MODELING INDIVIDUAL DIFFERENCES AMONG WRITERS USING READERBENCH

Laura Allen; Mihai Dascalu; Danielle S. McNamara; Scott Crossly; Stefan Trausan-Matu

doi:10.21125/edulearn.2016.2241

MODELING INDIVIDUAL DIFFERENCES AMONG WRITERS USING READERBENCH

Laura Allen, Mihai Dascalu, Danielle S. McNamara, Scott Crossly, Stefan Trausan-Matu

Research output: Chapter in Book/Report/Conference proceeding › Conference Article in proceeding › Academic › peer-review

75 Downloads (Pure)

Abstract

The current study builds upon a previous study, which examined the degree to which the lexicalproperties of students’ essays could predict their vocabulary scores. We expand on this previousresearch by incorporating new natural language processing indices related to both the surface- anddiscourse-levels of students’ essays. Additionally, we investigate the degree to which these NLPindices can be used to account for variance in students’ reading comprehension skills. We calculatedlinguistic essay features using our framework, ReaderBench, which is an automated text analysis toolsthat calculates indices related to linguistic and rhetorical features of text. University students (n = 108)produced timed (25 minutes), argumentative essays, which were then analyzed by ReaderBench.Additionally, they completed the Gates-MacGinitie Vocabulary and Reading Comprehension tests.The results of this study indicated that two indices were able to account for 32.4% of the variance invocabulary scores and 31.6% of the variance in reading comprehension scores. Follow-up analysesrevealed that these models further improved when only considering essays that contained multipleparagraph (R2 values = .61 and .49, respectively). Overall, the results of the current study suggest thatnatural language processing techniques can help to inform models of individual differences amongstudent writers.

Original language	English
Title of host publication	EDULEARN16 Proceedings
Subtitle of host publication	8th International Conference on Education and New Learning Technologies
Editors	L. Gómez Chova, A. López Martínez, I. Candel Torres
Publisher	IATED Academy
Pages	5269-5279
ISBN (Print)	978-84-608-8860-4
DOIs	https://doi.org/10.21125/edulearn.2016.2241
Publication status	Published - 2016
Externally published	Yes
Event	EDULearn16: 8th International Conference on Education and New Learning Technologies - Barcelona, Spain Duration: 4 Jul 2016 → 6 Jul 2016 https://library.iated.org/publications/EDULEARN16

Conference

Conference	EDULearn16
Abbreviated title	EDULearn16
Country/Territory	Spain
City	Barcelona
Period	4/07/16 → 6/07/16
Internet address	https://library.iated.org/publications/EDULEARN16

Keywords

writing skills
automated writing evaluation
comprehension prediction
vocabulary measures
natural language processing

Access to Document

10.21125/edulearn.2016.2241

Allen, Dascalu e.a.Accepted author manuscript, 236 KBLicence: CC BY-NC-ND

Rage: Realising an Applied Gaming Eco-system
Westera, W. (PI), Georgiadis, K. (CoI), Saveski, G. (CoI), van Lankveld, G. (CoI), Bahreini, K. (CoI), van der Vegt, W. (CoI), Berkhout, J. (CoI), Nyamsuren, E. (CoI), Kluijfhout, E. (CoI) & Nadolski, R. (CoI)
European Union
1/02/15 → 31/07/19
Project: Research

Cite this

Allen, L., Dascalu, M., McNamara, D. S., Crossly, S., & Trausan-Matu, S. (2016). MODELING INDIVIDUAL DIFFERENCES AMONG WRITERS USING READERBENCH. In L. Gómez Chova, A. López Martínez, & I. Candel Torres (Eds.), EDULEARN16 Proceedings: 8th International Conference on Education and New Learning Technologies (pp. 5269-5279). IATED Academy. https://doi.org/10.21125/edulearn.2016.2241

@inproceedings{fa19252681c84094bb2407b6d4d40dea,

title = "MODELING INDIVIDUAL DIFFERENCES AMONG WRITERS USING READERBENCH",

abstract = "The current study builds upon a previous study, which examined the degree to which the lexicalproperties of students{\textquoteright} essays could predict their vocabulary scores. We expand on this previousresearch by incorporating new natural language processing indices related to both the surface- anddiscourse-levels of students{\textquoteright} essays. Additionally, we investigate the degree to which these NLPindices can be used to account for variance in students{\textquoteright} reading comprehension skills. We calculatedlinguistic essay features using our framework, ReaderBench, which is an automated text analysis toolsthat calculates indices related to linguistic and rhetorical features of text. University students (n = 108)produced timed (25 minutes), argumentative essays, which were then analyzed by ReaderBench.Additionally, they completed the Gates-MacGinitie Vocabulary and Reading Comprehension tests.The results of this study indicated that two indices were able to account for 32.4% of the variance invocabulary scores and 31.6% of the variance in reading comprehension scores. Follow-up analysesrevealed that these models further improved when only considering essays that contained multipleparagraph (R2 values = .61 and .49, respectively). Overall, the results of the current study suggest thatnatural language processing techniques can help to inform models of individual differences amongstudent writers.",

keywords = "writing skills, automated writing evaluation, comprehension prediction, vocabulary measures, natural language processing",

author = "Laura Allen and Mihai Dascalu and McNamara, {Danielle S.} and Scott Crossly and Stefan Trausan-Matu",

year = "2016",

doi = "10.21125/edulearn.2016.2241",

language = "English",

isbn = "978-84-608-8860-4",

pages = "5269--5279",

editor = "{G{\'o}mez Chova}, {L. } and {L{\'o}pez Mart{\'i}nez}, {A. } and {Candel Torres}, {I. }",

booktitle = "EDULEARN16 Proceedings",

publisher = "IATED Academy",

note = "EDULearn16 : 8th International Conference on Education and New Learning Technologies, EDULearn16 ; Conference date: 04-07-2016 Through 06-07-2016",

url = "https://library.iated.org/publications/EDULEARN16",

}

Allen, L, Dascalu, M, McNamara, DS, Crossly, S & Trausan-Matu, S 2016, MODELING INDIVIDUAL DIFFERENCES AMONG WRITERS USING READERBENCH. in L Gómez Chova, A López Martínez & I Candel Torres (eds), EDULEARN16 Proceedings: 8th International Conference on Education and New Learning Technologies. IATED Academy, pp. 5269-5279, EDULearn16, Barcelona, Spain, 4/07/16. https://doi.org/10.21125/edulearn.2016.2241

MODELING INDIVIDUAL DIFFERENCES AMONG WRITERS USING READERBENCH. / Allen, Laura; Dascalu, Mihai; McNamara, Danielle S. et al.
EDULEARN16 Proceedings: 8th International Conference on Education and New Learning Technologies. ed. / L. Gómez Chova; A. López Martínez; I. Candel Torres. IATED Academy, 2016. p. 5269-5279.

Research output: Chapter in Book/Report/Conference proceeding › Conference Article in proceeding › Academic › peer-review

TY - GEN

T1 - MODELING INDIVIDUAL DIFFERENCES AMONG WRITERS USING READERBENCH

AU - Allen, Laura

AU - Dascalu, Mihai

AU - McNamara, Danielle S.

AU - Crossly, Scott

AU - Trausan-Matu, Stefan

PY - 2016

Y1 - 2016

N2 - The current study builds upon a previous study, which examined the degree to which the lexicalproperties of students’ essays could predict their vocabulary scores. We expand on this previousresearch by incorporating new natural language processing indices related to both the surface- anddiscourse-levels of students’ essays. Additionally, we investigate the degree to which these NLPindices can be used to account for variance in students’ reading comprehension skills. We calculatedlinguistic essay features using our framework, ReaderBench, which is an automated text analysis toolsthat calculates indices related to linguistic and rhetorical features of text. University students (n = 108)produced timed (25 minutes), argumentative essays, which were then analyzed by ReaderBench.Additionally, they completed the Gates-MacGinitie Vocabulary and Reading Comprehension tests.The results of this study indicated that two indices were able to account for 32.4% of the variance invocabulary scores and 31.6% of the variance in reading comprehension scores. Follow-up analysesrevealed that these models further improved when only considering essays that contained multipleparagraph (R2 values = .61 and .49, respectively). Overall, the results of the current study suggest thatnatural language processing techniques can help to inform models of individual differences amongstudent writers.

AB - The current study builds upon a previous study, which examined the degree to which the lexicalproperties of students’ essays could predict their vocabulary scores. We expand on this previousresearch by incorporating new natural language processing indices related to both the surface- anddiscourse-levels of students’ essays. Additionally, we investigate the degree to which these NLPindices can be used to account for variance in students’ reading comprehension skills. We calculatedlinguistic essay features using our framework, ReaderBench, which is an automated text analysis toolsthat calculates indices related to linguistic and rhetorical features of text. University students (n = 108)produced timed (25 minutes), argumentative essays, which were then analyzed by ReaderBench.Additionally, they completed the Gates-MacGinitie Vocabulary and Reading Comprehension tests.The results of this study indicated that two indices were able to account for 32.4% of the variance invocabulary scores and 31.6% of the variance in reading comprehension scores. Follow-up analysesrevealed that these models further improved when only considering essays that contained multipleparagraph (R2 values = .61 and .49, respectively). Overall, the results of the current study suggest thatnatural language processing techniques can help to inform models of individual differences amongstudent writers.

KW - writing skills

KW - automated writing evaluation

KW - comprehension prediction

KW - vocabulary measures

KW - natural language processing

U2 - 10.21125/edulearn.2016.2241

DO - 10.21125/edulearn.2016.2241

M3 - Conference Article in proceeding

SN - 978-84-608-8860-4

SP - 5269

EP - 5279

BT - EDULEARN16 Proceedings

A2 - Gómez Chova, L.

A2 - López Martínez, A.

A2 - Candel Torres, I.

PB - IATED Academy

T2 - EDULearn16

Y2 - 4 July 2016 through 6 July 2016

ER -

MODELING INDIVIDUAL DIFFERENCES AMONG WRITERS USING READERBENCH

Abstract

Conference

Keywords

Access to Document

Fingerprint

Projects

Rage: Realising an Applied Gaming Eco-system

Cite this