W1 EXPLAINING AND PREDICTING THE WORLD Flashcards
What are the 3 central tendencies?
Mean, Median, Mode
What is VARIANCE?
Average squared deviation from the mean
What is the variance of this set? {1,2,3,4,5} (Population)
Variance 1. Get mean (3) 2. For i in nums: total += (mean - nums[i])^2 total / len(nums) Variance = 10/5 variance = 2
What is the formula for Variance?
SUM[i-n] (y[i] - μ)^2 / n
What is the formula for Variance in a sample?
SUM[i-n] (y[i] - μ)^2 / (n - 1)
What is the formula for standard deviation?
√(SUM[i-n] (y[i] - μ)^2 / n)
What is σ and μ?
Standard Deviation and Mean
How to calculate Coefficient of variation?
CV = σ/μ x 100
What skew does this graph have?
Positive
What skew does this graph have?
Negative
What is kurtosis?
Kurtosis is a measure of whether the data are heavy-tailed or light-tailed relative to a normal distribution.
What is Mesokurtic?
A normal amplitude for a normal distribution curve
What is Leptokurtic?
A higher amplitude for a normal distribution curve
What is Platykurtic?
A shallower amplitude for a normal distribution curve
What are the approximate % within each standard deviation?
μ to 1σ: 34.1% 1σ to 2σ: 13.6% 2σ to 3σ: 2.1% 3σ to 4σ: 0.1%
What is the the null hypothesis?
A statement about a population parameter (e.g., population mean) that is assumed to be true. Where we start, and what we aim to falsify.
What is the alternative hypothesis?
Statement that directly contradicts the null hypothesis by stating that the actual value of a population parameter is less than, greater than, or not equal to the value stated in the null hypothesis. Hypothesis we are trying to prove.
How much significance is required to reject the null?
Usually 5%, but 1% and .5% depending on what it is. So long as you can justify your significance level.
What are one-tailed/two-tailed tests?
One tailed only counts one direction of the normal dist, and two tailed counts both.
What is a p-value?
A probability to compare to the significance level. Usually 0.05. A p-value is the probability of obtaining an effect at least as extreme as the one in your sample data, assuming the truth of the null hypothesis.
What would a p-value of 0.03 mean?
A p-value of 0.03. This P value indicates that if the intervention had no effect, we would obtain the observed difference or more in 3% of studies (because of sampling error)
What is the Type I error
When the researcher rejects the null hypothesis when it is actually true
What is the Type II error
When the researcher accepts the null hypothesis when it is actually false
What is POWER in hypothesis testing?
The probability of rejecting a null hypothesis that’s actually false. 1 - β
