Hello everyone, I need a bit help with statistics analitic methods.
My partner (MD) is conducting research as part of her residency exam on how years of occupation impact workers' hearing. Her known variables are years of employment, years of employment at current job position,age and percents of hearing loss (calculated with Fowler-Sabine formula, so in %).
She had a statistic working on her study and he did multivariate linear regression (explaining he used it because one variable is in %).
However one of her professors said she should use log regression analysis instead. WHY? Is multivariate linear not OK and is, why not?
Can anyone help explain which one should be used/ is better and why? We tried google but as we are not statistics or experienced researchers this is quite hard for us to understand. However, she need this done correctly as this study is a part of her residency exam.
Any help is much appreciated.
Many thanks!
Anze&Ana
Indeed, in most studies on hearing loss in working environments logistic regression is used. Linear regression predicts a continuous variable with infinite possible values (not in your work), while logistic regression predicts the categorical variable for one or more independent variables. See for more information on the difference of methods: https://learn.g2.com/logistic-regression.
You might pass the information to your statistic, as he might be more familiar with the matter.
Marcel Severijnen thank you. We did pass information you provided to statistic and he basically said he used continous variable (as delta Fowler, difference between last known Fowler and first known Fowler) but we now worry- should my partner find new statistic and do statistics all over again? Is there no way this method is OK? Thank you so much for your input.
We used multiple linear regression treating % hearing loss( as calculated in the Australian compensation system from audiogram decibel losses at certain frequencies with a binaural weighting) as a continuous variable with independent variables age (cts) , noise dose(cts) and use of hearing protection (categoric). See attached papers.
Choosing a statistical method is always difficult. Obviously (multiple) linear regression is used too in these kind of studies as for instance James Leigh explains, but rarely is the choice of model clearly articulated. Maybe you could directly ask why a choice for logistic regression is to prefer above linear regression in your case. I enclose two articles written in "normal" language for further understanding the difference between both approaches:
https://www.statology.org/logistic-regression-vs-linear-regression/
https://tinyurl.com/47vp7ahc
Wishing you good luck!
I believe there is a misunderstanding in this discussion. Could "log regression" not mean logarithmic rather than logistic? That makes more sense as % is still numeric.
- Badges
- Science topic
- Similar topics
- Geoscience
- Atmospheric Sciences
- Climate Science
- Climate Change