Civil War Resolution Data Set t-test: Quantitative Research Methods in Political Science - Statistics Project Example

Civil War Resolution Data Set t-test (Quantitative research methods in political science) Our beautiful earth since its ‘origination’ has been a ‘hotbed’ of activity. Activity in the sense, we humans has been involved in various activities, utilizing our physical as well as the mental part. The activities only transformed our lives which was bordering on primitiveness to all round development. But, these activities apart from development have also caused destructions. Destruction in the sense, people in the form of countries as well as groups has ‘assaulted’ other human lives, even killing many of them. That is, ever since humans originated, they have ‘segregated’ themselves into different nations or groups. Each country or groups want to top or dominate all spheres of human life. So, a group’s army in association with other armed groups only join ‘hands’ together, to protect the territorial integrity and to gain superiority. Among these two objectives, the wish to gain superiority over other groups has been the mentality of many groups and this has led to wars from time immemorial. When many nations or groups of the world have this objective and the other group of nations tries to protect their territory from these nations, it would result in civil wars. Along with these two major objectives, many other factors would also push groups to initiate Civil Wars. So, this paper will discuss and analyze the factors which contributed to the Civil War from a statistical point of view. This part of the paper explains how each of the hypotheses were presented and operationalized with a series of testing. The main aim is to make the testing as apparent as possible. So, that the abstract concepts such as identity, divisibility and agreement proportions were matched with concrete indicators and carried out with apt statistical analysis. This transparency gives more scope to replicate the statistical analysis presented here and fill in the holes that sometimes invariably happen and remain. Measuring variables is by nature an irregular and imperfect procedure as some of the indicators are close to tracing a particular variable, while others are inexact approximations. The literature on the statistical analysis of civil wars has generally assumed that negotiations fail because a conflict is not ripe for resolution or because the combatants cannot agree on the terms for peace. It does not consider that combatants may look down the road and realize that any terms they do accept may be unenforceable and therefore undesirable. This chapter builds a theory of civil war settlement that focuses on the problems combatants’ face enforcing and credibly committing to the terms of a peace agreement. The two specific factors—third party guarantees to protect combatants as they demobilize and power sharing guarantees in the first postwar government—ultimately deter mine whether groups will sign and implement settlements. Null Hypothesis: Mean regimetry between the two ethnic groups 0 and 1 do not differ significantly Alternative Hypothesis: Mean regimetry between the two ethnic groups 0 and 1 differ significantly Level of Significance: 5% level Descriptives ETHNIC = 0 N Minimum Maximum Mean Std. Deviation REGIMETY 37 -10 10 -2.27 6.560 Valid N (listwise) 37 ETHNIC = 1 N Minimum Maximum Mean Std. Deviation REGIMETY 35 -9 10 -.40 5.735 Valid N (listwise) 35 Independent Samples Test Levenes Test for Equality of Variances t-test for Equality of Means F Sig. T df Sig. (2-tailed) Mean Difference Std. Error Difference REGIMETY Equal variances assumed 1.073 .304 -1.285 70 .203 -1.870 1.456 Equal variances not assumed -1.290 69.581 .201 -1.870 1.450 The mean regimetry of the ethnic group 0 is -2.27 and that of the ethnic group 1 is -0.4. The mean difference between the two ethnic groups is -1.87 with p value 0.203 Considering the above the mean of ethnic group 0 is highly negative with -2.27 where as the mean of ethnic group 1 is little bit low with mean -0.40. Inference Since the probability value is 0.203 which is greater than 0.05, we say that there is no significant difference between the regimetry between the two ethnic groups at 5% level. Both the ethnic groups behave as the same with respect to regimetry. Null Hypothesis: Mean duration between the two ethnic groups 0 and 1 do not differ significantly Alternative Hypothesis: Mean duration between the two ethnic groups 0 and 1 differ significantly Level of Significance: 5% level Descriptives Group Statistics ETHNIC N Mean Std. Deviation Std. Error Mean DURATION 0 37 66.30 84.839 13.947 1 35 81.26 85.042 14.375 Independent Samples Test Levenes Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2-tailed) Mean Difference Std. Error Difference DURATION Equal variances assumed .418 .520 -.747 70 .458 -14.960 20.028 Equal variances not assumed -.747 69.759 .458 -14.960 20.029 The mean duration of the ethnic group 0 is 66.3 and that of the ethnic group 1 is 81.26. The mean difference between the two ethnic groups is -14.96 with p value 0.458 Considering the above the mean of ethnic group 0 is lower than the mean of ethnic group 1 is with mean 81.26. Inference Since the probability value is 0.458 which is greater than 0.05, we say that there is no significant difference between the duration between the two ethnic groups at 5% level. Both the ethnic groups behave as the same with respect to duration. Null Hypothesis: The two factors TotalG and Ethnic are not associated with each other. Both factors are independent. Alternative Hypothesis: The two factors TotalG and Ethnic are associated with each other. Both factors are dependent. Level of Significance: 5% level TOTALG * ETHNIC Crosstabulation ETHNIC Total 0 1 TOTALG 0 7 20 27 1 30 15 45 Total 37 35 72 Chi-Square Tests Value Df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 11.212a 1 .001 Continuity Correctionb 9.641 1 .002 Likelihood Ratio 11.568 1 .001 Fishers Exact Test .001 .001 Linear-by-Linear Association 11.057 1 .001 N of Valid Casesb 72 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 13.13. b. Computed only for a 2x2 table Inference Since the probability value is 0.001 which is less than 0.05, we say that there is an association between TotalG and ethnic. Both factors are dependent on each other. Null Hypothesis: The two factors Territor and Ethnic are not associated with each other. Both factors are independent of each other. Alternative Hypothesis: The two factors Territor and Ethnic are associated with each other. Both factors are dependent on each other. Level of Significance: 5% level TERRITOR * ETHNIC Crosstabulation Count ETHNIC Total 0 1 TERRITOR 0 34 15 49 1 3 20 23 Total 37 35 72 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 19.892a 1 .000 Continuity Correctionb 17.701 1 .000 Likelihood Ratio 21.581 1 .000 Fishers Exact Test .000 .000 Linear-by-Linear Association 19.616 1 .000 N of Valid Casesb 72 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 11.18. b. Computed only for a 2x2 table Inference Since the probability value is 0.001 which is less than 0.05, we say that there is an association between Territor and ethnic. Both factors are dependent on each other. Null Hypothesis: The two factors Ongoing and Ethnic are not associated with each other. Both factors are independent of each other. Alternative Hypothesis: The two factors Ongoing and Ethnic are associated with each other. Both factors are dependent on each other. Level of Significance: 5% level ETHNIC * ONGOING Crosstabulation ONGOING Total 0 1 ETHNIC 0 33 4 37 1 24 11 35 Total 57 15 72 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 4.636a 1 .031 Continuity Correctionb 3.470 1 .062 Likelihood Ratio 4.768 1 .029 Fishers Exact Test .043 .030 Linear-by-Linear Association 4.571 1 .033 N of Valid Casesb 72 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 7.29. b. Computed only for a 2x2 table Inference Since the probability value is 0.031 which is less than 0.05, we say that there is an association between Ongoing and ethnic at 5% level of significance. Both factors are dependent on each other. Null Hypothesis: The two factors Ongoing and Borderli are not associated with each other. Both factors are independent of each other. Alternative Hypothesis: The two factors Ongoing and Borderli are associated with each other. Both factors are dependent on each other. Level of Significance: 5% level BORDERLI * ONGOING Crosstabulation Count ONGOING Total 0 1 BORDERLI 0 47 9 56 1 10 6 16 Total 57 15 72 Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson Chi-Square 3.465a 1 .063 Continuity Correctionb 2.287 1 .130 Likelihood Ratio 3.145 1 .076 Fishers Exact Test .084 .070 Linear-by-Linear Association 3.417 1 .065 N of Valid Casesb 72 a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 3.33. b. Computed only for a 2x2 table Inference Since the probability value is 0.063 which is greater than 0.05, we say that there is no association between Ongoing and Borderli at 5% level of significance. Both factors are independent on each other. Read More