STAT13.1 Fall 2008: HW6 Solutions - 2-Sample T-test for Independent Means, Assignments of Statistics

Solutions to problem 1, 2 and 3 of homework 6 for the statistics 13.1 fall 2008 course at ucla. The problems involve using a 2-sample t-test to compare the means of two independent samples and testing the null hypothesis that the means are equal. The data, calculations, and conclusions for each problem.

Typology: Assignments

Pre 2010

Uploaded on 08/31/2009

koofers-user-xu9
koofers-user-xu9 🇺🇸

10 documents

1 / 4

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
www.stat.ucla.edu/~dinov/courses_students.dir/08/Fall/STAT13.1.dir/ 1
Stats 13.1 Fall 2008: Homework 6 Solutions
http://www.stat.ucla.edu/~dinov/courses_students.dir/08/Fall/STAT13.1.dir/
1. Problem 1
Here, we will use a 2-sample T-test to compare the means of 2 independent samples
H0: µ1 - µ2 = 0
HA: µ1 - µ2 0
494.0
5
19.7
5
73.8
22.2972.31
)(
0
22
21
21
0=
+
=
=yySE
yy
T
Now, we compare this to a Students T-distribution with n1 + n2 – 2 = 8 degrees of freedom using SOCR to
find p(T(df=8) > T0 = 0.494) = 0.32
In this case we have a 2-tailed test, and so our p-value is 2*0.32 = 0.64.
Conclusion: We fail to reject the null hypothesis since 0.64>0.10. We conclude that according to our data
the mean weight of thymus glands after being incubated for 14 days is not different than the mean weight after
being incubated for 15 days. In other words, there is no statistically significant mean difference in thymus
glands weights.
Although the mean weight of thymus glands after 15 days of incubation is lower, this is most likely due to
variability and small sample size.
Our result matches SOCR 2 independent sample T-test (see figure 1)
pf3
pf4

Partial preview of the text

Download STAT13.1 Fall 2008: HW6 Solutions - 2-Sample T-test for Independent Means and more Assignments Statistics in PDF only on Docsity!

Stats 13.1 Fall 2008: Homework 6 Solutions

http://www.stat.ucla.edu/~dinov/courses_students.dir/08/Fall/STAT13.1.dir/

  1. Problem 1
    • Here, we will use a 2-sample T-test to compare the means of 2 independent samples

H0: μ 1 - μ 2 = 0 HA: μ 1 - μ 2 ≠ 0

2 2 1 2

1 2 0 =

SE y y

y y T

Now, we compare this to a Students T-distribution with n 1 + n 2 – 2 = 8 degrees of freedom using SOCR to find p(T(df=8) > T 0 = 0.494) = 0.

In this case we have a 2-tailed test, and so our p-value is 2*0.32 = 0.64. Conclusion: We fail to reject the null hypothesis since 0.64>0.10. We conclude that according to our data the mean weight of thymus glands after being incubated for 14 days is not different than the mean weight after being incubated for 15 days. In other words, there is no statistically significant mean difference in thymus glands weights.

  • Although the mean weight of thymus glands after 15 days of incubation is lower, this is most likely due to variability and small sample size.
  • Our result matches SOCR 2 independent sample T-test ( see figure 1 )

Figure 1

  1. Problem 2
    • Here, we will use a 2-sample T-test to compare the means of 2 independent samples

H0: μ 1 - μ 2 = 0 HA: μ 1 - μ 2 ≠ 0

2 2 1 2

1 2 0 =

SE y y

y y T

Now, we compare this to a Students T-distribution with n 1 + n 2 – 2 = 13 degrees of freedom using SOCR to find p(T(df=13) > T 0 = 0.919) = 0.

In this case we have a 2-tailed test, and so our p-value is 2*0.19 = 0.38. Conclusion: We fail to reject the null hypothesis since 0.38>0.10.

  • We conclude that according to our data, there is no statistically significant mean difference in the number of bacteria colonies formed when using a “soap” solution or sterile water.
  • Our result matches SOCR 2 independent sample T-test ( see figure 2 )

23 13. 97 ( 36. 97 , 9. 03 ) 4

2 * 13 92 115 1. 52 *

82 % ( )

2

1 2

2

. 18 1 2 2 ,

1 2

= − ± =− ± = − −

= − ± −

CI x x t SE x x n n df

  1. Problem 4

    • 95% CI
  2. 059 0. 180 ( 0. 121 , 0. 239 ) 17

  3. 262

22

  1. 289
  2. 500 8. 441 2. 03 *

95 % ( )

2 2

1 2

2

. 05 1 2 2 ,

1 2

= − ± + = ± = −

= − ± −

CI x x t SE x x n n df

We can say that we are 95% confident that the true difference in mean head widths between female crickets that were successful and female crickets that were unsuccessful in mating is somewhere between -0.121 and 0.239. We might also suggest that since 0 falls in this interval, then there is evidence to suggest that there is no difference in head widths between the two groups, and therefore the body size does not play a role in mating success.