The use of statistics is constantly evolving. New metrics are continuously being used in areas such as politics and sports statistics—the discipline of speech, language, and hearing services is no different. Journal of Speech, Language, and Hearing Research (JSLHR) Editor-in-Chief for Speech Erick Gallun says the statistics that many professionals learned in graduate school may not be the most appropriate for answering research questions.
JSLHR’s latest forum will help address this new discrepancy, Gallun says. The forum, Advancing Statistical Methods in Speech, Language, and Hearing Sciences, contains “some really clear papers that explain what modern statistics looks like and [help] people move from the things they know to the methods that will actually answer the questions they want to answer.”
In the first article, “Essential Statistical Concepts for Research in Speech, Language, and Hearing Sciences,” Jacob J. Oleson, Grant D. Brown, and co-guest editor Ryan McCreery provide an overview of statistical concepts and methods used in the discipline. The authors write that although the traditional methods discussed have specific uses, often other tools and methods can answer the important research questions more accurately and with greater flexibility.
Katherine R. Gordon, one of the two guest editors for this issue, explored one of these new statistical methods in “How Mixed-Effects Modeling Can Advance Our Understanding of Learning and Memory and Improve Clinical and Educational Practice.” In the article, Gordon writes that traditional statistical methods may not be sufficient to answer questions about special populations, as the samples from these populations may be small and highly variable; however, mixed-effects modeling allows researchers to control for individual variability. She provides two clinical examples that illustrate mixed-effects modeling while also explaining additional advantages. “Researchers can improve their ability to discover accurate information about learning and retention through carefully planned methodologies and statistical techniques that make the best use of all the data that they have,” she writes.
In another article, Elizabeth A. Walker, Alexandra Redfern, and Jacob J. Oleson used linear mixed models to explore vocabulary depth and breadth in children who are hard of hearing. Whereas missing data points would be a problem developing data, linear mixed models allowed additional ways to examine change over time in language development.
The remaining articles continue to examine the use of both traditional and expanded statistical methods in research as diverse as word learning, auditory research, and neurophysiology.
We’d like to thank forum editors Katherine R. Gordon and Ryan McCreery for their work assembling this essential forum for JSLHR. Speech Editor Erick Gallun thinks these papers will be crucial for readers. “I expect that these papers will really help people,” he said. “When statistics are explained well, it’s just very obvious.”
Explore the Issue
Curran, M. K., Walker, E., Roush, P., & Spratford, M. (2019). Using propensity-score matching to address clinical questions: The impact of remote-microphone systems on language outcomes in children who are hard of hearing. Journal of Speech, Language, and Hearing Research, 62, 564–576. https://doi.org/10.1044/2018_JSLHR-L-ASTM-18-0238h
Gordon, K. (2019). How mixed effects modeling can advance our understanding of learning and memory and improve clinical and educational practice. Journal of Speech, Language, and Hearing Research, 62, 507–524. https://doi.org/10.1044/2018_JSLHR-L-ASTM-18-0240
McMillan, G. P., & Cannon, J. B. (2019). Bayesian applications in auditory research. Journal of Speech, Language, and Hearing Research, 62, 577–586. https://doi.org/10.1044/2018_JSLHR-H-ASTM-18-0228
Oleson, J. Brown, G. D., & McCreery, R. W. (2019). Essential statistical concepts for research in speech, language, and hearing sciences. Journal of Speech, Language, and Hearing Research, 62, 489–497. https://doi.org/10.1044/2018_JSLHR-S-ASTM-18-0239
Oleson, J. Brown, G. D., & McCreery, R. W. (2019). The evolution of statistical methods in speech, language, and hearing sciences. Journal of Speech, Language, and Hearing Research, 62, 498–506. https://doi.org/10.1044/2018_JSLHR-H-ASTM-18-0378
Paulon, G., Reetzke, R., Chandrasekaran, B., & Sarkar, A. (2019). Functional logistic mixed-effects models for learning curves from longitudinal binary data. Journal of Speech, Language, and Hearing Research, 62, 543–553. https://doi.org/10.1044/2018_JSLHR-S-ASTM-18-0283
Perry, L. K., & Kucker, S. C. (2019). The heterogeneity of word learning biases in late-talking children. Journal of Speech, Language, and Hearing Research, 62, 554–563. https://doi.org/10.1044/2019_JSLHR-L-ASTM-18-0234
Walker, E., Redfern, A., & Oleson, J. (2019). Linear mixed-model analysis to examine longitudinal trajectories in vocabulary depth and breadth in children who are hard of hearing. Journal of Speech, Language, and Hearing Research, 62, 525–542. https://doi.org/10.1044/2018_JSLHR-L-ASTM-18-0250
Xie, Z., Reetzke, R., & Chandrasekaran, B. (2019). Machine learning approaches to analyze speech-evoked neurophysiological responses. Journal of Speech, Language, and Hearing Research, 62, 587–601. https://doi.org/10.1044/2018_JSLHR-S-ASTM-18-0244