PART I
a. The hypotheses for testing if barking deer prefer to forage in
certain habitats over others can be stated as:
- Null Hypothesis (H0): Barking deer do not prefer foraging in
different habitats; the observed frequencies of foraging in specified
habitats are consistent with their availability.
- Alternative Hypothesis (Ha): Barking deer prefer foraging in
certain habitats; the observed frequencies of foraging in the
specified habitats are not consistent with their availability.
b. The type of test that can be used to answer this research
question is the Chi-squared test for independence.
c. To check if the assumptions and conditions required for the Chi-
squared test are satisfy.
1. Random sampling: The data should be based on a random
sample of foraging sites.
2. Expected frequencies: Each expected frequency should be
at least 5 for the Chi-squared test to be valid.
3. Independence: The observations should be independent of
each other.
Given the data provided, we can calculate the expected frequencies
as follows:
Total number of sites = 530.
Proportions of habitat types:
Woods: 0.048 × 530 = 25.44
Cultivated grassplot: 0.147 × 530 = 77.91
Deciduous forests: 0.396 × 530 = 210.48
Other: 530 − (25.44 + 77.91 + 210.48 )=216.17
The expected frequencies are approximately:
Woods: 25.44
Cultivated grassplot: 77.91
Deciduous forests: 210.48
Other: 216.17
Since all expected frequencies are greater than 5, the assumptions
are satisfied.
d. To conduct the Chi-squared test, we can calculate the Chi-
squared statistic as follows:
𝜒 2 =∑ ( 𝑂 𝑖 − 𝐸 𝑖 ) 2 𝐸 𝑖
Where 𝑂 𝑖 are the observed frequencies and 𝐸 𝑖 are the expected
frequencies.
We have the observed values:
Woods: 𝑂 1 = 6
Cultivated grassplot: 𝑂 2 = 18
Deciduous forests: 𝑂 3 = 71
Other: 𝑂 4 = 435
And the expected values:
Woods: 𝐸 1 = 25.44
Cultivated grassplot: 𝐸 2 = 77.91
Deciduous forests: 𝐸 3 = 210.48
Other: 𝐸 4 = 216.17
And the expected values:
Woods: 𝐸 1 = 25.44
Cultivated grassplot: 𝐸 2 = 77.91
Deciduous forests: 𝐸 3 = 210.48
Other: 𝐸 4 = 216.17
Calculating the Chi-squared statistics:
Where r is the number of rows (habitats) and c is the number of columns (preferences). In this
case, we have 4 habitats, so df=4-1=3
Given that the p-value < 0.001, we can conclude that the data provides strong evidence against
the null hypothesis. Therefore, we reject the null hypothesis and conclude that barking deer
prefer to forage in certain habitats over others.
PART II
a. To calculate the actual number of site visitors in each of the six
response categories, we can apply the percentages to the total
number of site visitors, which is 501.
1. Position 1, Download: Visitors = 501 × 0.16 = 80.16 ≈ 80
2. Position 1, No Download: Visitors = 501 × 0.209 = 104.659 ≈
105
3. Position 2, Download: Visitors = 501 × 0.148 = 74.748 ≈ 75
4. Position 2, No Download: Visitors = 501 × 0.212 = 106.812 ≈
107
5. Position 3, Download: Visitors = 501 × 0.119 = 59.619 ≈ 60
6. Position 3, No Download: Visitors = 501 × 0.152 = 76.152 ≈
76
Therefore, the number of visitors in each category is:
- Position 1, Download: 80 - Position 1, No Download: 105
- Position 2, Download: 75 - Position 2, No Download: 107
- Position 3, Download: 60 - Position 3, No Download: 76
b. To check if the groups were imbalanced, we set up our
hypotheses as follows:
Null hypothesis (𝐻 0): The distribution of visitors among the
groups is balanced.
Alternative hypothesis (H a): The distribution of visitors
among the groups is not balanced.
To perform a Chi-square test for goodness of fit, we use the
expected counts for each group if they were balanced.
With a total of 501 visitors and 3 groups, if balanced, each group
would have 501/3=167 visitors.
The conclusion states that the placement of the download link
significantly impacted on the number of downloads. The difference
in downloads observed between the different placements is large
enough to be considered statistically significant, meaning it's
unlikely to have occurred by random chance. Therefore, the groups
being assessed (different download link placements) were
imbalanced in their effectiveness.
References
Diez, D., Cetinkaya-Rundel, M., Barr C. D., & Barr, C. D.
(2019). OpenIntro statistics - Fourth edition. Open Textbook
Library. Please access it from the homepage.
Goss-Sampson, M. A. (2022). Statistical analysis in JASP: A guide
for students (5th ed., JASP v0.16.1 2022).
