Reference no: EM132263672
Quantitative Research Methods In Geography Assignment - Inferential Analysis
INTRODUCTION - The purpose of this assignment is to familiarize students with the procedure and interpretation of two- and three-sample hypothesis testing. This will be accomplished by using data and SPSS to run the analysis, and the student will be expected to interpret the output.
ASSIGNMENT -
Section I - Two-Sample Hypothesis Testing (Difference of Means)
Using the 'World Country-Level Data Set' available in UMLearn, I would like you to compare the infant mortality rates in African countries compared to Asian countries. For this part of the assignment, make sure to use all African and all Asian countries (do not use one of the samples - A through D).
Research Question: Is the infant mortality rate in African countries statistically different than Asian countries?
Conduct both parametric and non-parametric tests, testing for differences in infant mortality rates between African and Asian Countries. In order to complete this test, all the values for both samples have to be in one column with a second column containing the grouping variable (i.e. which group or sample each value corresponds to). The grouping variables must be numbers (they can be any number you would like - i.e. 1 for African countries and 2 for Asian countries). Make sure the grouping variable is nominal (i.e. change it in the Variable View - TYPE if necessary).
To run a two-sample difference of means test (parametric) in SPSS:
- Select ANALYZE > COMPARE MEANS > INDEPENDENT-SAMPLES T TEST.
- Select the Test Variable (the column with the sample values) and Grouping Variable.
- Select DEFINE GROUPS and enter the value that corresponds with the first sample as Group 1 and the value that corresponds to the second sample as Group 2.
- Make sure OPTIONS and enter an appropriate confidence interval.
- Select OK.
For the parametric test results, answer the following questions:
Question 1: Summarize and compare the results from the test. In your answer,
i. State your null hypothesis and alternative hypothesis.
ii. Specify your significance level and critical statistic (you will need to get this value from a table).
iii. Attach the test results from SPSS.
iv. State whether the null hypothesis should be rejected or accepted based on p-value and test statistic and explain why.
v. Finally conclude regarding the research question.
To run a two-sample difference test (non-parametric) in SPSS:
- Select ANALYZE > NONPARAMETRIC TESTS > LEGACY DIALOGS > 2 INDEPENDENT SAMPLES.
- If a message appears regarding the measurement level, make sure to close the window and select the appropriate measurement level for the data in Variable View. The data should be SCALE.
- Select the Test Variable (the column with the sample values) and Grouping Variable. Make sure Mann-Whitney U is selected under Test Type.
- Select DEFINE GROUPS and enter the value that corresponds with the first sample as Group 1 and the value that corresponds to the second sample as Group 2.
- Select RUN.
For the nonparametric test results, answer the following questions:
Question 2: Summarize and compare the results from the test. In your answer,
i. State your null hypothesis and alternative hypothesis.
ii. Specify your significance level and critical statistic (you will need to get this value from a table).
iii. Attach the test results from SPSS.
iv. State whether the null hypothesis should be rejected or accepted based on p-value and test statistic and explain why.
v. Finally conclude regarding the research question.
Section II - Two-Sample Hypothesis Testing (Matched Pairs)
Using the "World Country -Level Data Set" provided in UMLearn, select one of the two random samples ('C' or 'D'). Make sure to record the population estimate for all countries in the random sample.
Research Question: Is the population estimate for 1998 significantly different than that for 2025.
Conduct both parametric and non-parametric tests, testing for differences in population estimates. In order to complete this test, make sure to place the 1998 population in one column and 2025 in a second column. You should add a variable name and make sure the variables are scale and not nominal (i.e. change it in the Variable View - TYPE if necessary).
To run a two-sample matched-pairs test (parametric) in SPSS:
- Select ANALYZE > COMPARE MEANS > PAIRED SAMPLES T TEST.
- Select the First Variable (the column with the 1998 sample values) and the Second Variable (the column with the 2005 sample values). You should have the analysis set up as: (see in attached file).
- Select OPTIONS and enter an appropriate confidence interval.
- Select OK.
For the parametric test results, answer the following questions:
Question 1: Summarize and compare the results from the test. In your answer,
vi. State your null hypothesis and alternative hypothesis.
vii. Specify your significance level and critical statistic (you will need to get this value from a table).
viii. Attach the test results from SPSS.
ix. State whether the null hypothesis should be rejected or accepted based on p-value and test statistic and explain why.
x. Finally conclude regarding the research question.
To run a two-sample matched-pairs test (non-parametric) in SPSS:
- Select ANALYZE > NONPARAMETRIC TESTS > LEGACY DIALOGS > 2 RELATED SAMPLES.
- If a message appears regarding the measurement level, make sure to close the window and select the appropriate measurement level for the data in Variable View. The data should be SCALE.
- Select the First Variable (the column with the 1998 sample values) and the Second Variable (the column with the 2005 sample values). You should have the analysis set up as follows. Also make sure to select the Wilcoxon Test Type.
- Select RUN
For the nonparametric test results, answer the following questions:
Question 2: Summarize and compare the results from the test. In your answer,
vi. State your null hypothesis and alternative hypothesis.
vii. Specify your significance level and critical statistic (you will need to get this value from a table).
viii. Attach the test results from SPSS.
ix. State whether the null hypothesis should be rejected or accepted based on p-value and test statistic (1 mark) and explain why.
x. Finally conclude regarding the research question.
SECTION III - Three or More Sample Hypothesis Test
Using the 'World Country-Level Data Set' available in UMLearn, I would like you to compare the Per Capita GNP between Asian, Africa and Latin America. For this part of the assignment, make sure to use all of the countries listed under those continents (do not use one of the samples - A through D).
Research Question: Is there a significant difference in GNP between continents.
Conduct both an analysis of variance (ANOVA) and Kruskal-Wallis test to determine whether differences exist in the Per Capita GNP for the three continents. In order to complete this test, all the values for the three samples have to be in one column with a second column containing the grouping variable (i.e. which group or sample each value corresponds to). The grouping variables must be numbers (they can be any number you would like - i.e. 1 for African countries, 2 for Asian countries and 3 for Latin American countries). Make sure the grouping variable is nominal (i.e. change it in the Variable View - TYPE if necessary).
To run a three or more-sample difference of means test (parametric) in SPSS:
- Select ANALYZE > COMPARE MEANS > ONE-WAY ANOVA.
- Select the variable you want to compare and place in the Dependent List box. You need to select a FACTOR. For this, make sure to have all the data in one column (for all three cities) and then a factor in the next column. These factors should be 1, 2, or 3 which corresponds to the three cities. The FACTOR needs to be a nominal data. This is the grouping variable.
- Select OK.
For the parametric test results, answer the following questions:
Question 1: Summarize and compare the results from the test. In your answer,
i. State your null hypothesis and alternative hypothesis.
ii. Specify your significance level and critical statistic.
iii. Attach the test results from SPSS.
iv. State whether the null hypothesis should be rejected or accepted based on p-value and test statistic and explain why.
v. Finally conclude regarding the research question.
To run a three or more-sample difference test (non-parametric) in SPSS:
- Select ANALYZE > NONPARAMETRIC TESTS > LEGACY DIALOGS > K INDEPENDENT SAMPLES.
- Select the Test Variable (the column with the sample values) and Grouping Variable. Make sure Kruskal-Wallis H test is selected under Test Type.
- Select DEFINE GROUPS and enter the value that corresponds with the first sample as Group 1, the value that corresponds to the second sample as Group 2, and the value that corresponds to the third sample as Group 3. Enter 1 under Minimum and 3 under Maximum.
- Select OK
For the nonparametric test results, answer the following questions:
Question 2: Summarize and compare the results from the test. In your answer,
i. State your null hypothesis and alternative hypothesis.
ii. Specify your significance level and critical statistic.
iii. Attach the test results from SPSS.
iv. State whether the null hypothesis should be rejected or accepted based on p-value and test statistic and explain why.
v. Finally conclude regarding the research question.
Attachment:- Assignment Files.rar