SPSS Assignment 7 Instructions

# SPSS Assignment 7 Instructions

Part 1

From Blackboard, download the SPSS HW #7-1 data file. Dr. F is interested in finding out if there is a difference in proportions of students at her campus who wash their hands after using campus bathrooms compared to proportions found in other research articles. Dr. F has found in the research literature that roughly 10% of students wash their hands correctly after leaving the bathroom. She is curious if her campus has the same percentage. Dr. F and her research team made over 200 observations over the course of a few days and recorded their findings.

Before we can begin the chi-square goodness of fit test, there are a few assumptions we need to test for:

There is only 1 variable. The variable can have several nominal/ordinal categories.

Only 1 person placed into each category (the same person cannot be placed into several categories).

There must be at least 5 people within each category.

Since we have met these assumptions, we can run the analysis.

Perform a Frequencies on – Handwashing.

 Handwashing Frequency Percent Valid Percent Cumulative Percent Valid Washed hands 35 16.8 16.8 16.8 Did not wash hands 173 83.2 83.2 100.0 Total 208 100.0 100.0

Click on “Analyze”…“Nonparametric Tests”…“Legacy Dialogs”…“Chi-square.”

When this pop-up pops up…

Select the variable…and move it…to the “Test Variable List” box.

You will notice that “All categories equal” has been selected as a default in SPSS. This means that SPSS is assuming your null hypothesis includes that each category contains the same frequency of people.

Just for fun, let’s pretend that our imaginary Dr. F is only curious if the proportion of those who wash and do not wash their hands at her campus are not equal (in other words, she is not interesting in comparing her campus to research findings in the literature). In this case, we just click “OK.”

In our example, we have 35 people who washed their hands correctly and 173 who did not. According to the SPSS default, our null hypothesis is being tested that there would be an equal number of people who wash and do not wash their hands correctly.

This would be the Chi-square statistic…and if this value is less than .05, it is statistically significant.

In our example, this would mean that there is a statistically significant difference in the proportion of those who wash their hands correctly and those who do not.

Now, let’s return to our original study, in which Dr. F is curious if the same proportion of students on her campus wash their hands like what has been found in the research studies.

Since we are not interested in equal proportions (the SPSS default), we want to select the “Values” option.

Then, you must first type in the expected value of what you labeled as #1 in the variable view of SPSS. In this case, we labeled #1 as those who wash their hands correctly. Since we are interested in comparing to what has been found in the research literature, we would type .10 (which is 10%)

Then, you must type in the expected value of what you labeled as #2 in the variable view of SPSS. In this case, we labeled #2 as those who do not wash their hands correctly. Since we are interested in comparing to what has been found in the research literature, we would type .90 (which is 90%)

Then, click “OK.”

In our example, our null hypothesis is being tested that we expect 10% of students who wash their hands correctly and 90% who do not.

 Handwashing Observed N Expected N Residual Washed hands 35 20.8 14.2 Did not wash hands 173 187.2 -14.2 Total 208

And we are comparing the expectations from the research literature to what Dr. F observed on her own campus.

This is the Chi-square statistic…and if this value is less than .05, it is statistically significant.

 Test Statistics Handwashing Chi-Square 10.771a df 1 Asymp. Sig. .001 a. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 20.8.

If the chi-square value is significant, it means that the null should be rejected. It means that observed values are different than the expected values. In other words, there is a difference between the proportions expected from the research literature and what is observed at the study at hand.

If the chi-square value is not significant, it means that the null should not be rejected. It means that observed values are not different than the expected values. In our example, the chi-square is statistically significant; it means that Dr. F observed in her study was written about in the research literature. There are roughly 17% who wash their hands correctly in her study, and that is significantly greater than the 10% found in the research.

Example of how to write up the results:

A chi-square goodness of fit test was performed in order to compare the observed proportion (83.2%) of students who do not properly wash their hands after using campus bathrooms with the expected proportion (90%) of students who did not properly wash their hands after using campus bathrooms as found in the literature. There statistically significant difference between the observed and expected proportions, X2(1, N = 208) = 10.77, p = .001.

Finally, export your output to Word and submit it to Blackboard.

To export, click the export symbol

…click “Browse,” and then save your document with a name and location that you will remember.

Grading rubric: This assignment is worth a total of 30 points and will be graded by the following criteria:

 Category Unacceptable Mediocre Excellent Complete The SPSS HW is missing 2 or more outputs that were required in the instructions AND/OR the SPSS HW was not submitted on time (0 points) The SPSS HW is missing 1 of the outputs that was required in the instructions (1-4 points) The SPSS HW contains all of the outputs that were required in the instructions (5 points) Correct There are 2 or more errors in the SPSS HW outputs (0 points) There is 1 error in the SPSS HW outputs (1-24 points) There are zero errors in the SPSS HW outputs (25 points)

Part 2

From Blackboard, download the SPSS HW #7-2. Dr. Z is interested in researching if there is a difference between elementary, middle, and high school students’ frequency of hand washing after using a public restroom. Dr. Z has 3 groups of people (elementary school students, middle school students, high school students) who have a dependent variable of hand washing frequency (those who wash their hands every time after they use the bathroom, those who wash their hands most times after they use the bathroom, those who wash their hands sometimes after they use the bathroom, and those who never wash their hands after they use the bathroom).

Before we can begin the Kruskal-Wallis H test, there are a few assumptions we need to test for:

The independent variable has 3 or more categories, and they are nominal/ordinal.

The independent variables are independent from each other (only 1 person placed into each category).

The dependent variable’s categories are nominal/ordinal or scale (if it is scale, it should not be normally distributed).

We have met these assumptions because the independent variable has 3 or more categories that are nominal/ordinal, the dependent variable’s categories are nominal/ordinal, and the variables are independent from each other (only 1 person placed into each category).

Let’s run the analysis…

Click on “Analyze”…“Nonparametric Tests”…“Legacy Dialogs”…“K Independent Samples.”

When this pop-up pops up…

Select the dependent variable…and move it…to the “Test Variable List” box.

Select the independent variable…move it…to the “Grouping Variable” box

… and then click on “Define Range.”

When this pop-up pops up…

Type in the lowest code for your independent category (1…which is elementary school students)

…and then type in the highest code for your independent category (3…which is high school students).

…and then click on “Continue.”

Make sure that the “Kruskal-Wallis H” test is selected

…and then click “OK.”

The “Test Statistics” table shows us the chi-square value and if it is statistically significant. If this value is less than .05, it is significant.

NOTE: Here are your degrees of freedom.

The “Ranks” table contains the “Mean Rank” for each independent variable.

Example write-up:

A Kruskal-Wallis H test was performed to determine if there was a statistically significant difference in frequency of handwashing between school levels (elementary, middle, and high school students). We found a statistically significant difference, χ2(2, N=300) = 25.481, p = < .05, with a mean rank score of 184.63 for elementary school students, 138.04 for middle school students, and 128.83 for high school students.

Part 3

Follow the instructions below using the HW # 7-3 dataset to perform a Survival Analysis.

Define event as 1.

Click “Options” and select as shown below.

The Output should appear as shown below.

Finally, export your output to Word and submit it to Blackboard.

To export, click the export symbol

…click “Browse,” and then save your document with a name and location that you will remember.

Grading rubric: This assignment is worth a total of 20 points and will be graded by the following criteria:

 Category Unacceptable Mediocre Excellent Complete The SPSS HW is missing 2 or more outputs that were required in the instructions AND/OR the SPSS HW was not submitted on time (0 points) The SPSS HW is missing 1 of the outputs that was required in the instructions (1-4 points) The SPSS HW contains all of the outputs that were required in the instructions (5 points) Correct There are 2 or more errors in the SPSS HW outputs (0 points) There is 1 error in the SPSS HW outputs (10 points) There are zero errors in the SPSS HW outputs (15) points)

Page 21 of 21