English language assessment: unfolding school students’ and parents’ views

Avaliação em língua inglesa: revelando opiniões de alunos e pais

Abstract

The current study analyzes Chilean students’ and parents’ perceptions of English language assessment. 352 Chilean 10 ^th graders were surveyed using the Perceptions of the Assessment Tasks Inventory, the Perceived Classroom Assessment Environment Scale, and the Test Anxiety Scale in order to measure their perceptions of English language assessment. A semi-structured interview was also conducted to capture the perceptions of 74 parents about English and the way their children are assessed in the English subject. The quantitative data was analyzed using the Pearson Correlation Coefficient to measure how strong the relationship of the scores was in the three scales, and qualitative data was categorized using content and frequency analysis. The findings suggest that the subscales of the three surveys hold statistically significant relationships and students feel that a Learning-oriented assessment environment benefits them more than a Performance-oriented one. The former is a kind of environment that favors students’ learning and mastery of subject materials, whereas the latter is a type of environment that aligns with classroom assessment practices that emphasize the importance of grades over learning. The qualitative results also suggest that parents have a positive perception of the English language and believe it is useful for their children’s life even though they themselves might find it difficult to learn the language.

Assessment; Anxiety; Perceptions; Parents; Students

Resumo

O presente estudo analisa as percepções de alunos e pais chilenos sobre avaliação da língua inglesa. Trezentos e cinquenta e dois alunos de 2° ano do ensino medio no Chile foram pesquisados usando o inventário das tarefas de avaliação, a escala do entorno de avaliação percebida na aula e o teste de escala de ansiedade para medir suas percepções sobre a avaliação na disciplica de língua inglesa. Também foi realizada uma entrevista semiestruturada para capturar as percepções de setenta e quatro pais sobre o inglês e a maneira como seus filhos são avaliados na disciplina. Os dados quantitativos foram analisados pelo Coeficiente de Correlação de Pearson para medir a força do relacionamento dos escores nas três escalas, e os dados qualitativos foram categorizados por meio da análise de conteúdo e frequência. Os resultados sugerem que as subescalas das três pesquisas mantêm relações estatisticamente significativas e os alunos sentem que um ambiente de avaliação orientado à aprendizagem é mais benéfico para eles do que um orientado ao desempenho. O primeiro é um tipo de ambiente que favorece a aprendizagem e o domínio dos materiais de parte dos alunos, enquanto o segundo é um tipo de ambiente alinhado às práticas de avaliação em sala de aula que enfatizam a importância das qualificações sobre o aprendizado. Os resultados qualitativos também sugerem que os pais têm uma percepção positiva do idioma inglês e acreditam que é útil para a vida de seus filhos, embora eles mesmos possam achar difícil o aprendizado do idioma.

Avaliação; Ansiedade; Percepções; Pais; Alunos

Introduction

As the world becomes more and more globalized, the importance of English becomes more crucial and more effort is put into teaching the language at Chilean schools. However, the issue is not only how to teach a foreign language but also what students perceive when they are assessed on how they learn English. To learn a foreign language very often makes students feel anxiety about the assessment task they have to face and the environment in which assessment takes place. This study analyzes students’ perceptions towards assessment tasks, their classroom environment and the anxiety they experience as learners; it also addresses parents’ perceptions towards their children’s grades in the English subject. This paper is in the context of the research grants FONDECYT 1191021 under the name of Estudio correlacional y propuesta de intervención en evaluación del aprendizaje del inglés: las dimensiones cognitiva, afectiva y social del proceso evaluativo del idioma extranjero and VRID Enlace 218.003.002-10, La evaluación del inglés: conocimientos, percepciones, disposiciones afectivas y representaciones sociales. Una mirada sistémica del proceso evaluativo desde los profesores y futuros profesores de inglés, estudiantes, autoridades y apoderados.

Theoretical framework

Language assessment

Assessment is usually understood as an ongoing process that encompasses a wide variety of techniques and skills used in the classroom in order to evaluate a student’s performance in a certain area of study ( ABEYWICKRAMA; BROWN, 2010 ), and it is an essential part of teaching English as a foreign language. It is important to differentiate between assessment and testing, as both terms are commonly thought of as synonyms, when, in fact, they refer to different categories of evaluation. While assessment is a general and ongoing practice, testing is a prepared administrative procedure that occurs at identifiable times in a curriculum ( ABEYWICKRAMA; BROWN, 2010 ). The difference can be exemplified by saying that a test is an occurrence in which students collect all their knowledge in a given domain knowing that their responses are being evaluated, while assessment involves everything that a teacher evaluates of a student’s performance, such as when they answer a question during class, offer a comment, or try a new word or structure ( ABEYWICKRAMA; BROWN, 2010 ; ALEMI; TAVAKOLI, 2016 ; LIU; SHI, 2007 ). However, assessment, as many other disciplines, has been influenced by different paradigms that have drawn teachers´ and researchers’ attention to one or another component of assessment, as Behaviorism and Constructivism have done.

Behaviorist and constructivist views of assessment

There are two views to address language assessment ( ABEYWICKRAMA; BROWN, 2010 ). For behaviorism, assessment aims to check individual performances that demonstrate mastery. In the behaviorist perspective, the ground for assessment is that all students are the same; therefore, they should all receive the same instruction. They are all assessed in the same way through instruments and procedures aimed at grading and scoring their final performance. Assessment is centered on the final product, that is to say, on performing mechanic and repetitive actions that can be measured, observed, quantified, and reinforced or punished as suggested by Skinners’ psychology of behavior ( ONOMIA; ONOMIA, 2014 ). Teaching efficiently is highly valued and assessment should be aligned with the objectives that are formulated. This implies that mistakes should be avoided and grades are used as a mechanism to impose teachers’ authority. Thus, English is learned as a set of habits that have to be constantly rehearsed to achieve native-like language proficiency.

In a constructivist view of assessment, students play a central and active role in learning and their previous knowledge and competences are essential to understand their individual differences and interests. Students’ knowledge, skills and attitudes are part of their internal world so that attention focuses on the process rather than the final product. Language assessment under Constructivism becomes a process of data collection, analysis and communication that stresses the students’ construction of meaning and their ability to process information and be able to transfer what has been learned to different learning scenarios. English is then learned as a set of competences that have to be developed over time considering students’ individual differences and previous world knowledge ( VAVLA; GOKAJ, 2013 ).

Perceptions of language assessment

When literature addresses school students’ perceptions of language assessment, Valva and Gokaj (2013), for example, have identified that most students regard their assessments as mandatory tests, which uncovered high levels of anxiety in sitting for a test. This has a negative impact on overall language learning among students who fail to understand the test instructions and guidelines. In another paper about students’ perceptions towards self-assessment in the EFL classroom (MUÑOZ; ALVAREZ, 2007), conclusion was that most students regard self-assessment as a valuable tool leading to a better understanding of their learning processes. Participants also pointed out that self-assessment helped them regulate their own actions by becoming more objective, which contributed to reach their learning goals. Only few students saw self-assessment as taking responsibility for their own learning. It was also found that most students saw the teacher as a fair evaluator while they did not see themselves as very objective ( CZURA, 2017 ).

Regarding the role of parents in assessment, previous studies have observed that there is a strong connection between parental involvement and students’ academic achievement ( HAYES, 2012 ; SHUTE et al ., 2011). Research has also demonstrated that parental involvement can be a great contributor to student achievement and that an effective school-family partnership can be beneficial for parents as it may help them perceive their children’s school in a more positive way and increase their sense of efficiency as parents while changing their perceptions of their children as learners ( ÁVILA-DAZA; GARAVITO, 2009 ). There is often a substantial difference between the school’s expectations of the parents’ role in children’s education and the parents’ very expectations. Hoover-Dempsey et al . (2005) say that the parental role is a mixture of beliefs parents have when it comes to their children’s education and the different patterns that their cultural context has given them. Many parents bring to schools a completely different notion of what their responsibilities are in education, as this notion is derived from their cultural thinking, behaviors and social interactions ( BRUTON; ROBLES-PIÑA, 2009 ; MEDINA; GUZMÁN; WONG-RATCLIFF, 2015).

When it comes to learning a second language, excuses and lack of confidence increase in parents. Pino-Juste and Rodriguez (2007) and Aljure et al . (2010) conducted different research studies in Portugal and Colombia about parental’ involvement when it comes to students learning English. Their research showed that parents hold learning English in high esteem, that they expect their children to start learning it at an early age and expressed how it can be beneficial for their children’s future (PINO-JUSTE; RODRÍGUEZ, 2007). They also expressed that it can be helpful in aiding their children to develop their thinking and mentioned the positive outcomes learning English may have on their academic performance (ALJURE et al ., 2010). Parents do not participate or help much when it comes to English, as most of them believe that they lack education about second languages and do not want to give the wrong answers to their children ( ÁVILA-DAZA; GARAVITO, 2009 ).

Empirical studies

Perceptions of Assessment Tasks Inventory

Dorman and Knightley (2006) applied the Perceptions of Assessment Tasks Inventory (PATI) on a sample of 658 ninth- and tenth-graders in 11 schools in Essex, England in order to introduce it and use the data collected to group and describe students who hold similar views on assessment tasks. Participants were divided according to their results into four similarly-sized groups named A, B, C, and D. The results of this research were limited due to the fact that it was the first use of the Perceptions of Assessment Tasks Inventory with secondary school students (ALKHARUSI et al ., 2014).

Perceived Classroom Assessment Environment Scale

Cheng, Wu and Liu (2015) applied the Perceived Classroom Assessment Environment Scale together with the Perceptions of Assessment Tasks Inventory (PATI) to 620 university students from three universities in China in order to explore the relationship between students’ perceptions of assessment tasks and their classroom assessment environment. Their findings showed that congruence with planned learning and student consultation were both positive predictors of a learning-oriented classroom assessment environment and negative predictors of a performance-oriented one, showing the potential twofold benefits of aligning assessment tasks with the goals and objectives of the program and informing students about how they would be assessed ( BULDUR; DOĞAN, 2014 ).

Test Anxiety Inventory

Ali and Mohsin (2013) conducted a research study with the aim of translating the Test Anxiety Inventory into the Urdu language to explore its psychometric properties when it is applied on secondary and higher secondary school science students in Pakistan. Data was collected from 2,145 secondary and higher secondary school science students from 46 schools in five different districts of Pakistan. 1,280 students were from urban areas and 865 students from rural areas. The Alpha Reliability Coefficient was determined to be 0.893, which is close to the Alpha Reliability Coefficient of Spielberger’s original study (1980). In Ali and Mohsin’s study all the values of alpha reliability and discriminant validity replicated the results of previous research studies, leading to the conclusion that the Test Anxiety Inventory was valid and reliable when used in a Pakistani context among 10 ^th -grade science students (KAVANAGH; ZIINO; MESAGNO, 2016; SZAFRANSKI; BARRERA; NORTON, 2012).

Research design

Type of study

The present study is a cross-sectional and non-experimental, two-stage study because it contains two different approaches, a quantitative approach and a qualitative one. A quantitative approach was used to analyze 10 ^th -grade Chilean students’ perceptions of assessment tasks, their classroom environment and test anxiety related to the English language. A qualitative approach was then used to analyze parents’ perceptions of their children´s grades in the English subject expressed through the contents of their responses during interviews.

Research aims

- To identify the existing relationship of 10 ^th -graders’ perceptions of assessment tasks, assessment environment and test anxiety in the English subject.

- To analyze parents’ perceptions of their children’s grades in the English subject.

Participants

The participants of this study are both school students and parents from different Chilean schools.

The first group has 352 students in total from the 10 ^th grade, aged 14 to 15 years old. At the time of the study, they were attending a Chilean semi-private school that had a total of 1,230 students. In this school, students are expected to achieve an intermediate level of English by the end of their schooling years. 10 ^th -graders have 5 pedagogical hours a week. Out of the total of 352 students participating in the research, 154 were identified as male and 198 as female.

It is worth mentioning that these 352 students are mainly exposed to traditional tests that contain questions such as multiple choice, true or false, completion, gap-fill etc. 607 assessment instruments provided by their English teachers were analyzed and it turned out that 65% of them were language tests. However, this study does not focus on analyzing assessment instruments. Its aim is to collect relevant data on students and parents’ perceptions at one specific moment in time, following the cross-sectional nature of the study.

The second group, parents, included a total of 74 participants, aged from 25 to 55 years old. Participants were parents of students attending Chilean semi-private schools who had one or more children enrolled in any of the school grades (Kindergarten to 12th). Most participants had completed their high-school education, and some of them had, or were in the process of earning a university degree. Out of the total 74 parents participating in this study, 60 were female and 14 were male.

Instruments

Quantitative technique

With the purpose of gathering the students’ data, three different Likert scales were used. On every scale, students had to read a number of statements and mark each one with one (1) to five (5), where one (1) was a complete disagreement, and five (5) a total agreement.

The Perceptions of Assessment Tasks Inventory was formulated by Dorman and Knightley (2006) to measure students’ opinions on how they are assessed and if their results and what they are learning fulfill some of their own expectations when it comes to learning English. This scale has five dimensions and includes 35 statements in total: statement 1 to 7 cover the dimension of Congruence. Statements 8 to 14 measure Authenticity, statements 15 to 21 refer to Student consultation. Statements 22 to 28 are related to Transparency, and statements 29 to 35 are about Diversity.

The Perceived Classroom Assessment Environment Scale was created by Alkharusi (2015) and it determines both students’ perspectives and opinions on the different techniques used by their teachers during classroom assessment, and whether their teachers use these methods in an appropriate and fair manner. It also seeks to assess students’ perceptions of their teachers’ attitudes towards the class and the students. These scales are divided along two different dimensions: Learning-oriented Environment and Performance-oriented Environment, which are addressed by different sets of statements on the scale. This scale has two sections: 8 statements for Learning-oriented environment and 8 statements for Performance-oriented Environment, with a total of 16 statements.

The Test Anxiety Inventory was developed by Spielberger (1980) and it was especially designed to measure the anxiety high-school and college students experience when sitting for a test. This instrument contains two dimensions only: Total Anxiety and Emotional Anxiety. These dimensions assess the participants’ level of anxiety and feelings ranging from fear, nervousness, level of knowledge and confidence, and how these feelings interfere with their overall results in the English subject.

Since the three scales had not been used in the Chilean context before, their factor structure, internal consistency and validity were assessed through confirmatory factor analysis. In a confirmatory factor analysis, concurrent and discriminant validity examine the extent to which measures of a latent variable share their variance and how they are different from others. Average Variance Extracted (AVE) measures the level of variance captured by a construct versus the level due to measurement error: values above 0.7 are considered very good, whereas, 0.5 is an acceptable level. A ≥ 0.40 AVE may be still acceptable for first-time exploratory studies, as it is the case of this study, in which none of the scales has been used in the Chilean context before ( ALDAS, 2000 ; LAM, 2012 ).

The results for the Test Anxiety Inventory in the confirmatory factor analysis are shown below:

Table 1 Correlations, internal consistency, concurrent and discriminant validity across latent variables

Variables	1	2	FC	AVE	√AVE
1. Total	1.00		0.89	0.55	0.74
2. Emotional	0.68**	1.00	0.93	0.64	0.80

Note: ** p < .01

Source: own elaboration.

The reliability values of Alpha coefficient of the original version of the Test Anxiety Inventory ranged between 0.91 and 0.96 ( SPIELBERGER, 1980 ).

As for the case of the Perceived Classroom Assessment Environment Scale, the confirmatory factor analysis yielded the following:

At the original Perceived Classroom Assessment Environment Scale, internal consistency coefficients for perceived performance-oriented and learning-oriented assessment environment were .67 and .65 as measured by Cronbach’s alpha, respectively ( ALKHARUSI, 2015 ).

The results of confirmatory factor analysis for the Perceptions of Assessment Tasks Inventory can be seen below:

As reported by Alkharusi (2013) , internal consistency coefficients for the original measures of congruence with planned learning, authenticity, student consultation, transparency, and diversity were .71, .72, .65, .66, and .63 as indicated by Cronbach’s alpha, respectively.

Qualitative technique

For parents, a semi-structured interview was applied, to find out what they think about the grades their children obtain in the English subject and what their typical response is when they are shown their children’s grades. The interviews were carried out in Spanish and lasted approximately five minutes each. They were recorded and later transcribed for data analysis purposes.

Procedure

The students’ Test Anxiety Inventory, Perceived Classroom Assessment Environment Scale, and Perceptions of Assessment Tasks Inventory were applied during class time with the help of the head teacher in charge of the classroom. It was a total of 352 10 ^th -grade students. The semi-structured interview was carried out with 74 parents to find their views on the English subject, what is the meaning of child learn the language, the utility of it, and how they help in their children’s study.

Type of data analysis

Quantitative

The first step of the analysis was to test data normality through the Skewness and Kurtosis measures. Skewness measures the asymmetry of the probability distribution of a random variable around its mean. All Skewness measures of the data ranged from -0.92 to 0.19. Kurtosis, on the other hand, indicates the height and sharpness of the central peak, in relation to a standard bell curve. All Kurtosis measures of the data ranged from -0.89 and 0.60. These Skewness and Kurtosis values are considered acceptable in order to prove normal univariate distribution of the data (CURRAN; WEST; FINCH, 1996). Therefore, it is possible to use the parametric Pearson’s r test. The students’ responses were then analyzed through the Pearson’s Correlation Test to determine the strength of the relationship between the students’ responses on the various items of the scale and inventory. The Pearson’s Correlation Test measures the linear correlation between two variables, X and Y, with values ranging from 1 to -1. In this research, a value above 0.25 or below –0.25 determines the kind of relationship between the variables. A value above 0.25 indicates a direct relationship between X and Y, whereas a value below -0.25 implies an inverse relationship between the variables. If the value is 0, then there is no relationship between the variables. Data was later categorized and set in Microsoft Word tables.

Qualitative

For the parents’ responses, two techniques were used to analyze the data: The Content analysis method and Frequency analysis. Cavanagh (1997) mentions that content analysis is a method of collecting data about messages in a systematic and objective manner that can lead to the creation of inferences. This data was categorized through the software QDA Miner, a program created to assist researchers in managing, coding and analyzing qualitative data.

Results

Quantitative analysis

The following section of this research paper showcases the results obtained from the Perceptions of Assessment Tasks Inventory, the Perceived Classroom Assessment Environment Scale, and the Test Anxiety Inventory that were applied to the school students. A synthesis of the results will be firstly presented, followed by a detailed correlational analysis for each scale. Chart 1 details the subscales for each one of the instruments:

Chart 1
Scales and subscales of the instruments

Scales	Subscales
Perceptions of Assessment Tasks Inventory	Congruence Authenticity Student Consultation Transparency Diversity
Perceived Classroom Assessment Environment Scale	Learning-oriented environment Performance-oriented environment
Test Anxiety Inventory	Total Anxiety Emotional Anxiety

Source: own elaboration.

The results of the Pearson’s Correlation test are presented in Table 2 below, which includes the scores for the subscales in the Perceptions of Assessment Tasks Inventory (Congruence, Authenticity, Student consultation, Transparency and Diversity), the Perceived Classroom Assessment Environment Scale (Learning-oriented and Performance-oriented assessment environment) and the Test Anxiety Inventory (Total anxiety and Emotional anxiety).

Table 2
Correlations, internal consistency, concurrent and discriminant validity across latent variables

Variables	1	2	3	4	5	FC	AVE	√AVE
1. Learning	1.00					0.79	0.47	0.68
2. Performance	-0.40**	1.00				0.73	0.41	0.64

• Note: ** p < .01
Source: own elaboration.

The general results indicate that the subscales of each of the three scales had statistically significant relationships, as the double asterisks in Table 2 show below. The positive and significant correlations between the Perceptions of Assessment Tasks Inventory scores and those from the Perceived Classroom Assessment Environment Scale stand out. These correlation coefficients (r) oscillated between 0.35 and 0.48 on the Learning-Oriented Environment subscale, as shown by the blue circles in Table 2 . On the other hand, negative and significant relationships between the Perceived Classroom Assessment Environment Scale on the Performance-Oriented Environment subscale and the Perceptions of Assessment Tasks Inventory were observed. The correlation coefficients (r) fluctuated between -0.18 and -0.51, which means that there is an inverse relation between the variables, as shown by the red squares in Table 2 . Then positive and significant relationships were observed between the Perceived Classroom Assessment Environment Scale on the Performance-Oriented Environment Scale and the Test Anxiety Inventory, showing correlation coefficients (r) between 0.53 and 0.50, as shown by the inverse green triangles in Table 2 . This proves that there is a direct relationship among the variables studied. Furthermore, negative significant relationships were observed between the Perceived Classroom Assessment Environment Scale on the Learning-Oriented Environment Scale and the Test Anxiety Inventory, showing correlation coefficients (r) between -0.22 and -0.18, as can be seen in the green triangles in Table 4 .

Table 4
Pearson Correlation between subscales

	1.	2.	3.	4.	5.	6.	7.	8.	9.
1. Congruence	1.00	.63 **	.50 **	.49 **	.39 **	.42 **	-.45 **	-.40 **	-.35 **
2. Authenticity		1.00	.52 **	.50 **	.41 **	.48 **	-.51 **	-.37 **	-.27 **
3. Student consultation			1.00	.58 **	.39 **	.36 **	-.41 **	-.39 **	-.32 **
4. Transparency				1.00	.43 **	.35 **	-.48 **	-.41 **	-.33 **
5. Diversity					1.00	.41 **	-.28 **	-.28 **	-.23 **
6. Learning-oriented						1.00	-.40 **	-.22 **	-.18 **
7.Performance-oriented							1.00	.53 **	.50 **
8. Total Anxiety								1.00	.82 **
9. Emotional Anxiety									1.00

• Source: own elaboration.
• Correlational analysis for the Perceptions of Assessment Tasks Inventory

Source: own elaboration.

Correlational analysis for the Perceptions of Assessment Tasks Inventory

The Perceptions of Assessment Tasks Inventory measures the students’ perceptions of the way assessment tasks are used, the students’ level of involvement, the opportunities they have in the assessment process, and the relation that assessment tasks have with real life and course goals. These were categorized in five (5) subscales (Congruence, Authenticity, Students’ consultation, Transparency and Diversity), according to Dorman and Knightley (2006) .

Each subscale is represented by one or several items. Congruence is defined as the extent to which assessment tasks align with the goals, objectives or activities of the learning program. For example, item 4 in the Perceptions of Assessment Tasks Inventory states: “My assessment tasks are a fair indicator of what my class is trying to learn in the English class”. Authenticity is determined as the extent to which assessment tasks feature real life situations that are relevant to the learner. For example, item 8 states: “The teacher makes me apply what I have learned in the assessment tasks to real life situations”. Student consultation is determined as the extent to which students are consulted and informed about the forms of assessment tasks being employed. For example, item 17 says: “My teacher asks me about the types of assessment I would like to have in English class”.

Transparency is determined as the extent to which the purposes and forms of assessment tasks are well defined and clear to the learner. For example, item 22 states: “I know what is needed to successfully accomplish an English assessment task”. Diversity is determined as the extent to which all students have an equal chance of completing assessment tasks. For example, item 29 reads: “I complete English assessment tasks at my own speed”.

In order to conclude that a correlation is significant in the analysis, the results must be higher than 0.25 or lower than -0.25 as a rule of thumb. The scores obtained in the Perceptions of Assessment Task Inventory, outlined in Table 3 , oscillated between 0.39 and 0.63, which indicates positive and significant relationships among the five subscales, as shown by the circles in Table 5 . This means that all these variables were related to each other and if one variable is present, the possibilities of other variables being present are also high, showing a directly proportional relationship. The highest correlation obtained in the Perceptions of Assessment Tasks Inventory was between Congruence and Authenticity, with a score of 0.63, indicating that when assessment tasks are related to the goals and objectives of the program, they also tend to be relevant for the learner. The lowest scores obtained were also higher than 0.25, making them statistically significant. The correlation between Diversity, Congruence and Student consultation resulted in just 0.39. This may mean that when students perceive they are not consulted about the forms of assessment, there is also no diversity when it comes to learning opportunities and the assessment tasks may not be completely related to what they feel they learn in the classroom (See Table 5 ).

Table 3
Correlations, internal consistency, concurrent and discriminant validity across latent variables

Variables	1	2	3	4	5	FC	AVE	√AVE
1. Congruence	1.00					0.82	0.61	0.78
2. Authenticity	0.63 **	1.00				0.78	0.55	0.74
3. Consultation	0.50 **	0.52 **	1.00			0.80	0.57	0.76
4. Transparency	0.49 **	0.50 **	0.58 **	1.00		0.77	0.52	0.72
5. Diversity	0.39 **	0.41 **	0.39 **	0.43 **	1.00	0.66	0.41	0.64

• Note: ** p < .01
Source: own elaboration.

Table 5
Pearson’s Correlation for Perceptions of Assessment Tasks Inventory

	1.	2.		3.	4.	5.
1. Congruence	1.00	.63 **	.50 **		.49 **	.39 **
2. Authenticity		1.00	.52 **		.50 **	.41 **
3. Student consultation			1.00		.58 **	.39 **
4. Transparency					1.00	.43 **
5. Diversity						1.00

Source: own elaboration.

Correlational analysis of the Perceived Classroom Assessment Environment Scale

Alkharusi (2015) determines the Perceived Classroom Assessment Environment as an overall sense that students make out of the assessment practices that the teacher performs in the classroom. Based on the Achievement Goal Theory, Alkharusi (2011) created a scale to measure the students’ perception of the classroom assessment environment. This scale determined two different dimensions: Learning-oriented environment and Performance-oriented environment. The Learning-oriented assessment environment’s major objective is to enhance the students’ learning and mastery of subject materials through classroom assessment practices, e.g. providing chances for the student to improve their task performance, giving informative assessment feedback and asking them to perform meaningful assessment tasks of moderate difficulty. On the other hand, the Performance-oriented environment is characterized by classroom assessment practices that emphasize the importance of grades rather than learning comparative performances among them.

Each type of environment has different associations when it comes to academic self-efficacy and achievement. The Learning-oriented assessment environment tends to be positively associated with academic self-efficacy and achievement, whereas the Performance-oriented assessment environment tends to be negatively associated with academic self-efficacy and academic achievement ( ALKHARUSI, 2013 ).

The scores present in Table 6 show a correlation between both subscales of -0.40. This indicates an inversely proportional relationship between the two, which would mean that, whenever classroom assessment practices that enhance students’ learning are present, it can be expected that the presence of difficult and less meaningful tasks focusing only on grades will be low and vice versa as shown in Table 6 below:

Table 6
Pearson Correlation for the Perceived Classroom Assessment Environment Scale

	1.	2.
1. Learning-oriented	1.00	-.40**
2.Performance-oriented		1.00

Source: own elaboration.

Correlational analysis of the Test Anxiety Inventory

Reber (1995) defined anxiety as an emotional state under unpleasant conditions, from which we can infer that its effect is negative to students’ overall achievement and self-esteem. Dusek (1980) defined test anxiety as an emotional state experienced in formal testing and other evaluative situations; it has psychological and behavioral effects. Spielberger’s (1980) Test Anxiety Inventory (TAI) was designed to measure students’ anxiety when sitting for a test in high-school or college.

The results below in Table 7 show a statistically significant and directly proportional relationship between the two subscales with a score of .82. This indicates that a rise in Emotional anxiety feeds into a rise in Total anxiety in Chilean high-school students.

Table 7
Pearson Correlation for the Test Anxiety Inventory

	1.	2.
1. Total Anxiety	1.00	.82 **
2. Emotional Anxiety		1.00

Source: own elaboration.

Qualitative analysis

The main idea of qualitative research is to understand social phenomena in its natural setting. It mainly focuses on the reasons why certain social phenomena occur and it relies on human experiences to obtain the responses to a specific inquiry. As this research will both characterize the parents’ conceptions of their children’s English language assessment and interpret their perceptions and emotions towards the English language, it can be described as having both a descriptive and an interpretive focus. This research employed a semi-structured interview, which was applied to a target group consisting of both female and male parents. The total number of interviewees was 74, with ages ranging from 25 to 55.

Analysis of parents’ perceptions towards grades and their types of reinforcement

The question in this semi-structured interview was: “What’s your opinion of your child’s grades in the English subject? What do you tell them when they show you said grades?” In Chile, grades range from 1.0 to 7.0, being the range that goes from 1.0 to 3.9, failing grades, while 4.0 to 7.0 are passing grades. From this question, the interviewees’ views towards their child’s grades in the English subject were unfolded. In order to perform the analysis, the responses were categorized in dimensions and sub-dimensions.

Two dimensions were outlined: Dimension 1 addressed Parents’ reactions towards grades in the English subject while dimension 2 tackled Parents’ ways of reinforcing their children in the English subject, as presented in Figure 1 .

Figure 1
Research dimensions

Based on the responses, two sub-dimensions could be identified within Dimension 1, which were labelled as Positive reaction towards grades and Negative reaction towards grades (See Figure 2 ).

Figure 2
Sub-dimensions of Dimension 1

From the responses given to the interviewer, it was possible to determine the exact number of parents that fell into each of these sub-dimensions. On one hand, it was found out that 97% (68) of parents showed a positive reaction towards their children’s grades in the English subject, giving responses such as:

• [- My son has excellent grades in the English subject] (Participant N°1).

• [- Good, she has an average score of 6.9 or 7.0 sometimes] (Participant N°6).

On the other hand, and as shown in Figure 3 , 7% (5) of parents showed a negative reaction towards their children’s grades, some of them not saying that their children had low grades, but rather saying that their children had difficulties with the language, such as: [I know that he has a hard time with English] (Participant N°22), while others were more direct by making comments like [I think their grades are low because there is no interest in learning the language] (Participant N°19).

Figure 3
Parents’ reactions towards grades in the English subject

From Dimension 2, the five contributing ways parents use to reinforce their children were organized into two different sub-dimensions. Four of said ways were grouped as Positive reinforcement and one of them was labelled as Negative reinforcement. The sub-dimensions that were created can be seen in Figure 4 .

Figure 4
Sub-dimensions of Dimension 2

The five sub-dimensions were Acknowledging the child’s effort, Praising the child’s grades, Encouraging the child, Practicing English with the child and Punishment. Punishment was not a common method of reinforcement used by parents, totaling only 2% of the total responses (n:2). Positive reinforcements took the lead by a landslide. It was determined that Encouraging the child was the most common way of reinforcing their children regarding the English subject by 31% (n:32), followed closely by Praising the child’s grades by 30% (n:31), and Acknowledging the child’s effort reached 29% (n:30 participants).

Finally, it was noticed that only a small number of parents, just 8% of participants interviewed (n:8), used Practicing English with the child as a way of reinforcement, which is disheartening due to the fact that it could be a way of bonding with the child and encouraging them to continue putting effort into the subject (See Figure 5 ).

Figure 5
Parents’ way of reinforcing their children in the English subject

Discussion

In the present study students’ perceptions toward the assessment of English as a foreign language were investigated, using the Perceived Classroom Assessment Environment Scale, the Perceptions of Assessment Tasks Inventory and the Test Anxiety Inventory in order to gain an understanding of their views about the aforementioned topics. This research also covered parents’ perceptions towards their children’s grades, for which a semi-structured interview was used.

Discussion of the findings on students

It was determined that the main focus for students would be the assessment environment of the classroom, since the Perceived Classroom Assessment Environment Scale establishes two types: Learning-oriented environment and Performance-oriented environment. It was clear that there were differences in the results obtained in the Perceptions of Assessment Tasks Inventory and in the Test Anxiety Inventory based on these two environments. This section will be presented as follows (See Figure 6 ).

Figure 6
Framework for discussion

The subscale of Learning-oriented environment from the Perceived Classroom Assessment Environment Scale

As explained before, the main characteristics of a Learning-oriented Environment are classroom assessment practices that enhance student learning, such as providing the students with chances to improve their task performance, giving students meaningful assessment tasks with a moderate difficulty, and giving informative assessment feedback ( ALKHARUSI, 2015 ). The results obtained for the five subscales of the Perceptions of Assessment Tasks Inventory in relation to the Perceived Classroom Assessment Environment Scale in the Learning-oriented Environment range from 0.35 to 0.48; this indicates that these elements have a directly proportional relation.

Drawing from this data, it could be extrapolated that Chilean students, when immersed in a Learning-oriented assessment environment, perceive their assessment tasks as having an alignment with their learning goals (Congruence), that these assessment tasks are applicable to their own lives (Authenticity), that they are involved in their own assessment process (Student Consultation), that the tasks are well-defined and clear to them (Transparency) and that their learning opportunities are equal to all students (Diversity). In this context, it comes as no surprise that the results obtained confirm what was previously stated by Alkharusi (2015) , Dorman and Knightley (2006) and Czura (2017) , showing how, when classroom assessment techniques have a high degree of Congruence, Authenticity, Student Consultation, Diversity and Transparency, the students’ perception of the assessment environment is one in which learning is more important than grades, and the students’ opinion regarding their own learning process becomes relevant.

Regarding assessment anxiety, the scores from the quantitative analysis are less than the minimum of -0.25 required to be considered negatively correlated for this research, but we do not believe this makes the results less valuable; on the contrary, they are a clear indicator that when Chilean students are involved in a Learning-oriented Environment, the anxiety they feel about the language and the idea of failing tends to be reduced, minimizing the affective filter, which accounts for the influence of affective factors on second language acquisition. In other words, affective variables such as anxiety, fear, and nervousness may affect the acquisition of a second language by blocking comprehensible input (LIN, 2008). This also improves the overall learning process as their focus is primarily on the learning process rather than on the expectations of their grades or results.

The subscale of Performance-oriented environment from the Perceived Classroom Assessment Environment Scale

On the opposite side of the spectrum, a Performance-oriented Environment’s main feature is that of classroom assessment practices that are difficult and not meaningful to students because the main objective in this type of environment is getting good grades rather than learning. This environment is also negatively associated with academic self-efficacy and academic achievement ( ALKHARUSI, 2015 ).

The results from the five subscales of the Perceptions of Assessment Tasks Inventory in relation to the Perceived Classroom Assessment Environment Scale in the Performance-oriented environment range from -0.28 to -0.51, which shows an inversely proportional relation. This means that if the assessment environment’s main focus is achieving good grades, then the students’ perceptions of Congruence, Authenticity, Student Consultation, Transparency and Diversity are negatively affected. From this environment, the most affected perception is that of Authenticity by -0.5. This is a clear indicator that in an environment in which grades are the ultimate objective, the students’ perceptions of the usefulness of the assessment tasks and the relevance that these tasks have to the students’ lives are jeopardized.

These results were very much expected, and they clearly indicate that a Performance-oriented environment may negatively affect the students’ perception of language assessment. These results show that in a Performance-oriented environment, Chilean students perceive that the assessment tasks have no influence on their lives, that they have little to do with their learning goals, that their opinions are not relevant to their own assessment process, that some students have different learning opportunities than others and that the tasks are not well-defined nor are they clear to them.

The scores between the Test Anxiety Inventory and the Perceived Classroom Assessment Environment Scale in the Performance-oriented environment are 0.53 for Total Anxiety and 0.50 for Emotional Anxiety. This directly proportional relation is something to be expected; when the learning environment is more competitive and focused on grades rather than learning, the levels of anxiety tend to be higher. As Hashemi (2011) explained, the presence of second language anxiety is striking when it comes to second language learning. It can then be deduced that a Performance-oriented Environment causes high levels of anxiety on Chilean students because its focus is on grades rather than on students’ actual learning. Something that is similar to the Chilean context is what Valva and Gokaj (2013) discovered for 6th-graders in their EFL assessments. Their results indicated that most students regarded assessment as mandatory, while they also found high levels of test anxiety among students who, after the teacher had explained the test items, failed to understand some of the questions and found the test to be too difficult.

Discussion of the parents’ findings

Parents’ perceptions towards grades and their ways of reinforcing their children in the English subject

Drawing from the results in the sub-dimension of Parents’ reactions towards their children’s grades, 93% of the interviewees had a positive reaction towards their children’s grades and were pleased by the results that their children had shown in the English subject. However, 7% of parents indicated a negative reaction towards their children’s grades, with many of them expressing that they worried about their children’s grades, as these parents admitted that their children’s grades were low, and it was also very clear that their children struggled with the language but kept on making an effort. This goes in accordance with what Valva and Gokaj (2013) expressed in their research, in which they found out that students make an effort to learn but, due to the high levels of anxiety, they fail in comprehending and answering their assessment tasks. Chilean parents perceive the stress experienced by their children, which is one of the many characteristics that Medina, Guzmán and Wong-Ratcliff (2015) describe in their research; they explain that parents play home-based support roles for their children and engage in activities such as discussing school-related matters. Parents should discuss with teachers if they perceive that the levels of test anxiety in their children are affecting them in their learning process and assessment tasks.

Another surprising result was that of the Parents’ way of reinforcing their children in the English subject. These results were interesting due to the fact that 98% of the parents preferred to use positive reinforcement when being in presence of their children’s grades, with 31% stating that when their children showed them low grades, they preferred to encourage them to keep trying and studying. Only 2% explained that they use some form of negative reinforcement but that this was never of a physical nature and that they preferred to take away from them certain benefits that their children enjoyed, such as phones, computers, television and play time with friends. These results reveal parents’ behaviorist view of assessment as a process that has to be either positively or negatively reinforced reflecting that parents are not active agents in the development of their children’s English competence, but they just react to the grades they receive from their children. It cannot be ignored that these parents have probably received themselves very traditional and behaviorist schooling, in which English learning was conducted through the memorization of long lists of vocabulary and the translation of grammatical structures as Díaz, Alarcón and Ortiz (2015) suggest. It is then no surprise that parents are not likely able to produce English themselves, so they just offer their children positive or negative reinforcements based on the grades they bring from school.

It is suggested that parents should not only encourage their children to improve themselves, but also participate in their children’s process of learning; thus, 8% of parents helped their children to improve their results by practicing together and reviewing the assessment tasks in which the children had performed poorly. Parents are highly encouraged to make use of this practice not only to improve their children’s grades but also to engage and bond with them in their learning processes.

Conclusions

Firstly, the results were correlated using Pearson’s correlation to identify a relation between the variables, if any. Secondly, when reviewing the results from the scales, it was discovered that the type of assessment environment highly affects the levels of anxiety and the perceptions that students hold of assessment tasks.

It was eye-opening to find out that 97% of the parents interviewed had a positive perception of the English language. Moreover, all parents supported their children learning English as a foreign language because of the possibilities this could open in their future as they believe that English is important to keep up with the ever-changing world.

The findings of this study suggest the existence of a link between the type of assessment environment a classroom has, either Learning-oriented or Performance-oriented, the students’ perception of the assessment tasks and the levels of anxiety experienced by them. Findings also suggest an untapped opportunity for parents to learn and bond together with their children in an English language environment. It was also found out that the scales used here – the Test Anxiety Inventory, the Perceptions of Assessment Tasks Inventory and the Perceived Classroom Environment Scale – are reliable techniques for collecting students’ perceptions on language assessment in a Chilean context, and it might merit further use in EFL research in Chile.

It is believed that the current research and the findings reported might be of use to other researchers, especially to language practitioners who seek to understand their students’ perceptions of their language assessment environments and practices. Regarding students’ perceptions of language assessment as a foreign language, we believe we have provided evidence to suggest that teachers should strive to create an assessment environment that resembles what has been previously described as a Learning-oriented assessment environment, as it is perceived by students as more welcoming and fairer, while also creating lower levels of anxiety than a Performance-oriented environment. Learning-oriented constitutes an environment in which high levels of Congruence, Authenticity, Student consultation, Transparency, and Diversity can be found. Furthermore, we believe that teachers should try to actively involve students in their own assessment process because, as stated by Cheng, Wu and Liu (2015), the involvement of students, including their perceptions, is an indicator of quality classroom assessment.

Regarding parents’ perceptions of learning English as a foreign language, it seems incredibly important for teachers to find a way of including parents in the students’ learning process, as the bonding and support they can provide would be highly beneficial to the children’s improvement in learning English, and would even encourage parents who do not know the language to learn it too.

Finally, we must encourage teachers to look out for signs in their students that might show that the assessment techniques they are using may lead to a Performance-oriented environment, as this is highly damaging for students’ perception of the language and English assessment. We also encourage teachers to work with students’ families in order to improve the perceptions they have of language assessment and the language itself.

Further research may and should be conducted based on the findings of this study. We highly encourage other researchers to replicate this study to verify the findings reached so far in order to compare the results and, hopefully, add more. There are certain factors of this study that were not considered which might provide interesting insights, such as gender, economic background, and age differences. Furthermore, applying the Likert scales used in this study more widely at a regional or even national level could provide valuable data about the students’ perceptions and, perhaps, indicate if the perceptions of environment, test anxiety and assessment task would change depending on the region where the students are from.

We also believe that a more detailed investigation to be conducted over a longer period of time regarding parents’ perceptions of language assessment might help improve the relationship among parents, schools, and students. A larger-scale research including teachers and school managers would also yield further data on the different types of language assessment that have been applied and how participants have been affected by them.

References

• ABEYWICKRAMA , Priyanvada. ; BROWN , Douglas . Language assessment: principles and classroom practices . New York : Pearson Longman , 2010 .
• ALDAS , Joaquín . Problemas metodológicos de la evaluación de los modelos de exposición: especial referencia al caso español . In: BIGNÉ , Enrique . Temas de Planificación de Medios . ESIC : Madrid , 2000 . 89 - 114 .
• ALEMI , Maedeh ; TAVAKOLI , Ehteramsadat . Audio lingual method . In: INTERNATIONAL CONFERENCE ON APPLIED RESEARCH IN LANGUAGE STUDIES, 3 ., 2016 , Tehran , International... Tehran : University of Tehran , 2016 . p. 1 - 6 .
• ALI , Muhammad .; MOHSIN , Muhammad . Test anxiety inventory (TAI): factor analysis and psychometric properties . Journal of Humanities and Social Science , v. 8 , n. 1 , p. 73 - 81 , 2013 .
• ALJURE , Iliana et al . Concepciones y expectativas de la comunidad educativa con respecto al bilinguismo (español-inglés) en siete colegios de Bogotá y Cundimarca . Revista El Astrolabio , Bogotá , v. 9 , n. 2 , p. 45 - 57 , 2010 .
• ALKHARUSI , Hussain . An evaluation of the measurement of perceived classroom assessment environment . International Journal of Instruction , v. 8 , n. 2 , p. 45 - 54 , 2015 .
• ALKHARUSI , Hussain . Canonical correlational models of students’ perceptions of assessment tasks, motivational orientations, and learning strategies . International Journal of Instruction , v. 6 , n. 1 , p. 21 - 38 , 2013 .
• ALKHARUSI , Hussain . Development and datametric properties of a scale measuring students’ perceptions of the classroom assessment environment . International Journal of Instruction , v. 4 , n. 1 , p. 105 - 120 , 2011 .
• ALKHARUSI , Hussain et al . Modeling the relationship between perceptions of assessment tasks and classroom assessment environment as a function of gender . The Asia-Pacific Education Researcher , Philippines , v. 23 , n. 1 , p. 93 - 104 , 2014 .
• ÁVILA-DAZA , Nelly ; GARAVITO , Sandra . Parental involvement in English homework tasks: bridging the gap between school and home . Profile , Bogotá , v. 11 , n. 2 , p. 105 - 115 , 2009 .
• BRUTON , Ana ; ROBLES-PIÑA , Rebecca . Deficit thinking and Hispanic student achievement: scientific information resources . Problems of Education in the 21st Century , Lituania , v. 15 , n. 41 , p. 41 - 48 , 2009 .
• BULDUR , Serkan ; DOĞAN , Alev . Adaptation of the students’ perceptions of the Science and Technology course classroom assessment environment scale into Turkish . Education & Science/Egitim ve Bilim , v. 39 , n. 176 , p. 199 - 211 , 2014 .
• CAVANAGH , Stephen . Content analysis: concepts, methods and applications . Nurse Researcher , v. 4 , n. 3 , p. 5 - 16 , 1997
• CHENG , Lying ; WU , Yongfei ; LIU , Xiaoqian . Chinese university students’ perceptions of assessment tasks and classroom assessment environment . Language Testing in Asia , Thailand , v. 5 , n. 1 , p. 13 , 2015 .
• CURRAN , Patrick ; WEST , Stephen ; FINCH , John . The robustness of test statistics to nonnormality and specification error in confirmatory factor analysis . Psychological Methods , v. 1 , n. 1 , p. 16 - 29 , 1996 .
• CZURA , Anna . Adolescent learner perceptions of foreign language assessment: critical incident analysis. Glottodidactica . An International Journal of Applied Linguistics , v. 44 , n. 2 , p. 25 - 39 , 2017 .
• DÍAZ , Claudio ; ALARCÓN , Paola ; ORTIZ , Mabel . A case study on EFL teachers’ beliefs about the teaching and learning of English in public education . Porta Linguarum , Madrid , v. 23 , n. 20 , p. 171 - 186 , 2015 .
• DORMAN , Jeffrey ; KNIGHTLEY , Wendy . Initial use of the perceptions of assessment task inventory (PATI) in English secondary schools . The Alberta Journal of Educational Research , v. 52 , n. 3 , p. 83 - 94 , 2006 .
• DUSEK , Jerome . The development of test anxiety in children . In: SARASON , I. G. ( ed .) . Test anxiety: theory, research, and applications . Hillsdale : Lea , 1980 . p. 87 - 110 .
• HASHEMI , Masoud . Language stress and anxiety among the English language learners . Procedia - Social and Behavioral Sciences , v. 30 , n. 1 , p. 1811 - 1816 , 2011 .
• HAYES , De Marquis . Parental involvement and achievement outcomes in African American adolescents . Journal of Comparative Family Studies , v. 43 , n. 3 , p. 567 - 582 , 2012 .
• HOOVER-DEMPSEY , Kathleen et al . Why do parents become involved? Research findings and implications . The Elementary School Journal , v. 106 , n. 2 , p. 105 - 130 , 2005 .
• KAVANAGH , Bianca ; ZIINO , Sarah ; MESAGNO , Christopher . A comparative investigation of test anxiety, coping strategies and perfectionism between Australian and United States students . North American Journal of Psychology , v. 18 , p. 555 - 569 , 2016 .
• LAM , Long . Impact of competitiveness on salespeople’s commitment and performance . Journal of Business Research , v. 65 , n. 9 , p. 1328 - 1334 , 2012 .
• LIN , Grace Hui . Pedagogies proving krashen’s theory of affective Filter: Hwa Kang . Journal of English Language & Literature , v. 14 , p. 113 - 131 , 2008 .
• LIU , Qing-xue ; SHI , Jin-Fang . An analysis of language teaching approaches and methods effectiveness and weakness . US-China Education Review , v. 4 , n. 1 , p. 69 - 71 , 2007 .
• MEDINA , Monica ; GUZMÁN , Norma ; WONG-RATCLIFF , Monica . Latino parental involvement: myths, perceptions and inhibiting factors . Journal of Case Studies in Education , v. 7 , p. 1 - 15 , 2015 .
• MUÑOZ , Ana ; ALVARES , Martha . Students’ objectivity and perception of self-assessment in an EFL classroom . The Journal of Asia TEFL , v. 4 , n. 2 , p. 1 - 25 , 2007 .
• ONOMIA , Austin ; ONOMIA , Taiwo . Relevance of skinner’s theory of reinforcement on effective school evaluation and management . European Journal of Psychological Studies , v. 4 , n. 4 , p. 174 - 180 , 2014 .
• PINO-JUSTE , Margarita ; RODRÍGUEZ LOPÉZ , Beatriz . La importancia de la participación de los padres en la enseñanza del inglés en educación infantil . Didáctica Lengua y Literatura , Madrid , v. 19 , p. 187 , 2007 .
• REBER , Arthur . Dictionary of psychology . 2 . ed. Harmonsworth : Penguin , 1995 .
• SHUTE , Valerie et al . A review of the relationship between parental involvement and secondary school students’ academic achievement . Education Research International , v. 2011 . Retrieved from http://dx.doi.org/10.1155/2011/915326
  » http://dx.doi.org/10.1155/2011/915326
• SPIELBERGER , Charles . Test anxiety inventory: preliminary professional manual . Palo Alto : Consulting Psychologists Press , 1980 .
• SZAFRANSKI , Derek ; BARRERA , Terri ; NORTON , Peter . Test anxiety inventory: 30 years later . Anxiety, Stress & Coping , v. 25 , n. 6 , p. 667 - 677 , 2012 .
• VAVLA , Laureta ; GOKAJ , Rregjina . Learner’s perceptions of assessment and testing in EFL classrooms in Albania . Mediterranean Journal of Social Sciences , v. 4 , n. 11 , p. 509 , 2013 .

• ERRATUM
  Erratum: http://dx.doi.org/10.1590/s1517-9702202147226529E
  In the article English language assessment: unfolding school students’ and parents’ views DOI: 10.1590/S1678-4634202147226529 published in Educação e Pesquisa, v. 47, p. 01-25, e-location e226529, 2021, please change:
  Page 16:
  Where it is:
  

  Figure 3
  Parents’ reactions towards grades in the English subject
  
  Please, read:
  

  Figure 3
  Parents’ reactions towards grades in the English subject
  
  Page 16:
  Where it is:
  

  Figure 4
  Sub-dimensions of Dimension 2
  
  Please, read:
  

  Figure 4
  Sub-dimensions of Dimension 2
  

Publication Dates

• Publication in this collection
  16 Aug 2021
• Date of issue
  2021

History

• Received
  23 July 2019
• Reviewed
  12 Nov 2019
• Accepted
  24 Mar 2020