Foundations for Inference Flashcards
Sampling Distribution
a theoretical distribution of a sample statistic after taking repeated samples
Central Limit Theorem (for proportion)
allows us to make inferences about population proportions based on sample data and it helps us estimate how wrong the sample is likely to be
with a large enough sample size, the distribution of sample proportions will be approximately normally distributed.
Confidence Interval
A range around a sample statistic that is likely to include the population parameter.
Standard Error
The standard deviation of the sampling distribution of a given sample; a measure of how far off, on average, a given sample is likely to be from the population parameter.
Building a Confidence Interval
- decide how confident we want to be (standard is 95%)
- determine the alpha level (the probability of being wrong) (for a 95% confidence interval, alpha is 0.05)
- determine the Z-score value for 95% (For 95% confidence, the z score is 1.96)
- find the se (√((𝑃(1−𝑃))/𝑛)
- find margin of error: se x z value
- CI will be the sample proportion + or - that value
standard error
√((𝑝̂(1−𝑝̂))/𝑛)
margin of error
se x z
equation for confidence interval
𝑝̂ ± moe
aka: 𝑝̂ ± se x z
explain the equations for
1. find se
2. find moe
3. find CI
find se: √((𝑝̂(1−𝑝̂))/𝑛)
find moe: se x z
find CI: 𝑝̂ ± moe
When you find the two values of CI (𝑝̂ ± moe) what are you saying?
we are 95% confident that the true proportion is within these two values
Parameters vs Statistics
parameters are the features of a population while statistics are features of a sample
Population mean (μ)
Population standard deviation (σ)
Population proportion (p)
parameters: greek lettters
Sample mean (𝑥̅, x-bar)
Sample standard deviation (s)
Sample proportion (𝑝̂, p-hat)
statistics: latin letters
the parameter/statistic represents what?
parameter: population
statistic: sample
Will your CI become narrower or wider if your confidence level goes from 95% to 80%?
narrower with 80%.: confidence is lower, but possibly more precise
wide with 95%. you are more confident with 95% but less precise