Skip to main content
. 2025 Mar 26;75(2):313–338. doi: 10.1007/s11881-025-00326-1

Table 3.

Measures considered in each of the language tasks

Reading comprehension
Number of correct answers Participants were asked to answer four open-ended questions (two for each reading) by reformulating the information in the text—that is, without copying it verbatim. Each question was scored between 0 and 2: 0 if the answer was incorrect, 1 if the answer was correct but not reformulated, and 2 if the answer was correct and reformulated. Participants were warned that spelling errors would not be taken into account. The maximum score was 8 points
Number of times the text was consulted The screen recordings made it possible to note how often the participants scrolled up to consult the text again while answering the questions
Reading time Time (in seconds) needed to read the texts. The participant’s screen was recorded to calculate the time spent viewing the text (see the Procedure section)
Total task time Total time (in seconds) taken to read the text and answer the questions
Oral comprehension
Number of correct answers One point for each correct answer. The maximum score possible was 8 points
Number of times the audio was consulted Number of times the audio files were listened to (noted by the researcher)
Total task time Total time (in seconds) taken to complete the task (reading the questions, listening to the audios, and answering the questions)
Written production
Percentage of errors The number of errors in relation to the number of written words (see Table 10 in the Appendix). Initially, five major blocks of errors (spelling, grammatical, semantic, punctuation, and self-correction) were considered, which included different sub-types of error. The percentage of each subtype within the different categories was calculated. The different types of spelling error are letter substitution, letter omission/addition, missing/unnecessary space between words, letter inversion, errors resulting from a clear “spelling by ear,” and incorrect use of upper/lower case letters. Grammatical errors include incorrect verb tense, misuse of the -s in the third person singular, incorrect singular/plural agreement, incorrect word order, missing/unnecessary grammatical word, substitution of a grammatical word, incorrect derivation of words, and incorrect use of’s in possessives. Semantic errors are substitution of a content word, invention of a word, and missing/unnecessary content word
Mean length of utterances in words (MLU) A measure of syntactic complexity calculated by dividing the number of words by the number of utterances
Subordination index (SI) A measure of syntactic complexity that gives the ratio of the total number of clauses to the total number of c-units (independent clauses with their modifiers) (Miller & Iglesias, 2012)
Lexical diversity Guiraud index of lexical diversity (type/√token) was used to know how many different words appear in a text, controlling for differences in the length of the narratives
Writing productivity Number of words written per minute
Word types Percentage of determiners, adjectives, nouns, personal pronouns, other pronouns, auxiliary modals, auxiliary operators, verbs, copula forms, verb particles, adverbs, intensifiers, prepositions, existential there, coordinators, subordinators, infinitives, and negation words
Oral production
Percentage of errors Pronunciation errors were added to grammatical and semantic errors (see Table 10 in the Appendix)
Mean length of utterances in words (MLU) See written production
Subordination index (SI) See written production
Lexical diversity See written production
Oral productivity Number of words spoken per minute
Word types See written production
Percentage of mazes Series of words or unattached fragments which do not contribute to the meaning in the ongoing flow of language
Number of interjections per minute Any kind of filler with no linguistic content was considered an interjection
Pauses The number of pauses per minute, the average pause duration, and the percentage of pause time (silence time in relation to total speech duration) were measured. Any silence longer than 100 ms was considered a pause. Praat software was used to perform these analyses (Boersma & Weenink, 2024)