Essential statistical concepts for research in speech, language, and hearing sciences

Jacob J. Oleson, Grant D. Brown, Ryan McCreery

Research output: Contribution to journalReview article

1 Citation (Scopus)

Abstract

Purpose: Clinicians depend on the accuracy of research in the speech, language, and hearing sciences to improve assessment and treatment of patients with communication disorders. Although this work has contributed to great advances in clinical care, common statistical misconceptions remain, which deserve closer inspection in the field. Challenges in applying and interpreting traditional statistical methods with behavioral data from humans have led to difficulties with replication and reproducibility in other allied scientific fields, including psychology and medicine. The importance of research in our fields of study for advancing science and clinical care for our patients means that the choices of statistical methods can have far-reaching, real-world implications. Method: The goal of this article is to provide an overview of fundamental statistical concepts and methods that are used in the speech, language, and hearing sciences. Results: We reintroduce basic statistical terms such as the p value and effect size, as well as recommended procedures for model selection and multiple comparisons. Conclusions: Research in the speech, language, and hearing sciences can have a profound positive impact on the lives of individuals with communication disorders, but the validity of scientific findings in our fields is enhanced when data are analyzed using sound statistical methods. Misunderstanding or misinterpretation of basic statistical principles may erode public trust in research findings. Recommendations for practices that can help minimize the likelihood of errors in statistical inference are provided.

Original languageEnglish (US)
Pages (from-to)489-497
Number of pages9
JournalJournal of Speech, Language, and Hearing Research
Volume62
Issue number3
DOIs
StatePublished - Mar 2019

Fingerprint

Hearing
Language
statistical method
communication disorder
language
science
Research
Communication Disorders
scientific findings
field of study
psychology
medicine
Statistical Methods
Patient Care
Medicine
Psychology
Statistical Inference
Effect Size
Misconceptions
Real World

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language
  • Speech and Hearing

Cite this

Essential statistical concepts for research in speech, language, and hearing sciences. / Oleson, Jacob J.; Brown, Grant D.; McCreery, Ryan.

In: Journal of Speech, Language, and Hearing Research, Vol. 62, No. 3, 03.2019, p. 489-497.

Research output: Contribution to journalReview article

@article{6bd7dbefc5ac4dc1a6965ed9f3f32332,
title = "Essential statistical concepts for research in speech, language, and hearing sciences",
abstract = "Purpose: Clinicians depend on the accuracy of research in the speech, language, and hearing sciences to improve assessment and treatment of patients with communication disorders. Although this work has contributed to great advances in clinical care, common statistical misconceptions remain, which deserve closer inspection in the field. Challenges in applying and interpreting traditional statistical methods with behavioral data from humans have led to difficulties with replication and reproducibility in other allied scientific fields, including psychology and medicine. The importance of research in our fields of study for advancing science and clinical care for our patients means that the choices of statistical methods can have far-reaching, real-world implications. Method: The goal of this article is to provide an overview of fundamental statistical concepts and methods that are used in the speech, language, and hearing sciences. Results: We reintroduce basic statistical terms such as the p value and effect size, as well as recommended procedures for model selection and multiple comparisons. Conclusions: Research in the speech, language, and hearing sciences can have a profound positive impact on the lives of individuals with communication disorders, but the validity of scientific findings in our fields is enhanced when data are analyzed using sound statistical methods. Misunderstanding or misinterpretation of basic statistical principles may erode public trust in research findings. Recommendations for practices that can help minimize the likelihood of errors in statistical inference are provided.",
author = "Oleson, {Jacob J.} and Brown, {Grant D.} and Ryan McCreery",
year = "2019",
month = "3",
doi = "10.1044/2018_JSLHR-S-ASTM-18-0239",
language = "English (US)",
volume = "62",
pages = "489--497",
journal = "Journal of Speech, Language, and Hearing Research",
issn = "1092-4388",
publisher = "American Speech-Language-Hearing Association (ASHA)",
number = "3",

}

TY - JOUR

T1 - Essential statistical concepts for research in speech, language, and hearing sciences

AU - Oleson, Jacob J.

AU - Brown, Grant D.

AU - McCreery, Ryan

PY - 2019/3

Y1 - 2019/3

N2 - Purpose: Clinicians depend on the accuracy of research in the speech, language, and hearing sciences to improve assessment and treatment of patients with communication disorders. Although this work has contributed to great advances in clinical care, common statistical misconceptions remain, which deserve closer inspection in the field. Challenges in applying and interpreting traditional statistical methods with behavioral data from humans have led to difficulties with replication and reproducibility in other allied scientific fields, including psychology and medicine. The importance of research in our fields of study for advancing science and clinical care for our patients means that the choices of statistical methods can have far-reaching, real-world implications. Method: The goal of this article is to provide an overview of fundamental statistical concepts and methods that are used in the speech, language, and hearing sciences. Results: We reintroduce basic statistical terms such as the p value and effect size, as well as recommended procedures for model selection and multiple comparisons. Conclusions: Research in the speech, language, and hearing sciences can have a profound positive impact on the lives of individuals with communication disorders, but the validity of scientific findings in our fields is enhanced when data are analyzed using sound statistical methods. Misunderstanding or misinterpretation of basic statistical principles may erode public trust in research findings. Recommendations for practices that can help minimize the likelihood of errors in statistical inference are provided.

AB - Purpose: Clinicians depend on the accuracy of research in the speech, language, and hearing sciences to improve assessment and treatment of patients with communication disorders. Although this work has contributed to great advances in clinical care, common statistical misconceptions remain, which deserve closer inspection in the field. Challenges in applying and interpreting traditional statistical methods with behavioral data from humans have led to difficulties with replication and reproducibility in other allied scientific fields, including psychology and medicine. The importance of research in our fields of study for advancing science and clinical care for our patients means that the choices of statistical methods can have far-reaching, real-world implications. Method: The goal of this article is to provide an overview of fundamental statistical concepts and methods that are used in the speech, language, and hearing sciences. Results: We reintroduce basic statistical terms such as the p value and effect size, as well as recommended procedures for model selection and multiple comparisons. Conclusions: Research in the speech, language, and hearing sciences can have a profound positive impact on the lives of individuals with communication disorders, but the validity of scientific findings in our fields is enhanced when data are analyzed using sound statistical methods. Misunderstanding or misinterpretation of basic statistical principles may erode public trust in research findings. Recommendations for practices that can help minimize the likelihood of errors in statistical inference are provided.

UR - http://www.scopus.com/inward/record.url?scp=85064317344&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064317344&partnerID=8YFLogxK

U2 - 10.1044/2018_JSLHR-S-ASTM-18-0239

DO - 10.1044/2018_JSLHR-S-ASTM-18-0239

M3 - Review article

C2 - 30950745

AN - SCOPUS:85064317344

VL - 62

SP - 489

EP - 497

JO - Journal of Speech, Language, and Hearing Research

JF - Journal of Speech, Language, and Hearing Research

SN - 1092-4388

IS - 3

ER -