High Dictation, Low Comprehension in Kannada
In a Kannada-medium government high school in Mysuru district, a Class 8 Kannada teacher has 42 students and six 45-minute periods a week. A mid-term oral reading check shows 35 students reading a Class 8 passage aloud at about 60 correct words per minute with few decoding errors; a 20-word dictation gives a class average of 17. But on an unseen 80-word passage, 34 of 42 answers were lifted verbatim from the text, only 3 students could say why a character behaved as she did, and 30 could not infer the meaning of an unfamiliar word from its sentence. The head teacher, pointing to the SA-1 blueprint where grammar carries 20 marks and comprehension 15, wants the next three weeks spent on vyakarana drills and dictation. Twelve students speak Urdu at home, and the school owns only one set of the supplementary reader for 42 students.
1. Diagnose what is and is not the problem, and say what a single false inference from these data would be.
Read the worked answer
The data separate two components that the school is treating as one.
Word recognition and fluency are not the bottleneck: 35 of 42 read at roughly 60 correct words per minute with few errors, and dictation is strong at 17 of 20. Comprehension is the bottleneck: 34 of 42 copied verbatim, only 3 could infer motive, and 30 could not use context for an unknown word. The weakness is at the level of vocabulary depth, inference and monitoring of meaning.
High dictation scores are therefore not evidence of reading ability; dictation tests the encoding of known words. Vyakarana drills label forms, they do not build meaning. The false inference to avoid is that strong spelling and fluent reading mean the class can comprehend, so more of the same practice will fix it; more drill will raise the same scores and leave comprehension untouched.
One caveat before grouping: check whether the 12 Urdu-home-language students were included in the oral reading sample, and re-ask one inferential question orally to ten students. Poorly worded questions also produce copying, so confirm the failure is in meaning, not in understanding the question.
2. Allocate the next 18 periods between grammar and comprehension, and justify the split against the blueprint.
Read the worked answer
Proposed: 5 periods grammar, 11 comprehension, 2 assessment and feedback.
Rationale: grammar carries 20 marks, so it cannot be dropped, but grammar taught through the comprehension text earns both sets of marks. The week's passage supplies the sentences for identifying vibhakti or tense, and students then apply the same label to a fresh sentence. This is not less grammar coverage; it is grammar coverage that transfers.
Comprehension periods use gradual release: model the strategy aloud on one paragraph, then whole class, then pairs, then individual, on prediction, question generation and summarising. Because there is one supplementary reader for 42 students, write the text on the board or a chart, or use a jigsaw where each group holds one paragraph and reports it. The Urdu-home-language students get a bilingual glossary of eight to ten key words.
Assessment periods: a short unseen passage with one literal, one inferential and one evaluative question, marked on whether the answer is in the student's own words. Tell the head teacher the split in writing, with the diagnostic numbers attached; a coverage argument is answered with evidence, not assertion.
3. Design a three-item formative assessment that separates copying from inference.
Read the worked answer
Three items, each with a different demand:
- Literal: What did the farmer do first? The answer is in the text. This tests retrieval only, and copying earns the mark.
- Inferential, not copyable: The text does not say the boy was afraid. Which two words tell you he was? This requires selecting evidence and naming an inference.
- Evaluative and transferable: Would you have done what the boy did? Give one reason from the story and one from your own life. This requires a stance plus textual support.
Marking rule, stated in advance: a verbatim sentence scores full marks only on item 1. On items 2 and 3, credit the reason, not the length. Keep an oral version for students whose writing is slow. Record the class pattern (how many move from item 1 to 2 to 3), not only a total; the pattern is what tells you whether the teaching worked, and it gives the head teacher a number that is about comprehension rather than coverage.