Error Analysis: Activity 2
Classify errors in learner language
Watch part of the comparison task while you read the corresponding transcripts.
- Mark all the errors you hear on a photocopy of the transcript.
- Fill out the table below for Sebastian’s and Maximus’ errors.
-- In Column 1, indicate the line number where you found the error.
-- In Column 2, copy the error and the words that immediately preceded and followed the error, and make a note on whether it is an error of phonology, morphology, syntax, or vocabulary.
-- In Column 3, indicate your reconstruction(s) of the learner’s erroneous utterance. Try to list more than one possible reconstruction.
-- In Column 4, indicate possible causes of the error. To find out if the cause might be NL transfer, you may need to consult a resource providing contrastive analyses.
- Now reflect on what you see in your data analysis by considering the following questions:
-- Compare the learners’ errors. Did one of these learners seem to make more errors (in terms of either types or tokens) than the other, or make different types of errors than the other? Explain.
-- What psycholinguistic processes might have caused their errors? Where their errors were different, was this because of the influence of NLs, or for some other reason? Where their errors were the same, why do you think this was?
-- Is number of errors a good measure of proficiency? How about type of error? -- What did you learn from this error analysis that you might be able to use if you were Sebastian’s or Maximus’ teacher?
Line number | Phrase with error (phonology, morphology, syntax, vocabulary) |
Target language reformulation(s) |
Cause of error |
|---|---|---|---|
|
|
|
|
Maximus
Line number |
Phrase with error (phonology, morphology, syntax, vocabulary) |
Target language reformulation(s) |
Cause of error |
|---|---|---|---|
|
|
|
|
Comparison Task:
Please type your answers to the questions in the box below.
When you have finished typing your answer, click to compare your response with the Learner Language staff response.
2.
SebastianLine number |
Phrase with error (phonology, morphology, syntax, vocabulary) |
Target language reformulation(s) |
Cause of error |
|---|---|---|---|
8 |
migi no shashin wa |
- migi no shashin no ie wa |
A performance error (a slip of the tongue) |
8 |
tabun chotto atarashi soo desu ne (lexicon: “tabun [may be/probably] with the suffix “soo [seem/look/sound like]” does not make sense in this context) |
- tabun atarashii desu ne |
A performance error (a slip of the tongue) |
13 |
garaaji (lexicon) |
gareeji |
NL borrowing (NL lexicon, TL phonology) |
14 |
migi shashin (morphology: no particles) |
migi no shashin |
- NL Transfer: There are no particles and no equivalent syntactic structures in English, or |
25 |
naze wa (morphology: no particles) |
- naze ka wa |
- NL transfer: There are no particles in English and thus, the interrogative is used with no particle, or |
28 |
cháneru (phonology) |
channeru |
NL transfer |
29-30 |
mado no futatsu ga arimasu kara (morpho-syntax: wrong particle and unnecessary particle) |
mado ga futatsu arimasu kara |
Developmental, not NL transfer since there are no equivalent syntactic structures in English. This error may be systematic (see the same type of error below in line 36) |
30 |
kazoku no hoo ga (morphology: unnecessary particle or lack of demonstrative pronoun) |
- kazoku ga |
A performance error (there are instances where he correctly uses a combination of demonstrative pronoun + noun) |
36 |
doa no yottsu kara (morpho-syntax: wrong particle and no predicates(e.g., verb or auxiliary verb) |
- doa ga yottsu da kara |
Developmental, not NL transfer since there are no equivalent syntactic structures in English. This error may be systematic (see the same type of error above in lines 29-30) |
46 |
higashi toka nishi arimasu ne (morphology: no particles) |
- higashi toka nishi toka ni arimasune |
Developmental, possibly overgeneralization of particle omission typical in informal conversations, or A performance error |
48 |
roohoo, roohoo shashin wa (phonology and syntax) |
- kono/korerano shashin wa ryoohoo tomo |
- NL transfer (e.g., both pictures) and/or |
Maximus
Line number |
Phrase with error (phonology, morphology, syntax, vocabulary) |
Target language reformulation(s) |
Cause of error |
|---|---|---|---|
5 |
onaji toshi gurai (lexicon: word choice -- ‘toshi (age)’ can not be used for inanimate objects in Japanese) |
- dochira mo furusa wa onaji gurai |
NL transfer since ‘age’ can be used for inanimate objects such as ‘house.’ |
5 |
kitanaide: |
- kitanaishi |
Developmental: |
11 |
tsuna (lexicon: wrong word) |
- kusa |
Developmental |
11 |
tsuite (lexicon: word choice) |
haete ite |
Developmental |
17 |
America ni yoru to (lexicon: word choice) |
- America dewa |
Developmental |
17 |
furui machi ni (morphology: wrong particle) |
furui machi wa |
Developmental |
18 |
ookiide: |
- ookiishi |
Developmental: |
18 |
old grove suburb (lexicon) |
furuku kara aru koogai |
NL Borrowing |
18 |
kono hoo (lexicon: word choice? NS may use this phrase) |
- kocchi wa |
Developmental |
19 |
sonna ni ookii janai desu ne (morphology: negation for i-adjectives) |
sonna ni ookiku nai desu ne |
Developmental: Overgeneralization of the negative form for na-adjective |
20 |
sonnani kitanai janai kara (morphology: negation for i-adjectives) |
sonna ni kitanaku nai kara |
Developmental: Overgeneralization of the negative form for na-adjective |
27 |
terebi ga daisuki kara (morphology: missing the auxiliary verb ‘da’ for na-adjectives) |
terebi ga daisuki da kara |
Developmental: |
33 |
16 sai koukousei ga (morphology: no particles) |
16 sai no koukousei ga |
NL transfer: |
34 |
ii kuruma kedo (morphology: missing the auxiliary verb ‘da’ with nouns) |
ii kuruma da kedo |
Developmental? |
42 |
Aribama (phonology) |
Arabama |
NL? |
3.
-- Compare the learners’ errors. Did one of these learners seem to make more errors (in terms of either types or tokens) than the other, or make different types of errors than the other? Explain.
The learners both seem to make roughly the same number of errors. Maximus makes slightly more errors than Sebastian (note that a few of Sebastian’s errors are performance errors), but then he produces longer sentences. Both Sebastian and Maximus make similar kinds of errors -- lexical and morphological. More specifically, Sebastian consistently has issues with particles while Maximus has issues with different forms including i-adjective conjugations, particles, and auxiliary verbs.
-- Where their errors were different or the same, why do you think this was?
Since both Sebastian and Maximus speak English as their first language, similar errors they make could be due to NL transfer. Dropping particles is one salient example in the selected excerpt. Differences in specific types of errors such as Maximus’ errors on i-adjectives may be due to multiple factors. As we discussed in the learners section, their individual personalities and backgrounds may have come into play. For instance, as activity 2 in the learners section reveals, Sebastian spoke positively about his Japanese classes while Maximus said he learned the language most in the informal setting in Japan. Although both took the same Japanese courses that seem to focus on accuracy, the learners’ preferences as to the way they learn Japanese might have resulted in their differences in grammatical accuracy. In fact, as we examine in the next activity, Maximus systematically makes errors on i-adjectives across the different tasks. Sebastian, on the other hand, produces the forms more correctly across the tasks (one of his few errors include omission of ‘da’).
-- Is number of errors a good measure of proficiency? How about type of error?
In this task, Maximus takes the initiative and mostly leads the conversation. Your impression may be that he is more fluent than Sebastian. In fact, his turns are generally longer than Sebastian’s although Sebastian speaks about the same amount overall. Maximus also helps Sebastian with vocabulary, helps him complete sentences. If you initially felt that Sebastian makes more or the equal amount of errors, your error analysis may have contradicted that impression. As we discussed in the “learners” section, the number of errors (i.e., accuracy) as well as the number of complex sentences (i.e., complexity) and fluency need to be considered to measure one’s proficiency.
-- What did you learn from this error analysis that you might be able to use if you were Sebastian’s or Maximus’ teacher?
The error analysis might have helped you see certain errors that initially you did not notice, and that could benefit from some instructional input. In addition, in doing this activity, you may also have identified potential causes of these errors that you did not consider earlier.
