Revisiting the role of breadth and depth of vocabulary knowledge in reading comprehension

Abstract

The purpose of this study was to investigate the role of breadth and depth of vocabulary knowledge in reading comprehension at Debre Markos University. A quantitative approach was taken to gather and analyze the data. Out of 235 students learning at the college, 61 samples were taken randomly. To investigate their knowledge of vocabulary breadth, the Vocabulary Levels Test (VLT) was employed. The Depth of Vocabulary Knowledge (DVK) test was utilized to investigate the depth of vocabulary knowledge. The reading section of the Test of English as a Foreign Language (TOFEL) was used to determine the reading comprehension performance of the students. Pearson Product-moment correlation was used to examine the relationship between vocabulary knowledge (breadth and depth) and reading comprehension. In addition, to find out which aspect of vocabulary knowledge best explains reading comprehension, Standard Multiple Regression was employed. The data were analyzed using SPSS (version 21). The findings suggest that there was a significant strong positive relationship between knowledge of vocabulary breadth and reading comprehension (r = .73, n = 61, P, =.000 < 0.05). Besides, the result reveals that there was a significant strong positive relationship between knowledge of vocabulary depth and reading comprehension (r = .60, n = 61, P, =.000 < 0.05). The finding also shows that vocabulary breadth and depth together were able to predict respondents’ reading comprehension. However, vocabulary breadth (Beta =.58) had a more unique explanatory power than knowledge of vocabulary depth (Beta =.315).

Keywords:

• Vocabulary knowledge
• breadth
• depth
• reading comprehension

1. Introduction

Reading comprehension in a foreign language is a convoluted process. It is a demanding and challenging skill to develop. As Bilikozen and Akye (2014) and Graesser (2007) state, reading is a multifaceted, complex process that involves the interplay of a wide range of components.Among the various antecedents, vocabulary knowledge plays a huge role in comprehending reading texts. As Grabe (2009), I. S. P. Nation (2000), and Schmitt (2000) contend, vocabulary knowledge strongly influences students’ reading comprehension. According to Khosravi and Rashidi (2010), vocabulary has been considered a prerequisite and strong determinant of reading achievement for years.As Laufer (1997) contends, no text comprehension is possible, either in one’s native language or in a foreign language, without understanding the text’s vocabulary. This could mean that lexical problems hinder reading comprehension a great deal. According to Haynes and Baker (1993) cited in Laufer (1997), the most significant handicap for L2 readers is not a lack of reading strategies but insufficient vocabulary in English. As demonstrated by Grabe (2009), Qian (1998, 2002), and Schmitt (2000), vocabulary knowledge plays a significant role in enabling students’ reading comprehension.Although a significant number of studies have been undertaken to find out the role of vocabulary knowledge on reading comprehension, most of them have focused on first language reading comprehension (L1). Furthermore, a range of studies have conceptualized vocabulary knowledge as one-dimensional, mainly in terms of size (the number of words a reader knows). In many studies undertaken so far, the tests used to measure knowledge of vocabulary breadth appear limited in terms of incorporating representative words.Contrary to this, the multidimensional and complex nature of vocabulary knowledge (Daller & Milton, 2007; I. S. P. Nation, 2000; Schmitt, 2000) remains unexplored. Hence, this study adds three contributions to the existing literature.

Firstly, a vast majority of previous studies focused on vocabulary knowledge and reading comprehension in L1. It might be difficult to generalize the findings to foreign languages. L2 readers in many contexts read texts that are usually reasonably challenging. They read materials that require higher linguistic knowledge than their level. Vocabulary knowledge is the most significant problem in the reading comprehension of second language students. As a result, this study attempts to provide a comprehensive picture of the role of vocabulary knowledge in reading comprehension in an L2 context.

Secondly, this study conceptualizes vocabulary knowledge as a multidimensional construct. This is because both breadth and depth of vocabulary knowledge are central to reading comprehension. It helps me understand vocabulary knowledge and its relation to reading comprehension deeply. It also allows for the determination of the relative explanatory power of each aspect of vocabulary knowledge on reading comprehension.

Lastly, in terms of methodology, the current study employs comprehensive tests that measure vocabulary knowledge fully so that a dependable conclusion can be drawn. Therefore, this study aims to answer the following research questions.

1. What is the relationship between students’ knowledge of vocabulary breadth and depth and their reading comprehension?

2. Which is the best predictor of reading comprehension: knowledge of vocabulary breadth or depth?

2. Review of related literature

2.1. Vocabulary knowledge in English

In second language research, various proposals have been made as to what is meant by knowing a word. For many years, lexical researchers have developed various criteria for understanding what is involved in knowing a word. As Daller and Milton (2007) describe, on the basis of how knowing is defined, there are very different ideas about what involves knowing words. The counts of learners’ vocabulary size also vary based on the definition of knowledge utilized. As Qian (2002) points out, vocabulary knowledge is divided into two main categories: knowledge of word meaning and levels of accessibility to this knowledge. The former includes generalization (being able to define the word), breadth of meaning (recalling the different meanings of the word) and precision of meaning (applying the word correctly to all possible situations). The latter comprises availability (being able to use the word productively) and application (selecting an appropriate use of the word). However, this does not include other aspects of lexical knowledge, such as spelling, pronunciation, morpho-syntactic properties, and collocation.According to Meara (2009), word knowledge include rate of recurrence (frequency), constraints on use, linguistic function, derivation (the underlying form and the derivations that can be made from that form), associations (antonym, synonym), semantic features, and polysemy. Although this conceptualization appears more comprehensive than the previous one, there are still some missing elements, such as pronunciation, spelling, and collocation.In a more comprehensive way, I. S. P. Nation (2000) proposes the aspects of knowing a word by making distinctions between receptive/productive dimensions. As he states, receptive means that learners receive language input from others through listening or reading and attempt to comprehend it. On the other hand, productive means that learners produce language forms by speaking and writing to deliver messages to others. As the words receptive and productive include all the aspects of what is involved in knowing a word, knowing a word encompasses form, meaning, and use (see table one).

2.1.1.

Although Nation’s framework appears more comprehensive, according to Schmitt (2014), there is a lack of an accepted conceptualization of what receptive and productive knowledge of vocabulary entails. Daller and Milton (2007) also criticize the relationship between the two types of knowledge as lacking clarity. They might also be conceptualized differently depending on a variety of individual learner characteristics and the type of test used.Because vocabulary knowledge is complex and multi-faceted, to illuminate the way words are learned and stored, a different conceptualization is forwarded by Daller and Milton (2007) which is used in this study as a theoretical framework. This is called lexical space where learners’ vocabulary knowledge is described as a three-dimensional space where each dimension represents an aspect of knowing a word.

In the figure 1, the horizontal axis represents the concept of lexical breadth, which is intended to define the number of words a learner knows regardless of how well he or she knows them. This might entail the Form and the form and meaning elements of Nation’s conceptualization. The vertical axis represents the concept of lexical depth which is intended to define how much the learner knows about the words he or she knows. This might embrace the elements of concepts and referents, associations, grammatical functions, collocations, and constraints on use from Nation’s explanation. The final axis is fluency which is defined as how readily and automatically a learner is able to use the words he/she knows and the information he/she has on the use of these words. This would involve the speed and accuracy with which a word can be recognized or called to mind in speech or writing. This dimension is not used in this study, as the research focuses on the breadth and depth of vocabulary knowledge and their association with reading comprehension.

Figure 1. The lexical space: dimensions of word knowledge and ability (p, 8).

3. The role of vocabulary knowledge in reading comprehension

Students’ vocabulary knowledge should reach a level appropriate to reading and comprehending written texts intended for them. They should reach the threshold level of vocabulary for successful reading comprehension. In this regard, I. S. P. Nation (2000) describes at least two ways of defining the meaning of threshold. The first way to understand a threshold is as an all-or-nothing phenomenon. This implies that when learners do not cross the threshold, they cannot adequately comprehend written texts. When the learners cross the threshold, other things remain constant, they can comprehend written texts. The second perspective is to look at threshold as a probabilistic boundary. This means that when learners do not cross the threshold, there are low chances of adequately comprehending written texts.In both perspectives, Laufer and Sim (1985a) cited in I. S. P. Nation (2000) draw a conclusion that vocabulary knowledge is the most demanding need of foreign language learners. I. S. P. Nation (2000) notes that threshold level studies indicate that, the majority of text words (98%) must be known for comprehension to occur during unassisted reading. Schmitt (2007) also concludes that knowledge of around 3,000 word families is the threshold that could help learners start to read authentic texts. More importantly, Schmitt (2007) argues that knowledge of the most frequent 5,000 word families could provide sufficient vocabulary to allow learners to read authentic texts. Although several vocabularies can still be unknown, this amount of knowledge could enable learners to infer the meaning of many of the new words from context and to understand most of the communicative content of the text. L2 learners with knowledge of the most frequent 10,000 word families in English can be considered to have a wide vocabulary, and Schmitt (2007) a vocabulary of this amount may be needed to succeed in university study in a second language.For this reason, vocabulary knowledge is essential for successful reading comprehension. Vocabulary knowledge is conceptualized in this study as having breadth and depth. Lexical breadth refers to the number of words a learner knows, regardless of how well he or she knows them. This might entail the word form and the form and meaning elements. Depth means how much the learner knows about the words he or she knows. This appears to embrace the elements of concepts and referents, associations, grammatical functions, collocations, and constraints on use.Both aspects of vocabulary knowledge play tremendous role in explaining students’ reading comprehension. According to Khosravi and Rashidi (2010), learners who have higher breadth of vocabulary knowledge can perform better in reading comprehension. Accordingly, students’ reading comprehension can be disrupted by deficiency of their vocabulary knowledge. Besides, they assert that the growth in vocabulary knowledge matches with more reading comprehension. Thus, differences in students’ vocabulary knowledge are salient in explaining the perceived differences in their reading comprehension. Both students’ vocabulary size and depth are powerful in predicting their reading performance. Their knowledge in both dimensions has significant correlation with their reading performance (Choi, 2013; Fatemi & Shadman, 2012; Moinzadeh & Moslehpour, 2012; Sen & Kuleli, 2015). Qian (1998) also studied the relationships between vocabulary size, depth of vocabulary knowledge, and reading comprehension in English as a second language (ESL) in 74 adult Chinese and Korean speakers’ comprehension of general academic texts in English. The results indicate that vocabulary size, depth of vocabulary knowledge, and reading comprehension were positively, and closely, related. Depth of vocabulary knowledge made a unique contribution to the prediction of reading comprehension. Rahman and Iqbal (2019) also undertaken a research on the same topic and found out that there was a high correlation between depth of vocabulary knowledge and reading comprehension. The result also showed that there was a moderate correlation between breadth of vocabulary knowledge and reading comprehension. Regression analysis showed that vocabulary depth has high predictive power than vocabulary breadth. Therefore, the current study was carried out to investigate the role of breadth and depth of vocabulary knowledge on reading comprehension. By finding out which aspect of vocabulary knowledge better predicts reading comprehension, this research makes contributions new knowledge to the existing literature.

4. Methodology

4.1. Participants of the study

The participants in this study were first-year students at Debre Markos University. As could be guessed, the students at this university have been learning English for at least 12 years. All of them are supposed to have relatively similar socio-economic and educational backgrounds. This is to mean that they were all nonnative speakers of English; were taught by nonnative teachers, and were taught using textbooks prepared by the government. They did not also have exposure to the target language outside the classroom. Hence, they speak English as a university-level nonnative speaker. More specifically, they were students learning in the College of Social Science and Humanities. This college was selected because this researcher observed relatively severe reading problems in this college. There were 139 males (59.15%) and 96 females (40.85%) in the population. This brings the total number of students in the college to 235. Their age range was 19–22.

4.2. Sample size and sampling technique

The sample size was determined based on Tabachnick and Fidell (2007) that recommend the following formula, taking into account the number of independent variables that are used in research. Here is the formula: N > 50 + 8 m (where m = number of independent variables). Therefore, N > 50 + 8 × 2 is 66. However, five students quit participating at the start of the data gathering process.From the sampling frame (a list of all elements in the target population) in particular, the samples were selected using the lottery method. The total number of students who were learning at the college (235) was taken from the registrar. Then, each student’s name was written on a separate piece of paper. After that, each paper was folded. Then, 66 pieces of paper containing each student’s name were drawn. However, out of this number, five of them quit participating at the start of the data gathering process. Therefore, 61 participants were included in the study.

4.3. Instruments

To answer the research questions of the study, three tests, were used. The tests were employed to find out participants’ level of knowledge of vocabulary breadth, depth and reading comprehension.

The first test was employed to measure vocabulary breadth. There are various standardized tests of English vocabulary breadth in the literature. However, a test which meets the purpose of this study seems to be the Vocabulary Levels Test (VLT). It was originally designed by Nation (1983), and updated by Schmitt et al. (2001) as a means to determine the extent to which test takers could recognize the form-meaning connections of words at four word frequency levels (2000, 3000, 5000, 10000) and an academic vocabulary level. It was designed to give an estimate of receptive vocabulary size for L2 learners of general or academic English. Its separate levels measure learners’ knowledge of words from a number of distinct frequency levels. As Schmitt et al. (2001) state, the Levels Test derives its name from the fact that separate sections measure learners’ knowledge of words from a number of distinct frequency levels. In this way, it can provide a profile of a learner’s vocabulary, rather than just a single-figure estimate of overall vocabulary size.

At each vocabulary size level, the words are presented in 10 clusters of six words (three keys and three distractors) and three definitions. There are 10 test items, each comprising six words, and three definitions. The test taker is required to match the three definitions with three of the six words provided by writing the corresponding number of the word beside its definition as in the example below. Each level contains 30 correct choices. Therefore, the maximum total score is 150. Reflecting the distribution of these word classes, the words from the stratified sample tended to fall into a 5 (noun): 3 (verb): 2 (adjective) ratio. This ratio was maintained in the test, with each section containing five noun clusters, three verb clusters, and two adjective clusters. As Webb, Sasao and Balance (2017) contend, the proportion of nouns, verbs, and adjectives is representative of their proportional occurrence in English although it should be noted that this may vary within frequency bands. The following illustrates the format of a noun cluster:

You must choose the right word to go with each meaning. Write the number of that word next to its meaning.

1. concrete

2. era ____ circular shape

3. fiber ____top of a mountain

4. hip ___a long period of time

5. loop

6. summit

Taken from (Schmitt et al., 2001, p 58)

In this VLT, Schmitt et al. (2001) suggested a cutting point threshold for mastery of a level. They recommend that if test takers’ scores were 26/30 (87%) or higher, they had achieved mastery of that level and might then focus on learning words from the next level.

The appropriate test for the purpose of this study to measure depth appears to be The Word Associates Test (WAT), currently renamed as Depth of Vocabulary Knowledge (DVK), which was first developed by (Read, 1993, 2000; Carter, 1998). The DVK mainly measures two aspects of depth of vocabulary knowledge: meaning (synonymy and polysemy) and collocation, or the paradigmatic and syntagmatic relationships of words which are the central components of the construct. Although the DVK taps only knowledge of adjectives, given the design of the measure, which requires the identification of nouns that collocate with the adjectives tested, nouns are actually indirectly tested as well (Qian, 1998).

In it, the target word is followed by eight options, four of which have some relationship with the target word and four of which do not. This DVK version contains 40 items. The maximum possible score, therefore, is 160. In the example below, the words in the left box have a paradigmatic relationship (sudden - quick, surprising) with the target word, and the ones in the right box have a syntagmatic relationship (sudden change, sudden noise). Participants were told that there are four associates and are asked to find them (Schmitt, 2000).

4.3.1. Sudden

Table

beautiful quick surprising thirsty

change doctor noise school

(Schmitt, 2000, p, 176)

The Internet-Based Test (IBT) reading section of Test of English as a Foreign Language (TOFEL) (2009) developed by Educational Testing Service (ETS) was used to determine the reading comprehension performance of the students. This test was also developed for nonnative university level students, which appears to be appropriate for participants of this study. Based on their scores, the students are grouped into three levels: high, intermediate, and low. Students who score (22–30) are labeled as high level; (15–21) are categorized as intermediate level readers. Students who get scores (0–14) are at low level.

4.4. Methods of data analysis and interpretation

Quantitative methods were used to analyze the data as tests were used to gather data. As Pallant (2010) and Kenny (1987) point out, raw data should be transformed into a standard score to investigate relationships between variables. Transformations have three purposes: interpretability (increase clarity of interpreting numbers), comparability (to compare results with other samples studied anywhere), and symmetry (symmetric distribution of scores) (Kenny, 1987). Inferential statistical techniques also assume these purposes (Kenny, 1987 & Pallant, 2010). Accordingly, in this study, the raw scores were transformed into standards. Percentile rank (Kenny, 1987) was done to analyze the relationship between vocabulary knowledge (breadth and depth) and reading comprehension. The percentile rank was found out using the following formula: 100 (R-.5)/n where n is the sample size and R is the rank order of the score (Kenny, 1987).

Regarding the statistical tests, to examine the relationship between vocabulary knowledge (breadth and depth) and reading comprehension, Pearson Product moment correlation was used as the level of measurement in both variables is continuous. In addition, to analyze how well students’ knowledge of vocabulary breadth and depth predicts their reading comprehension and to find out which aspect of vocabulary knowledge best predicts reading comprehension, regression analysis was employed. This method was utilized as the two independent and one dependent variables are continuous. More specifically, Standard Multiple Regression was used as it involves the entry of all of the independent variables into the model at once. The data was analyzed using SPSS (version 21) (See Table 1).

Table 1. Aspects of vocabulary knowledge

Form	Spoken	R P	What does the word sound like? How is the word pronounced?
	Written	R P	What does the word look like? How is the word written and spelled?
	Word parts	R P	What parts are recognizable in this word? What word parts are needed to express the meaning?
Meaning	Form and meaning	R P	What meaning does this word meaning signal? What word form can be used to express this meaning?
	Concepts and referents	R P	What is included in the concept? What items can the concept refer to?
	Associations	R P	What other words this make us think of? What other words could we use instead of this one?
Use	Grammatical functions	R P	In what patterns does the word occur? In what patterns must we use this word?
	Collocations	R P	What words or types of words occur with this one? What words or types of words must we use with this one?
	Constraints on use (register, frequency …)	R P	Where, when, and how often would we expect to meet this word? Where, when, and how often can we use this word?

In column 3, R = receptive knowledge, P = productive knowledge. Taken from Nation (2000:39)

5. Results and discussion

5.1. The role of breadth of vocabulary knowledge on reading comprehension

As can be seen in Table 2, the relationship between students’ knowledge of vocabulary breadth and their reading comprehension was investigated using Pearson product moment correlation. The finding indicates that there was a significant relationship between the two variables P, =.000 < 0.05. As already indicated, Muijs (2004) describes whether or not the relationship is statistically significant (unlikely to exist in the sample if it does not exist in the population), the standard cut-off point is <0.05. As the significance level or probability value (p-value) in this research is .000, the relationship is significant.

Table 2. Correlations between respondents’ vocabulary breadth and reading comprehension

Correlations
		Respondents’ Vocabulary Size	Respondents’ Reading Comprehension
Respondents’ vocabulary size	Pearson Correlation	1	.735**
	Sig. (2-tailed)		.000
	N	61	61
Respondents’ reading comprehension	Pearson Correlation	.735**	1
	Sig. (2-tailed)	.000
	N	61	61

Note: **. Correlation is significant at the 0.01 level (2-tailed).

The relationship was positive r = .74.This means when students’ knowledge of vocabulary breadth decreases, their reading comprehension level also decreases. A low score in breadth is associated with a low score in reading comprehension.

Moreover, the strength of the relationship was strong(r = .74). This is because Muijs (2004) states that some rules of thumb on effect size are: <0. ± 1 weak, <0. ± 3 modest, <0. ± 5 moderate, <0. ± 8 strong and ≥=±0.8 very strong. Accordingly, the strength of the relationship in the current research is strong (r = .74).

Although their relationship is positive, the result could mean that as the students’ vocabulary breadth decreases, their reading comprehension also decreases because students’ achievement in both variables is low. Hence, it would suggest that as vocabulary breadth decreases, so does reading comprehension. It could also implicate that when reading comprehension increases, knowledge of vocabulary breadth increases.

The shared variance in the relationship between students’ knowledge of vocabulary breadth and their reading comprehension is .74 × .74 = 0.55 = 55%. This would seem relatively large. Therefore, the practical significance would be seen as high. Students’ knowledge of vocabulary breadth helps to explain 55 percent of their reading comprehension. This can be considered as a highly regarded amount of variance explained.

This result is consistent with previous assumptions. For example, I. S. P. Nation (2000), Qian (1998), and Schmitt (2000) contend that knowledge of vocabulary breadth and reading comprehension has positive and strong associations.

5.2. The relationship between depth of vocabulary knowledge and reading comprehension

As shown in Table 3, the relationship between students’ knowledge of vocabulary depth and reading comprehension was analyzed. As with breadth, the relationship of depth and reading comprehension was investigated using Pearson product moment correlation coefficient. The result indicates that there was a significant relationship between the two variables P, =.000 < 0.05. Taking Muijs’s (2004) suggestion for a standard cut-off point of statistical significance (<0.05), the probability value (p-value) in this research (.000) is significant.

Table 3. Correlations between respondents’ vocabulary depth and reading comprehension

		Respondents’ vocabulary depth	Respondents’ reading comprehension
Respondents’ vocabulary depth	Pearson Correlation	1	.601**
	Sig. (2-tailed)		.000
	N	61	61
Respondents’ reading comprehension	Pearson Correlation	.601**	1
	Sig. (2-tailed)	.000
	N	61	61

Note: **. Correlation is significant at the 0.01 level (2-tailed).

The relationship was positive r = .601.This indicates that students’ knowledge of vocabulary depth decreases as students’ reading comprehension level also decreases. A low score in depth is associated with a low score in reading comprehension. This would mean that when students’ knowledge of vocabulary depth decreases, their reading comprehension also decreases. It could also suggest that as students’ knowledge of vocabulary depth increases, their reading comprehension level increases.

Besides, the effect size of the relationship was strong (r = .601). Assessing it against the rule of thumb suggested by Muijs (2004) on effect size, that is, <0. ± 1 weak, <0. ± 3 modest, <0. ± 5 moderate, <0. ± 8 strong and ≥=±0.8 very strong, the strength of the relationship between students’ knowledge of vocabulary depth and their reading comprehension is strong.

Shared variance was also calculated to know the practical significance of the relationships. The shared variance of the relationship between students’ knowledge of vocabulary depth and their reading comprehension is .601 × .601 = 0.36 = 36.12%. The variance would seem relatively moderate. Therefore, the practical significance would also be seen as moderate. Students’ knowledge of vocabulary depth helps to explain nearly 36 percent of their reading comprehension. This can be considered as an adequate amount of variance explained.

This result is also consistent with previous conclusions. For example, as Qian (1999) states, knowledge of vocabulary depth has a strong and positive correlation with reading comprehension. More recently, Rahman and Iqbal (2019) also undertaken a research on the role of knowledge of vocabulary breadth and depth in reading comprehension and found out that there was a high positive correlation between depth of vocabulary knowledge and reading comprehension.

5.3. Vocabulary knowledge that best predicts reading comprehension: breadth or depth

The relationship between students’ knowledge of vocabulary breadth and their reading comprehension was already examined. Similarly, the association between respondents’ knowledge of vocabulary depth and their reading comprehension level was analyzed. However, it is still unknown whether or not how well both variables (breadth and depth) together predict their reading comprehension. Furthermore, which aspect of respondents’ vocabulary knowledge best predicts their reading comprehension remains unknown. For this reason, standard multiple regressions was employed.

As can be seen in Table 4, the model summary demonstrates the measure of how well the overall model, i.e. the two predictors together is able to predict respondents’ reading comprehension. The R gives the amount of variance in respondents’ reading comprehension explained by the two-predictor variables together. This is a measure of how well the predictors predict the outcome. However, the square root of R is taken to get a more accurate measure. “Adjusted R Square” is a correction to R square, which takes into account that a sample is being looked at rather than the population. As the model is likely to fit the population less well than the sample, R square is adjusted downwards to give us a measure of how well the model is likely to fit in the population. R square and Adjusted R square vary between 0 and 1(Muijs, 2004).

Table 4. Model summary

Model Summary
Model	R	R Square	Adjusted R Square	Std. Error of the Estimate
1	.785^a	.616	.603	10.48574

Predictors: (Constant), respondents’ vocabulary size, respondents’ vocabulary depth.

According to Muijs (2004), the rule of thumb to see how well model fits the data (<0.1: poor fit, 0.11–0.3: modest fit, 0.31–0.5: moderate fit and >0.5: strong fit). Accordingly, the model in this research is strong fit (0.603). This is to mean that the model (vocabulary breadth and depth) explains 60.3 percent of the variance in respondents’ reading comprehension.

From the model shown in Table 5, it is seen that 60.3 percent of the variance in students’ reading comprehension is explained by their knowledge of vocabulary breadth and depth together. As shown in Table 5, this result is statistically significant; P, =.000 < 0.05. The ANOVA indicates that the value in the model is statistically significant.

Table 5. Anova

ANOVA ^a
Model		Sum of Squares	Df	Mean Square	F	Sig.
1	Regression	10226.002	2	5113.001	46.503	.000^b
	Residual	6377.137	58	109.951
	Total	16603.139	60

Note: a. Dependent Variable: respondents’ reading comprehension

Predictors: (Constant), respondents’ vocabulary depth, respondents’ vocabulary size

As can be seen in Table 6, the Beta tells us how much of the total variance in the dependent variable is uniquely explained by each variable. Betas vary between 0 and 1, with, as usual, 1 being the strongest effect. The Beta value for breadth is .58 and vocabulary depth is=.315. The sig (p) value for breadth and depth is .000 and .001 respectively. Therefore, vocabulary breadth has more unique explanatory power than knowledge of vocabulary depth. Its explanatory power is strong (.000), and sig (p) value is .000 < 0.01.

Table 6. Regression Coefficients^a

Coefficients^a
Model		Unstandardized Coefficients		Standardized Coefficients	T	Sig.
		B	Std. Error	Beta
1	(Constant)	.407	3.760		.108	.914
	Respondents’ vocabulary size	.683	.110	.580	6.198	.000
	Respondents’ vocabulary depth	.391	.116	.315	3.366	.001

Note: a. Dependent Variable: respondents’ reading comprehension

Standard Multiple Regression was employed to examine how well students’ knowledge of vocabulary breadth and depth is able to predict their reading comprehension. It was also used to analyze which variable, from breadth and depth, is the best predictor of their reading comprehension.

The model, i.e. the two-vocabulary knowledge (breadth and depth) together strongly fits the data. According to Muijs (2004), the rule of thumb to see how well a model fits the data is (<0.1: poor fit, 0.11–0.3: modest fit, 0.31–0.5: moderate fit and >0.5: strong fit). Accordingly, the model in this research is a strong fit (0.603).

As this is a measure of how well the two predictors predict the outcome, i.e. reading comprehension, the amount of variance in students’ reading comprehension explained by the two-predictor variables together is strong fit (0.603). This means 60.3% of the variance in reading comprehension can be predicted from the combination of knowledge of vocabulary breadth and depth. This amount of variance in reading comprehension was explained by the model. This is statistically significant, P, =.000 < 0.05 (how likely it is that we would have found a relationship this strong in our sample if there was not one in the population). This finding is consistent with previous studies whose results indicated that there was a significant correlation between both aspects of vocabulary knowledge with reading performance (Choi, 2013; Fatemi & Shadman, 2012; Moinzadeh & Moslehpour, 2012; Sen & Kuleli, 2015).The Beta value for students’ knowledge of vocabulary breadth is (.58). The Beta value for the students’ knowledge of vocabulary depth is (.315). The sig (p) value for breadth is .000, and the sig (p) value for depth .001. Therefore, vocabulary breadth has stronger unique explanatory power than knowledge of vocabulary depth. Its explanatory power is strong (.58), and sig (p) value is .000 < 0.05. Therefore, knowledge of vocabulary breadth has a more explanatory role in students’ reading comprehension. This result is inconsistent with Qian (1999) and more recently studied research by Rahman and Iqbal (2019). Although the current study supports Qian’s and Rahman and Iqbal’s (2019) findings in the role of both knowledge aspects, breadth and depth, it contrasts with their conclusions in the sense that knowledge of vocabulary depth has a stronger unique explanatory power.In the current research, the degree of predictability increased when both knowledge of vocabulary breadth and depth are considered as a predictor variable (0.603). Separately knowledge of vocabulary breadth predicted (.58). Knowledge of vocabulary depth predicts (.315). Although breadth exceeds to depth, their combination is stronger in predicting reading comprehension.

6. Conclusions

The purpose of the study was to investigate students’ breadth of vocabulary knowledge, depth of vocabulary knowledge and their association with reading comprehension. Based on the results, the following conclusions can be drawn.Based on the results, the following conclusions can be drawn. There was a significant positive strong relationship between knowledge of vocabulary breadth and reading comprehension. Similarly, there was a strong significant positive relationship between knowledge of vocabulary depth and reading comprehension. It was also found out that students’ knowledge of vocabulary breadth and depth together predict well their reading comprehension. However, students’ knowledge of vocabulary breadth has a more unique explanatory effect on their reading comprehension level.

7. Implications

This study has two helpful implications. Firstly, in this research, the degree of predictability increased when both knowledge of vocabulary breadth and depth are taken together as a predictor variable. Although breadth exceeds depth, their combination is stronger in predicting reading comprehension. Hence, students will be benefited when they develop both aspects of vocabulary knowledge instead of learning only the single aspect such as either breadth or depth. Both aspect of vocabulary knowledge should be emphasized in the instruction process. Secondly, although both aspects of vocabulary knowledge are found to be critical, breadth of vocabulary knowledge deserve huge attention and predominant emphasis over depth of vocabulary knowledge.This research gives a comprehensive understanding of the role of vocabulary knowledge on reading comprehension. However, the finding in this study is inconsistent with Qian’s (1999) and Rahman and Iqbal’s (2019) studies regarding the relative role of breadth and depth of vocabulary knowledge on reading comprehension. In the current research, breadth of vocabulary knowledge has more explanatory power than depth. However, Qian (1999) and Rahman and Iqbal (2019) found out that depth of vocabulary knowledge has more predictive power than breadth. Thus, future researches should be carried out to determine the inconsistencies.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

There is no funding to report.

Notes on contributors

Animut Tadele Dagnaw

Animut Tadele Dagnaw is an Assistant Professor of English at Bahir Dar University, Ethiopia. He earned his bachelor’s degree in English Language and Literature from University of Gondar. He did his MA and PhD degrees in Teaching English Language as a Foreign Language from Addis Ababa University. He has over fifteen years of experience in teaching English language in universities. He has written a book on ‘Communicative English Grammar: A self-study Reference and Practical book for high School Students.’ Currently, he is also serving as a Director of Information and Strategic Communication at College of Medicine and Health Sciences, Bahir Dar University.