Critical Values and Confidence Intervals
What you’ve been doing… Gathering Data Calculating Data Interpreting Data With surveys, experiments, and simulations, you only sample the population.
With a census, the entire population is used (to the best of the census taker’s ability).
Do you remember? Population: the entire group Parameter: an unknown proportion which describes the population (called p) Sample: a smaller group which will describe the population Statistic: a value calculated of the sample, generalized to describe the population (called p-hat)
So how can we use this??? If you are given the example: The 2001 Youth Risk Behavioral Survey questioned a nationally representative sample of 12,960 students in grade Of these, 3340 said they had smoked cigarettes at least one day in the past month. How can you represent the data listed above?
High School Students and Smoking Population:Students in grades 9-12 in the United States. Parameter: Proportion of students in grades 9-12 who have smoked cigarettes at least once in the past month. Sample: 12,960 students randomly selected to represent students in grade 9-12 in the United States Statistic: 3,340 out of 12,960 students said they have smoked cigarettes at least once in the past month (this is p-hat)… Which ≈ or 25.77%
So we can say… In the United States, among high school students, grade 9-12, there are 25.77% of students who say they have smoked cigarettes at least once in the past month. However, we can add to this statement to make it more accurate…
Since we cannot be that EXACT, we need to give a range or “spread” of data… In order to do this, we need to use a new formula: z* is the critical value (coming up) n is the number of items in the sample.
Remember, the Rule? This is better!
So, now… O p-hat = O 1-p-hat = O n=12960 O And for a 95% confidence interval (the standard), z* is 1.96 O Plug it all in…
The Confidence Statement O We are 95% confident that the true proportion of high school students who have smoked cigarettes at least one day in the past month is between 25.03% and 26.51%.
What’s different??? This statement now includes the margin of error (since nothing’s perfect).
Now, in your groups… Write a scenario. Include in your write up the population, parameter, sample, and statistic. Solve the problem, identifying the value of each variable. Show your work. Make a generalized statement at the end.