# Free Statistical analysis Dissertation Example

0 / 5. 0 Category: Dissertation results

Subcategory: Dissertation Topics

Level: Masters

Pages: 3

Words: 825

Statistical Analysis

The table below shows the provision of the data in tabular form. The table has two variables.
Group
Population Group i Group ii Group iii Group iv
Oestrogen Level
1.3416
4.7104
2.0937
39.734
Progesterone Level
31.31
136.51
25.685
16.455
no plug
3 2 4 1
Restatement of Hypotheses in Symbol Form
From the data given above, the chi-square analysis gives the distribution of the variables and the statistical significance of the data. Planned number of animals and no plug does not affect group size is represented in the percentage form hence giving the fractional value for the effect of group size. The confidence level of the data is assumed to be 95% in the calculations. Planned number of animals means good group size is expected to be independent of the no plug results. Alternative hypothesis (Ha) shows that the high group size means the good planned number of animals depends on the size population being tested. Level of significance for the analysis is taken to be 0.05 while the degree of freedom as 4. Null hypothesis (Ha) assumed that planned number of animals means good group size independent of the plug results.
Probability test involves rejection region which is taken to have an area of 5.643 square units. Value of the level of significance for the data analysis is therefore to be less than the rejection region. There is not enough evidence that the evidence for the 5% significance. The variables group size and planned number of animals are assumed to be either dependent or independent before the hypothesis test. The degree of freedom for the data is taken as 4.Chi-square independence is assumed to give the correct dependency of the two variables (Rice, 224). Test statistic and critical values for the data gives the evidence for the hypothesis to be either null or alternative depending on the results of the chart.
Significance tests are necessary for the analysis of any data set with two variables. Level of significance is directly proportional to the degree of dispersion. Common methods of data analysis such as the median, average and standard deviation are related to the significance ratio. Rejection region gives the overview of the independence of variables. A large rejection region shows that the variables are not correlated (Rice, 223). Comparison of the variable and the rejection region gives the validity of the hypothesis. Chi-square records the frequency of variables and gives the critical value for the given set.
The degree of freedom gives provisional results on how the chosen variables relate. For example, how a given set of statistical data is tailed. The Chi-square produces contingency table to analyze the data easily. Significance level chosen should correspond to the type of the variables and their correlation.
Significance Test Calculations
A degree of freedom is given as (3-1) (3-1) = 4
Level of significance is given as 0.05
The rejection region = X> 5.643
X^2 = 12.692
12.692 > 5.63
The above calculation shows that the null hypothesis is rejected. Therefore, alternative hypothesis is valid. Level of significance is a tool that is used to rank parameters according to their closeness to the median. High ranked parameters are above median while the low ranked parameters fall below the median. Planned number of animals and plug results in each group does not affect the distribution of the results depending on the level of significance concept. Assignment of the parameters to various respondents is done in a manner that ensures the assignment remains optimum.
ANOVA Test

The table below shows the summary of the results for the four groups.
estrogen Progest Cycle
Group1 1.3416 31.31 596908445500
3.7631 71.69
17.774 14.311
6.216 8.7231
3.6214 84.718
4.6725 155.62
3.7082 22.264 no plug
2.9835 70.874
6.2084 6.9319
7.6024 12.136
8.8616 13.635
5.1157 15.13

Group2 4.7104 136.51

3.7971 171.43
11.947 73.702 146050-16637000
36.504   no plug
5.1935 3.8277
7.8263 7.3828
4.47 42.787
3.4837 93.425
8.7799 13.035
4.6549 83.913
9.1517 14.945
4.7129 150.12

Group 3 2.0937 25.685
7.4254 13.893 62865-11049000
10.511 14.097
5.088 136.51
6.5469 10.367
2.7406 13.174
9.0573 14.816
2.0036 121.31
2.7176 12.477
1.0174 8.0899
3.8752 158.66
5.8571 14.469

Group 4 18.551 11.693
no plug
15.121 60.173   no sample
14.799 4.9625   no plug
36.621 25.774 62230-58102500
16.713 82.325
24.719 4.5617
36.093 14.391   no plug, bleeding
31.618 5.8213
4.6296 10.753
16.547 3.72
18.658 57.715
39.734 16.455   no plug
Conclusion
Ways to improve your project: Experiment can be improved by having planned number of animals with a wide range of variables to record the accurate data. Participants should differentiate between estrogen and progesterone and how they affect the accuracy of the results. Again the project can be improved by use of the ARENA software gives the simulation of the planned number of animals and the group size. The simulation shows how the adopted parameters work and give the results regarding the cost incurred.
Calculations to show the order in which the parameters should be assigned to a different planned number of animals and group size. Allocations are done to the cells with the least improvement index.
Works Cited
Rice, William R. “Analyzing tables of statistical tests.” Evolution 43.1 (2015): 223-225.

All Examples