Download Learning More about Statistics 201 Students - Project | STAT 201 and more Study notes Statistics in PDF only on Docsity!
Statistics 201 Project 2 – Learning More About Statistics 201 Students By: Seth Williams April 29, 2010
1. Differences with respondents and others: A One fundamental difference in the respondents and college students as a whole is that the respondents were strictly students at the University of Tennessee- Knoxville. B One fundamental difference in the respondents and UT students as a whole is the fact that only UT-Knoxville students were represented and there are 5 campuses statewide, four of which were not represented. Statistics 201 is also a class that is only required by certain majors, which would lead other majors to be excluded from participating. C The most obvious answer to why this survey might not represent all ~ students in Statistics 201 is because no one was forced to take it, leading some students to not participate. A second possibility is that not everyone that took it, took it seriously and were completely honest with their answers. 2. Confidence Interval for a proportion: I chose to analyze the categorical variable “Where do you sit in the class?” (#14). I chose students who sit “Near Back” during Stats 201. The proportion is 11.90%. A
14 Where Do You Sit In Class?
Frequencies
Level Count Prob Front Row 39 0. Near Front 281 0. Middle 347 0. Near Back 99 0. Back Row 31 0. Different All The Time 35 0. Total 832 1. N Missing 0 -- 6 Levels B My student ID# ends with a 0, therefore my sample size is 73+0 = 73. What the following graph shows about the interval is that you can be 90% confident that the true proportion of students that classify their seats as “Near Back” will be between 4.31% and 15.12%.
14 Where Do You Sit In Class?
Frequencies
N Missing 0, Levels Level Count Prob Front Row 3 0. Near Front 21 0. Middle 36 0. Near Back 6 0. Back Row 2 0. Different All The Time 5 0. Total 73 1.
- the book says that this test alone can be used to check the Nearly Normal Condition. Which this graph meets. C Using my sample set again, and the 98% confidence interval for the population mean is (14.39, 15.48), meaning that I am 98% Confident that the population mean is between 14.38 and 15.48. Being that the (shown below) population average value (μ) is 14.93, it proves the confidence interval is correct.
4. Hypothesis Test Regarding the Difference in Means for Independent Samples: I chose “Gender”(#2) as my Categorical Variable, and “Texts Sent Per Day” (#21) as my Quantitative Variable. A
A (Continued) In both the male and the female results there were two outliers but only one of them were extreme, the other one lies closer to the results and can be looked at along with the other results. (Graph with excluded points below.) After looking at the graph that was produced after the extreme outlier was deleted, they appear normal enough to perform a 2-sample test. B My observations of the results of the distribution reveal that females more than double (on average) the number of texts that males send per day. I want to know if females, on average, double the texts per day that males send. My Hypothesis Pair would be… Null Hypothesis: The number of texts sent per day is independent of gender. Alternative Hypothesis: There is an association with the number of texts sent per day and gender. The difference when comparing the female’s mean (80.36) to the male’s mean (40.09) is 40.27 texts per day. The standard error of difference = 13.58, The P- Value= .0019.
A
B The largest Cell Chi^2 Value is students who had at least one parent graduate from college and whose parents are still married. A large Chi^ value usually means that the null hypothesis is to be rejected. C Null Hypothesis: The fact that the students’ parents are married is independent of the fact that at least one of the parents graduated from college. Alternative Hypothesis: There is an association between the parents being married and at least one of the parents graduating college. P-Value: .0001 (Because α=.05, and the P-Value is lower than that at .0001, my null hypothesis is proven false .)
Conclusion: There is an association (in the students that participated) between their parents being married and at least one of the parents graduating college. D This would be a Type I Error. The probability of making a mistake is equal to .05 (α) for this problem.
6. Interferences Regarding a Regression: Using Question #20 and #5 as directed. A
Before Exclusions After Exclusions
speeds period because it is not a significant difference, men just have higher speeds traveled and it distorts their line with a slightly more positive slope.
