Testing is inevitable although not desirable. It is necessary in order to keep the track of overall progress of language teaching programme. Debates have been going on for and against the testing. However, the important point to note here is that it is the faulty process of testing that is being criticized not the concept of testing itself. In fact, such criticism is necessary as it can help improve the system. The sphere of language testing in Nepal is also not free from criticism. Therefore, we decided to test the testing system of Nepal in this interactive article. We have attempted to explore the existing problems in the field of language testing and possible solutions to them after an interaction with experts and readers. We believe such interactive can play a significant role to reform the system. A thematic question was asked to language experts as well as Choutari readers. The question was ‘What is a major problem in language testing system of Nepal and what can be the solution to it?’ Among the responses collected, we have presented the opinions of eight respondents here:
There are many problems with current language testing regime (as well as some good things). One issue that’s come up in our conversations is how testing practices typically ignore multilingual competencies. At first, this may seem like an impossible ideal, but if you look deeper, the question becomes why not. Ours is a multilingual society and students’ language proficiencies are not isolated; their English is a part of a complex sociolinguistic tapestry; their other languages don’t “hamper” English; languages aren’t just mediums but rich epistemological resources; and, humans have always spoken multiple languages without seeking a monolingual standard. So, when we face the task of teaching and testing students’ English abilities in isolation, we shouldn’t act like helpless slaves of the system; when discussing the roots and stems and branches and bitter fruits of the current regimes, there’s no need to surrender to the “reality.” The reality includes politics, power, and possibilities beyond their grips, and thus, we must broaden the base of our discussions so we can see testing as a broader phenomenon than, well, testing. Scholarly conversations under the tree here can and should help the community rethink the fundamentals.
Shyam Sharma is an Assistant Professor in the Program in Writing and Rhetoric at Stony Brook University (State University of New York)
I call it an ‘issue’ rather than a ‘problem’; why do we still ‘test’ monolingual ability (although our students have bi-/multilingual ability)? Another issue embedded within this issue is: How can we test students’ multilingual ability? First, we must be clear that ‘testing’ is not a ‘fixing-shop’ where you can fix a ‘problem’ rather it is a complex discipline which needs a critical scrutiny from multiple perspectives for a valid evaluation of students’ ability. Our assumption that ‘language testing’ should only test ‘monolingual ability’, meaning that multilingual testing is impossible, is the major challenge for reforms in language testing. This dominant assumption decontextualizes language testing from students’ cultural, linguistic and educational contexts. So, the major issue is: our tests are not context-sensitive. For example, I still remember that we were often asked to write an essay in SLC (School Leaving Certificate) exam about different highways in Nepal but I had never seen any highways (when I was in school). We were asked to memorize their lengths, construction dates and so on. I could not even conceptualize what a ‘highway’ was. However, I could write more and better when I had to write about ‘my village’ or ‘my school’.
The issue of contextualization is closely associated with testing multilingual abilities; locally-contextualized test items require students to work with their abilities in more than one language. For example, when I had to write an essay about my village I used to think in Limbu, Nepali and English. I (and my friends) could not think about the topic in only one language – no separation of languages! But the tests did not allow me to use my Limbu and Nepali abilities while writing essays in English. This is the major issue, right? If language tests are meant to test ‘language ability’, why don’t we test students’ functional abilities in multiple languages? This applies to Nepali language tests as well. For example, when students speak Nepali they simultaneously use English as well (and/or other local languages if their first language is other than Nepali); one cannot create the fixed boundary of a language. Suppose a bilingual student writes “आजको class मा कस्तो frustrate भएको…” (I had frustration in today’s class) for her Nepali essay (it can be more complex than this in the case of Maithili and Newari children, for example), how do we evaluate her Nepali language ability? The first reaction could be ‘असुद्द” (incorrect –literally impure). However, she is expressing her views fluently by using both Nepali and English in her repertoire. She cannot separate one language from another. This means that monolingual tests do not test students’ bilingual or multilingual abilities. Unfortunately, the students who show their bi-/multilingual abilities in language tests are considered ‘deficient’ and ‘poor’. However, the above example represents the use of language in the real-life (authentic) context.
There are ways to test multilingual abilities. For example, an inquiry-based formative assessment, which engages students in doing research and working with teachers to receive qualitative feedback on their work, can be one way to help them fully utilize their multilingual abilities. Such assessments encourage students to translanguage (use multiple languages to perform different tasks) to achieve the goals as specified by the test criteria. However, any kind of so-called ‘standardized test’, which are guided by the monolingual assumption, cannot test bi-/multilingual abilities. We should say a big ‘NO’ to the standardized tests if we truly believe in developing equitable language testing.
Prem Phyak is an MA (TESOL), Institute of Education, University of London, UK, M.Ed., Tribhuvan University, Nepal
Tirth Raj Khaniya:
Lack of professionalism is the main problem of English Language Testing in the context of Nepal. Professionalism is known as ability of applying fairness, ethics and standards in exam related issues. While dealing with exam related matters we need to be fair. We assume that we are professional but in reality we are not professional thus the test is not testing what it is supposed to test.
In language testing for teachers’ to be professional they require both necessary skills and abilities and application of those skills and abilities in a proper manner. To maintain professionalism it is necessary to have wide discussion among teachers and therefore all those who are involved in exams will have clear understanding.
Tirth Raj Khaniya has a Ph. D. in Language Testing from University of Edinburgh, UK. Currently, a Professor of English Education, he teaches language testing in the Department of English Education, TU.
Ganga Ram Gautam:
The main problem of language testing in Nepal is that the test itself is faulty. It does not test the language skills but test the memory of the text materials given in the textbook. There are also other several problems that include the issues with the test writers, test item construction, test administration and validation of the tests.
One solution of this problem could be to develop standardized tests and administer them in the various key stages such as primary level, lower secondary level and secondary level. In order to do this, we need to train a team of experts to develop the test and the test should be standardized by going through the reliability and validity testing. Once the tests are developed, they should be administered in a proper way so that the real language proficiency of the students can be obtained.
Ganga Ram Gautam is an Associate Professor at Mahendra Ratna Campus, Tribhuvan University and former president of NELTA.
There is no need to reiterate that the aim of the learning a foreign language is to be able to communicate in it. In order to find out whether English language learners in the Nepalese schools have developed communicative skills in this foreign language, there is a provision for the testing of listening and speaking at the SLC level. I feel that this test is not serving the purpose. The lowest marks students get in speaking is 10 out of 15, which is 66%. However, when we communicate with the SLC graduates (let alone who fail the examination), most of them perform very poorly. There are two reasons for this inflated marking: the speaking test includes predictable questions for which the responses can be rehearsed: personal introduction, picture description and one function-based question (which is repeated so often that students can prepare a limited set of responses and be ready of the test). Secondly, there is a kind of extreme leniency in the examiners; they just award marks irrespective of the quality if the responses.
Two interventions could improve the situation. Firstly, the examiners should be trained to ask very simple everyday realistic questions which students cannot respond without knowing the language. Secondly, each test should be video recorded so that inflated marks can be easily scrutinised. Administrative issues should not come in the way of quality testing which has far-reaching consequences.
Laxman Gnawali is an Associate Professor at Kathmandu University and Former Senior Vice President of NELTA
Laxmi Prasad Ojha:
I think we are giving too much priority to examinations and tests in our education system. We do not understand the purpose of testing and evaluation. We don’t test the comprehension and understanding of students. This is the main cause of the failure of our education system in many cases, including the language teaching programmes.
I think “formative” should be the key word here. Laxmi ji, pointed out an important bottleneck we have experienced due to lack of purpose of testing and evaluation. If we think of a typical Nepali school, we do give more importance on summative tests than the formative ones. What we seriously lack (and that’s why we have a tremendous opportunity to work on) is systematic feedback for student.
Uttam Gaulee is Graduate Research Fellow, University of Florida College of Education, Gainesville, Florida
Bal Krishna Sharma:
Yeah, one way would be to introduce and practice more formative type of assessment. This will evaluate and test students’ ongoing progress and learning outcomes.
Ph.D. student, University of Hawaii at Manoa
Although the issue was one, the thematic question unbelievably raised so many genuine issues. The respondents highlighted the issue of testing multilingual competencies apart from only testing monolingual ability and also suggested some ideas on how to test students’ multilingual abilities. In the same way, the interaction raised the issue of lack of professionalism in language testing. Similarly, the respondents also urged that our memory-driven testing system itself is faulty. Furthermore, there is problem in test construction and administration and suggestion is put forward to develop and practise standarized tests to minimize the problems. In relation to the problem in testing listening and speaking in SLC exam, it emphasized that the test items are predictable and examiners are lenient and award marks irrespective of quality. The solution proposed is to train the examiners properly and introduce the system of video recording students’ performance. On the other hand, overemphasizing exams and not testing what it should test is characterized as a problem. The solution discussed over such problem is to give more importance to formative test rather than summative test, which helps keep the track of students’ achievement.
Now the floor is open for you. Share what you think is the problem of testing system in our context and what can be the solution. We believe such interaction contributes in the development of innovative ideas in ELT.