Book a Demo!
CoCalc Logo Icon
StoreFeaturesDocsShareSupportNewsAboutPoliciesSign UpSign In
YStrano
GitHub Repository: YStrano/DataScience_GA
Path: blob/master/april_18/lessons/lesson-04/code/solution-code/lab-solution-code-4.ipynb
1905 views
Kernel: Python 2

Lesson 4 Independent Practice

For this exercise, we are going to take a look at several studies and outputs in order to get comfortable reading outputs and practicing hypothesis testing.

Data output 1

This dataset examines the association between an outcome and generic variables x1, x2, x3

Question 1.1 X6 has an association with the outcome of -0.2. Is this finding statistically significant?

Answer: No, it has a p-value >0.05 and the CI crosses 0.

Question 1.2 Which of the variables X1, X3, and X6 have a statistically significant finding?

Answer: X1

Question 1.3 What does a 95% CI of 0.38-0.866 indicate?

Answer: That if we repeated this study 100 times our point estimate would lie within that range 95 times

Data Output 2

This is a data set of different animals. We measured the brain size and body size of each animale in our data set. We aim to determine if there is an association between the body size and brain size of animals in our dataset.

Question 2.1 What is the outcome in this model?

Answer: brain size

Question 2.2 What is the predictor in this model?

Answer: body size

Question 2.3 What is the association body and brain size?

Answer: 0.97

Question 2.4 Is this relationship statistically significant? How did you decide?

Answer: yes p < 0.05

Question 2.5 is this relationship biologically significant? What other information would you need to decide this?

Answer: Unclear. We would need to understand more baout the dataset, the animals in the dataset, how the dataset was collected etc.

Data Set 3- A study about french fries and diabetes

Take a look at the abstract

Table 1 - Review Table 1

Question 3.1 Is there a statistically significant diffence by age among the different levels of fry or potato intake?

Answer: No

Question 3.2 What are the numbers in the parentheses?

Answer: 95%CI intervals

Question 3.3 Is the risk of diabetes statistically significant for women who eat the highest mean servings of potatos (quintile 5) compared to the lowest group (quintile 1)?

Answer: Yes

Question 3.4 Is the risk of diabetes statistically significant for women who eat the second lowest mean servings of potatos (quintile 2) compared to the lowest group (quintile 1)?

Answer: No