Subject: MEL 208: Language Testing and Evaluation

Teacher: Ms. Analie L. Lomocso

Submitted by: Kleene B. Sison

Standardized Proficiency Tests in a Campus-Wide English Curriculum: A Washback Study

Shao-Ting Alan Hung & Heng-Tsung Danny Huang

Objectives of the Study Framework Methodology Results/Findings Implications/ Recommendations

The current study bridged this Common European Participants and Research Results to research question 1: in The results of this study confirm to
research gap by exploring the Framework of Instrument what aspects do washback exist language testing researchers and
washback effects of a Reference for in student learning? classroom teachers that washback
proficiency test on student Languages (CEFR) The current study recruited a total of effects on learning do exist. To be
learning in a campus-wide 753 participants from different The results not only indicated that more specific, washback effects on
English curriculum, academic disciplines. For the washback existed in various aspects such aspects as personal image,
uncovering relationships questionnaire development, nine of students’ English learning, such learning motivation, emotion, and
between washback and students were invited to participate in as personal image, learning future job opportunities were
learner characteristics such focus group interviews for item motivation, emotion, and future job especially salient. In addition, such
as major, gender, and generation, and 50 students were opportunities, but also revealed that changes as improving English ability,
proficiency level. invited to fill out the questionnaire for the proficiency test induced such studying harder, and performing better
piloting purposes. For the official changes on student learning as on tests were also induced. Among
The study aimed to answer study, 694 students were invited to improving English ability, studying these, the most influenced aspect
the following questions: respond to the finalized, official harder, and performing better on relates to future careers. Students
questionnaire. Among these 694 tests. become aware that higher scores on
participants, 390 were recruited from English proficiency tests increase their
engineering- related departments However, compared with the job opportunities as job markets
1. Are there identifiable (hereafter Engineering cohort), 231 abovementioned washback effects become more competitive.
washback effects of a were from business-related majors on various learning aspects, the
proficiency test on (hereafter Business cohort), and 73 ones on relationship with teachers Particularly noteworthy is that less
student learning in were from the foreign language and peers tended to be weaker. For washback was found on the
tertiary institutions in department (hereafter foreign instance, item 3: My performance on interaction with the teachers and
Taiwan? language cohort). The percentages of standardized tests affects my peers. It was believed that tests would
2. What is the learner distribution in Engineering, relationship with teachers and item affect test-takers’ interactions with
relationship between Business, and Foreign Language 4: My performance on standardized teachers and peers (Cheng, 2004;
washback effects and cohorts are 56%, 33%, and 11%, tests affects my relationship with my Cheng, Klinger & Zheng, 2007).
proficiency levels? respectively peers received mean scores lower However, in the current study, such
3. Is there a significant than 3, suggesting that external aspects of washback were not salient.
difference in the The research data for the current factors seemed less salient as
strength of washback study came from questionnaires and compared with internal factors, such The findings also revealed that the
experienced by male students’ test scores. as personal image and learning relationship between the strength of
and female students? motivation. washback and proficiency levels was
4. Is there a significant To investigate the washback effects of statistically significant: the high-
difference in the the GEPT on student learning, a Results to research question 2: is proficiency group experienced a
strength of washback questionnaire was adapted from the strength of washback related significantly stronger washback effect
experienced by Cheng (2005). The questionnaire to proficiency levels? than the low-proficiency group.
examined students’ reactions toward
learners of different the impact of proficiency tests on The second research question However, the study revealed no
academic majors (i.e., various aspects of learning. Cheng’s investigated whether the strength of significant difference between
Engineering, (2005) questionnaire was designed to washback is related to learners’ washback and test- taker
Management, and explore students’ attitudinal and English proficiency. The grand mean characteristics. This finding suggests
Foreign Languages)? behavioral changes in relation to of 9 items on washback was used that the influences generated from
class- room in the context of the Hong because it represents all aspects of standardized tests have little
Kong College Entrance Exam washback being examined in the relationship with gender and academic
(HKCEE). It also explored the role of current study. Furthermore, after background.
the public examination in student ranking all of the participants’ scores
learning. Hence, the aims of Cheng’s from highest to lowest, the top Finally, the results from the study
questionnaire well served the quartile (high-proficiency group) and carry some implications for language
purposes of the current study, which bottom quartile (low-proficiency teachers, policymakers, and school
also aimed to explore how students’ group) scores were used. These two administrators. Derived from the
learning was influenced by a sets of scores were then compared current study, the significant
proficiency test. using the t test for independent relationship between the strength of
samples. The relationship be- tween washback and proficiency levels
The Test Scores the strength of washback and suggests that low-proficient students
proficiency levels was found to be tend to be less influenced by and
Students’ scores on GEPT- statistically significant. Specifically, devote fewer efforts to standardized
intermediate, which are composed of the high-proficiency group tests. Hence, a mechanism should be
two sub-scores on listening experienced a significantly stronger established to encourage low-
comprehension and reading proficient students to take greater
comprehension, were obtained to be washback effect than the low- responsibility for their learning as well
compared with their responses from proficiency group as better prepare for standardized
the questionnaires. After the university tests. Moreover, since standardized
administered one simulated GEPT- Results to research question 3: is proficiency tests such as the GEPT
intermediate at the end of the there a significant difference in influences students of different majors
semester, the researcher requested the strength of washback and genders no differently, these
the scores and ensured confidentiality experienced between male and variables need not be considered
by replacing students’ numbers and female students? when formulating foreign language-
names with codes. related policies, such as graduation
benchmarks.
The third research question explored
the differences in washback
experienced between male and
female students. The independent t
test performed on the collected data
revealed that male and female
students did not differ statistically in
the strength of the washback
experienced

Results to research question 4: is

there a significant difference in
the strength of washback
experienced among different
academic majors (Engineering,
Management, and Foreign
Languages)?

A one-way ANOVA disclosed no

statistically significant difference in
the strength of washback
experienced among Engineering,
Management, and Foreign
Languages majors
Synthesis

On the topic washback, Hung & Huang’s study in 2019, “Standardized Proficiency Tests in a Campus-Wide English Curriculum: A Washback Study”. Showed significant

results on language tests’ washback effect. The researchers’’ study is significant in the context in language testing since it revealed important aspects of washback, reinforced the

existing theory and beliefs on washback, and raised relevant questions regarding language tests’ washback effect to learners or test-takers.

In their study, the researchers utilised a quantitative descriptive-correlational research design. They aimed to explore “the washback effects of a proficiency test on

student learning in a campus-wide English curriculum, uncovering relationships between washback and learner characteristics such as major, gender, and proficiency level.” The

study revealed “aspects of students’ English learning, such as personal image, learning motivation, emotion, and future job opportunities, but also revealed that the proficiency

test induced such changes in student learning as improving English ability, studying harder, and performing better on tests.” It was also found out that the higher the proficiency

level of the students, the greater the washback effects. In addition, the study highlights the need for considerations for washback in creating educational policies related to

language testing.

The researchers of this study were able to draw out results that are important to support the concept of washback. Based on the major findings of the study, washback

has influenced several aspects of students’ learning. The study has also contributed to the existing body of knowledge since this topic is unexplored in their context. Hung and

Huang (2019) also mentioned the Common European Framework of Reference for Languages (CEFR) which is a substantial basis or reference for English language testing.

Furthermore, the research questions raised in this study are valid and valuable in addressing the general objectives. However, it must be noted that the study only included a

single institution. Hence, the study’s results may not be applicable to the schools.

The study has a great contribution to the literature supporting the existence of washback in language testing. Considering that English has become the “international

tongue” and the “lingua franca”, Hung and Huang’s study is beneficial in terms of policy making for language testing, decision making of teachers and administrators, seeing the
relevance and effect of English language testing of the students, and quality assurance (at par with the international standards for English language testing). However, since the

study has its limitations especially on its scope, it will be more worthwhile if another study will be conducted.