Measurement and Evaluation in Health Science - Chi-Square Data and Analysis - Assessment Answer

Measurement and Evaluation in Health Science Assignment

Assignment Task

You will complete a series of statistical analyses (Chi-Square, t-Test and ANOVA) on a set of variables. Some data sets will be provided (see below) but you are also encouraged to search out your own data via online databases (such as Australian Bureau of Statistics, Centre for Disease Control and Gapminder) or develop a real world data set. You are encouraged to use a data set that reflect your own professional interests.

Different aspect of the data set will be required for each statistical analysis.You must create a research question/scenario that can be answered using your data set as well as the required statistical analyses (Chi-Square, t-Test and ANOVA). Again, you are encouraged to develop scenarios that reflect your own professional interests.

The data set must contain a minimum of 30 cases (individuals) per variable and must be organised appropriately in an Excel spreadsheet. The data for the Chi Square can be derived from the same data set by organising the cases and/or variables into categories. Each data set and accompanying statistical analysis should be completed on a clearly labelled Excel sheet within one Excel file. The test results should then be reported in APA style in an accompanying Word document under the required headings (Chi-Square, t-Test and ANOVA).

You must upload 2 files (1 Excel File and 1 Word Document):

Excel File must contain:

1. Complete Data Set
2. Chi-Square Data and Analysis
3. t-Test Data and Analysis
4. ANOVA Data and Analysis

Word File must contain:

PART A: A brief written introduction to the scenario (max. 100 word) that reflects the data contained in the Excel File. The submission must include a concise explanation of the data, and the measures being used in the analysis reported in Parts B, C and D. Students are encouraged to use data that reflect their own professional interests.

PART B: You must submit a brief (max. 150 word) summary reporting the research question and the statistical findings of the Chi-Square relating to their scenario in APA style.

PART C: You must submit a brief (max. 150 word) summary reporting the research question and the statistical findings of the t-Test relating to their scenario in APA style.

PART D: You must submit a brief (max. 150 word) summary reporting the research question and the statistical findings of the ANOVA relating to their scenario in APA style.

You may use tables and/or figures to supplement their written report. Any tables and figures presented must be reported in APA style.

NOTE: Simply copying data from the Excel spreadsheet and embedding into the word documents will NOT suffice for this assessment piece. You are encouraged to provide the written data in a format that best highlights the story being told. If text is used to explain the data set then providing tables and graphs should not be used as they would represent a repetition of the same information. If a table is used then text and a graph of the same information would be repetitive, etc. You should only include tables and figures into the summary report if it adds to the information already being presented and does not repeat information. Tables and figures should also follow APA format.

Part A

The study mainly focuses on the health report of people living in Australia. The data was taken from Australian Institute of Health and Welfare. The data taken into consideration measures the health status of Australians, expenditure on health safety, major causes of ill health and treatments, Health precautions taken through life course, prevention and treatment on ill health and health system performance. Here, we wish to determine whether the Australian government has taken necessary steps to decline the trend of death rates over the years. The research hypotheses for this study are given below

Research Hypotheses

The research hypotheses are

• There is an association between disease and gender
• There is a significant difference in mean deaths per 100,000 population between male and female counterparts
• There is a significant difference in mean number of clients by principal drug of concern and age group

Part B – Chi Square Test

In order to determine whether there is an association between disease and gender, chi – square test for independence is performed. The null hypothesis states that there is no association between gender and disease against the alternate hypothesis that there is a significant association between gender and disease. The chi – square test statistic workings is given below

From the above table, we see that the value of chi – square test statistic (with 4 degrees of freedom) is 2346.756 and its corresponding p – value is 0.000 < 0.05. Since the p – value of chi – square test statistic falls well below 0.05, there is sufficient evidence to reject the null hypothesis at 5% level of significance. Therefore, we conclude that there is an association between gender and disease. About 40.2% of the males in Australia die due to Coronary heart disease and 18.22% of the males in Australia die due to Lung Cancer. on the other hand, 30.69% of the females in Australia die due to Coronary heart disease and 25.53% of the females in Australia die due to Dementia and Alzheimer disease. Thus, we see that Coronary heart disease is the common cause of death for many people in Australia. For males, Lung cancer also turns out to be a vital cause of death and for females, Dementia and Alzheimer disease turns out to be second vital cause of death almost equaling coronary heart disease

Part C – Independent Sample t test

In order to determine whether there is a significant difference in mean deaths per 100,000 populations between male and female counterparts, we perform independent sample t test. The null hypothesis states that there is no significant difference in mean number of deaths per 100,000 population between male and female counterparts against the alternate hypothesis that there is a significant difference in mean number of deaths per 100,000 population between male and female counterparts. The table given below shows the workings of independent sample t test statistic

t-Test: Two-Sample Assuming Equal Variances

From the above table, we see that the value of t test statistic is 7.021449 and its corresponding p – value is 0.000 < 0.05. Since the p – value of t test statistics falls well below 0.05, there is sufficient evidence to reject the null hypothesis at 5% level of significance and conclude that there is a significant difference in mean number of deaths per 100,000 population between male and female counterparts. Going through the mean values, it is found that the mean number of male death (1557.94 ± 468.71) is high when compared with female counterparts (1124.90 ± 432.80). The data was taken from 1907 to 2013 and it was found that, the number of male deaths is high in 1907 and it started to decline and reached 645.9 deaths per 100,000 population in 2013. Even though the death rates decreased considerably over the years, the proportion of male death is alarmingly high when compared with female counterparts

Part D – ANOVA

In order to determine whether there is a significant difference in mean number of clients by principal drug of concern and age group, we perform two way ANOVA. The ANOVA output is given below

ANOVA: Two-Factor Without Replication

Going through the ANOVA table, we see that the value of F test statistic for the main effect principal drug of concern is 5.24 and its corresponding p = value is 0.00354 < 0.05, indicating that there is a significant difference in the mean number of people affected with various types of principal drug of concern

Going through the ANOVA table, we see that the value of F test statistic for the main effect Age is 6.08 and its corresponding p = value is 0.0006 < 0.05, indicating that there is a significant difference in the mean number of people affected among different types of age groups

Going through the mean values, it is found that the mean number of people using Alcohol as principal drug of concern is high when compared with other categories of principal drug of concern followed by Cannabis and Amphetamines

Conclusion

The study mainly focuses on the health report of people living in Australia. The data was taken from Australian Institute of Health and Welfare. Regarding the distribution of age, about 54% of the people aged between 20 years and 39 years, 32% of the people aged over 40 years and 14% of the people aged between 10 years and 19years. Here, we see that there is a significant difference in the mean death rates between the male and female counterparts. Regarding the association between gender and disease, Coronary heart disease is the common cause of death for many people in Australia. For males, Lung cancer also turns out to be a vital cause of death and for females, Dementia and Alzheimer disease turns out to be second vital cause of death almost equaling coronary heart disease. Thus, we see that the most common disease that causes death to Australian population is Coronary heart disease.