Classroom assessment is an integral part of the language learning process and a powerful informed decision-making tool. Unfortunately, not many language teachers in Colombia are trained to make assessment decisions that will engage and motivate students and, as a result, enhance learning. In this article, we present the results of a study that examines teachers' perceptions about language assessment and the way they use language assessments in their classroom. The findings suggest that there is a significant difference in the perceptions that teachers have depending on the level of training they have in language assessment. Thus, we highlight the importance of providing adequate training in language assessment for all prospective language teachers in Colombia.

Key words: Teachers' perceptions about classroom assessment, assessment use, assessment training

Palabras clave: Percepciones de los docentes sobre la evaluación en el aula, uso de evaluaciones, formación en evaluación

Reynolds, Livingston, & Willson (2006) argue that while many teachers love teaching, many are not very interested in assessing students. As a result, teachers tend to have a negative view of assessment. More often than not, this negative view stems from personal experiences. Terms such as assessment, testing and evaluation usually have a negative connotation as they are associated with anxiety, stress, pressure or failure. Moreover, tests play a powerful role in the lives of language learners (Hamp-Lyons, 2000; Shohamy, 2001). They provide information about both student achievement and growth, but tests are also used to provide rewards or sanctions for schools, teachers, and students. For instance, tests are used to determine who passes or fails a course, to control discipline, to threaten students, among other things (López, 2008a). This is in part why so many people have a negative view of assessment.

Something that could help minimize this negative perception is to understand the differences found in assessment, testing and evaluation. Assessment is "a term often used interchangeably with testing; but also used more broadly to encompass the gathering of language data" (Davies et al., 1999, p. 11). In other words, an assessment is any systematic procedure to collect information about students. This information is then interpreted and used to make decisions and judgments about the teaching-learning process. Testing, on the other hand, is simply one way to assess, so it can be described as a procedure to collect and interpret information using standardized procedures (American Educational Research Association [AERA], American Psychological Association [APA], & National Council on Measurement in Education [NCME], 1999). Finally, evaluation can be described as a "systematic gathering of information in order to make a decision" (Davies et al., 1999, p. 56). All these terms combined describe the classroom assessment process. Teachers gather information about what students know and can do; they interpret this information and make decisions about what to do next. Sometimes they quantify this data to assign grades and then make judgements based on them (e.g. pass/fail). What we, the authors, have learned from our experiences is that some teachers usually collect information at the end of the process and therefore the assessment cannot be used to enhance learning. Furthermore, what some teachers lack the most is the ability to use and interpret this information to guide the decision-making process.

Another aspect that needs to be mentioned here is that the assessment component is recognized as an essential part of the curriculum, but it is the area in which many teachers express a lack of confidence and claim the least knowledge (Nunan, 1988). Moreover, teachers commonly conceive assessment as an isolated activity (separate from teaching); equate assessment to simply giving a grade or score, and view assessment as a summative process rather than an ongoing process (Pérez, Guerra, & Ladrón, 2004).

In our experience we have found that in some language classrooms, assessment is not a continuous process and it tends to be more summative than formative, in the sense that the only feedback students get is their grades (López, 2008b). When we observe foreign language classrooms, more often than not we notice that assessment is generally not used appropriately. Likewise, we find that language testing is not given the importance it should have. An example of this is that some teacher education and teacher-training programs in Colombia do not offer extensive training in language assessment. As a result of this lack, tests and testing systems are often subject to abuse because test scores and test interpretations are put to a host of different uses (Hamp-Lyons, 1997). Thus, tests are frequently used unethically for purposes other than those they were intended for originally and do not facilitate the language learning process. Previous studies about language testing in Colombia have highlighted the need for more research as regards the use of assessment practices in the Colombian context (e.g. Arias & Maturana, 2005; Rodríguez, 2007).

Therefore, we feel we need to begin a conversation about the role of language testing in the classroom and in the language learning process. This is why it is critical to examine the perceptions that English language teachers have about the purpose of assessment, the use and interpretation of assessments and the impact that these have on the educational system and individuals. Research about teachers' perceptions of assessment is important because teachers' conceptions of teaching, learning, and curricula have a strong impact on how teachers teach and what students learn or achieve (Brown, 2002). The main goal of this article is to create awareness among the language teaching community in Colombia about the importance of adequately and effectively using assessments in the classroom to promote language learning. In particular, we want to focus on these two research questions:

1. What perceptions do Colombian English language teachers have about classroom assessment?
2. How are Colombian English teachers currently using language assessment in the classroom?

Classroom Assessment

Classroom assessment refers to a wide variety of strategies employed by teachers to get feedback from students about how they are experiencing the learning process (McMillan, 2003). Classroom assessments are also known as teacher-made assessments or alternative assessments (Hughes, 2003). As the name implies, teacher-made assessments are assessments made by a teacher or group of teachers for a specific set of instructional outcomes for a particular group of students. Alternative assessments, on the other hand, are broadly defined as any assessment method that is an alternative to traditional paper-and-pencil tests and requires students to demonstrate the skills and knowledge that cannot be assessed using a multiple-choice or true-false test (McNamara, 1997). Classroom assessment seeks to reveal students' critical-thinking and evaluation skills by asking them to complete open-ended tasks that often take more than one class period to complete. Some examples include portfolios, experiments, interviews, oral presentations, demonstrations, projects or exhibitions.

Assessment practices are currently undergoing a major paradigm shift mainly because of the emphasis on standardized testing and its perceived shortcomings (Hamayan, 1995). Alternative assessments were proposed as a response to large-scale assessment instruments with the idea that they would enable educators to attend to differences in learners, address learning over a period of time, and include communicative performances in a variety of ways. Some of the most commonly used alternative assessment instruments or procedures are writing samples, journals, portfolios, classroom projects, and interviews (Brown & Hudson, 1998).

Among some of the advantages we find in the use of alternative assessments are that they are more integrative than traditional tests, are more easily integrated into the classroom, provide easily understood information, are more responsive to each individual learner, promote learning and enhance access and equity in education (McNamara, 1997). Hamayan (1995) also points out that alternative assessments usually are low-stakes in terms of the consequences and supposedly have beneficial washback effects. Alderson & Wall (1993) define washback as the effects that tests have on teaching and learning. And unlike scores on large-scale assessments, alternative assessments are useful with English language learners because they can provide a multidimensional perspective of student progress and growth over time (O'Malley & Valdez Pierce, 1996). Alternative assessments also help make assessment an important component of the teaching-learning process (Cárdenas, 1997). Among some of the disadvantages we find in the use of alternative forms of assessments are that they are not easy to administer and score, are time consuming, and lack consistency in scoring (Hamayan, 1995). So, their use does not guarantee that these assessment procedures are necessarily valid and reliable. By valid we mean that the interpretations we make based on test scores are appropriate and by reliable we mean that tests are scored consistently (Bachman & Palmer, 1996).

Brown & Hudson (1998) present a critical overview of alternative assessment approaches. They point out that most of the research on alternative assessments are simply descriptive and persuasive in nature and are based on research on empirical studies examining the advantages and disadvantages of the alternative approaches to assessment. They claim that many studies, which advocate for the use of alternative assessments, present their value and validity without providing any evidence to support their claims. Their main point is that these alternative assessment instruments need to also be reliable and valid. Therefore, there is also a need for more research examining how these alternative assessment instruments are used and interpreted. More research is also needed to examine how alternative assessment procedures can be used more consistently and how we can use them to enhance teaching and learning.

Uses and Consequences of Tests

According to Shohamy (2001), tests are very powerful instruments Tests are powerful because they have the power to inform and the power to influence (Li, 1990). They have the power to inform because they provide feedback and they also have the power to influence because they often force teachers and students to do things they would not otherwise do. But tests are even more powerful when they are used as the only indicator for determining the future of English language learners (Spolsky, 1997).

Tests also serve a number of functions in society (Wall, 2000). For instance, Shohamy (1998) explains that tests are used, among other things, to define membership; to classify people; for developing curricula and textbooks, to determine criteria for success and failure; for power and control; and to influence teaching and learning. Now there is a widespread use of language assessment as an instrument in government policy (Davies, 1997; Shohamy, 2001). Assessment reform is sometimes used as a means for external control of schools and stems from a distrust of teachers (Darling-Hammond, 1994). Language tests are also used as gatekeeping instruments (Spolsky, 1997). That is, tests are often used as a means of political and social control. Potentially, tests can provide valuable data for gatekeeping decisions, but they should not be used as the only instrument to achieve these decisions (Spolsky, 1997). Spolsky urges that "we must make sure that gatekeeping processes are under human and not automatic control" (p. 6).

There has also been, in the language testing community, an increase in recognizing the social and political context of testing. Hawthorne (1997) claims that the main purpose of many tests is largely political. In fact, many of the testing systems in the world are mostly political activities and show that there is a close relation between testing and politics and that there are often political reasons behind education reform initiatives. Messick (1989) claims that tests are closely connected to a whole set of political and social values that affect the teaching, learning, curriculum, materials, politics, social classes and knowledge. But these political reasons are often in conflict with using tests to provide feedback about the learning process (Brindley, 1998). Messick (1989) argues that politicians need to consult with educators about such initiatives. This means that policy makers need to collect information about schools' needs and realities directly from teachers before they impose new educational policies.

Unfortunately, tests and testing systems are subject to abuse because test scores and test interpretations are put to a host of different uses (Hamp-Lyons, 1997). For this reason, Shohamy (2001) has developed a notion of critical language testing (CLT) which "implies the need to develop critical strategies to examine the uses and consequences of tests, to monitor their power, minimize their detrimental force, reveal the misuses, and empower the test takers" (p. 131). Shohamy (2001) uses the term 'test takers' to refer to any stakeholder group that is directly affected by the outcome of a test (e.g. schools, teachers, students, parents). Critical language assessment looks at the social, cultural and political context of assessment and challenges the fairness of language assessment (Pennycook, 2001).

Therefore, it is very important to examine how teachers use and interpret language tests in their classrooms and the consequences that they have. We believe that the lack of adequate training in language testing is one of the reasons some Colombian English language teachers are not able to monitor the consequences (intended or unintended) of their tests. Thus, it becomes particularly important in understanding how classroom assessments are going to be used (or misused) and interpreted (or misinterpreted).

Participants

Eighty-two English teachers participated in this study. We used two sampling techniques to select our participants. First, we used purposeful sampling to select key participants. According to Patton (1990), "the purpose of purposeful sampling is to select information-rich cases whose study will illuminate the questions under study" (p. 169). We established the following criteria to select key participants: 1) teachers currently teaching English in a Colombian institution, and 2) teachers who were willing to participate in the study.

We made a list of teachers who we thought would be willing to participate in this study by completing an online qualitative survey. Once we had identified some key potential participants, we contacted them via e-mail. The rest of the participants were selected through a snowball sampling technique (Patton, 1990). This technique "identifies cases of interest from people who know people who know people who know what cases are information-rich, that is, good examples for study, good interview subjects" (Patton, 1990, p. 182). We asked all the teachers to identify other key potential participants that they felt would be willing to participate in this research study.

All the participants currently teach English at different levels (primary school, secondary school, university, technical institutes or language institutes). Teachers' experience ranges from 3 to 17 years. Only 32 of the participants have had previous training in language assessment. Twentyseven of them had formal training in graduate programs (specialization, master's or doctoral program). The other five teachers had received training through certificate programs such as ICELT, TKT or IELTS, or through workshops and seminars. More information about teachers is presented in Table 1 below.

Online Survey

This research study adheres to the assumption that "the perspective of others is meaningful, knowable, and able to be made explicit" (Patton, 1990, p. 278). Therefore, we conducted a qualitative online survey to gather information from teachers in Colombia (See Appendix 1). Through the survey we wanted to obtain the participants' perspectives, experiences and concerns (Lincoln & Guba, 1985), in this case about language assessment. This survey consisted of two parts. The first part was designed to elicit background information (e.g. educational background, teaching experience) about the participants and information about the teachers' training (pre-service and in-service) in language testing. The second part was designed to elicit information about how they felt about assessment, how they used assessments, how they scored the assessments and how they provided feedback to their students. All of this information allowed us to answer the two research questions we stated earlier.

We downloaded curricula from 27 undergraduate programs and seven graduate programs aimed at training English language teachers in Colombia. We selected public and private institutions all over the country to represent the diversity in programs. Undergraduate programs included programs such as "Licenciaturas" in Modern Languages, English, Spanish and English, English and French, Philology, or Bilingual Education. Graduate programs included both specialization programs and master's programs in areas such as Autonomous Learning in EFL, Applied Linguistics, Didactics in English Teaching or Bilingual Education. These documents provided information about the number of language assessment courses offered in Colombian institutions.

The responses to the open ended questions were analyzed through a process of coding. These codes were not pre-set and emerged from the data collected as issues and ideas, which were important and relevant to the study. According to Dey (1993), a natural creation of codes occurs with "the process of finding a focus for the analysis, and reading and annotating the data" (p. 99). Consequently, we looked at all the data collected to search for meaningful patterns (Patton, 1990). As part of the data analysis, we examined similarities and differences between perceptions from teachers with training in language assessment and teachers without training. We were also able to find tendencies in language assessment practices. Conversely, the university programs were analyzed using enumeration. That is, we simply read each program and identified courses related to language assessment and counted the number of courses and number of institutions.

Teachers' Perceptions about Language Assessment

In this study, we found that there was a significant difference between the perspective of teachers who have had formal training in language assessment and those who have not. Trained teachers tend to view assessment more as an integral part of instruction and as a powerful tool to guide the learning process. For instance, one of the teachers with training in language assessment stated that he uses assessments "to keep track of the process, to measure achievement and to provide feedback". Likewise, another participant stated that she uses "assessment for learning". These statements suggest that some teachers have a positive view of assessment. In this perception, assessment is used to gather information about what the students know and can do. Then this information is interpreted to make decisions about teaching and learning. But the most important aspect is that this information is shared with the students so that they can all take steps to improve language learning. Among the most important positive views we found about language assessment are the following:

1. Assessment as a tool to align learning and instruction – "I also use the assessments to redirect my classroom practice" "I use assessments to evaluate my methodology".
]]> Assessment as a tool to monitor learning – "to see the process of my students, their strengths and weaknesses".
2. Assessment as a tool to aid in communicating with students – "to test students to know if they are improving or not in class and be able to give them better feedback".
3. Assessment as a tool to empower students – "to encourage learners to study material covered in the course".

We also found that teachers with no training (pre-service or in-service) in language assessment tend to have a more negative view of language assessments. In this view, assessment is simply used a means to give a grade or to make judgments about the students, but not as a strategy to enhance learning. For example, some of the teachers, who have not had formal training in language assessment, stated that assessments are used "to get grades that I have to submit to the institution". On a similar note, another teacher expressed that assessments are "tools to determine passing or failing". From these statements, we can infer that assessments are simply equated with grades. This implies that grades are the only feedback students get. Also, many teachers do not see the added value of assessment and only assess because they are required to do so. Among the most noticeable negative views we found about language assessment are the following:

1. Assessment as a summative process – "to generate a quantitative grade".
2. Assessment as a mandate – "to get grades that I have to submit to the institution".
3. Assessment as an instrument of power and control – "to force students to study what I teach in class".

We believe that the negative view of assessment that some teachers hold stems from a lack of adequate training in language assessment. For instance, 45 out of 48 teachers (93.8%) who had a negative view of assessment indicated that they had neither taken an assessment course in college nor had received any type of training in assessment (e.g. workshops, conferences, in-service training). On the contrary, 32 out of 34 teachers (94.1%) who had a more positive view of assessment had taken at least a course in assessment, had received language teaching training at work, or had attended a workshop on language testing.

From Table 2, we can see that only two public universities with education programs for teachers offer courses on language assessment. This is worrisome because the majority of the English teachers in Colombia are trained in these kinds of institutions. On the other hand, we found that five private universities with education programs for teachers offered a course on language assessment. This is a very promising finding in the sense that they are preparing prospective teachers to design, use and interpret assessments, and could contribute to creating a more positive view of assessment as well as starting a culture of using assessments to improve instruction. But in general, these results highlight the need for more training in language assessment in Colombian education programs for teachers.

At the graduate level, we found that all the master's programs related to English teaching are offering at least a course in language assessment (See Table 3). This is also promising because all the students taking these courses will have an opportunity to apply this knowledge in their teaching and, hopefully, help their students and fellow colleagues develop a more positive view of assessment. Unfortunately, the number of teachers who actually complete a master's program in Colombia is not very high. So we need to emphasize the importance of offering language assessment courses at the undergraduate level. Language assessment courses are a professional development space for in-service teachers' critical reflection upon their beliefs and practices regarding testing, assessment and evaluation (Quintero, 2003). Moreover, from our analysis of programs, we found three specialization programs related to English teaching in Colombia, but none of them offer a course in language assessment. As specialization programs become more popular, especially for English teachers in primary and secondary public institutions in Colombia, it is imperative that these programs also focus on language assessment.

We were also interested in gathering information about how English teachers in Colombia use language assessments in the classroom. From the online qualitative survey we found that 71 teachers usually assess at the end of a unit or course. The other 11 teachers argued that they tend to assess regularly, at least once a week. Sixty-four teachers reported that they tend to use traditional tests (e.g. paper and pencil, multiple-choice, fill in the blank types of tests). They also tend to use final exams and quizzes. Only a few teachers (18) reported using authentic classroom assessments such as essays, presentations, interviews and others.

In terms of scoring, 59 teachers explained that they scored tests by assigning numbers or letters. The other 23 teachers used qualitative descriptors to score their students' tests. Most of the feedback is provided at the end of the teaching-learning process and is usually given by assigning a grade. The feedback is usually provided by returning the marked exams. Sixty-three teachers reported using mostly objective scoring in the form of scoring keys. A few teachers (19) reported using subjective scoring methods such as scoring rubrics. In general, assessment seems to be more summative than formative. It is important for teachers to learn to provide better feedback to students. In order for classroom assessment to be effective, teachers need to provide immediate, relevant and ongoing feedback in their assessments. The feedback needs to be descriptive and it should focus on students' strengths and limitations and inform them about possible ways to reinforce or enhance learning (López, 2008b).

Language teachers should also be concerned with issues of ethics (Davies, 1997). Ethical issues deal with how tests are used and how tests results are interpreted. Language tests generally used ethically questionable and unstated political purposes that are often quite distinct from their stated purposes (Shohamy, 2001). For example, tests are sometimes used as gatekeepers or instruments to exercise power and control (Spolsky, 1997).

In the last two decades or so there has been a rise in ethical awareness in language testing. This has resulted in an increased interest in considering all the participants in the testing process. McNamara (1998) explains that the purpose of ethical language testing is to examine the role of language testers, the power they hold, the principles and structures in the use of that power, and the limits of that power. In a way, ethical language testing puts the burden of responsibility onto the tester (Hamp- Lyons, 1997).

Language teachers should set themselves high standards when they assess their students and take every step to ensure that these standards are upheld. By standards we refer to "a code of professional practice or set of professional guidelines which could cover all stages of test development, from initial construction, through trialing and on to operational use" (Davidson, Turner & Huhta, 1997, p. 303). This definition is similar to Stansfield's (1993) suggestion that language testers need to define ethics as a standard of appropriate professional practice and as a set of moral obligations. Similarly, Davies (1997) calls for a professional morality among language testers (i.e. English language teachers) to protect both the individuals from misuse and abuse of tests and to protect the profession's members.

Corson (1997) argues that ethical principles for testing should be concerned with three important issues: That everyone is treated equally; that everyone is respected; and that everyone benefits from the test. As a reaction to all these ethical concerns, the Code of Ethics of the International Language Testing Association (ILTA, 2000), developed nine principles. The last principle states that "language testers shall regularly consider the potential effects, both short and long-term, on all stakeholders of their projects, reserving the right to withhold their professional services on the grounds of conscience" (p. 6). But Davies (1997) argues that language testing professionals "have a hard task to influence other stakeholders, particularly the contracting stakeholders since the only real influences on them are their own prejudices and personal experiences" (p. 338). Language testers should, to some extent, be at least accountable for ensuring that the information they gather is used for ethical purposes. For instance, when people use language tests to exercise control rather than to provide information about the language learning process, they are being unethical (Shohamy, 2001).

Moreover, Shohamy (1997) claims that language tests which contain content or employ methods which are not fair to all test-takers are not ethical, and discusses ways of reducing various sources of unfairness. She also claims that tests should be used to provide information on proficiency levels and not to exercise control and manipulate stakeholders. It is crucial that we examine ethical issues in the assessment of English language learners in Colombia. We need to examine the ways assessment instruments are used and the consequences that are brought about with such uses.

In this study, we presented information about teachers' perspectives on language assessment. We found that there seems to be a correlation between language assessment training and perceptions about language assessment. We believe that proper education and training of teachers will help change teachers' perceptions about language assessment. If teachers have a positive view of assessment, they will be able to select or design appropriate assessment procedures for their context and students that will allow the assessments to provide useful information.

We also presented information about how teachers use language assessment in the classroom. We found that there is a tendency to use traditional assessment instead of alternative assessment. Moreover, we found that the majority of the feedback provided is in the form of a grade and is usually done at the end of the process. So from the finding in this study, we can argue that classroom assessment in English teaching in Colombia tends to be more summative than formative.

We believe that the outcomes of research studies, such as the one we present here, may stimulate administrators, pre-service and in-service teachers, and the educational community as a whole, to update their professional development and improve their assessment practices to enhance the quality of language education and students' motivation for learning. For now, it is important to remember that assessment is not simply measuring or assigning grades. We feel that it is more motivating and less threatening for language teachers to begin talking about assessment for learning rather than assessment of learning. We also think more research is needed on how tests are developed and how all the stakeholders are involved in this process, especially when this research takes into consideration the uniqueness of the Colombian context. Finally, we want to raise the issue of professionalization of the field of language assessment in Colombia. This implies that both teachers and prospective teachers need more training in language assessment. We feel that the responsibility to train language teachers in how to develop, use, score and interpret language assessments lies in higher education institutions that have education programs for teachers, in the institutions that have language programs and in the language teachers themselves. It is imperative that all prospective teachers take at least a course in language testing before they start teaching, and should strive to better themselves through in-service training, conferences, workshops and so forth to create a language assessment culture for improvement in language education.