Title III provided a blanket of NCLB guidelines and support for ELs.Guidelines within this legislation allow state educational agencies, in conjunction with local educational agencies, to have the discretion to choose the type of ESL programs that are provided to insure language acquisition achieved for ELs.Hauck et al. (2013) stated that No Child Left Behind (NCLB) legislation established the requirement that all ELs should be included in all state mandated assessment that are administered for accountability.These accountability measurements are identified as Annual Measureable Achievement Objectives (AMAO's) and are directly linked to available Title III funding as stated by August et al. (2015).As suggested by Hopkins et al. (2013), the districts must determine whether their ESL program meets the needs of all ELs in order to reach a proficiency level of English on these assessments.Hakuta (2014) concluded that significant relationships have been established between English language proficiency assessment and content assessment, and in response to that conclusion, the data of this study provided further knowledge that may serve as an opportunity to better develop an effective assessment system by determining if differences exists between English Learner and non-English Learner student achievement.Additionally, as accountability is tied to language and content assessment scores (Hakuta, 2014), determining the relationship between the achievement levels of ELs as compared to non-ELs offered substantial insight into developing a consistent, ongoing, and academically enriched ESL program to ensure their success.

Research Design and Hypotheses
This research incorporated a quantitative method approach to examine the differences in achievement between English Learners who had achieved proficiency levels on the WIDA-Access and non-English learners as it relates to TCAP achievement, specifically in English language arts and math.This research also explored whether there was a significant difference in EL achievement as compared to the achievement of their non-EL peers.This quantitative research involved collecting existing data in order to compare and measure the relationship and/or differences between two or more variables (Creswell, 2015).Research questions were designed to guide the researcher to determine if English Learner students, who obtained the appropriate proficiency score on the state adopted English language assessment (WIDA-Access) in order to test out of ESL services, performed as well as non-ELs on academic content assessments.Intellectus Statistics was used to conduct all quantitative analysis of the data.The following research questions were utilized to guide the study and data collection.1).Is there a statistically significant difference in the English Language Arts TCAP proficiency levels of non-English learners and English learners who achieved a proficient level on WIDA Access? 2).Is there a statistically significant difference in the Math TCAP proficiency levels of non-English learners and English learners who achieved a proficient level on WIDA Access?

Participants
The target population represented for this study came from one rural school district in Tennessee consisting of eight elementary and middle schools within the district.English learners from the 3 rd to 8 th grade who took the WIDA-Access in this district during the 2014-2015 academic year, along with native English speakers, served as the target student population.Two hundred (200) English Learner participants who achieved the state mandated English proficiency score on the WIDA-Access were included in the sampling.A random sampling of 102 non-English learners who participated in TCAP alongside the ELs, within the same school and grade level as the ELs, were randomly selected for the study.Table 1 provides an overview of the demographics for the target population.

Procedures
The collected data was obtained from 2014-2015 academic year.The archived data collection of the WIDA-Access and TCAP in the content areas of English language arts and math of the population sample was compared to determine if there was a statistically significant difference in English language arts and math TCAP achievement scores between ELs and non-ELs.

Sampling Procedures
Upon approval from the Institutional Review Board at the researcher's university, and the director of schools, the researcher collected district data from the local educational agency.The archived data collection from the WIDA-Access and TCAP in the content areas of English language arts and math was compared to determine if there was a statistically significant difference in TCAP achievement between ELs, who obtained the appropriate proficiency score on the state adopted English language assessment (WIDA-Access) in order to test out of ESL services, and non-ELs.All ELs that met the specified exit criteria on the WIDA-Access scores were included in the study, and as a result, their TCAP scores were recorded.A random sampling of non-ELs' TCAP scores was then selected from the district database.Data from each school was protected for privacy as outlined under the Family Educational Rights and Privacy Act (FERPA).

Sample Size, Power, and Precision
In this quantitative study, data from 200 ELs were collected and used to answer the chosen research questions.Specifically, WIDA-Access data was used to separate students who had scored greater than or equal to 5 on the English proficiency assessment (WIDA-Access).Furthermore, the researcher collected the TCAP data of these identified EL students, in addition to 102 non-ELs.Descriptive statistics demonstrated the percentage of students who performed below basic, basic, proficient, and advanced on TCAP in the prospective schools included in the study.Independent sample t-tests were utilized to determine if significant differences existed between EL WIDA-Access proficiency levels and scale scores and TCAP English language arts and Math achievement levels and scale scores.This sample size provided a medium effect size according to Fual, Erdfelder, Buchner, and Lang (2008).Power analysis for an independent sample t-test was conducted in G-POWER to determine a sufficient sample size using an alpha of 0.05, a power of 0.80, a medium effect size (d = 0.5), and two tails (Faul et al., 2008).Based on the aforementioned assumptions, the desired sample size for a medium effect size includes at least 128 participants.

Measures and Covariates
The instruments used to collect data for this study consisted of the Tennessee Comprehensive Assessment Program (TCAP) assessments, specifically the achievement scores from the English language arts and math content assessments.Along with Tennessee Comprehensive Assessment Program assessments (TCAP), the World Instructional Design and Assessment Access (WIDA-Access) English language proficiency assessment was utilized to determine the level of English language acquisition growth of ELs.

Tennessee Comprehensive Assessment Program
This instrument utilized criterion-referenced assessments administered annually in grades three through eight in order to measure student knowledge of Tennessee standards (TDOE, 2015).The results of TCAP assessment help educators identify and analyze the strengths and weaknesses of a student's knowledge of the content material and further provide districts with valuable information in order that informed administrative decisions can be made to improve instruction and achievement.
Reliability of TCAP is determined by the use of a base score of student knowledge, the projected expectation of growth, and a comparability of raw scores of annual achievement (TDOE, 2016).The content validity of TCAP, along with the construct variance and construct validity, is in the technical report published by the Tennessee Department of Education (TDOE, 2016).

World Instructional Design and Assessment Access
WIDA-Access is an English language proficiency assessment that is administered to determine the language proficiency level of English learners.The proficiency indicators of language acquisition that are assessed within the administration of WIDA-Access range between "entering" and "reaching," with a scaled score range of 1 and 6.WIDA-Access identifies students who have attained English proficiency, meets the state and federal mandates that require EL student assessment, and provides valuable information to local school districts when developing meaningful ESL programs.
To ensure the reliability and validity of the WIDA-Access, the assessment is annually field-tested to validate its ability to equitably assess English proficiency.The tested reliability and validity is noted in the annually released technical WIDA manual (WIDA, 2015).

Results
Research question 1.Is there a statistically significant difference in the English Language Arts TCAP Proficiency levels of non-English learners and English learners who achieved a proficient level on WIDA Access?
An independent samples t-test was conducted to examine whether the mean of English Language Arts TCAP scale was significantly different between the ELs that achieved a proficient level on WIDA Access and non-ELs.
The result of the independent samples t-test was significant, t (160) = -2.31,p = .022,suggesting that the mean of the English Language Arts TCAP scale scores was significantly different between the ELs and non-ELs.The mean of the English Language Arts TCAP scale scores for ELs was significantly lower than the mean of the English Language Arts TCAP scale scores for non-ELs.Table 2 presents the results of the independent samples t-test.Research question 2. Is there a statistically significant difference in the Math TCAP proficiency levels of non-English learners and English learners who achieved a proficient level on WIDA-Access?
An independent samples t-test was conducted to examine whether the mean of Math TCAP scale scores was significantly different between the ELs and non-ELs.The result of the independent samples t-test was significant, t(160) = -2.95,p = .004,suggesting that the mean of the Math TCAP scale scores was significantly different between ELs and non-ELs.The mean of the Math TCAP scale scores in for ELs was significantly lower than the mean of the Math TCAP scale scores for non-ELs.Table 3 presents the results of the independent samples t-test.

Discussion
The data revealed that the mean scores of English language arts TCAP scale scores were significantly different between the EL and non-EL students.Specifically, students who tested out of ESL services had significantly lower scores on the English language arts TCAP assessment than did their non-EL peers.Both the complex linguistic and grammatical structure of English language arts, as well as the amount of reading that is required on the English Language Arts TCAP, may serve as viable reasons that ELs showed a deficiency as compared to their English speaking peers on this assessment.Although ELs receive additional time on the English Language Arts assessment, the amount of test items, depth of knowledge needed in the writing prompt, and the overall length of this assessment can all further add to the plausible causes of their lower achievement levels.Figure 1 presents the mean of the English Language Arts TCAP scale scores (ELs) and English Language Arts TCAP scale scores (non-ELs).
Figure 1.The mean of English Language Arts TCAP scale scores by EL and non-ELs Analyses of data suggested that the mean of Math TCAP scale was significantly different between ELs and non-EL.English learners Math TCAP scale scores were significantly lower than that their non-EL peers.These results could have been a result of the fact that ELs can provide the correct response to math problems when they are presented in numerical form, i.e. when the question is presented as 3 + 3.However, when presented in the form of a word problem, the complexity of the vocabulary may result in an inaccurate response from ELs.The data supports the alternative hypothesis given, which was a statistically significant difference in the Math TCAP scale sores of non-English learners and English learners who achieved a proficient level on WIDA-Access.Figure 2 presents the mean of the Math TCAP scale scores (EL) and Math TCAP scale scores (non-EL).The findings in this study found that although English proficiency levels rise for ELs, the level of achievement growth on academic content is at a lower level, both individually and as compared to their English-speaking peers.Studies have shown that acquisition of academic language skills, known as CALP, can take between five to ten years to fully develop (Cummins, 1999), and therefore, the expectation that ELs should perform at the same level of proficiency as their peers the first year they enter into a school is unattainable.Research has shown that it takes three to five years to develop communicative language skills, known as BICs, (Cummins, 1999), so it is unrealistic to expect ELs to perform at a high level of proficiency both socially and academically in a limited amount of time.The data analysis results in this study have provided valid research that implies that ELs need additional time, instruction, and resources to reach proficiency in the target language to ensure success in all their academic endeavors.
Also, for ELs who were deemed English proficient based on their WIDA-Access scores, these findings demonstrated that there may be a discrepancy between the academic language understanding of ELs and the language skill set that is needed to reach proficiency on English language proficiency assessments.Methods and strategies that content teachers have relied upon when working with ELs have not been working (Molle, 2013), and ELs achievement gap on content standard-based assessments is broadening as a result.Arkoudis (2006) stated that, in order for schools to effectively meet both the academic and linguistic needs of ELs there must be a collaborative effort between ESL teachers and content teacher.ESL and content teachers must consistently work in conjunction with one another to develop beneficial teaching methods to ensure that all ELs academic needs are to be met.The results of this study may have an impact upon how local educational agencies provide content area teachers with adequate training for effectively teaching diverse students.
As local educational agencies have the discretion to choose the types of English language arts curriculum and program models that are implemented in their districts, these programs should also be chosen with the EL sub group in mind to ensure that reading proficiency is achieved for ELs as well.These results may be significantly lower due to the lack of training of content area teachers, as it pertains to their exposure to teaching diverse student populations.Based upon the previous research of López and McEneaney (2012), teachers that were provided with professional development and strategies to effectively work with ELs, improved the reading and language arts proficiency of ELs.
This study has shown that although ELs improve English language proficiency, as demonstrated by the WIDA-Access proficiency level descriptors (WIDA, 2007), their overall performance on academic content assessments are lagging behind those of their English speaking peers.Once ELs meet the required English proficiency level, they are reclassified as transitional students, monitored for a 2-year period (Saunders & Marcelletti, 2012) and no longer monitored by school districts for academic success.Perhaps additional procedures should be implemented to accurately measure the long term growth of ELs in language acquisition as well as in academic achievement (U.S.Department of Education, 2016).
Currently, accountability guidelines require that ELs participate in state mandated content assessments the first year of arrival into a school (TDOE, 2013).Although ELs are exempt from the English language arts portion of the assessments, they are required to participate in the math, science, and social studies with non-ELs (TDOE, 2015).Although the achievement scores of these students are not factored into school accountability the first year of their participation, their achievement levels are included upon their second year of participation in content assessments.There have been limited studies that have addressed the accountability that local educational agencies are required to meet when serving ELs, and an implication of the findings of this study suggest that further consideration should be given to the revision of these requirements.
These accountability measures are identified as Annual Measureable Achievement Objectives (AMAO's) and are directly linked to available Title III funds (August et al., 2015).AMAO's are a progress monitoring system of measuring ELs growth in English proficiency assessments and academic growth on academic content standards.
In order that school districts meet adequate yearly progress, ELs achievement is assessed each year.As the results of this study have shown, the proficiency achievement of ELs on WIDA Access and their achievement on TCAP content assessments are not equitable.Districts that do not meet their AMAOs for 2 or more consecutive years are required to submit an improvement plan to address the lack of growth in ELL achievement (Babal, Cao, Filer, Hedtke, & Lo, 2015).Had the growth or progress of the student populations on these 2 assessments been considered for this study, the data may have produced different findings.
Many variables in this study were not controlled, such as the growth or progress monitoring of the EL population during prior years.The English language proficiency assessment used prior to the adoption of WIDA-Access (TDOE, 2013) had been used in the years prior to the 2014-2015 academic year, therefore, the prior growth of ELs was not considered in this study.The previous achievement growth of both non-ELs and ELs on the TCAP content assessments was also not analyzed in this study, and a better understanding of the previous growth data would have contributed to the study.Perhaps another way to show growth of ELs outside academic content assessments is needed to accurately measure language growth of this group of students.

Conclusion
The schools are in the necessity of meeting the academic needs of the growing population of ELs, and criteria; and hence an accountability measure concerning this population is ever evolving.These federally mandated guidelines are directly aligned to the accountability of state and local educational agencies to meet the academic needs of ELs (Cook, Linquanti, Chinen, & Jung, 2012).Although these measures are established by state educational agency, implementation of these mandates is required at the local level.Therefore, school districts must determine whether their ESL program is meeting the needs of all ELs in order to reach a proficiency level of English for meaningful accountability to occur.
Policies concerning ELs impact the use of English language proficiency assessments as a measurement of assessing English language proficiency, and there exists an assumption or idea that English proficiency equates academic language proficiency (WIDA, 2007).ELs are held to the same level of accountability on state mandated content assessments as their English speaking peers, regardless of their proficiency level of English, and data from this study documented that ELs were not doing as well as their non-EL peers.The achievement of ELs on content assessments do not report a clear, concise, and timely understanding of student's progress and academic gains in the target language (TDOE, 2015).
Although ELs receive ESL services, they are not always given effective support to develop academic language skills for content knowledge.All teachers who work with ELs should receive specialized training on how to work with this diverse population of students.Professional development is needed for school district personnel and educators in gaining an understanding of how to analyze and interpret the data from English language proficiency assessments.
There is a gap between ELs ability to achieve language proficiency on English language proficiency assessments while simultaneously achieving academic proficiency on content standard-based assessments.There is a direct correlation between effective and equitable assessments and EL achievement.When constructing future English language proficiency assessments, and requiring content assessment participation, collaboration is needed to develop a systematic, standard-based, and improved assessment that is truly beneficial for all ELs (Hauck et al., 2013).
The purpose of this study was to determine the connection between ELs proficiency on the WIDA-Access and the English language Arts and Math subtests on TCAP assessments.The data revealed that although ELs achieved the exit criteria on WIDA-Access, their subsequent performance on TCAP was consistently behind the achievement levels of their native English-speaking peers.The data in this research revealed that, for this population of ELs, the currently adoped ELP did on accurately predict success on content standard assessments, especially when compared to non-ELs.

Figure 2 .
Figure 2. The mean of the Math TCAP scale scores by ELs and non-ELs

Table 1 .
Demographics of student sample

Table 2 .
Independent Samples t-Test for the Difference between English Language Arts TCAP Scale Scores (EL) and English Language Arts TCAP Scale Scores (non-EL) Note.Degrees of Freedom for the t-statistic = 160.d represents Cohen's d.

Table 3 .
Independent Samples t-Test for the Difference between Math TCAP Scale Scores (EL) and Math TCAP Note.Degrees of Freedom for the t-statistic = 160.d represents Cohen's d.