Estimation and Confidence Intervals Flashcards
Definition of statistical estimation
Estimate the parameter being investigated by finding the mean value from a random sample
Definition of sample/point estimation
Use of sample data to calculate an estimate of an unknown parameter
Definition of confidence interval
A range of plausible values for an unknown calue
Can quantify uncertainty and imprecision
Definition of sampling error
Different samples give different estimates
Definition of sample distribution
Sample estimates calculated from multiple ampler from the same population will have varying distribution values
Definition of standard error
Indication of the extent of the sampling error
Tells us how much a sample mean tends to cary from the population mean
Provides and estimate of the precision of the sample mean
How would you estimate a value from a large population
Impossible to measure everyone in a population
- Take a random sample, find the sample mean
- Different samples => different point estimates
- Sample estimates vary around a true value
- Can use this data to work out a confidence interval
What is the confidence interval and what is it used for
What can it not tell you
Cannot deduce exact population value from a sample
Can use sample to obtain a confidence interval
-Range within which a population value is likely to be
Describe the pathway in an investigation and show how you’d use the info found on the population parameter
Design a study based on the population parameter
Take a sample and estimate the parameter
Take the results and deduce that it applies to the whole population
What is sampling error
How does it arise
Different samples => different estimates
Provide an incomplete picture of the population
What 2 values help you find the true mean
Bigger sample size => estimate closer to true mean
Smaller SD => estimate closer to true mean
What is the standard error
What is the equation for standard error
Tells you how much a sample mean tends to vary from the true population
Estimate of precision of sample mean
SE = SD/√N
How do you increase precision if you can change the SD or the sample size
Decreased SD or increased sample size => smaller SE => increased precision
What does it mean to have a 95% confidence interval
True mean expected to lie within range in 95% of calculations
Lower confidence limit = -1.96 SD of sample mean
Upper confidence limit = +1.96 SD of sample mean
Confidence interval = range between lower confidence limit and upper confidence limit
What are the 3 assumptions in calculating the confidence interval
Normal data or large sample ≥60)
Sample chosen at random from population
Observations are independent of each other