Assessing Children’s Inference Generation:
What Do Tests of Reading Comprehension Measure?*
What Do Tests of Reading Comprehension Measure?*
Background. Previous research suggests that children with specific comprehension difficulties have problems with the generation of inferences. This raises important questions as to whether poor comprehenders have poor comprehension skills generally, or whether their problems are confined to specific inference types.
Aims. The main aims of the study were (a) using two commonly used tests of reading comprehension to classify the questions requiring the generation of inferences, and (b) to investigate the relative performance of skilled and less-skilled comprehenders on questions tapping different inference types.
Sample. The performance of 10 poor comprehenders (mean age 110.06 months) was compared with the performance of I0 normal readers (mean age 112.78 months) on two tests of reading comprehension.
Method. A qualitative analysis of the NARA II (form I) and the WORD comprehension subtest was carried out. Participants were then administered the NARA II, WORD comprehension subtest and a test of non-word reading.
Results. The NARA II was heavily reliant on the generation of knowledge-based inferences, while the WORD comprehension subtest was biased towards the retention of literal information. Children identified by the NARA II as having comprehension difficulties performed in the normal range on the WORD comprehension subtests. Further, children with comprehension difficulties performed poorly on questions requiring the generation of knowledge-based and elaborative inferences. However, they were able to answer questions requiring attention to literal information or use of cohesive devices at a level comparable to normal readers.
The percentage of the questions tapping each of the six inference types is shown for each test in Fig. 1. In total, there were 44 questions in the NARA II, and 38 questions in the WORD. The results are based on the mean number of questions classified into each category by the seven raters. The mean inter-rater reliability was acceptable at .84 for the NARA II and .80 for the WORD tests. These figures were arrived at by correlating the mean rating across judges for each item with the ratings assigned by the first author.
Aims. The main aims of the study were (a) using two commonly used tests of reading comprehension to classify the questions requiring the generation of inferences, and (b) to investigate the relative performance of skilled and less-skilled comprehenders on questions tapping different inference types.
Sample. The performance of 10 poor comprehenders (mean age 110.06 months) was compared with the performance of I0 normal readers (mean age 112.78 months) on two tests of reading comprehension.
Method. A qualitative analysis of the NARA II (form I) and the WORD comprehension subtest was carried out. Participants were then administered the NARA II, WORD comprehension subtest and a test of non-word reading.
Results. The NARA II was heavily reliant on the generation of knowledge-based inferences, while the WORD comprehension subtest was biased towards the retention of literal information. Children identified by the NARA II as having comprehension difficulties performed in the normal range on the WORD comprehension subtests. Further, children with comprehension difficulties performed poorly on questions requiring the generation of knowledge-based and elaborative inferences. However, they were able to answer questions requiring attention to literal information or use of cohesive devices at a level comparable to normal readers.
The percentage of the questions tapping each of the six inference types is shown for each test in Fig. 1. In total, there were 44 questions in the NARA II, and 38 questions in the WORD. The results are based on the mean number of questions classified into each category by the seven raters. The mean inter-rater reliability was acceptable at .84 for the NARA II and .80 for the WORD tests. These figures were arrived at by correlating the mean rating across judges for each item with the ratings assigned by the first author.
Both tests required the reader to make a high proportion of inferences in order to answer the questions posed. However, this was particularly true in the case of the NARA II in which 14% of the questions could be answered on the basis of literal information provided in the text. In comparison, almost a third of the questions following the WORD could be answered in this way. Another notable difference between the tests was the relative proportions of elaborative, cohesive, and knowledge-based inferences. It is clear that the NARA II taps the ability to make cohesive and knowledge-based inferences more than the WORD does, perhaps because comprehension of the narrative structure requires the development of integrated text-level representations. In contrast, the WORD questions demand more use of elaborative inferences that are appropriate following expository texts.
Conclusions. Different reading tests tap different types of inferencing skills. Some components of this study reveal data that suggest less-skilled readers have particular difficulty applying real-world knowledge to a text during reading, and this has implications for the formulation of effective intervention. The findings reported here suggest that the reading comprehension difficulties experienced by some children may be mediated by a difficulty with specific inference types. Furthermore, practitioners need to be aware that a single test of comprehension may not be adequate to assess a child’s specific educational needs.
* Adapted from full article (DOI: 10.1348/000709904X22674):
Bowyer-Crane, Claudine, and Margaret J Snowling. “Assessing Children's Inference Generation: What Do Tests of Reading Comprehension Measure?” British Journal of Educational Psychology, vol. 75, no. 2, 2005, pp. 189–201*
This text has 40 sentences, with 599 words (15.36 per sentence) with 1.76 syllables per word.
Flesch-Kincaid Grade Level 11.2
Bowyer-Crane, Claudine, and Margaret J Snowling. “Assessing Children's Inference Generation: What Do Tests of Reading Comprehension Measure?” British Journal of Educational Psychology, vol. 75, no. 2, 2005, pp. 189–201*
This text has 40 sentences, with 599 words (15.36 per sentence) with 1.76 syllables per word.
Flesch-Kincaid Grade Level 11.2
Comprehension and Reading Skills Quiz
If taking the online version of this quiz, please click through using the button below. It will automatically score itself when you finish. If you miss two or more questions, you should consider taking a Reading Skills lab to enhance your reading and study skills proficiency.