the objective of this assignment is to use methods we have studied in hypothesis
ID: 3322481 • Letter: T
Question
the objective of this assignment is to use methods we have studied in hypothesis testing to test claims about the skyline population. you may choose to investigate a printed claim about a population and desigh and conduct a survey to challenge (or fail to challenge) the claim in it's application to skyline students. you may instead choose to compare opinions or attributes between two subpopulations.
your project should be conducted as follows:
1) find an article in a newspaper, magazine or on the internet that makes a claim about one population mean or one population proportion. the claim may be based upon a survey that the article was reporting on. decide wheather this claim is the null or alternative hypothesis.(2) or propose a comaprison between two sub-population using sample means or sample proportion.(3) detail your sampling procedure. (4) conduct your survey. you must have more than 100 responses in your sample. (5) state the ststistics that are a result of your data collection : sample sizes, sample means or proportion and sample standard deviation. (7) create a graph that illustrtaes your data. this may be pie chart or bar char or may be histogram or box plot, depending on the nature of your data. (8) record the hypothesis test. make your decision using 5% level of significance. (9) identify the type of errors you are potentially commiting based on your hypothesis decision. provide at least one consequence of this error. (10) construct a 95% confidence interval with your results and interpret your meaning. include the graph showing the bounds and sample mean/ proportion or their difference.
Explanation / Answer
(a) Generally while validating claims we conduct a lower tail test
H0 : U>=U0, Ha: U<U0
(b) If we have about a 100 repsonses then this is our sample, we can compute the sameple mean and sd in excel or R using funtions like mean() and sd()
(c) Similarly, charts like histograms and boxplots can be created using the hist() and boxplot() commands
(d) if we are doing the lower tail test as stated above then we need to calculate a z statistic where z= x-U0/ std dev
We then need to find the area to the left of this statistic on the std normal curve which will be p value
At a 5% level of significance alpha = 0.05, if p<alpha then we reject the null hypothesis
(e) Confidence intervals can be set up by doing mean (+-) z(alpha/2)* std.dev/sqrt(n) where n is total no. of observations This gives the interval around which the population mean will lie.
Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.