Oxy Biostats

[oxybiostats]

I think the most relevant thing I learned in this class is how to make honest data visualizations and how to critique graphs when they distort data or intend to mislead. Here is a bar graph that goes from 80million to 82million and tries to claim that it’s a “massive” increase (which doesn’t even make numerical sense lol)

[oxybiostats]

Most important Biostats Concept: Deciding between Frequentist and Bayesian Stats #biostats #Laursen #ilovebiostats

[oxybiostats]

I learned about publication bias through our discussion of the Yaphe et al. 2001 paper. When analyzing statistical results, one should consider the possibility that these results were published based on the author’s / publisher’s preferences.

#Ringmayr

[oxybiostats]

I think Bayesian statistics may have been the most important concept I learned because it made me question the way we typically look at statistics. Also, its use on five thirty eight has me excited about my team’s odds of making the playoffs. https://projects.fivethirtyeight.com/2017-college-football-predictions/ #Brown

[oxybiostats]

The most important thing I learned about was ANOVAs and how to use R as I used it for two other research papers this semester! #bowers

[oxybiostats]

The most important concept I learned in biostats is the importance of the normality test. Because the normality is the base of all statistical test. If the data is not normally distributed, and cannot be ignored, the results of the statistical analysis would be very inaccurate.

#zhang

[oxybiostats]

[oxybiostats]

Variation of bacterial concentrations of Pseudomonas aeroginosa colonizing lung tissue of mice after treatment with furanone c30 (Wu et. al 2004)

#Yao

[oxybiostats]

Variation in travel rate (cm/min) of Nassarius pullus towards bait among different levels of litter on sandy intertidal. The different litter treatments are control (0% litter), treatment 1 (25% litter), treatment 2 (50% litter), and treatment 3 (75% litter). (Aloy, Vallejo Jr. , & Juinio-Meñez, 2011).

#Ringmayr

[oxybiostats]

Actors were given random sentences to read in different emotional states. Actors were supposed to read in neutral, sad, angry or happy. The authors measured for different factors of speech to see differences between emotions. They measured for utterance duration, silence between words, energy in speech, and speaking rate.

[oxybiostats]

Efficiency of AloeVera to treat Paederus Dermatitis in Mice (Khaghani, Mirzaii-Dizgah, & Ghasemi, 2017). 

#biostats #Bowers #ANOVA #graphs

[oxybiostats]

Variation in mean OHIP scores #laursen #ANOVA #ilovebiostats

[oxybiostats]

FIGURE. Box plots with quartiles (upper values75%,median50%, and lower values 25%) for the attach slope and spectral centroid in each stimulus category. Outliers and extreme values are excluded.  

Happy music with lyrics had significantly faster attack slopes than sad music with and without lyrics(T >3, p <0.002). However, happy or sad music with lyrics did not differ in attack slopes from happy or sad music without lyrics (p >0.07), and happy music without lyrics did not differentiate in the attacks lopes from sad music with or without lyrics (p >0.1). 

Brattico, E., Alluri, V., Bogert, B., Jacobsen, T., Vartiainen, N., Nieminen, S., & Tervaniemi, M. 2011. A Functional MRI Study of Happy and Sad Emotions in Music with and without Lyrics. Frontiers in Psychology, 2: 308. 

#zhang

[oxybiostats]

Variation in depth within stream habitats in streams where salamanders reside (Yeiser & Richter 2015).

[oxybiostats]

1. Although one‐tailed hypothesis tests are commonly used, clear justification for why this approach is used is often missing from published papers.2. Here we suggest explicit questions authors should...

Good reminder to err on the side of using a two-sided hypothesis test unless you have really good reason to use one-sided. If you do use one-sided, make sure you very clearly justify why the opposite side is not a plausible outcome!

[oxybiostats]

What to do when your data are not normal?

Can you draw a flow chart to illustrate which analyses to choose depending on the normality of your data?

Tests to include:

Two-sample t-test

Welch’s two-sample t-test

Levene’s test

F-test

Shapiro-Wilk test

Mann-Whitney U/Wilcoxon Ranked Sum test

Paired t-test

Sign test

Wilcoxon Signed Rank test

Permutation test

[oxybiostats]

Midterm practice question. Which statistical test would you use? Post answers below.