Aryan Marwah, Studentid: S223162057 Address:- Aryan Marwah 2/17 Johnston st Burwood VIC 3125 Australia
Make sure to delete the instructions!!
Introduction:The data taken into consideration represents the bat population counts. The bat count normally represents the number of bats found in the roost sites. Normally, the bats was found to be most commonly seen in abandoned houses, church steeples and also their common occupied structures. The regression model was used to predict the bat population
Section 1:
Q1: EXPLANATION OF THE Hypothesis Testing
Null Hypothesis: H0: µ = 100
That is, the mean bat counts for Good tree health do not differ from 100
Alternate Hypothesis: Ha: µ ≠ 100
That is, the mean bat counts for Good tree health differ from 100
Null Hypothesis: H0: µ = 100
That is, the mean bat counts for Fair tree health do not differ from 100
Alternate Hypothesis: Ha: µ ≠ 100
That is, the mean bat counts for Fair tree health differ from 100
Null Hypothesis: H0: µ = 100
That is, the mean bat counts for bad tree health do not differ from 100
Alternate Hypothesis: Ha: µ ≠ 100
That is, the mean bat counts for bad tree health differ from 100
The one sample z test was used to test the claim and the p - value of z test statistic falls above 0.05 for the three tests, indicating that there is no sufficient statistical evidence to conclude that the average Bat Population is not 100 bats per location of the three locations
Q2: EXPLANATION OF THE Correlation Analysis
The correlation coefficient was performed for dependent variable bat population count against all numerical independent variables. It was found that location size and tree numbers tend to have significant and strong relationship with bat population count. In turn, it was observed that both location size and tree numbers have strong relationship between them. Therefore, any one of the two variables which has strong relationship with bat population count will be included in the regression analysis. Therefore, tree numbers tend to have strong relationship with bat population count and therefore it was included in the regression analysis. The other independent variables included in the regression analysis are predators low, predators high and sprinkler yes
Q3: EXPLANATION OF THE Regression Analysis
It was found that location size and tree numbers tend to have significant and strong relationship with bat population count. In turn, it was observed that both location size and tree numbers have strong relationship between them. Therefore, any one of the two variables which has strong relationship with bat population count will be included in the regression analysis. Therefore, tree numbers tend to have strong relationship with bat population count and therefore it was included in the regression analysis. The other independent variables included in the regression analysis are predators low, predators high and sprinkler yes
The regression equation is
Bat Count = 73.88 + 5.64 * Tree Numbers - 10.27 * High Degree of Predators
+ 7.07 * Low degree of predators + 7.482 * Sprinkler Yes
The f test statistic for overall model significance was tested and the p - value was found to be less than 0.05, indicating that regression model derived is good fit in predicting bat count.
The regression coefficient Tree number is 5.64, indicating that on holding other independent variables constant, the bat count increases by 5.64 bats per location for every one additional tree count and statistically significant
Section 2 - recommendations TO MAXIMISE Bat population
The regression analysis found that
1. To increase the number of trees in the bat location
2. To increase the number of low degree of predator presence in the location
3. Presence of the Irrigating system which helps to cool the location
Recommendation 1
Explanation of Recommendation 1
It has been observed that there exists strong positive linear relationship between tree number and bat population (correlation coefficient is 0.582, p < 0.05), it was highly recommended to plant maximum number of trees that certainly helps in increasing the bat count
Recommendation 2
Increasing the presence of low predators in the location certainly increase the bat population
Explanation of Recommendation 2
The coefficient of low predators was found to be 7.07, certainly shows that the bat population increases by 7.07 bats in the location when there are low degree of predator presence in the location, provided other independent variables hold constant. Thus, low degree of predators certainly increases the bat population count
Conclusion:
Thus, the study findings concludes that the presence of sprinkler system, low degree of predators presence and increasing the number of trees in the location certainly have a positive impact on the bat population in the location