Assignment: Apply Statistical Tests

Assignment: Apply Statistical Tests
Assignment: Apply Statistical Tests
As a current or future health care administration leader, how will you decide which statistical test is most appropriate for the goal of monitoring, tracking, or overseeing operations in your health services organization?
Each statistical test is dependent on a given set of research parameters and expectations. As you have examined throughout this course, understanding why and when to use each statistical test is necessary in conducting the tests for process control and comparison. Thus, chi-square tests are useful for table data, ANOVA for a single quantitative dependent variable and categorical independent variables, and ANOM for analysis of mean differences. Regression modeling is useful for determining the influence of independent variables on a quantitative dependent variable.
For this Assignment, review the resources for this week regarding chi-square, ANOVA, ANOM, and regression. Pay particular attention to the examples shown in the textbook.
The Assignment: (4– pages)
Using SPSS and Microsoft Word, complete problems 1 through 10 on pages 367–370 in the Ross textbook. Show all work. Submit both your SPSS and Word files for grading.
Book to use
https://ereader.chegg.com/#/books/9781118603642/cfi/362!/4/[email protected]:0.00
Will send login details through the chat.thank you
What is the purpose of a statistical test?
A test statistic is a number that describes how much the relationship between variables in your test differs from the null hypothesis of no association.
The p-value is then calculated (probability value).
If the null hypothesis of no link were true, the p-value estimates how probable it is that you would see the difference specified by the test statistic.
You can infer a statistically significant association between the predictor and outcome variables if the test statistic’s value is more extreme than the null hypothesis’s statistic.
If the test statistic’s value is less extreme than the null hypothesis’s, you can conclude that there is no statistically significant association between the predictor and outcome variables.
When should you use a statistical test?
Statistical tests can be performed on data that has been obtained in a statistically valid manner, such as through an experiment or observations made using probability sampling methods.
Your sample size must be large enough to approximate the true distribution of the population being investigated for a statistical test to be valid.
To figure out the statistical test to use, you’ll need to know the following:
whether your data is consistent with particular assumptions
the several types of variables you’re dealing with
Statistical hypotheses
Statistical tests make a few assumptions about the data they’re looking at:
No autocorrelation (observation independence): The observations/variables you consider in your test are unrelated (for example, multiple measurements of a single test subject are not independent, while measurements of multiple different test subjects are independent).
Homogeneity of variance refers to the fact that the variance within each of the groups being compared is similar across all of them.
The test’s usefulness will be limited if one group has significantly greater variation than the others.
Data normality: the data has a normal distribution (a.k.a. a bell curve).
This assumption only applies to numerical data.
If your data do not meet the normality or homogeneity of variance assumptions, you may be able to use a nonparametric statistical test to make comparisons without making any assumptions about the data distribution.
If your data does not meet the assumption of observation independence, you might be able to perform a test that takes into consideration the structure of your data (repeated-measures tests or tests that include blocking variables).
Variables of various types
What type of statistical test you can employ is usually determined by the types of variables you have.
Quantitative variables are measurements of quantities (e.g. the number of trees in a forest).
The following are examples of quantitative variables:
Continuous variables (also known as ratio variables) represent measures that can be divided into smaller units than one (e.g. 0.75 grams).
Discrete variables (also known as integer variables) indicate counts that can’t be split into smaller units than one (e.g. 1 tree).
Categorical variables are a type of variable that represents groups of objects (e.g. the different tree species in a forest).
Categorical variables include the following:
Ordinal data is data that is represented in a specific order (e.g. rankings).
Nominal: used to express the names of groups (e.g. brands or species names).
Binary: a yes/no or 1/0 outcome is used to represent data (e.g. win or lose).
Select a test that is appropriate for the types of predictor and result variables you have gathered (if you are doing an experiment, these are the independent and dependent variables).
To see which test best suits your variables, look at the tables below.