Journal of English Language Studies

The current study set out to investigate the alignment of alternative assessment (AA) with the communicative English skills course (CESC) curriculum design and the CESC teaching practices at three Ethiopian universities. Employing a mixed-methods design, it was guided by three research questions. The study utilised summative content data and quantitative content data, and subjected them to descriptive statistics, and Mann-Whitney U Test and Kruskal-Wallis Test. Some of its main results are worth mentioning. Firstly, out of a total of 632 assessment items in CESC, only 30.22% of them were found to be communicatively using AA, while 69.78% of the items in the module were found to be using traditional assessment (TA). Secondly, the instructors’ assessment practices in CESC were inconsistent with AA, and varied across the three universities and from instructor to instructor. The assessment items were disproportionate to the language objectives in CESC because the instructors mainly devoted more time to assessing reading skills, grammar, and vocabulary knowledge. For example, the correlation coefficients of the assessment items for grammar (0.142), vocabulary (0.139), and reading (0.115) were better aligned with the content items in the CESC module than the assessment


INTRODUCTION
On the theoretical basis, the principles and applications of communicative language teaching (CLT) have globally been employed by textbook writers and by language teaching specialists to match the assessment techniques with course design, and language teaching and learning activities (Al-Mamari, Al-Mekhlafi & Al-Barwani, 2018;Kibbe, 2017).The alignment between teaching and assessment methods in the context of a communicative English skills course (CESC) is evaluated using a CLT lens.To this end, a CLT theoretical framework was reconstructed to illustrate how the design of the curriculum, and methods of instruction and assessment are aligned in CESC (see Motuma & Chaka, forthcoming).The assumptions of CLT are based on its origin, its evaluation and its influence on language curriculum design, and teaching and assessment practices (Dames, 2012;Garuana & Mcpherson, 2015;Kibbe, 2017).Likewise, CLT, which is a recent phenomenon in the field of Teaching English as a Foreign Language (TEFL) at higher education in Ethiopia, has been understood as a set of principles of language teaching and assessment as argued by Morrow (2018).These principles describe how students learn a language, and the types of classroom tasks to facilitate learning, and the role of instructors and students in a communicative language classroom (Morrow, 2018;Motuma, 2018;Reyes-Chua, 2013).
The term 'communicative' embodies and aligns three interrelated theories: the theory of communicative course design, the theory of active learning method (ALM), and the theory of alternative assessment (AA) in CESC as employed by Al-Mamari et al. (2018), Motuma (2018), Forutan (2014), and Motuma and Chaka (forthcoming), and as implied in Figure 1.
Figure 1: The alignment of the communicative language teaching (CLT) components (also see Motuma & Chaka, forthcoming) In the context of CESC, the communicative course design underpins CLT and assessment practices.It also informs: content-based activities; theme-based learning tasks; experiential learning-based tasks; episode hypothesis-based tasks; and integrated learning tasks (see Ansarey, 2012;Brown;2010;Davies;2013;Motuma & Chaka, forthcoming;Smith;2000;Ur, 2010).The reason for incorporating these types of activities and tasks is that CESC focuses on diverse real-world, well-integrated, interactive, and practical tasks that demand CLT-oriented assessment approaches intended to help students develop and master language competence.
From the point of view of CLT, certain studies equate the principles of AA with the principles of ALM in the context of CESC, as described by Benzehaf (2017), Davies (2013), Kibbe (2017) and Morrow (2018).Hence, the fundamental principles of CESC teaching and assessment practices can precisely be conceptualised as the alignment of ALM and AA with the communicative course design (CCD).These principles inform the purpose of ALM and AA, the roles of the instructors and learners in the context of CESC (Benzehaf, 2017;Kibbe, 2017;Wood, 2011).Both ALM and AA include the use of multiple strategies, a variety of tools, comprehensive, progressive, and relevant teaching, and assessment activities to meet the components of language domains in CESC (Benzehaf, 2017;Morrow, 2018;Smith, 2000).Thus, the alignment of AA and ALM is essential for the communicative nature of CESC because it cannot be taught and assessed by traditional approaches (TA), as argued by Brown (2012).If an instructor teaches or assesses CESC using TA, certainly, there is a mismatch between how the course is taught and how it is assessed.
Therefore, the teaching and assessment practices in CESC should be consistent with its intended learning outcomes.
The main principles of CLT that make up a communicative language course design, and that inform teaching and assessment practices are related to meaningful communication, authentic situation, creative language input, unpredictable language output and integrated skills in the course design, and teaching and assessment activities as identified by Trines 2018) and as recommended by Benzehaf (2017), Kibbe (2017), andMorrow (2018).In this study, these five requirements, which are not exclusive of one another, are used as alignment strategies for the inseparable components of CESC.
One of the five requirements for the communicativeness of a course design is that teaching and assessment need to be associated with the meaningfulness of the tasks and/or practices in CESC.A meaningful practice can meet learners' learning styles, interests and desires, thereby, stimulating them to creatively use language skills (Benzehaf, 2017;Kibbe, 2017).Such a practice needs to be based on real-life and contextual tasks as argued by WENR (2018).These could include greetings, introductions, descriptions of places, ideas and persons, and drawing and labelling objects and maps (Ansarey, 2012).The second requirement is related to the authenticity of the course design and the teaching and assessment practices in CESC.Ansarey (2012) argues that the alignment of ALM and AA provides learners with the opportunities to receptively and productively use language in authentic situations.This situation begins with what students know and move on to a creative language use.For this reason, instructors ought to extract authentic materials to design tasks and implement them in the teaching and assessment activities to create authentic, interactive and lively atmosphere for students to progressively respond to the curriculum as emphasised by Devi (2019).These can be achieved through information gap activities, jigsaws, information gathering, and information transfer as well as reasoning gap activities, role plays, and pair and group work (Christiana, 2019;Kibbe, 2017).Supporting this idea, Brown (2010) contends that multiple methods are necessary to teach and assess learners' multiple talents and multiple abilities.
Unpredictable language input is the third requirement for the communicativeness of teaching and for corresponding assessment practices in CESC, particularly as explained by Devi (2019), Krashen (1985), and Liu (2015).In reality, language input is usually what an instructor can provide (Christiana, 2019).
This comprises interpretation of new information, beliefs, views, opinions, real-world tasks, opinion sharing, and ordering activities.Unpredictable language input encourages learners to engage in explorations and data gathering through puzzles and games, and through map-reading and picture-labelling activities (Fraooq, 2015;Morrow, 2018).The fourth requirement for the communicativeness of tasks is students' creative language output (Devi, 2019;Liu, 2015;Swain, 1993).Language output is largely dependent on language input.Creative language output can be realised through integrated, contextual, and progressive activities, which involve visual presentations, virtual reality, real-world problem solving, and language quest projects (Sidek, 2012;Wood, 2011).The fifth and last requirement for communicativeness focuses on the integrated language skills teaching and assessment practices in CESC, as in real-life communication.Every activity is expected to integrate at least two or more language skills for it to be able to foster some form of communicativeness.
The implementation of ALM and AA in CESC can be looked at from the point of view of "freedom factor" as defined by Benzehaf (2017), which is an amalgam of a number of components, such as initiative, choice, vision, self-discipline, compassion, trust, and spontaneity.This means, students set goals for themselves, select the work that demonstrates what they have learned, choose what they want to learn, and are able to explain why they chose it.As a key aspect of the abovementioned theoretical framework, one cannot imagine the implementation of AA strategies without ALM in CESC (Benzehaf, 2017;Kabouha & Elias, 2015;Wood, 2011).
In relation to the application of ALM and AA in the Ethiopian context, the concepts of CA and ALM became officially operative after 1994 (Getachew & Deribe (2006), following the implementation of the Ethiopian educational and training policy (ETP) (MOE, 1994).However, ETP does not explicitly state the implementation of AA.Prompted by the drawbacks of TA, MOE has been urging universities, in particular, to implement CLT in English language courses to enhance students' communicative competence in the country (MOE, 2018).To this end, the government has developed a policy reform known as one-to-five education quality development army, which aims to link some of the elements of AA to ALMs in order to enhance communicative language teaching practices in the country (Getachew & Deribe, 2006;Temesgen, 2017).Likewise, CESC was first designed and developed in 2013, and it was revised in 2018 by a committee nationally established by MOE, which used it across universities.This nationally harmonised curriculum is known as Communicative English Skills Course I, and is intended for undergraduate students.It is a three-credit bearing or a five European credit transfer system (ECTS) course.
The current study focused on three Ethiopian universities, namely, Addis Ababa University (AAU), Ambo University (AU), and Wollaga University (WU).Its research questions were: (1) To what extent are the communicative English skills course (CESC) materials designed communicatively?
(2) To what extent do the instructors orient assessment practices to the CESC curriculum materials and to the CESC teaching practices?
(3) Are there any differences between the three universities in aligning their assessment approaches with the CESC teaching approach?

METHOD
This study employed a mixed-methods design.The reason it did so is that its data sets comprised quantitative data and qualitative data.In particular, the study utilised a dialectics aspect of the mixed-methods approach.A dialectic perspective combines two or more paradigms in a meaningful and complementary way.Its emphasis is on trying to harmonise divergent data (see Shannon-Baker, 2016).While the study is aware of the ongoing and the unresolved debate related to the mixedmethods approach (see, for example, Creswell & Creswell, 2018;Greene, Caracelli, & Graham, 1989;Sandorova, 2014), it deliberately steers clear of this debate as it is not its major focus.The quantitative and qualitative nature of the data sets of this study is described below.The three government universities, AAU (the largest and preeminent), WU and AU at which the study was conducted, were purposively selected from first, second and third generation universities, respectively.

Data Sources and Data Analysis
A centrally designed and developed CESC guidebook, which is commonly used by all Ethiopian universities, and assessment materials including continuous assessment (CA) and summative assessment (SA) items, were selected as the main sources of the data.This is because the communicativeness of the teaching and the assessment materials are supposed to imply the alignment between AA and ALM in the context of CESC.Based on this assumption, the whole units of the teaching materials, three sets of CA and two sets of SA of CESC, which were used by the observed instructors from 2019-2020, were purposively selected at each university for the present study.
To determine the alignment between ALM and AA in CESC, summative content analysis approach was employed using a checklist and a protocol which were adapted from Christiana (2019) and Hashemnezhad (2015).This checklist and this protocol are commonly used for both the teaching and assessment materials because the communicativeness of the all items in the teaching materials and in the assessment, ensure the alignment between AA and ALM in CESC (see Christiana, 2019;Hashemnezhad, 2015;Kabouha & Elias, 2015) and Shannon-Baker (2016).
Concerning a mixed-methods approach, a summative content analysis method is not a mere counting of items in a certain topic in a text: it is the process to determine the existence and frequency of the explicitly and implicitly stated concepts related to the communicativeness of items in the teaching and assessment documents of CESC.The steps of a summative content analysis method adhered to in this study included identifying the communicative-related items; developing categories; determining the unit of analysis and coding systems; counting the occurrence frequencies; and determining the meaning and the relationship between the items (Sandorova, 2014;Shannon-Baker, 2016; also see Griffiths, 2016;Hsieh & Shannon, 2005).
Language items, such as instructions, methods, and questions in every activity in the teaching and assessment documents were considered and clustered separately under teaching and assessment items.Every item as a unit of analysis was coded as a CIT and as a CIA to signal the communicativeness of the items in both the teaching and assessment materials.In both cases, six digits were accorded to each code.The digits in each code imply whether an item satisfies the five fundamental requirements of CLT: (1) meaningful communication; (2) authentic situation; (3) unpredictable language input; (4) creative language output; and (5) integrated language skills.The last digit indicates the frequency of occurrence in both cases.For instance, CIT123451 and CIA123451 imply that the first item in both cases satisfies all the requirements of CLT.Therefore, in this case, items are communicative, and there is an alignment between the two items.If an item fulfills three or more requirements of CLT, then, the item is considered communicative.If not, it is regarded as uncommunicative or conventional.Summative content analysis was supplemented by qualitative content analysis to represent underlining deeper meaning of the data and to enhance the quality of the findings of the study (see Hashemnezhad, 2015;Sandorova, 2014).
Quantitative content analysis was conducted to determine the level of the alignment between the course design, teaching, and the assessment practices.
Percentages, means, ranking orders, standard deviation, Pearson Sidney Siegel's correlation contingency coefficient (C), Mann-Whitney U Test, and Kruskal-Wallis Test were computed based on the value of the: • observed communicative value in the module (OCVM); • observed communicative value in the assessment (OCVA); • average observed items value in the assessment (AOIVA); • overall average value of the module (OAVM); • average of items in the assessment materials (AIAM); • expected assessment value (EAV); • Chi-square (x 2 ); and • Correlation coefficient (see Table 7).
Quantitative content analysis was conducted using SPSS ® 26.0.The quantitative content analysis results were interpreted against a predetermined criterion to make the findings as meaningful as possible (see Shannon-Baker, 2016;Sidek, 2012).Qualitative data was compared with its quantitative data counterpart with a view to providing a descriptive summary of the surface meaning of the data.

RESULTS
The alignment of ALM and AA is relevant for teaching and assessing CESC.
Thus, the analysis of the communicativeness of teaching and assessment materials was conducted in order to determine the alignment of AA with the course design and ALM in CESC.

Communicativeness of CESC Teaching Materials
The key objective of this section is to determine the level of the communicativeness of the CESC teaching materials used during 2019-2020.To determine this objective, the course materials were analysed before the assessment materials.The items analysed are displayed in Table 1.
Table 1.A matrix of language domains and contents in each unit in CESC As depicted in Table 1, the CESC module incorporates five units, each of which commonly includes six major language domains: speaking, reading, writing, listening, vocabulary, and grammar.The sequence of the six language domains in each unit of the module varies from unit to unit for no apparent reason.However, Davies (2013) and Demir & Pismek (2018) argue that the sequence of the language domains in such a course should contribute to the communicativeness of the language skills.Regarding the language domains, CESC has six general objectives that students are expected to attain to complete the course (see Table 2).There are still authentic objectives in CESC module which require students to use contextualized language as in a real-life situation, such as giving instruction, asking and giving direction, introducing oneself and others, asking for help, making requests and giving response to requests, giving advice, making excuses, describing processes, comparing and/or contrasting things or ideas, defending an argument, classifying objects or ideas, narrating stories and so forth.
Table 3 presents a summary of the communicative and the uncommunicative items in CESC materials.skills received more attention than other skills.This means that 372 (66.43%) of the items in the module were evaluated to be communicative (see Davies, 2013).

Communicativeness of Assessment Materials in CESC
This section presents the analysis of the items related to the assessment materials used by instructors in CESC at the three universities.Based on the Ethiopian universities' assessment policy, English as a foreign language (EFL) instructors employed 50% CA and 50% SA or final examination (FE) in CESC as summarised in Table 4. Table 5 shows that EFL instructors employed 191(30.22%)communicative items out of 632 items to assess language domains in CESC at the three universities.
Specifically, 73(31.47%),61(30.81%)and 57 (28.22%) of the items used by the instructors were communicative at AAU, AU and WU, respectively.Using the same teaching module, 522 (82.6%) of the instructors' assessment items were dominated by reading, grammar, and vocabulary at the three universities.Reading activities received greater attention at AU and WU, followed by grammar and vocabulary items.At AAU, however, grammar, 66 (28.45%), received more attention than reading items in CESC.Vocabulary assessment items were the third emphasised language domain in CESC at the three universities.Nevertheless, listening 9(1.4%), writing 45 (7.1%) and speaking (8.9%) skills were overlooked by the instructors at the three universities.

Alignment between Teaching and Assessment Items
This section presents the results of the relationship between the teaching of CESC items and the assessment of the CESC items to determine whether CLT was assessed communicatively or not as summarised in Table 6.
Table 6.The communicativeness of the teaching items in CESC and of the assessment items in CESC by percentages and rank order Table 6 illustrates that more than two-third or 372 (66.43%) of the items in the CESC, and less than one-third or 191 (30.22%) of the items in the assessment, were designed based on a communicative approach.Inversely, less than one-third or 188 (33.57%) of the items in the CESC module, and more than two-third or 441 (69.78%) of the items in the assessment materials, were designed based on the traditional assessment approach.To determine the level of alignment, the communicativeness of the items in the module and the items in the assessment documents of CESC was summarised using Sidney Siegel's correlation coefficients (C) (see Gibbons, 1967;Savage, 1957) as illustrated in Table 7. value of Chi-square (X 2 ), which is 7.397.Thus, comparing each language domain in CESC, the correlation coefficients, c = 0.142, c = 0.139 and c = 0.115, show that the items in grammar, vocabulary and reading in the module were respectively found to have a relatively better alignment with the AA items than the items in the speaking (0.106), listening (0.017) and writing (0.060) skills.Table 8 summarises the results of Mann-Whitney U Test in relation to the alignment between the teaching and assessment items in CESC across the three universities.The overall summary of Sidney Siegel's correlation coefficients (0.1291) in Table 8 demonstrates that the alignment between the communicativeness of the items in the assessment and the communicativeness of the items in the CESC module was very low at the three universities during the study period.However, AU (0.1361) and WU (0.1320) look better than AAU (1063) as the level of the alignment at the first two universities was more than the overall results of the coefficient.
Likewise, Kruskal-Wallis Test result was also computed to determine the differences between three universities in aligning their assessment approaches with the CESC teaching approach (see Table 9).R 2 represents the sum of the mean square of all the observations at the three universities; df was 2.
As indicated in Table 9, 632 assessment items were designed and used at the three universities, with 232 items at AAU, 198 items at AU, and 202 items at WU.The result of Kruskal-Wallis Test (t = 1.203 with p = 2.460 at p < 0.05) reveals that there was not a significant difference between the three universities in relation to the alignment between their assessment items and their teaching material items in CESC.The following section presents the discussion of the findings.

DISCUSSION
The current study set out to investigate the alignment of alternative assessment (AA) with the CESC curriculum design and the CESC teaching practices at three Ethiopian universities.It did so by framing three research questions mentioned under the introduction.As illustrated in Tables 1 and 2, CESC has five units comprising six language domains.These units consist of the activities that are intended to be communicative (e.g., public speaking, debating, and interviewing) and the activities (e.g., scanning, skimming, summarising, working out meanings from context, conditional sentences, collocations, modals, and active and passive voice) that are hardly communicative in nature.In fact, the last set of activities are traditional language learning activities not aligned to any form of communicativeness.But in the actual CESC module content itself, there are more communicative activities than the communicative activities as will be highlighted below.This means that CESC needs to explicitly categorise its activities under each language domain as communicative activities and uncommunicative activities to avoid any confusion and conflation between the given activities.
Importantly, the overall CESC content should be structured in such a manner as to align both the active learning method (ALM) and AA, and to have this alignment tailored to the needs of communicative language teaching (CLT).For instance, researchers like Kibbe (2017) contend that language objectives need to be designed in a manner that should help learners develop communicative competence rather than simply learning elementary and mechanical cognitive knowledge related to language forms.Therefore, mechanical activities in CESC such as greetings and introductions as well as those focusing on converting some forms of language to other forms, cannot enforce communicative and interactive teaching and assessment.This finding is inconsistent with the findings by Al-Mamari et al.
Regarding the comparison between the communicativeness of the teaching by instructors and the assessment materials, from a total of 632 assessment items, only 30.22% of the items in CESC were assessed communicatively using AA components.
But more than two-third (69.78%) of the items in the module were assessed using TA components.This reveals that most of the instructors tested their students on elementary cognitive knowledge using TA strategies rather than testing students' language skills through AA strategies.This result contradicts the results of Davies' (2013), Nasab's (2015), Okeeffe's ( 2013) and Sandorova's (2014) studies which showed that instructors were able to align their assessment with their teaching strategies within the framework of the intended learning outcomes of CESC.
As mentioned in the first paragraph of this section, when all the items in CESC were analysed individually beyond those depicted in Tables 1 and 2, more than twothird (66.43%) of them satisfied the communicativeness requirements (see Table 3).
The remaining 33.57% of the items in CESC were considered uncommunicative, and as such, not liable to be assessed using AA components as featured in Brown (2012), Davies (2013) and Herdiawan (2018).The instructors at the three universities slightly employed more (53.17%)continuous assessment (CA) items than summative assessment (SA) items (46.83%) during the study period.Although most of the CA items were not communicative, the use of more CA than SA seemed to be reasonable in the context of CESC.The instructors used continuous test instead of using continuous and progressive AA strategies to measure the language objectives in CESC.
The results of the current study highlight that the instructors' assessment practices in CESC were inconsistent with AA, and varied across the three universities and from instructor to instructor.The assessment items were also disproportionate to the language objectives in CESC because the instructors mainly devoted more time to assessing reading skills, grammar, and vocabulary knowledge.This result is inconsistent with what Hashemnezhad (2015) and Nasab (2015) call "deep" and "surface" assessment approach to fit an assessment to its teaching purpose in CESC.
Supporting this, Brown (2012) and Iyer (2015) argue that the range of variation in assessment can cause harmful effects on the teaching-learning of the course.
Nonetheless, very few of the instructors used elements of AA strategies in CESC.
Based on the correlation coefficients, the overall alignment between the instructors' AA items and the communicativeness of the items in CESC (0.1291) was generally very low at the three universities during the study period.For example, studies by Brown (2012), Okeeffe (2013) and Sandorova (2014) found that a very low correlation between the teaching and assessment materials unacceptable in the CLT era in Ethiopia.The results of the correlation coefficient for each language domain in CESC also indicate that the assessment items for grammar (0.142), vocabulary (0.139) and reading (0.115) had a better alignment with the items in the CESC module than the assessment items for speaking (0.106), listening (0.017) and writing (0.060).
Considering each university, the correlation coefficients (0.1063) at AAU, (0.1361) at AU and (0.1320) at WU, show that the alignment between the items in the CESC module and AA used by the instructors was very low.The Kruskal-Wallis Test result (t = 1.203) reveals the same findings, that there was not a statistically significant difference between the three universities in aligning their assessment items with the items in the CESC teaching material.This result contradicts the results of the studies conducted by Brown (2012), Nasab (2015), Okeeffe (2013) and Sandorova (2014) in which instructors were able to align their assessment and their teaching strategies with the intended learning outcomes in CESC.

CONCLUSION
The level of the communicativeness of the CESC module seems to meet most of requirements of EFL teaching and assessment practices.However, the implementation of the mechanical and disproportionate instructor-based assessment in CESC at the three universities is a grave concern that these universities need to urgently address.In all, most of the instructors at the three universities were unsuccessful in aligning their assessment with the communicativeness of the curriculum and with teaching requirements of CESC.The alignment between the instructors' AA items and the communicativeness of the items in the CESC module (c = 0.1291) is very low and regrettable in the era of CLT in Ethiopia.To this end, there was no statistically significant difference between the three universities in relation to the alignment between their assessment items and their teaching material items in CESC.Importantly, CESC should be designed in such a way as to incorporate complex authentic and challenging learning activities.
Finally, the alignment between dimensions of CESC in the Ethiopian higher education context is a new area of engagement.Therefore, there is evidently wide room for further studies to investigate this issue without which the teaching and the assessment of CESC are likely to be compromised.The constructive alignment strategies, integrated language skills teaching and assessment, the role of instructors and the students and the challenges to promote the alignment between the curriculum, and the teaching and assessment practices in CESC are some of the urgent areas that need to be addressed in Ethiopia.

Table 2 .
The description of the language domains and language objectives in CESC

Table 3 .
The communicativeness of the Items in CESC by percentages and rank order

Table 4 .
Frequency of items in the assessments at the three universities

Language domains Frequency of the items at each university Total frequency Rank Addis Ababa Ambo Wollaga
items, 296 (46.83%).This seems to be reasonable in the context of CESC as CA can be interpreted as a part of the AA strategies.More specifically, 120 (51.7%), 114 (57.58%) and 101(50.5%)CAitemswere set at AAU, AU and WU, respectively, whereas 112(48, 3%), 84 (42.42%) and 100 (49.5%)SA items were set at AAU, AU and WU, correspondingly.By contrast, Table5summarises the level of the communicativeness of the items designed by the instructors at the three universities during the study period.
and used at the three universities during the period of the study.Considering the type of assessment, the number of CA items, 336 (53.17%), is slightly greater than the number of SA

Table 5 .
Summary of the instructors' communicative assessment items by percentages and rank order at the three universities

Table 7 .
The Alignment between the Items in AA with the Items in the CESC module

Table 8 .
Results of Mann-Whitney U Test among the three universities

Table 9 .
The differences between three universities in aligning their assessment approaches with the CESC teaching approach as determined by Kruskal-Wallis Test