drill

Results:

The purpose of this study was to determine if the mean percent of time (in seconds) Lyle spent in self-aggression was significantly less on enrichment days compared to non-enrichment days. The purpose was also to determine if there was a significant difference in the mean percent of time Lyle spent in self-aggression on days with natural enrichment and days with unnatural enrichment. The hypotheses that were tested stated that enrichment would have a decreasing effect on the mean percent of time Lyle spent in self-aggression, and that there was a significant difference in the mean percent of time Lyle spent in self-aggression between days with natural enrichment and days with unnatural enrichment.

The data gathered for the Lyle were broken into two core groups: the mean percent of time Lyle spent in self aggression on days with enrichment, and the mean percent of time Lyle spent in self-aggression on days without enrichment. The data on days with enrichment was broken up even further, into two other groups: the mean percent of time Lyle spent in self-aggression on days with natural enrichment, and the mean percent of time Lyle spent in self-aggression on days with unnatural enrichment. A Z-test for the comparison of two independent sample means was run to determine if the mean percent of time Lyle spent in self-aggression was significantly less on days with enrichment compared to that of days without enrichment, and an F-test for variance was run to determine if there was a significant difference in the variances of those two sets of data. An F-test was also run on the natural enrichment vs. unnatural enrichment data to determine if there was a significant difference between those variances. From that, a T-test for two independent sample means assuming unequal variances was run to determine if there was a significant difference in the mean percent of time Lyle spent in self-aggression on days with natural enrichment compared to days with unnatural enrichment.

Figure 1: Difference Between Percent of Time Spent In Self-Aggression: Enrichment Vs. Non-Enrichment

This study was conducted at an alpha level of 0.05. Figure three illustrates the difference in the means of percent of time Lyle spent in self-aggression on enrichment days compared to the percent of time he spent in self-aggression on non-enrichment days. The figure shows that there was a difference in the two means, but the difference was not statistically significant according to the Z test.

Table 1: Results of Z test for percent of time Lyle spent in self-aggression: Enrichment Vs. Non-enrichment

	Enrichment Days	Non-Enrichment Days
Mean	0.16%	0.81%
Variance	0.00266	3.43 %
Sample Size	33	43
Hypothesized Mean Difference	0
Z Stat	1.22
Z Critical one-tail	1.645	Fail to Reject

The Z Statistic at an a=.05, which was 1.22, did not exceed the critical value of 1.645, therefore the null failed to be rejected, rendering the study inconclusive. This means that there was not enough evidence to support the claim that on enrichment days, the percent of time Lyle spent in self-aggression was significantly less than the percent of time Lyle spent in self-aggression on non-enrichment days.

Table 2: Results of an F-Test for Variance

F-Statistic	44.23
Critical Value	2.01
Alpha	0.05
	Reject Null

Table four contains the results from the F test for the comparison of two variances. The F Statistic at an a=.05 was 44.23, and the Critical Value was 2.01 . The F Statistic surpassed the Critical Value therefore the null was rejected. This means that there is enough evidence to support the claim that the variances are not equal. Pertaining to this study, this means that the data of the data for non-enrichment days is spread apart more than the data for enrichment days because the variance is significantly larger. This means that Lyle the male Drill did not act consistently on non-enrichment days with his self-aggression acts, leaving the study inconclusive.

Summary
For hypothesis one (enrichment days vs. non-enrichment days), an F-Test for variances and a Z-Test for means were ran. On the F-test, the F-Statistic did exceed the critical value; therefore the null hypothesis could be rejected. That meant that there was enough evidence to support the claim that the variances for the percent of time Lyle the male Drill spent in self-aggression on days with enrichment and on days without enrichment was not equal. On the Z-test, the Z statistic did not surpass the critical value; therefore the null hypothesis couldn’t be rejected. That meant that there was not enough evidence to support the claim that the mean percent of time Lyle the male Drill spent in self-aggression on days with enrichment is significantly less than the mean percent of time Lyle spent in self-aggression on days without enrichment.

Figure 2: Difference Between Percent of Time Spent In Self-Aggression: Natural Vs. Unnatural

This study was conducted at an alpha level of 0.05. Figure 2 illustrates the differences in the mean percent of time spent in self-aggression on days with natural enrichment compared to days with unnatural enrichment. The difference in the two means was not found to be statistically significant.

Table 3: Results of T-Test for Two Independant Sample Means: assuming unequal variances

	Natural	Unnatural
Mean	0.52%	0.23%
Variance	0.11%	450.2 %
Number of Eggs	14	19
Hypothesized Mean Difference	0
t Stat	1.1667
t Critical one-tail	2.09	Fail to Reject

The T Test for the comparison of two means assuming unequal variances results are provided in table eight. The T Statistic at alpha =.05 , which was 1.1667, did not exceed the critical value, which was 2.09, therefore the null hypothesis could not be rejected. This means that there was not enough evidence to support the claim that the average percent of time Lyle the male Drill spent in self-aggression was significantly less on days with natural enrichment when compared to days with unnatural enrichment. The study is inconclusive, as the null was not rejected.

Table 4. Results of F-Test for Variance

F-Statistic	39.005
Critical Value	3.18
Alpha	0.05
	Reject Null

Table seven contains the results from the F test for the comparison of two variances. The F Statistic at an alpha = .05 was 39.005, and the Critical Value was 3.18. The F Statistic surpassed the Critical Value therefore the null was rejected. This means that there is enough evidence to support the claim that the variances are not equal. Pertaining to this study, this means that the data for unnatural enrichment varies more than the data for natural enrichment. Lyle the male Drill did not act consistently on days with unnatural enrichment. This also means that the T-Test assuming unequal variances can be used.

Summary
For hypothesis two (natural enrichment vs. unnatural enrichment), an F-Test for variances and a T-Test for means assuming unequal variances were ran. On the F-test, the F-statistic exceeded the critical value; therefore the null hypothesis was rejected. That meant that there was enough evidence to support the claim that the variances for the percent of time Lyle the male Drill spent in self-aggression on days with natural enrichment and days with unnatural enrichment were not equal. On the T-Test, the T statistic did not exceed the critical value; therefore the null hypothesis couldn’t be rejected. That meant that there was not enough evidence to support the claim that the percent of time Lyle the male Drill spends in self-aggression is significantly less on days with natural enrichment compared to days with unnatural enrichment. In other words, natural enrichment did not seem to have a greater effect on decreasing Lyle’s self-aggression habits. In the population, the percent of time Lyle the male Drill spends in self-aggression will be between 0% and 0.11% on days with natural enrichment, and -0.07% and 0.54% on days with unnatural enrichment according to the confidence interval.