Interaction between Language Testing Research and Classroom Testing Practice

The existing gap between research and practice in language testing has posed a huge challenge to language teachers. In particular, this study intended to examine language testing research and classroom testing activities for their degree of interaction from Iranian EFL teachers’ points of view. The analysis drew on the questionnaire developed by Nassaji (2012), which consists of quantitative and qualitative sections. The data were collected from 200 language institute teachers in Bushehr, Iran. The findings indicated that most of the teachers confirmed their familiarity with language testing research, while only a few believed that they were able to conduct or publish testing research. Although the participants acknowledged that the information gained from reading language testing could be valuable for their testing performances, they found their experience more relevant than knowledge gained from language testing research. This study also illustrated that there is a growing consensus among teachers in that there should be collaboration between researchers and teachers in testing


Introduction
The aim of this research paper is to address EFL teachers' views on the interaction between language testing research and classroom testing practice. Before, we discuss the purpose of this paper; let us look briefly to the role of English teaching throughout the world. Khansir (2013) argued that "teaching of English language develops learners' ability to enhance their international communication and improve their cultural quality so as to meet the needs of their country's social development and international exchanges" (p. 1141). Today, countries use the English language in their educational systems from primary school through to universities, and many countries use it as a second or foreign language. In the Iranian education system, English is used as a foreign language, Khansir and Gholami Dashti (2014) mentioned that in Iran, the English language is taught as a subject from the middle school and that Iranian learners should accept it as a foreign language in order to pass their examinations.
A long-standing concern has been evidenced between the interaction of research and practice in almost all fields of study. There has always been a debate among practitioners as to whether or not research contributes to professional practice (Blank & de las Alas, 2009;McNamara, 2001). This has also been excessively reflected in key speakers' and lecturers' talks in seminars and academic meetings. Additionally, the extent to which research and practice are interrelated has become a point of dilemma among practitioners. In order to minimize the gap between research and practice, the idea of "evidence-based practice" was adapted or adopted in a variety of professions, including education (Heilbronn, 2001(Heilbronn, , 2008. As this idea was derived from the hard sciences, many attempts have been made to apply it to the social sciences such as within the educational setting. Although it has been tried to bridge the gap between language testing research and classroom testing practice, warnings and questions have also been levelled against this attempt. Although the primary objective in language testing research is to improve the testing practice within the instructional setting, researchers such as Tarone, Swain, and Fathman (1976) warn language testers of "hasty pedagogical applications" (p. 29). Hatch (1978) also warned teachers against using research results in language pedagogy. More specifically, McNamara (2001) argued that assessment research has not well served the needs of classroom-based testing and that there is an urge to just conduct research that meets the students' and teachers' needs.
Different reasons underlie the existing gap between language testing research and practice. One of the major reasons is researchers' and teachers' distinctive outlooks and objectives. They, in fact, belong to two opposing trends. Researchers tend to grow technical and theoretical knowledge; whereas, teachers are oriented towards practical knowledge (Nassaji, 2012). More clearly, Ellis (2001) points out that researchers try to systematically evaluate theories through empirical studies, while teachers are more dependent on intuitive findings and experiences. The second reason, which is closely related to the first, pertains to the language of report and methodological paradigm. It is thought that the majority of language testing research is conducted in the postpositive paradigm with findings that are context free, while teachers seek practical answers within a specific context. Furthermore, the style and language of research can make findings unintelligible to teachers (Crookes, 1997).
Despite the present concerns about the relevance of language testing research to the testing practices of the classroom, there is a growing body of research that has a noticeable influence on classroom testing practices, such as alternative assessments (Ross, 1998), Computer Assistant Language Testing (CALT) (Attali, Lewis, & Steier, 2013), and different test method facets (Lee & Winke, 2012). Hence, although some researchers downplay the importance of language testing research under the pretext that research has theoretical objectives inherently inconsistent with teachers' practical needs and that experimental studies have little, if any, impact on the real actions in classes, it is considered as a narrow and unsupported view by others. Some, such as Daniel and King (1998), indicate that teachers have become aware of the contribution of testing to learning contexts. One of the ways to intensify the relevance of testing research results to testing practice and to fill the gap between the two is to let teachers themselves undertake the research. This type of research, which is also called action research or teacher research, has come to the center of attention (see Burns, 1999;Crookes & Chandler, 2001;Nunan, 1997). The principal aim of teacher research is to identify and resolve the practical problems in classrooms and to improve the educational performances. However, this is not the only way; teachers can also employ other relevant techniques including diary and survey studies, observational studies, and profiles to improve their classroom practices.
One of the basic requirements for assessing the practicality and relevance of testing research findings is the researchers' acquaintance with related testing practices in the classroom. Since language testing researchers are not familiar with classroom language testing practices, they might not necessarily qualify to present and interpret their findings in a way deemed useful for classroom practices; else, they might not have an exhaustive understanding or may interpret findings based on their own views. Therefore, it seems necessary for researchers to consult teachers to render a reliable and common interpretation of their findings. Pica (2005) recommended interaction between teachers and researchers, as it can not only help teachers understand the researchers' wording, but thereby "researchers hear what teachers are saying" (Lightbown, 2000, p. 453). Such interaction can make the outcomes of research meaningful to the practitioners and thereby facilitates application of the results within real-world educational settings.
Previous studies have been adequately attentive to testing issues such as validity and reliability (Davies, 2011;Kane, 2010), testing methods, techniques, and design (see Lee & Winke, 2012;Pellicer-Sánchez & Schmitt, 2012;Xie & Andrews, 2012). However, it could be argued that in the current assessment studies, there is a lack of concern about classroom assessment practices and language testing research. In fact, there have been very few, if any, investigations on what EFL teachers know about language testing research, and how useful they might find research results to their professional assessment activities.
The current study aims to investigate testing research and classroom testing practice by examining teachers' perceptions of the relationship between the two in terms of their familiarity with language testing research, their involvement in language testing research, and the effectiveness of testing research for their classroom testing activities. The study also aims at investigating the issues of teacher-researcher interaction, teachers' beliefs about the interaction between teachers and researchers, and what teachers can gain from testing research. The current study therefore attempts to answer the following five research questions:

Methodology
This study is a survey analysis to investigate Iranian EFL teachers' perceptions, views, and experiences of the relationship between language testing research and classroom testing practices. Through the use of an in-depth adapted questionnaire applied to 230 EFL teachers, the study draws upon both quantitative and qualitative data analysis in order to achieve proper findings. Table 1 summarizes the EFL teachers' background information. Their ages range from 20 to 42 years, with the mean of 28.17 years. Their teaching experience ranges from five to 26 years, with the mean of 8.96 years. Teachers with experience of less than five years were excluded from the study as a potential source of invalid and unreliable judgment. The participants of the study were selected from among language institute teachers as, unlike school teachers, they do not receive specific testing training and therefore design and implement tests on their own which are mostly based on their own testing experiences.
Given the particular sample of language institute teachers in Iran, there is no expectation of generalizing the study's findings. However, it is hoped to discuss some thought-provoking testing issues among teachers and those who interact with them. It is hoped that the findings will provide EFL teachers with what Bassey (2001) considers "a powerful and user-friendly summary which can serve as a guide to professional action" (p. 5). Regarding the participants' gender, 70 (33.7%) are male and 130 (66.3%) female. The majority of the participants were undergraduates (or held a Bachelor's degree) (59.8%), while 39.1% were postgraduates (or held a Master's degree). Only one participant (1.1%) held another type of degree. As for the age of the participants' students, most taught adult students (42%), whereas 39.8% taught children, and 18.2% taught both student groups. A high percentage of the teachers possessed no additional teaching certificate (81.1%), while a small number (n = 17, 18.9%) had additional teaching certificates such as TTC (Teaching Training Course) certificates.
The participants were 230 EFL teachers from language institutes in Bushehr, Iran. They voluntarily participated in the study and received the questionnaires through personal contact, e-mail, or from students or colleagues who gave them the questionnaire if they expressed interest in participating in the current study. The participants were asked to complete the questionnaire anonymously and were informed that their responses would be kept strictly confidential. Of the 230 questionnaires issued, a total of 200 were completed and returned; thus, creating a response rate of 86%.
The data were collected by an adaptation of a questionnaire designed by Nassaji (2012). The questionnaire included qualitative and quantitative sections. The qualitative sections included open-ended questions, while the quantitative sections comprised of close-ended questions such as Yes/No or Likert-type scale items. The questionnaire was piloted with 15 teachers who were asked to complete the questionnaire and to add their comments and suggestions. Then, the questionnaire was modified based on the comments and suggestions provided by the teachers. For instance, some items of the questionnaire were rewritten to fit the SLA and EFL context of the study. Furthermore, analysis of the pilot data suggested that some parts of the questionnaire such as teachers' expectation of testing research and journal names might not be relevant to the purpose of the current study, and were therefore removed.
Following the administration of the questionnaire, the Yes/No and Likert-type item responses were analyzed for their frequencies and response percentage. In some of the results tables it can be seen that the totals do not add up to 200, since some teachers avoided answering all of the items. Finally, qualitative analysis was also performed on the responses to the open-ended questions.

Results
As can be seen in Table 2, over half of the participants indicated that they had taken courses in second language research (58%) and language testing (60%). Table 2 shows a large number of respondents indicated that they found research and testing courses to be useful. Unlike the first two parts, a high percentage of the teachers (72%) stated that they had not conducted any research. When asked for their reasons as to why they had not done any research, more than two-thirds indicated "lack of time" (39.4%) and "interest" (32.3%). Surprisingly, none of the respondents pointed to uselessness of research as their reasoning (see Table 3). EDUPIJ • Volume 8 • Issue 1 • 2019  Teachers' access to language testing research, sources for teachers' consultation, extent to which teachers read language testing research articles, and their reasons for not reading LTRA (Language Testing Research Articles) In this section, the teachers were asked to note the degree of their accessibility to testing research articles. They were also asked to state the testing research sources that they consulted, and to indicate the extent to which they read or not read testing research articles. They were also asked to give their reasons for that. As Table 4 shows, more than half of the teachers (59.4%) agreed that testing research articles were "easily accessible to them." Those who gave a positive answer to this question pointed to their sources of testing articles. Of the given options, "Internet" and "books" (39.3% and 35.7%) respectively saw the largest response. A relatively lower percentage of the teachers (21.4%) noted "journals" as the main source of their consultation. This question was followed by one appertaining to the sources teachers used to obtain information about testing. The sources in order of their use by the participants were "reading journal articles" (31.8%), "reading books" (30.4%), "talking to colleagues" (22.3%), "attending conferences and workshops" (8.1%), and "conducting empirical research" (6.1%) (see Table 4). With regard to the extent to which the teachers read testing research articles, just over half of the teachers indicated that they "rarely" (42.1%) or "never" (10.5%) read such articles. Only 11.6% of the teachers cited that they "often" read articles about language testing research; with 33.7% indicating that they "sometimes" read such articles (see Table 5). When asked to put forward their reasons for not reading LTRA (cases of rarely and never), above one-third of the teachers attributed that to "lack of interest" (34.4%) and 29.3% pointed to a "lack of time" as the main reason for not reading testing research. However, although a much lower number (n = 10), some teachers indicated that they found reading LTRA to be "useless" (see Table 5). Some teachers added comments along with their response, such as the following:

Practicality of testing research to classroom testing practice
As Table 6 displays, a noticeable percentage of respondents (49.5%) agreed or strongly agreed (14.4%) with the statement that knowing about language testing research improves language testing practice. On the other hand, very few of the teachers (n = 14) disagreed (4.1%) or strongly disagreed (2.1%) with that statement. Likewise, a majority of the teachers agreed (54.2%) or strongly agreed (17.7%) with the idea that teachers can benefit from the practical tips of language testing research, while only a few (n = 8) showed they "disagreed" or "strongly disagreed". Over a half of the teachers "disagreed" (43.8%) or "strongly disagreed" (13.5%) that language testing research could be related to language class testing. EDUPIJ • Volume 8 • Issue 1 • 2019 As for the last item, 41.7% of the teachers "somewhat agreed" and 22.9% "agreed" that their experiences were more practically relevant to their testing practices than the suggestions offered by language testing research. Contrastively, 15.26% of the teachers "somewhat disagreed" and 11.5% "disagreed" with the statement, thus indicating that language testing research is more contributive to their testing practices.

Researcher-teacher relationship
As can be seen in Table 7, more than half of the teachers "strongly disagreed" (21.9%) or "disagreed" (30.2%) with the statement that researchers should be university professors or academics and not teachers, while very few "agreed" (9.4%) or "strongly agreed" (4.2%) with this statement. As for the second statement, there was a relative balance in the distribution of data, with over one-quarter of the teachers (27.8%) having "disagreed" with the statement and 21.6% "somewhat agreed." As expected, nearly all the teachers (combined total of 91.7%) "strongly agreed" (22.7%), "agreed" (44.3%), or "somewhat agreed" (24.7%) with the statement that teachers and researchers should work together. None of the teachers "strongly disagreed" with this statement. Similarly, a small proportion of the teachers (8.4%) "disagreed" or "somewhat disagreed" with the idea that teachers and researchers must consult on testing issues. On the contrary, nearly all the teachers (combined total of 91.6%) "strongly agreed" (15.6%), "agreed" (45.8%), or "somewhat agreed" (30.2%) to the idea that teachers and researchers must consult on testing practices. As for the last statement, the majority of the teachers "agreed" (45.8%) or "strongly agreed" (16.7%) that researchers should consult teachers for advice on issues they want to research (62.5%). Similar to the previous item, a small percentage of the teachers (8.3%) indicated that researchers do not need to consult teachers for advice on the issues they want to research. The analysis of the findings for the last two items showed that the participants commonly believed that researchers and teachers should be involved in educational assistance. Some teachers highlighted this point through their written comments: I think that we, as teachers, should work with researchers if we want to improve our teaching and testing knowledge. (Additional response from an EFL teacher)

Discussion
The investigation of Iranian EFL teachers' testing research and testing practices, and the way the two can be related showed that a large percentage of the participant language institute teachers were familiar with language testing research and had attended research method courses. However, very few of the teachers stated that they were able to conduct and publish research themselves, referring to a lack of time and interest as the main reasons for this. Most of the teachers agreed on the practicality of testing research and courses for classroom testing practice.
Regarding the resources that the teachers consulted, it was revealed that the Internet and journal articles were the main resources. The frequent use of Internet and research articles could be attributable to the Internet and such technologies being very much part of modern life which therefore facilitates quick and easy access to research articles. Today, teachers are more eager to have access to immediate and quick information. More than half of the teachers noted that they rarely or never read testing research articles. However, it should be pointed out that the exclusive reliance on research as the solitary source of change in the classroom assessment activities would not be sufficient. Other relevant mediating factors are also involved in the process of change (Tierney, 2006). Similar to the reasons mentioned for the avoidance of conducting research, most of the participants in the current study asserted lack of time and interest as the chief reasons for not reading language testing research. Teachers mostly have too busy a schedule to find time for reading or for conducting testing research. Teachers are primarily busy with teaching and classroom management, and are therefore reluctant to read or conduct any studies. This is a common problem in teacher education (e.g., Cambone, 1995;Collinson & Cook, 2001 In addition to time limitation, lack of interest was also found to constrain teachers from reading testing research articles. This lack of interest to read testing research articles can be highly attributed to the difficulty associated with reading them. Since the majority of testing articles are quantitative in nature with considerable psychometric data analysis, their registers and discourse could seem quite mind-numbing to EFL teachers. Scholars have offered suggestions for solutions to this problem. First, Crookes (1997) suggested that researchers should report their results in a simpler way in order to be more understandable to teachers. Although some one-third of the teachers in the current study were graduates, few had conducted research themselves or published papers. Second, Gass (1995) suggested that teachers be familiarized with discourse and format of scientific research. Language institutes could hold workshops to familiarize teachers with the innovative trends and issues in language testing and technical knowledge in research.
The suggestion from Gass (1995) could be directly linked to the finding in the current study that the majority of teachers confirmed the contributive and constructive role of testing research in testing activities; although most of them indicated that their practical experience seemed more beneficial in this regard. However, filling the gap between testing research and practice through training courses could raise problems in that teachers' understanding and evaluation of research findings might be quite different from those of the researchers. Teachers might draw on different sources and strategies reading research articles (Bartels, 2003). On the other hand, researchers' attention to comparability and interpretability of assessments has somewhat blurred the boundary between language testing research and classroom testing (McNamara, 2001).
Furthermore, the idea of familiarization highlights a top-down model in teacher education (Nassaji, 2012), which positions researchers in a higher position than that of teachers. This view that researchers are producers of information and teachers are consumers could certainly discourage teachers from reading testing research articles. Accordingly, teachers prefer to rely upon their experience more than research findings. An alternative suggestion is that teachers should play a more active role in meeting their testing requirements through their involvement in testing research.
As for the last part of the findings, a significant proportion of the teachers advocated close cooperation between researchers and teachers. This level of cooperation can be accomplished in two ways; "teacher-researcher collaboration" and "teacher research." As for teacher-researcher collaboration, they can be jointly involved in one single research activity (Lightbown, 2000). In mutual activities they have a better chance of being aware of each other's concerns and problems. This would also strengthen the reciprocal interaction between research and practice (Ellis, 1997;Pica, 2005). They can collaborate through personal talks, seminars and conferences, as well as research projects etc.
The second suggestion is for language institute teachers themselves to conduct studies in order to resolve their practical concerns in the classroom. Although some of the challenging issues raised in class could be addressed via the teachers' personal experiences, there are others which need a detailed action research plan to be satisfactorily dealt with. In these cases, teachers are the only ones who can find an effective solution to the given EDUPIJ • Volume 8 • Issue 1 • 2019 classroom challenges. It has been discussed that action research can improve educational practice (Mills, 2011;Stringer, 2008). In fact, teacher research is one of the stronger tools to fill the gap between research and practice (Hine, 2013;Mills, 2011). It should be argued that these studies are strong assets for meeting class-based issues. In testing practices, in particular, teachers can not only have the initiative and innovation to introduce new insights into their profession, but they can also testify as to the practicality of researchers' hypotheses in testing.
The given suggestions could improve the current state of language testing in Iranian language institutes. Language institute teachers may not feel the necessity of upgrading their testing techniques and strategies; there also might not be enough room for innovation in language testing (probably due to lack of interest and time). Additionally, teachers might not realize the importance of acquiring the essential skills for designing tests, preferring to be mere followers and consumers. This traditional norm of language testing in language institutes in Iran has resulted in a lack of originality in testing performances. As a result, teachers are forced to take the traditional constructs in language testing for granted. Altogether, they are regarded as a negative byproduct derived from a specific educational system. Raising teachers' awareness of alternative ways to approach language testing (teacher-researcher collaboration, and action research) can to a large extent improve the contemporary condition.

Conclusions
Consistent cooperation between teachers and researchers is highly suggested as being able to be operationalized in two distinct ways. First, teachers can inform researchers of the possible challenges in testing performance (e.g., test methods, test qualities, test design), and thereby jointly work towards solving a common problem. Since researchers are more involved with scientific studies, teachers, on the other hand, are more involved with practice. Therefore, each group can offer inspiring ideas to the other. It seems that identifying practical testing problems is the teachers' responsibility, while finding scientific answers to such problems is the researchers' duty. Successful interaction requires that both sides contribute.
Action research was also mentioned as another possibility which empowers teachers to overcome problems. Action research, in fact, was introduced as an important method for teacher researcher (Mills, 2011). Researchers can share their activities with teachers by holding workshops and classes to help teachers master the vital and necessary techniques of conducting scientific research. As suggested by the findings in the current study, teachers have either lost interest in or face considerable issues with lack of available time to conduct research. In order to improve the situation, educational organizations and institutes can include research courses as part of their in-service programs. By highlighting the importance of such classes, teachers are more likely to regain motivation to conduct research. Additionally, this would lead to teacher's independence from testing researchers.

Notes
Corresponding author: ALI AKBAR KHANSIR