Assignment 5B

docx

School

University of Washington *

*We aren’t endorsed by this school

Course

317

Subject

Statistics

Date

Apr 3, 2024

Type

docx

Pages

5

Uploaded by JusticeGuanacoPerson753

Report
Assignment 5B 1. Why is the following statement not a proper null hypothesis? H o : M = M 0 a) Correct this null hypothesis. H o : μ = μ 0 b) State the appropriate non directional (two tailed) alternative hypothesis. H 1 : μ μ 0 μ > μ 0 or μ < μ 0 2. When might it be justifiable for a researcher to conclude from a non significant hypothesis test decision that a) there is no effect, and b) the effect is likely too small to be important? a. Is justifiable for a researcher to conclude that there is no effect in a non- significant hypothesis test, where we fail to reject the null hypothesis, it suggests that the observed data isn't significantly different from what's expected under the null hypothesis. If the effect size is minimal or close to 0, it may be reasonable for a researcher to conclude there's no meaningful effect. b. When there's insufficient evidence to suggest an effect, often indicated by failing to reject the null hypothesis, the observed effect may be too small to be practically meaningful. 3. Is a statistical result more likely to lead to rejecting a null hypothesis (i.e., be statistically significant) at a .01 alpha level or at a .05 alpha level? A more “conservative” statistical test is one in which it is harder to reject a null hypothesis. Which alpha level is more conservative, .05 or .01? An alpha level of .01 is more conservative compared to .05. Making it harder to reject a null hypothesis at alpha .01 then .05. An alpha of .01 requires more evidence against the null hypothesis to rejct it than an alpha of .05.
4. Suppose you choose an alpha level of .01 for your null hypothesis significance test (NHST). What is the probability that you will make a Type I error: a) if the null hypothesis is true, and b) if the null hypothesis is false? a. The probability of making a Type I error is .01. b. If the null hypothesis is false then there is no way to commit Type I error because Type I error can only happen when a null hypothesis is true and rejected. 5. How do the Type I and Type II error rates differ between one- and two tailed hypothesis tests? 6. IQ scores among college students are thought to be nearly normally distributed with mean 116; the standard deviation of the population of college student IQs is not known. UW President Cauce believes that IQ scores of UW students are higher than for the average college student. She has asked you to obtain a random sample of UW students, measure their IQs, and report back to her. Your data from 100 students yield the following: M = 124.2 s = 12.8 You know that this sample mean, by itself, is not sufficient evidence to support President Cauce’s claim about UW students generally, so you decide to carry out a null hypothesis significance test. As a researcher, President Cauce commonly used a .01 alpha level, so you will, too. a) What is your null hypothesis? What is your alternative hypothesis? (Be sure to use useful subscripts!) H o : μ = 116 H 1 : μ 116 μ > 116 or μ < 116 b) Show your computation of the test statistic.
t obt = 124.2 116 12.8 100 = 6.406 c) Determine either a two tailed p value for that test statistic or a critical value. t ( df = 99 = 0.01 , 2 tail ) = ± 2.626 | 6.406 | < 2.626 We reject the null hypothesis. d) Can you reject the null hypothesis of no difference? Why or why not? We can reject the null hypothesis of no difference the computed test statistic 6.406 is far beyond the critical value of 2.626 for two-tailed test. e) Can you conclude that President Cauce’s belief is supported? Explain. Yes, President Cauce’s belief is supported since we can reject the null hypothesis UW students IQ scores are equal to average college students. T obt of 6.406 is significantly greater than the critical value of 2.626 for a two-tailed test. IQ of UW students (124.2) is higher than the population mean, this supports President Cauce's belief that UW students have higher IQ scores on average compared to the average college student. 7. Return to question 6. If the sample size were actually n = 16 (M = 124.2, s = 12.8), what will be the numerical value of the test statistic? How will your answers to c), d), and e) change? t obt = 124.2 116 12.8 16 = 2.563 c . ( df = 15 = 0.01 , 2 tail ) = ± 2.947 | 2.563 | < 2.947 Fail to reject the null hypothesis. This doesn’t support president Cauce’s belief.. There is not enough evidence to conclude that there is any difference between the IQ scores of UW student to the average college students.
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
8. Return again to question 6. If the sample mean was actually found to be M = 118.1 (n = 100, s = 12.8), what will be the numerical value of the test statistic? How will your answers to c), d), and e) change? t obt = 118.1 116 12.8 100 = 1.641 ( df = 99 = 0.01 , 2 tail ) = ± 2.626 | 1.641 | < 2.626 Fail to reject the null hypothesis. This would not support president Cause’s idea. There is not enough evidence to conclude that UW students IQ scores and the average college students are different. 9. T F A 99% CI for data in question 6 will cover 116.3 T critical = ± 2.626 99% = 124.2 2.626 ( 12.8 100 ) ≤ μ≤ 124.2 + 2.626 ( 12.8 100 ) 120.839 ≤μ≤ 127.561 10. T F A 99% CI for data in question 7 will cover 116. T critical = ± 2.947 99% = 124.2 2.947 ( 12.8 16 ) ≤μ≤ 124.2 + 2.947 ( 12.8 16 ) 114.770 ≤ μ≤ 133.631
11. T F A 99% CI for data in question 8 will cover 116. 99% = 118.1 2.626 ( 12.8 100 ) ≤ μ≤ 118.1 + 2.626 ( 12.8 100 ) 114.739 ≤μ≤ 121.461 (Be sure you could support your answers to 9, 10, and 11!)