Week 4 Flashcards
What is statistical inference?
Making estimates about a population based off of data collected from the sample.
How is statistical inference measured?
Using correlation coefficient
Analysis of continuous data: looking for relationship between two variables
What are the four steps to making statistical inference?
- Having a question about a population
- Gather data to answer question
- Describe the data
- Estimation and testing
List and describe the two parts of statistical inference.
- Estimation: estimating the unknown population values using point and interval estimates
- Null hypothesis significance testing
What are the two types of estimation in statistical inference?
- Point estimate: a single value, sample mean, sample prevalence about the data
- Interval estimate: a range (interval) with lower and upper limits that contains the true population value
What does the length of an interval indicate in an interval estimate?
Length of an interval indicates the size of the sampling error.
Describe the difference between a narrow and a wide interval estimate.
- Narrow interval estimate is a good estimate for true value (e.g., between 15 to 20% off the true value)
- Wide interval estimate is a bad estimate for true value (e.g., 10 to 40% off the true value)
How does sample size affect the confidence interval?
A bigger sample size creates a more narrow (good) confidence interval than a smaller sample size.
What is the general rule of confidence intervals?
Higher confidence interval means less room for error:
1. 99% CI is better than 95%
2. 95% CI is better than 90%
What is a null hypothesis?
The suspicion that the results of the data are due to chance; there is no relationship between the variables.
What are the five steps to performing Null Hypothesis Significance Testing?
- Set up/write Null & Alternative hypothesis
- Set the significance level (alpha) - already preset
- Choose suitable test
- Make decision = reject null or don’t reject null
- Assess practical significance
What is the symbol for the null hypothesis?
H0
What is the symbol for the alternative hypothesis?
HA
What is the hypothesis selected when the Null hypothesis is rejected?
Alternative hypothesis
What is the first test (we looked at in week 4) that you can use to obtain the probability value (P value) to provide the probability of the data being due to random chance?
Pearson’s correlation analysis
What is the first decision you need to make when attempting the Null Hypothesis Significance Testing?
Reject or do not reject null
According to the Pearson’s correlation analysis, what should you do if P value is less than 0.05?
Reject null hypothesis because results have statistical significance.
According to the Pearson’s correlation analysis, what should you do if P value is more than 0.05?
Don’t reject null hypothesis because the results have no statistical significance.
