1690 HW 7
docx
School
University of Texas *
*We aren’t endorsed by this school
Course
PHM1690
Subject
Statistics
Date
Apr 3, 2024
Type
docx
Pages
5
Uploaded by marshallmalaysia
Part A
1.
Part I
a.
The hypotheses for the research question can be described as follows:
i.
H
0
: There is not an evident difference in the average IBS-GIS between the “Open
Placebo” group and the “No treatment” group
ii.
H
A
: There is an evident difference in the average IBS-GIS between the “Open Placebo” group and the “No treatment” group
b.
The independent two-sample t-test would be the most appropriate to answer the research question.
c.
The conditions required to complete this test are:
i.
The two samples are independent
ii.
Each sample came from a (nearly) normal distribution with mean µ
k
and population SD σ
k (where k = 1,2)
d.
The test statistic was calculated to be 3.48
x
1
−
x
2
=
5.0
−
3.9
=
1.1
(
1.5
)
2
37
=
0.060810811
(
1.3
)
2
43
=
0.039302326
t
=
1.1
√
0.100113137
=
1.1
0.316406601
=
3.476539353
e.
The degrees of freedom were computed to be 71 and the p-value 0.0009.
f.
Using the t-distribution function, the p-value was 0.00086591. With rounding, this is the same p-value that was found in (e). g.
We have enough evidence to support that the difference in the average IBS-GIS between the “Open Placebo” group and the “No Treatment” group is different from zero. h.
The alternative test that could be used to answer the research questions is the Wilcoxon
Rank-Sum Test
2.
Part II
a.
The confidence interval for the true mean difference of IBS-GIS between the two groups is 0.252 and 1.948.
b.
We are 99% confident that the true mean difference of IBS-GIS falls between 0.252 and 1.948.
c.
The confidence interval does not provide convincing evidence that there is a real (average) IBS-GIS difference between Open Placebo and No Treatment groups because the confidence interval does not contain the null value (0). 3.
Part III
a.
The power of the test for the difference in mean IBS-GIS between the two groups at 0.01 significance level is 0.7939.
b.
The minimum sample size needed to attain he desired power is 102 (51 samples per group).
4.
Diamond Prices
a.
Based on the graphs and summary statistics, it does appear that the 1 carat diamonds are more expensive on average than the 0.99 carat diamonds.
b.
The hypotheses can be defined as follows:
i.
H
0
: the mean price of 1 carat diamonds are the same as the price of 0.99 carat diamonds.
ii.
H
A
: the mean price of 1 carat diamonds are more than the price of 0.99 carat diamonds. c.
The most appropriate test to answer the research question above is the independent two-sample t-test. d.
t
=
(
x
1
−
x
2
)
−
null
√
s
1
2
n
1
+
s
2
2
n
2
=
(
56.81
−
44.51
)
−
0
√
16.13
2
23
+
13.32
2
23
=
12.3
√
11.31
+
7.71
=
2.82
e.
The degrees of the freedom were calculated to be 42 and the p-value 0.0036. f.
We have enough evidence to support that the average price of 1 carat diamonds are more than the average price of 0.99 carat diamonds. Part B
a)
H0: there is no difference in ERG levels between dominant and recessive genes of RP.
H0: μ
dominant
=
μ
recessive
HA: there is a difference in ERG levels between dominant and recessive genes of RP.
HA: μ
dominant
≠μ
recessive
b)
The summary statistics for each group are listed in the table below:
Group
Mean
SD
Median
IQR
N
Dominant
0.4594894
0.281343
0.429
0.511
47
Recessive
0.3385
0.210293
0.3025
0.3135
32
c)
The p-value is 0.0325. We have enough evidence to support that the difference of true means of
ERG levels between dominant and recessive genes of RP is different from 0 at significance level 0.05. d)
Assessing Normality for Each Group
i.
Histogram
The graph for the dominant group shows a very loosely defined normal distribution while the histogram for the recessive group shows a slightly right-skewed distribution. ii.
Boxplot
Both boxplots show an approximately normal distribution with no extreme outliers. iii.
QQ Plot
The QQ plots above show a relatively normal distribution. There are a few outliers in each plot, but it can still be approximated as normal. iv.
Shapiro-Wilks Test
Obs.
w
v
z
Prob > z
Dominant
47
0.94931
2.271
1.743
0.04069
Recessive
32
0.95273
1.577
0.945
0.17221
The results of the Shapiro-Wilks Test yield a large p-value for the Recessive group, but a small p-value for the dominant group. The data of the dominant group has a p-value that is too small to properly assess normality. This data would need to be transformed in
order to establish normality.
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
e)
i.
Using ladder
1/Cubic 1/(erg^3) 69.73 0.000
1/Square 1/(erg^2) 67.52 0.000
Inverse 1/erg 54.87 0.000
1/(Square root) 1/sqrt(erg) 37.34 0.000
Log log(erg) 13.97 0.001
Square root sqrt(erg) 3.61 0.165
Identity erg 9.56 0.008
Square erg^2 5.51 0.064
Cubic erg^3 10.10 0.006
Transformation Formula chi2(2) Prob > chi2
. ladder erg if gene_type == "dominant"
1/Cubic 1/(erg^3) 49.86 0.000
1/Square 1/(erg^2) 49.46 0.000
Inverse 1/erg 43.61 0.000
1/(Square root) 1/sqrt(erg) 30.48 0.000
Log log(erg) 9.74 0.008
Square root sqrt(erg) 0.43 0.807
Identity erg 2.45 0.294
Square erg^2 11.81 0.003
Cubic erg^3 20.21 0.000
Transformation Formula chi2(2) Prob > chi2
. ladder erg if gene_type == "recessive"
The results of the latter transformation show large p-values for the square root function of both groups. This is the only transformation that both groups have in common. ii.
Using gladder
Using the gladder transformation, we see that the dominant group (left) best demonstrates normality using the square root transformation. This is also the case for the recessive group (right).
iii.
Using qladder Using the qladder transformation, the data for the dominant group (left) best fits the square root transformation. The recessive group (right) also shows a pretty good fit for the square root transformation. f)
The p-value was computed to be 0.0779. We do not have enough evidence to support that the difference of true means of ERG levels between dominant and recessive genes of RP is different from zero at significance level 0.05. g)
The p-value that was obtained using the Wilcoxon Rank-Sum test was 0.0741. Our data does not
present evidence that the mean ERG levels differ by genetic type. h)
T-test with Original Data
T-test with Transformed Data
Nonparametric Test
Pros
Cons
Pros
Cons
Pros
Cons
Easier to interpret
Cannot guarantee validity of the test if not normally
distributed
More efficient than nonparametri
c method
Difficult to find
a proper transformation
Does not require normality assumption
Less efficient than the original t-
test
Difficult to interpret the results
Easy to use and interpret
Robust to outliers
Ultimately, I would use the t-test with transformed data because it is more efficient than the nonparametric method. The results of the nonparametric method test are invalid because it is based off the original data and not the transformed data.
Related Documents
Related Questions
According to a Health of Boston report, female residents in Boston have a higher
average life expectancy as compared to male residents (The Boston Globe, August 16,
2010). You collect the following sample data to test the following hypotheses at 0.01
level of significance: Ho: H1s H2 versus H1: H1 >H2, where females represent group 1 and
males represent group 2.
Females
81.1
Males
Мean
74.8
Sample Size
32
32
Assume the population standard deviations of life expectancy are 8.2 and 8.6 for
females and males, respectively. What is(are) the critical value(s)?
Select one:
O None of these are correct
2.390
O +2.575
O 2.33
O +2.660
arrow_forward
You recently completed an experiment concerning the effects of carbs on happiness. One group of 11 people are assigned a high carb diet, another group of 11 people are assigned a moderate-high carb diet, another group of 11 people are assigned a moderate-low carb diet, and a final group of 11 people are assigned a low carb diet. If I find that I fail to reject the null hypothesis, what might my next step be?
Group of answer choices
Change my analysis and alpha to make the finding significant and then run post-hoc tests.
1. Run multiple dependent measures t-tests to see if there are any significant differences between particular groups
2. Run multiple independent samples t-tests to see if there are any significant differences between particular groups
3. Change my hypothesis to see if I can find something significant with a different hypothesis for this study
4. Run home and binge-watch Bridgerton because there is nothing else that should be done statistically
arrow_forward
A comparison between species: Biologists wish to compare the prevalence of a genetic mutation across
two species of space aliens. In a sample of 6 space aliens of Species I, 0.6% had the mutation while in a
sample of 14 space aliens of Species II, 1.1% showed the mutation. Is there significant evidence to suggest a
difference in the prevalence of the mutation across the two species? Conduct a hypothesis test at a
significance level of a = 0.01.
a) The hypotheses for this test are:
O Họ: PII = 0.006
Hạ: PII = 0.006
O Họ: PI - PII = 0
Hạ: PI - PII 0
What is the test statistic for this test?
test statistic =
(Report answer accurate to 2 decimal places.)
What is the p-value for this test?
p-value =
(Report answer accurate to 4 decimal places.)
The decision reached is
O Fail to reject Họ
O Reject Ho and accept Ha
As such, the final conclusion is that
O The prevalence of the mutation is the same for both species.
O There is significant evidence that suggests no difference in the prevalence…
arrow_forward
SEE MORE QUESTIONS
Recommended textbooks for you
Glencoe Algebra 1, Student Edition, 9780079039897...
Algebra
ISBN:9780079039897
Author:Carter
Publisher:McGraw Hill
College Algebra (MindTap Course List)
Algebra
ISBN:9781305652231
Author:R. David Gustafson, Jeff Hughes
Publisher:Cengage Learning
Related Questions
- According to a Health of Boston report, female residents in Boston have a higher average life expectancy as compared to male residents (The Boston Globe, August 16, 2010). You collect the following sample data to test the following hypotheses at 0.01 level of significance: Ho: H1s H2 versus H1: H1 >H2, where females represent group 1 and males represent group 2. Females 81.1 Males Мean 74.8 Sample Size 32 32 Assume the population standard deviations of life expectancy are 8.2 and 8.6 for females and males, respectively. What is(are) the critical value(s)? Select one: O None of these are correct 2.390 O +2.575 O 2.33 O +2.660arrow_forwardYou recently completed an experiment concerning the effects of carbs on happiness. One group of 11 people are assigned a high carb diet, another group of 11 people are assigned a moderate-high carb diet, another group of 11 people are assigned a moderate-low carb diet, and a final group of 11 people are assigned a low carb diet. If I find that I fail to reject the null hypothesis, what might my next step be? Group of answer choices Change my analysis and alpha to make the finding significant and then run post-hoc tests. 1. Run multiple dependent measures t-tests to see if there are any significant differences between particular groups 2. Run multiple independent samples t-tests to see if there are any significant differences between particular groups 3. Change my hypothesis to see if I can find something significant with a different hypothesis for this study 4. Run home and binge-watch Bridgerton because there is nothing else that should be done statisticallyarrow_forwardA comparison between species: Biologists wish to compare the prevalence of a genetic mutation across two species of space aliens. In a sample of 6 space aliens of Species I, 0.6% had the mutation while in a sample of 14 space aliens of Species II, 1.1% showed the mutation. Is there significant evidence to suggest a difference in the prevalence of the mutation across the two species? Conduct a hypothesis test at a significance level of a = 0.01. a) The hypotheses for this test are: O Họ: PII = 0.006 Hạ: PII = 0.006 O Họ: PI - PII = 0 Hạ: PI - PII 0 What is the test statistic for this test? test statistic = (Report answer accurate to 2 decimal places.) What is the p-value for this test? p-value = (Report answer accurate to 4 decimal places.) The decision reached is O Fail to reject Họ O Reject Ho and accept Ha As such, the final conclusion is that O The prevalence of the mutation is the same for both species. O There is significant evidence that suggests no difference in the prevalence…arrow_forward
Recommended textbooks for you
- Glencoe Algebra 1, Student Edition, 9780079039897...AlgebraISBN:9780079039897Author:CarterPublisher:McGraw HillCollege Algebra (MindTap Course List)AlgebraISBN:9781305652231Author:R. David Gustafson, Jeff HughesPublisher:Cengage Learning
Glencoe Algebra 1, Student Edition, 9780079039897...
Algebra
ISBN:9780079039897
Author:Carter
Publisher:McGraw Hill
College Algebra (MindTap Course List)
Algebra
ISBN:9781305652231
Author:R. David Gustafson, Jeff Hughes
Publisher:Cengage Learning