Primary school students’ reading comprehension skills Tuula Merisuo-Storm

Introduction Reading skills are an important tool that students need to become academically successful. When studying almost any area in the school curriculum they use texts to acquire new information. Therefore, when reading a text the goal is to understand its content. It is a process that exceeds decoding, and includes comprehension processes of the word, sentence, and text level. A child who does not learn to read and comprehend in the early school years has severe difficulties also in studying other school subjects (Bowyer-Grane & Snowling, 2005, p. 190; McGee & Johnson, 2003, p. 49). An important goal of literacy teaching is to awaken pupils’ interest in language and literature and also give them a lasting positive attitude towards reading. Children who are motivated to read spend more time reading than their less motivated peers. Unfortunately, poor readers are often unmotivated to read. Consequently, the difference between their respective levels of reading skills becomes even more significant. The aim of literacy teaching is to support the development of pupils’ ability to read, interpret and use different texts. Each pupil should adopt a habit of evaluating and observing himself or herself as a reader. In addition, he or she should learn to select appropriate reading material for different purposes and to find information in various sources (Merisuo-Storm, 2006).

Reading comprehension

Reading comprehension is a complex process: the reader constructs meaning by interacting with text using his or her previous knowledge and experience and the information that can be found in the text. The more background information related to the text the reader possesses, the easier it is for him or her to understand the text. When reading the text he or she activates the previous knowledge about the topic. Moreover, each text is unique as regards the structure of the text, its genre, vocabulary, and language. Several factors influence a reader’s interaction: how easy the text 1

is to read, how accurately it follows the conventions of its genre or structure, the language it is written in, and even the type and the size of font (Blair-Larsen & Vallance, 2004, p. 37; Pardo, 2004, p. 272–275). Reading is thinking cued by written language. A skilful reader finds – when reading fiction as well as non-fiction – several pieces of information in the text that make the understanding easier. Effective readers locate the basic facts from the text. Literal comprehension is needed when reading non-fiction but also when reading fiction. In fiction, a reader has to identify the characters and follow the events of the story. In non-fiction, a reader needs to comprehend the topic, learn new facts related to it, and be able to find and remember important information (Scharer, Pinnell, Lyons, & Fountas, 2005, p. 25). Different types of texts are organised in different fashion. Most narrative texts follow a general structural pattern, which is often called story grammar. In contrast, expository texts have greater structural complexity. They include several patterns for instance, description, sequence, compare– contrast, cause–effect, and problem solution. Already before entering school children develop sensitivity to narrative structure. It helps them to understand stories. They take note of the setting (time, place, and characters), problems or conflicts, actions and reactions of the characters, and the story resolution as they read. However, mature comprehension includes generalisation beyond the characters and the events of the story to the people and the events in real life. Expository texts are generally more demanding to comprehend because they contain many different structures. In addition, they often introduce many new concepts. Therefore, a reader needs previous knowledge on the topic to generate accurate inferences. Moreover, many children are not acquainted with nonfiction because most parents mainly read storybooks with young children (Williams et al., 2005, p. 538; Williams, 2005, p. 7). Snow, Sweet, Alvermann, Kamil, and Stricland (2002) suggest that reading comprehension is, in part, made up of nine cognitive components: fluency, vocabulary, world knowledge, discourse knowledge, linguistic knowledge, motivation, purposes and goals, cognitive and meta-cognitive strategies, and integrating non-print information with text. They are used when a reader aims to create comprehension of the text. Good comprehenders are fluent readers. Non-fluent reading demands mental capacity that would be needed for comprehension process. Fluent readers read with ease and expression. Fluency involves accuracy and automatic word identification, decoding with facility, and expressiveness in reading. It demands exact word identification and decoding ability. However, a reader may have problems in comprehending a text although he or she is able to read single words correctly. This is due to difficulties located in particular in the higher order skills of making conclusions, forming a logical meaning of a text and monitoring his or her comprehension (Hurry & Doctor, 2007, p. 44). 2

In addition, skilful comprehenders need an extensive vocabulary. A reader cannot comprehend a text if he or she does not understand the individual words it includes. There is evidence that vocabulary differences affect comprehension, and that increasing vocabulary acquisition has a positive influence in reading comprehension. Furthermore, increasing reading practice can increase vocabulary and comprehension. Research has also shown that often children with reading disabilities have lower-than-average vocabularies. This can be a consequence of the effects of linguistic problems or limited exposure to print. It appears that vocabulary growth is largely dependent on parental practices, especially before third school year (Biemiller, 2003; Pressley, 2002, p. 297; Snow et al., 2002, p. 95, 98). Children’s learning of word meanings through content can be minor in comparison to the learning of word meanings through direct vocabulary instruction. However, there is a difference between written and spoken language. Written language is lexically richer than spoken language and it can, therefore, provide more learning opportunities than a spoken context. Consequently, a student who is an eager reader encounters more words than a student who reads less. He or she has many opportunities to acquire and improve vocabulary knowledge through inference from context (Cain, Oakhill, & Lemmon, 2004, p. 672). A child acquires world knowledge from experience and from texts he or she reads. Skilful comprehenders use discourse knowledge to connect text elements in a consistent fashion and to connect the content to the messages of the writer. Discourse knowledge includes for instance, the structure of different genres, the discrimination of old and new information, the main messages the author intends to convey, and the topic structure. Linguistic knowledge includes oral language capacities – production and comprehension – and the ability to reflect one’s knowledge of language. Skilful reading requires knowledge of all linguistic levels – phonology, morphology, syntax, and higher-level discourse structures. In addition, reading comprehension often involves integrating non-print information for instance, graphs and pictures, with text. Electronic texts often include multimedia components and the reader has to integrate them with text to comprehend the contents of the text (Snow et al., 2002, p. 95–98). Skilful readers are purposeful. They read to obtain information, to enjoy literature, to locate specific information, and to learn from text. Reading motivation consists of several aspects. The students with strong sense of competence are usually eager readers. In addition, when children are inherently motivated to read, value reading highly, and are mastery-goal oriented, they are more likely to engage in reading (Snow et al., 2002, p. 95). Good readers use comprehension strategies to support the construction of meaning. A reader may use these strategies consciously or they may have become automatic with practice. The use of a meta-cognitive strategy is activated by the reader’s assessment of his or her own cognitive state. 3

For instance, when reading a text on an unfamiliar topic the reader slows down. Skilful comprehenders often skim the text before reading it. They make predictions about it based on their previous knowledge. When reading, they pay special attention to the sections of the text that are essential for their purpose or particularly difficult. The other parts of the text they read more quickly and with less care. While reading they respond to the text, ask questions, create mental images representing the meaning of the text, and interpret the text. After reading they continue to reflect on the text. They review it and possibly reread the text parts that seem especially important to their purpose or that they did not understand sufficiently well (Pressley, 2002, p. 297). Parker and Hurry (2007, p. 311–313) argue that many teachers do not have explicit knowledge of the key reading comprehension strategies. Direct oral questioning seems to be the dominant strategy for teaching reading comprehension. Often this kind of directive questioning produces predictable correct answers. More infrequent are teachers’ questions that assist children to develop more elaborated ideas. Even if the range of teachers’ questions is wide and appropriate children’s role is too passive.

Comprehending non-fiction Good comprehenders read fiction and non-fiction. They use their knowledge of text structure to effectively and strategically process the text. They have developed this knowledge from reading texts of different genres. Reading different genres and text formats affords opportunities for strategy use, enhances understanding of how words work, and provides bases for discussion and meaning negotiation. Good readers construct and revise meaning while reading and also monitor their comprehension. If they have difficulties in understanding the text because of insufficient background information, difficulty of words, or unfamiliar text structure, they know many strategies to use, and are also able to select appropriate strategies (McLaughlin, 2006, p. 6–7). Because, there are a great variety of written materials available the act of comprehension is very sophisticated. Introducing pupils to many different texts prepares them for understanding more complex texts and complex issues in the future. Comprehension strategies are necessary for a reader because they can provide access to knowledge that is beyond his or her personal experience. Students that use cognitive strategies, such as previewing, activating prior knowledge, predicting, making connections, monitoring, organising, summarising, questioning, and visualising, are likely to comprehend and to be able to remember more of what they read. Researchers suggest that using such strategies helps pupils to become meta-cognitive readers. To become a skilful comprehender, a reader needs to decide which strategies to use depending on the content of the text and its level of difficulty. When reading a difficult text, the reader needs to consciously direct the process of 4

meaning construction (Barton & Sawyer, 2003, p. 334–336; Bimmel & van Schooten, 2004, p. 86; Dougherty Stahl, 2004, p. 598; McLaughlin, 2006, p. 6). Children are often more successful in comprehending narrative texts than expository texts. Expository texts are challenging for the reader as a result of their structural complexity, density of information, and demands of prior knowledge. They may contain abstract and logical relations that are difficult to interpret. For instance, the level of reader’s prior knowledge has great influence on his or her comprehending a science text focusing on the needs of plants (Best, Floyd, & McNamara, 2008, p. 153). Research has shown that during the first school years the children do not get many opportunities of reading and writing information texts in school. However, integrating the teaching of comprehension strategies to learning information in content areas could be an effective manner of proceeding (Reutzel, Smith, & Fawson, 2005, p. 277–278). Because world knowledge is so closely related to comprehension of informational texts, one should teach children strategies that match their knowledge of texts. They may benefit from learning strategies that help them to integrate their prior knowledge with the information in a text (Best et al., 2008, p. 154–155). When reading informational texts there are four key elements that are central to comprehension processes: prior knowledge, inferential reasoning, self-regulation, and affective variables connected to efficacy and motivation. The readers with prior knowledge of the topic and the structure of an informational text are most able to mentally organize and remember the ideas the text provides. Inferential reasoning refers to a reader’s ability to read between lines. He or she makes connections that are not clearly expressed in the text. Inferential reasoning is an essential factor of skilled reading. A reader with sufficient prior knowledge of the topic makes more inferences than a less knowledgeable reader to make his or her comprehension easier. The term selfregulated reading refers to self-questioning and repair processes. Skilful readers are consciously aware of efficient information-seeking processes. They control these processes by choosing alternative strategies when others do not work. In contrast, less skilful readers are unsure of other strategies that might work when one strategy fails. Reader’s aims, beliefs, and attitudes towards reading may also influence how he or she uses cognitive reading strategies (Coiro & Dobler, 2007, p. 218–219). Van der Schoot, Vasbinder, Horseley, and van Lieshout (2008, p. 220) argue that more and less skilful comprehenders differ also in their use of reading strategies. More successful comprehenders spend more time in processing important than unimportant text elements. In contrast, less successful comprehenders use equally long time processing important and unimportant elements of text. They suggest that this shows that skilful comprehenders construct meaning actively and less skilful comprehenders are more passive readers. 5

Summarising is a very difficult comprehension strategy. However, good readers summarise while reading, and it helps them to remember and connect important ideas of the text. Nevertheless, many students struggle with determining the main ideas and themes of the text as well as combining similar ideas, and synthesising them into a coherent whole. Often they just repeat most of the text or give a very vague statement. Still a good summary should give a whole picture of the story, and include only the important parts in same order as the text as well as knowledge of how they are related (Diehl, 2005, p. 65). The older the pupils get the more they need their comprehension skills to acquire new information. They are, after reading a text, frequently expected to remember main ideas and concepts from the assigned passage. In addition, they are often asked to explain their decisions. A pupil can help the summarising process by asking himself or herself for instance, following questions: What does the paragraph seem to be about? Which sentence gives the main idea? Does this sentence tell anything new or more important than the main idea? Does this sentence repeat what has already been said (Swanson & De La Paz, 1998)?

Deriving the meaning of an unknown word Written language is lexically richer than spoken language, and gradually young readers start reading texts that contain many words that are not part of their oral vocabulary. It is a demanding task for a child to derive the meaning of an unknown word from the written context. The complexity of the word is a crucial factor. If the word is related to a known concept it may be a simple synonym. However, the word can refer to unfamiliar concepts. Additionally, the difficulty of deriving the meaning of a word from the context is connected to the concreteness of the unfamiliar concept. Furthermore, the complexity of deriving word meaning from context is also influenced by the nature of the context. Contexts do not necessarily reveal the full meaning of a word, even when explicit clues are present. Sometimes the context is even misleading (Fukkink, 2005, p. 24; Goerss, Beck, & McKeown, 1999, p. 153). A pupils’ ability to derive the meaning of a new word from context is also affected by the distance between the word and its cue(s). In addition, in non-fiction texts, a reader often has to integrate information from several idea units, which can be found in various places throughout a passage. The longer the distance between the pieces of information to be integrated the more difficult the process. If a pupil’s working memory is weak the process is even more demanding (Cain et al., 2004). Strategies that a reader uses when deriving the meaning of a new word from written context are defined as purposeful and situated (contex-related) sequences of activities. Which strategies a reader adopts depends on interactions among four factors. The first of these factors is the number of 6

unfamiliar words in the text. The second is reader’s interest: does he or she want to gain a more extensive vocabulary or only find out what the words in this particular text mean. The third is related to the task in hand: whether the reader’s goal is to comprehend the text or learn new words. The fourth factor is related to the learning context: for instance, if the reader studies in language or science class. On average, skilful readers and readers of high verbal ability tend to use more advanced strategies than young readers and readers of low verbal ability (van Daalen-Kapteijns, Elshout-Mohr, & de Glopper, 2001, p. 147–148). Fukkink (2005) argues that we do not yet know what activities untrained primary-school students use when deriving word meaning from written context. Therefore, there is only limited knowledge of how strategy teaching can fit in their natural routines. Another issue is how to encourage children to actually use these strategies during their normal reading. He found out in his studies that primary school children employ many activities and encounter several problems in deriving the meaning of conceptually challenging words from a written context. He asked second, fourth, and sixth grade students to think aloud when reading a text that included words that they were supposed to explain. Students from all grades were able to infer one or more meanings. They also checked them and evaluated their success in explaining the words. Fukkink suggests that young children can use advanced strategies even without instruction. Their natural process seems flexible. Therefore, teaching them complex strategies may not be the best approach.

Method The study was initiated with two goals in mind. One purpose was to investigate how pupils’ reading comprehension skills develop during the first six school years. The second goal was to determine if there are differences in the development of girls’ and boys’ reading comprehension skills. The study observed the development of the pupils in six classes in Southern Finland from the beginning of first grade, when the children were six or seven years old, to the end of the sixth grade. In the beginning of first grade in these classes there were 132 pupils – 73 girls and 59 boys. Four measurements were conducted during this period, and four tests were used as indicators. The initial test was divided into five sections; each of these sections consisted of several items. It was used to measure the pupils’ general level of school readiness, auditory and visual perception, mathematical skills and memory. The items in the general section were aimed at measuring the pupil’s understanding of number, of sentences and of phoneme-grapheme correspondence, his or her ability to continue phrases and lines of patterns, and his or her ability to find synonyms and rhyming words.

7

Phonological awareness is essential for learning to read and write. It was measured by, for instance, the following items of the initial test: finding rhyme words, in the general section of the test, and building words out of sounds and segmenting words into syllables, in the auditory section of the test. The success in tasks that require phonemic awareness is considered the most important prerequisite for learning to read. In the auditory section of the initial test such tasks were, for instance, the identification of the first or last phoneme in a word, and the phoneme synthesis task. Cronbach’s alpha, a measure of internal consistency, was for the initial test .88. The pupils’ reading skills were measured at the end of the first, the second and the sixth school year. The reading tests for first and second grade included reading aloud, soundlessly and reading comprehension tasks, and the reading test for sixth grade measured pupils’ ability to comprehend fiction and non-fiction texts.

Results FIRST GRADE After the first school year there proved to be great differences in pupils’ reading skills. In the reading aloud test the time that the pupils needed for the reading task varied from 50 seconds to 350 seconds, and while reading they made 0–29 errors. In the reading comprehension test the pupils’ scores varied from six to eighteen. The correlation between the pupils’ speed of reading and reading comprehension was significant (r= .55, p= .000). Those pupils who were fluent readers were also best comprehenders. As was mentioned above, non-fluent reading demands mental capacity that would be needed for comprehension process. Because the boys were significantly less fluent readers than the girls (t= -3.23, p= .002) it was predictable that they also performed less successfully in reading comprehension. In the reading comprehension test the pupils read a short story soundlessly. After that they did three tasks that measured their understanding of the text. Most pupils succeeded well in finding six given words from a text. About 80 per cent of them located all the words in the text. However, there were six pupils who found only half or less of the words. The second task was to fill gaps in six sentences with a word keeping in mind the events of the story. This was a more demanding task. Only 39 per cent of the pupils – 43 per cent of the girls and 34 per cent of the boys – filled all the gaps with appropriate words. In 23 per cent of the papers only half or less of the gaps were filled with suitable words. In the third part of the test the pupils were asked to continue sentences. This was the most difficult task. Five per cent of the pupils performed exceptionally poorly and could not complete a single sentence. Almost half of the boys managed to complete only half or less of the sentences. The girls were slightly more successful in the task than the boys. 8

SECOND GRADE After the second school year there were still great differences in the pupils’ reading skills. In the reading aloud test the pupils spent time in reading the text from 62 seconds to 290 seconds, and made 0–22 errors. The boys were still less fluent readers than the girls but the difference between the two genders was not significant. In the reading comprehension test finding words from the text was easier than in first grade, and 75 per cent of the pupils found them all. In filling the gaps with words 64 per cent of the pupils made no mistakes. As in first grade continuing sentences was more difficult. Only 40 per cent of the pupils completed the sentences correctly. In addition, in second grade the pupils were asked six questions about the story. Half of the pupils answered all questions acceptably. However, there were three pupils who gave only one correct answer. In the second grade test there were also tasks that measured pupils’ vocabulary. The pupils were asked to find from the story synonyms for six words and opposites for six words. It was mentioned above that vocabulary differences affect comprehension, and the results of this study support that. The correlation between the scores of reading comprehension and the scores of finding synonyms and opposites correlate significantly (r= .64, p= .000). Although the correlation between reading fluency and finding synonyms and opposites was not as high (r= .46, p= .000), one has to keep in mind that children who were non-fluent readers may not have had energy enough to read the text again and search synonyms and opposites. Furthermore, the results of the study showed that the level of a school starter’s school readiness had a strong effect on his or her reading comprehension skills still in second grade. This was clearly apparent when the results of the second grade reading comprehension test were compared with the results of the initial test. The initial test as a whole was a good predictor of the development of pupils’ reading comprehension skills. There was a strong correlation between the performance in the initial test and in the second grade reading comprehension test (r= .65, p= .000, r2= .43). According to the results of a stepwise regression analysis, the auditory section of the initial test was the best predictor of the pupil’s reading comprehension skills at the end of second grade (r= .56, p= .000). However, as the R Square (r2= .31) shows, in second grade it accounted only for 31 per cent of the variance.

SIXTH GRADE At the end of the sixth school year the pupils’ reading comprehension skills were measured with a test that consisted of three different texts. At that time there were 132 pupils (64 boys and 68 girls; aged 12–13) who took part in the study. About 60 per cent of them were the same pupils who took part in the initial test six years earlier. The first text was a story, the second text was a newspaper 9

article, and the third text a non-fiction text. Cronbach’s alpha, a measure of internal consistency for the test was .85. The first text was a two pages (A4) long story written by H.C. Andersen. The title of the story was ‘What the old man does is always right’. It is a story about a man who goes to the market place to sell his horse. On the way to town he meets several people and does many exchanges with them. Firstly he gives his horse to a farmer who in exchange gives him his cow. After that he changes the cow to a sheep, the sheep to a goose, and the goose to a hen. When he arrives to an inn he does his last exchange and changes the hen to a sack full of rotten apples. In the inn he meets three rich Englishmen. When they hear about the silly exchanges that the man has done they are certain that his wife will be angry when he comes home with a bag of rotten apples. However, the man says that he knows that his wife will be happy and will reward him with a kiss. The Englishmen say that if this happens they will give the man a basket full of gold coins. At the happy end of the story the man gets the kiss and the golden coins. After reading the story the pupils answered 20 questions. To some questions it was easy to find answers in the text. For instance, How did the story end? and To whom did the man tell his story? However, for instance, 35 per cent of the boys could not answer the question Why did they want to change the horse to some other animal? correctly. It is possible that after reading the story several pupils had already forgotten the reason because it was mentioned at the beginning of the story. Especially many boys did not want to reread the story and invented an incorrect explanation. Nevertheless, 85 per cent of the girls had found the right answer. Another question to what the boys significantly more often than the girls (t= -2.67, p= .000) gave an incorrect answer was What happened to the apples in the inn? It was a small detail in the story, and only nine per cent of the boys gave the correct answer: They were near the fire and started to cook. It seemed that the girls were more often than the boys willing to read the story again in order to find the correct answers. There were questions that proved to be rather difficult, such as What were the main characters like? and What is your opinion: why was the wife not angry? Also to these questions the girls gave significantly more often than the boys correct answers (t= -3.20, p= .000; t= -3.25, p= .000). It appears that the girls were more sensitive and understood the characters’ personalities and feelings more easily than the boys. This was apparent also two years earlier when the same pupils took part in another study by the author. Then the focus of the study was on the pupils’ creative writing skills. The girls described more often the characters and their personality in their stories than the boys. In addition, they also paid attention to the surroundings in which the characters lived. Similarly, in sixth grade the girls could more easily answer the question What kind of place did the couple live in? The boys seemed to focus their attention more to the events of the story.

10

The most difficult questions to answer were those in which the pupils had to put together more than one peace of information from the text. A very demanding question proved to be the question about the title (‘What the old man does is always right’). The pupils were asked: What is the title based on? To be able to answer this question a reader needed inferential reasoning. That is, he or she had to make connections that are not clearly expressed in the text – read between lines (Coiro & Dobler, 2007). Although most pupils understood that the man did very foolish exchanges, many of them accepted the title as a fact. Only 30 per cent of them mentioned that it was only an expression that his wife used or that it was her opinion about the matter. They understood that the wife loved her husband and did not want to criticise his actions. They wrote that she did not like to argue with him or that she was a positive, sweet person who was always able to find something good in her husband’s actions. Explanations of this kind were more common in the girls’ (38 %) than in the boys’ (22 %) papers. In addition, the girls proved to have heard or read more Andersen’s stories than the boys. They could significantly more often produce a title of another story by Andersen. The difference between the two genders was significant (t= -2.65, p= .009). The aggregated scores of the fiction section of the test varied from 6–20. The girls’ scores were significantly higher than the boys’ scores (t= -4.15, p= .000), and the four pupils who managed to answer all the questions correctly were all girls. In fourth grade also the pupils’ reading attitudes were measured (Merisuo-Storm, 2006). Then the girls proved to be significantly more motivated to read books (t= -2.77, p= .006) and to visit a library (t= -2.80, p= .006) than the boys. At that time most of the boys were especially reluctant to read poems but also stories and fairytales were unpopular amongst them. Consequently, one can assume that in sixth grade the boys were not as motivated as the girls to read texts like this story by Andersen and in addition, caused by the lack of experience they had more difficulties in understanding it beyond the story events. The second text was a newspaper article about H.C. Andersen. The title of the article was ‘In the large footsteps of H.C. Andersen’. The article described the H.C. Andersen route in Copenhagen where people could see and hear in several places something connected to Andersen’s life or work. The route was a part of the H.C. Andersen’s bicentenary celebrations. After reading the article the pupils’ task was to answer ten questions and explain the meaning of ten words picked from the text. There were no significant difference in the aggregated scores of the boys and the girls in this section of the test although 40 per cent of the girls’ marks were excellent and only 23 per cent of the boys succeeded equally well. However, ten per cent of the girls and only three per cent of the boys had poor marks. The answers to most of the ten questions could be found in the article. For instance, the pupils were asked What name has been given to H.C. Andersen? and What was Danish money called when 11

Andersen lived?. It proved to be more difficult for the boys than for the girls to find small details in the article. It was apparent that the girls put more effort in searching them. One example was the question: How many stops there are along the route?. The number 62 was remarkably more often found in the girls’ papers. Nevertheless, the easiest questions were those, which could be answered with a single word. However, to be able to answer some questions the pupils had to put together several different peaces of information from the text, which was a more demanding task. For instance, the correct answer to the question What kinds of texts one can hear on the route? included four different kinds of texts. The last question, Which two things does the expression ‘in the large footsteps’ in the title of the article refer to? was even more demanding. It was mentioned in the text that Andersen’s shoe size was 47. In addition, the pupils had to understand that the phrase in question could also refer to the fact that Andersen was a great author, who is famous all over the world. Consequently, that question proved to be the most difficult one in the test. Only 28 per cent of the pupils (32 per cent of the boys and 25 per cent of the girls) produced the right answer. This was the only question to which the boys managed to give the correct answer more often than the girls. The words that the pupils had to explain could be found in the article. It was stressed that they should write what the words meant in this particular text. Only three pupils, one boy and two girls, could explain all the words correctly. In this section of the test 22 per cent of the girls’ marks and only three per cent of the boys’ marks were excellent. However, eleven per cent of the girls and eight per cent of the boys had poor marks. Some words proved to be especially difficult. As was mentioned above, deriving the meaning of an unknown word from the written context is a demanding task. In addition, the difficulty of deriving the meaning of a word is connected to the concreteness of the unfamiliar concept. (Fukkink, 2005.) It is also obvious that some pupils could not take the advantage of the clues in the text. They gave an explanation that the word has in some other context. The most difficult word appeared to be välähdys. In this text it means ‘a glimpse of something’ but in a different context the same word means ‘a flash’ or ‘a gleam’. Although the word refers to the previous sentence, several pupils (30 per cent of the boys and 35 per cent of the girls) did not pay attention to the context and gave a false explanation. It seems that the concrete meaning of the word was so strong in the pupils’ minds that they ignored the clues in the text. Another word that has a concrete as well as an abstract meaning is the verb valaista. It can be translated to English ‘light up’ and ‘enlighten’ or ‘illustrate’. The word was used in a sentence: ‘... their purpose was to illustrate the history of Copenhagen.’ Again many pupils produced the concrete meaning of the word although in this context the word was used in its abstract meaning.

12

The word kohde can be translated to English as ‘a target’ but in this text it meant a place where people could see or hear something connected to Andersen’s life or work. As was mentioned above, the pupils were asked how many such places there were along the route. In addition, the word appeared twice in the text. However, about 50 per cent of the pupils did not connect the word to the text and the question and gave another explanation. The word kortteeri is an old expression that one does not often hear nowadays. It means the rooms that someone lives in. Although the text provided explicit clues, only 35 % of the pupils could derive the meaning of this word from the text. It seems that although the instructions stressed that the pupils should write what the words meant in this particular text many of them did not want to look for the words from the text and gave the explanation that was familiar to them. As was suggested above, good comprehenders need an extensive vocabulary. There was a strong correlation between the success in explaining the words and answering the questions connected to the newspaper article (r= .54, p= .000). In addition, there was a strong correlation between the results of the explaining the words section of the test and the results in the fiction text section (r= 51, p= .000). Although it seemed that the boys’ performance in the newspaper article section was slightly better than in the story section, the correlation (r= .51, p= .000) between the two sections shows that the same pupils had good scores in both sections. Furthermore, the results of the second grade reading comprehension test correlated significantly with the newspaper article section (r= .59. p= .000) of the sixth grade comprehension test. It may suggest that several pupils who had difficulties in reading comprehension in the end of second grade were still struggling comprehenders in sixth grade. The third text was a one-page-long (A4) non-fiction text about orang-utans and their living conditions in today’s world. Pupils were asked to read the text and then write a five-sentence summary of it. In the instructions it was stressed that the summary should include the most essential ideas of the text. This part of the test proved to be the most difficult. Although the difference in the boys’ and the girls’ results was not significant the girls performed better than boys also in summarising. As eleven per cent of the girls and only three per cent of the boys had excellent marks, ten per cent of the boys and only two per cent of the girls had poor marks. Many pupils could not determine what were the main ideas in the text. Some of them repeated all the facts, although how trivial, from the text. Consequently, their sentences were very long and included many ideas – or there were more than five sentences in their papers. Some pupils concentrated in the first two passages of the text and described in several sentences what orangutans look like and what they eat. They had not been able to combine similar ideas in one sentence. In some papers there were same ideas repeated in several sentences. Although, the most important ideas in the text were that the orang-utans are an endangered species and that we should protect 13

their environment, some pupils did not mention them in their summary. The results are consistent with Diehl’s (2005) findings: Many students struggle with determining the main ideas and themes of the text as well as combining similar ideas, and synthesising them into a coherent whole. Often they just repeat most of the text or give a very vague statement. It seemed that the pupils had not had practice enough in summarising and therefore many of them failed in doing it. Several pupils who had performed considerable well in previous sections of the test had succeeded poorly in summarising the text. Consequently, while the correlation between the results of the story text comprehension and the newspaper text comprehension section was strong (r= .51, p= .000), the correlations between them and the results of the summarising section (r= .34, p= .000; r= .39, p= . 000) were not as high. It was obvious that the pupils had more experience in answering questions than in finding main ideas in a text. The results support the findings of Parker and Hurry (2007, p. 311–313) who argue that many teachers do not have explicit knowledge of the key reading comprehension strategies. Direct oral questioning seems to be the dominant strategy for teaching reading comprehension. The pupils scores in sixth grade were also compared with the scores they had in the initial test. Although at the end of the sixth school year only 60 per cent of the pupils were the same pupils as the pupils at the beginning of the first school year, the results suggest that the level of a school starter’s school readiness still in many cases had a notable effect on his or her reading comprehension skills in sixth grade. There was a considerable correlation between the performance in the initial test and in the sixth grade reading comprehension test (r= .55, p= .000, r2= .30). According to the results of a stepwise regression analysis, the general and the auditory section of the initial test were the best predictors of the pupil’s reading comprehension skills at the end of sixth grade (r= .51, p= .000). However, as the R Square (r2= .26) shows, in sixth grade they accounted only for 26 per cent of the variance.

Conclusion The results of the study show that during the first two school years there were great differences in the pupils’ reading fluency, reading comprehension skills, and vocabulary. Furthermore, it seemed that the level of a school starter’s school readiness had a strong effect on his or her reading skills still after two school years. After six years in school some pupils still had great difficulties in comprehending different texts. The results also suggest that several pupils who had poor reading comprehension skills in second grade were still struggling comprehenders at the end of sixth grade. However, next autumn they moved to secondary school and there they need good reading comprehension skills to be able to acquire new information independently. 14

In sixth grade the girls succeeded in every reading comprehension task better than the boys. Their aggregated scores of the fiction section were significantly higher than the boys’ scores, and the four pupils who managed to answer all the questions correctly were all girls. In the newspaper article section of the test 40 per cent of the girls’ marks were excellent and only 23 per cent of the boys succeeded equally well. In addition, the girls’ scores in the explaining the meaning of the words task were higher than the boys’ scores. The results of the fourth grade reading attitude test had shown that the girls read more than the boys. As was mentioned above, increasing reading practice can increase vocabulary. A pupil who is an eager reader encounters more words than a pupil who reads less. In addition, regular reading is likely to lead to more efficient access of word meanings. Furthermore, the texts the girls liked to read represented a wider range of genres than the texts that the boys were interested in. Consequently they were in sixth grade more experienced in reading different texts than the boys. It was also very interesting to notice how different readers the boys and the girls were. It appeared that when reading a story the girls were more sensitive and understood the characters’ personalities and feelings more easily than the boys. In addition, they also paid attention to the surroundings in which the story characters lived. The boys seemed to focus their attention more to the events of the story. Although the girls, as a group, succeeded better than the boys in all the tests there were also girls whose marks were poor. Especially these girls and several boys need support in learning reading comprehension strategies. In first and second grade a lot of time is spent in teaching children to read and write. It is also essential to start to teach reading comprehension strategies as early as possible. In addition, a considerable time should be spent in teaching vocabulary. Moreover, the pupils need to be taught how the text context helps to understand the meaning of an unfamiliar word. They should learn what kind of clues a text can provide and how to find those clues. The results of the study showed that deriving the meaning of an unknown word from the written context is very difficult. It seems that the pupils had not had enough practice in it because they often produced a meaning for a word that it has in some other context. In addition, more time should be spent in summarising texts. Summarising is a demanding task and needs a lot of practice. Many pupils could not discriminate the main ideas and the trivial ideas in the text. Several pupils who had performed considerable well in other sections of the test had succeeded poorly in summarising the text. However, in school the pupils are, after reading a text, frequently expected to remember main ideas and concepts from it. Also comprehending is easier if a pupil is able to find the main ideas from the text.

15

References Barton, J., & Sawyer, D.M. (2003). Our students are ready for this: Comprehension instruction in the elementary school. The Reading Teacher, 57 (4), 334–347. Best, R.M., Floyd, R.G., & McNamara, D.S. (2008). Differential competences contributing to children’s comprehension of narrative and expository texts. Reading Psychology, 29 (2), 137–164. Biemiller, A. (2003). Vocabulary: Needed if more children are to read well. Reading Psychology, 24 (3/4), 323–335. Bimmel, P., & van Schooten, E. (2004). The relationship between strategic reading activities and reading comprehension. L1 – Educational Studies in Language and Literature 4 (1), 85–102. Blair-Larsen, S.M., & Vallance, K.M. (2004). Comprehension instruction in a balanced reading classroom. In S.M. Blair-Larsen & K.A. Williams (Eds.), The Balanced Reading Program: Helping All Students Achieve Success (pp.37-52). Newark, DE: International Reading Association. Bowyer-Grane, C., & Snowling, M.J. (2005). Assessing children’s inference generation: What do tests of reading comprehension measure? British Journal of Educational Psychology, 75 (2), 189–201. Cain, K., Oakhill, J., & Lemmon, K. (2004). Individual differences in the inference of word meanings from context: The influence of reading comprehension, vocabulary knowledge, and memory capacity. Journal of Educational Psychology, 96 (4), 671–681. Coiro, J., & Dobler, E. (2007). Exploring the online reading comprehension strategies used by sixth-grade skilled readers to search for and locate information on the Internet. Reading Research, Quarterly, 42 (2), 214-257. Diehl, H.L. (2005). Snapshots of our journey to thoughtful literacy. The Reading Teacher, 59 (1), 56–69. Dougherty Stahl, K.A. (2004). Proof, practice, and promise: Comprehension strategy instruction in the primary grades. The Reading Teacher, 57 (7), 598–609. Fukkink, R.G. (2005). Deriving word meaning from written context: a process analysis. Learning and Instruction, 15 (1), 23–43. Goerss, B.L., Beck, I.L., & McKeown, M.G. (1999). Increasing remedial students’ ability to derive word meaning from context. Reading Psychology, 20 (2), 151–175. 16

Hurry, J., & Doctor, E. (2007). Assessing literacy in children and adolescents. Child and Adolescent Mental Health, 12 (1), 38–45. McGee, A. & Johnson, H. (2003). The effect of inference training on skilled and less skilled comprehenders. Educational Psychology, 23 (1), 49–59. McLaughlin, M. (2006). Guided comprehension in the primary grades. Newark, DE: International Reading Association. Merisuo-Storm, T. (2006). Girls and boys like to read and write different texts. Scandinavian Journal of Educational Research, 50 (2), 111–125. Pardo, L.S. (2004). What every teacher needs to know about comprehension. The Reading Teacher, 58 (3), 272–280. Parker, M., & Hurry, J. (2007). Teachers’ use of questioning and modelling comprehension skills in primary classrooms. Educational Review, 59 (3), 299–314. Pressley. M. (2002). Metacognition and self-regulated comprehension. In A.E. Farstrup & S.J. Samuels (Eds.), What Research Has to Say About Reading Instruction (3rd ed.) (pp. 291–309). Newark: International Reading Association. Reutzel, D.R., Smith, J.A., & Fawson, P.C. (2005). An evaluation of two approaches for teaching reading comprehension strategies in the primary years using science information texts. Early Childhood Research Quarterly, 20, 276–305. Scharer, P.L., Pinnell, G.S., Lyons, C., & Fountas, I. (2005). Becoming an engaged reader. Educational Leadership, 63 (2), 24–26. Snow,C. , Sweet, A.P., Alvermann, D.E., Kamil, M.L., & Stricland, D.S. (2002). Formulating a research agenda about reading for understanding. In A.M. Roller (Ed.), Comprehensive Reading Instruction Across the Grade Levels. A collection of papers from the Reading Research 2001 Conference (pp. 88–110). Newark, DE: International Reading Association. Swanson, P.N., & De La Paz, S. (1998). Teaching effective comprehension strategies to students with learning and reading disabilities. Intervention in School & Clinic, 33 (4), 209–218. Van Daalen-Kapteijns, M., Elshout-Mohr, M., & de Glopper, K. (2001). Deriving the meaning of unknown words from multiple contexts. Language Learning, 51 (1), 145–181. Van der Schoot, M., Vasbinder, A.L., Horsley, T.M., & van Lieshout, E.C.D.M. (2008). The role of two reading strategies in text comprehension: An eye fixation study in primary school children. Journal of Research in Reading, 31 (2), 203–223.

17

Williams, J.P. (2005). Instruction in reading comprehension for primary-grade students: A focus on text structure. The Journal of Special Education, 39 (1), 6–18. Williams, J.P., Hall, K.M., Lauer, K.D., Stafford, K.B., DeSisto, L.A., & deCani, J.S. (2005). Expositary text comprehension in the primary grade classroom. Journal of Educational Psychology, 97 (84), 538–550.

18