Psychology

profileStephDiaz
Wee5Day2.pptx
Home>Psychology homework help>Psychology

Week 5, Day 2

PSY 3215 U01B 1215, Lab

Lab Overview

Part I: Data & Analysis Relevant to your Paper

Starting an SPSS File

Importing Data from Excel

Selecting the Right Data for your Study

Running your Analysis

Part II: More Advanced Data Analysis

One-way ANOVAs

Tukey’s Test

Part I: Data & Analysis Relevant to your Paper

Part I

For your final paper (Experimental Paper II), you will need to take data Dr. Bonnin has created for you in an excel sheet and:

1: Select which data will be appropriate for your study

2. Transfer this data to SPSS

3. Run the appropriate analyses on it.

To this end, this section of the lab slides will cover how to accomplish this.

Part I Starting an SPSS File

To begin, you’re going to have to create an SPSS file so you have something to move the data from excel into.

To begin, when you launch SPSS, go to the [New Dataset Option] under the New files*

*For this example, I’ll be working with SPSS on a mac computer. If you’re having trouble replicating my steps on Windows, please reach out to me.

Part I The blank SPSS file

Your new SPSS file should look like this; blank and empty

We’ll come back to this later. For now, we’re going to look at the experimental data for your final paper provided by Dr. Bonnin

Part I Accessing and inspecting the Paper data

To begin, you will need to download the “Experimental Paper II – Data” under Canvas Modules. You can find this under the Paper Data module.

Once you’ve downloaded the data, open it up. Your view should look like this

Part I Get the data file to the right view

Now go to the bottom of the excel page

Click on the “DV Data” tab to switch to the first panel

Part I Understanding the layout of the data

The data from the study is broken up into two parts, with one part having two sub-parts.

In the ”DV Data” tab, on the left there is the two-group comparison data

This is the data you will select if your study was looking at two groups (experimental/quasi-experimental where you had a treatment group and either a control group or a different group – possible a standardized treatment group)

To the right there is the end, or DV, of the a pre-post test data set

This is the part with two sub-groups

If you are doing a pre-post test dataset, then this data on the right will be your post-test data

Part I Understanding the layout of the data

If you go to the second tab on the excel sheet ( ) then you will see two other selections of data related to the pre-post test data:

Both of these are pre-test data

The left selection is for dependent variables that are supposed to decrease after treatment

The right selection is for dependent variables was supposed to increase after treatment

Part I Working with a hypothetical study

For the purpose of explaining the data, I will be proposing a hypothetical study

This hypothetical study has the following parameters:

IV -Two levels; depressed people before adopting a cat and depressed people six months after adopting a cat

DV – Depression measure

Part I Selecting the right data for your study

For your own study that you report on in Experimental Paper II, you will need to select the correct dataset for your study.

This starts by understanding the design of your study and selecting the correct data from the Experimental Paper II – Data excel file

You’ll need to ask your self the following questions:

Does my study consist of two groups?  Yes/No

If no, then you should have a pre/post-test study. If you think you do not, please reach out to me to ask about your study.

If you study does not have two groups and is a pre/post-test design, ask yourself: Should the measure of my study go up from pre-test to the post-test?  Yes/No

Part I Selecting the right data for your study

If you said yes to “Does my study consist of two groups?”, then you’ll be using the left data on the first tab (DV Data) for your experiment. This the data from columns A, B & C

If you said no, then your study will use the data from the two columns on the left in the first tab (column H) , and either:

The left data from column B & C on tab 2 (PRE-POST Data); Only if your data is supposed to decrease from the pre-test to the post-test

The right data from column G & H on tab 2 (PRE-POST Data); only if your data is supposed to increase from the pre-test to the post-test

Part I Selecting the Data for our hypothetical study

Our hypothetical study had: a pre/post-test dataset that is supposed to decrease (depression scores in the face of treatment using cat adoption)

That means we select the left data from tab 2 (pretest) & the right data from tab 1 (posttest)

Pre-test Data

Post-test Data

Part I Selecting the right data for your study

NOTE: YOU ARE RESPONSIBLE FOR SELECTING THE RIGHT DATA FOR YOUR OWN STUDY

THE DATA SELECTED FOR THIS HYPOTHETICAL STUDY IS AN EXAMPLE AND MAY NOT MATCH THE TYPE OF STUDY YOU ARE DOING

PART I PREPARING SPSS TO RECEIVE THE EXCEL DATA

Now that you’ve figured out which data you’ll be using, it’s time to go to SPSS to get it ready to receive your data

You can’t just copy data from excel into an empty SPSS file; you have to set up the variables first

To do this, you need to go to the ”Variable View” tab at the bottom of the blank SPSS page and select it

PART I PREPARING SPSS TO RECEIVE THE EXCEL DATA

When you go to Variable View, you will see a blank set of variables at the top; you will need to fill these in

Start by filling in the first variable, calling it ”SubjID”, short for Subject ID

Once you fill in the first variable name, the rest of the parameters should auto-fill. These are fine, leave them as is

PART I PREPARING SPSS TO RECEIVE THE EXCEL DATA

Next, fill in the next two variables as preResults (Short for “Pretest Results”) and postResults (Short for “Prosttest Results”)

These names are for the HYPOTHETICAL STUDY examining adopting cats treating depression. Your own study may vary (if you are doing two groups, use the two groups)

PART I PREPARING SPSS TO RECEIVE THE EXCEL DATA

You’ve set up your variables!

Now, switch back to Data View in SPSS

You’re ready to copy and paste the excel data into SPSS!

Part I Moving the data from SPSS to Excel

Select the appropriate columns in excel and copy and paste them into SPSS

For the hypothetical study, we’ll be copying and pasting from the left data column on the PRE-POST DATA tab & the right columns on the DV Data tab

Part I Moving the data from SPSS to Excel

This is what a sample of the data will look like in SPSS

Part I Comparing the data

Now, in order to analyze this data, whether it is a pre-post test study like in the hypothetical example or a group1 group 2 study, you’ll need to use the:

Paired-Samples T-test

Part I Comparing the data in SPSS

To do this in SPSS, do the following

First, go to Analyze  Compare Means  Paired-Samples T-test

Part I Comparing the data in SPSS

Once you’ve started, proceed to select your two groups and move them to the paired variables section by clicking on the arrow in the center.

Then hit OK to run!

Part I Analyzing the data in SPSS

When you hit run, a new window from SPSS will pop up; the output window which looks like…

Part I Analyzing the data in SPSS

From this window, you’ll get important numbers to put in your results. Things like:

The means and standard deviation of each group

The mean difference between each group (important to talk about in your results & interpret in your conclusion!)

Your t value!

Your p-value!

Your degrees of freedom (df)

All of these will be very important when you’re writing your results & conclusion!

Part II: More Advanced Data Analysis

Part II New analyses

In addition to the paired-samples t-test which can compare the results of a dependent variable between two groups, there are other tests that can be run to analyze 2 or more groups

This section of the lab will cover some of those tests. I will provide you with the test, an explanation of what it does and the theory behind it, and then finally how to practically implement the test in SPSS

Part II Getting started

To begin, download the in class practice data under canvas modules

Modules  Example Data  In-Class Practice Data

Once downloaded, open the file in SPSS

Part II Understanding the dataset

IT IS IMPORTANT THAT YOU READ THE EXPLAINATION TO UNDERSTAND THE DATA

NOT READING THIS EXPLAINATION MAY RESULT IN AN INABILITY TO COMPLETE THE LAB ACTIVITY

In the examples in Part II of this lecture, you’ll be looking a hypothetical study that examined how people rated the flavor of ice cream from two different stores in a city

Part II Understanding the dataset

Looking at the data in SPSS in Variable View, you will see the following

These represent the variables in the example study for Part II, of which you will be using three

Part II Understanding the dataset

subj represents the subject’s number in the study, and isn’t useful for our analyses

flavor represents the flavor of ice cream our subject ate and then rated. There were three flavors

1 represents Chocolate

2 represents Vanilla

3 represents Strawberry

Part II Understanding the dataset

store represents one of two stores the subject bought the ice cream at: store 1 or store 2

Rating represents how much the subject responded when asked “How much did you enjoy the ice cream?”

1 – Not at all

2 – I enjoyed it a little

3 – I enjoyed it a moderate amount

4 – I enjoyed it

5 – I enjoyed it a lot!

Part II Looking at more than two groups

In the dataset provided for Experimental Paper II, you are selecting data to do a comparison with two groups; either a pre-test & a post-test of the same groups or a group A & B, most likely in an experimental study condition

However, in science, it is common to have more then two groups

We need statistics that let us analyze more than two groups

So in this case, when we are trying to measure One DV against an IV with Three or More Groups, we will use the One-Way ANOVA

Part II One-way ANOVAs

ANOVA is short for) ANalysis Of VAriance. The ANOVA compares the means & variances (a measure of the average difference between the mean of the group and any given number from that group) to answer the question:

Is there a difference in the dependent variable depending on the IV group?

Part II One-way ANOVA for our example study

For our study, the one-way ANOVA will help us answer the question: Was there a difference in how much people enjoyed ice cream based on the flavor?

For this question, how much people enjoyed the ice cream is the DV, measured by rating

flavor is the IV with three groups (Chocolate, Vanilla and Strawberry)

Part II One-way ANOVAs: Running in SPSS

In order to run the ANOVA in SPSS, follow these instructions using the downloaded In-Class Practice Data

Go to Analyze  Compare Means  One-Way ANOVA and click

Part II One-way ANOVAs: Running in SPSS

In the window that pops up, move your variables in the following way

rating (how much people liked their ice cream, the DV) goes in the Dependent List box

flavor (the IV) goes in the Factor box

Part II One-way ANOVAs: Running in SPSS

When you hit OK and run the ANOVA, you get the following result

Part II One-way ANOVAs: Running in SPSS

This tells us that yes, we can reject the null hypothesis

Null hypothesis: There was no difference between the flavor groups in how people rated their enjoyment of ice cream

There were differences between our groups (flavors)!

Part II Beyond the ANOVA

We did it! We can now see that there are differences between the groups!

But what kind of differences?

Was chocolate better liked than strawberry? Than vanilla?

And were all the groups different from each other? Maybe just some? Maybe just two?

These kinds of important questions are why it’s important to run post-hoc tests!

Part II Beyond the ANOVA

Post hoc – Latin for “after this”

In our case, we use “after this” tests to examine what the significance of our ANOVA means

To do this, we will change some settings in SPSS and run a test called a Tukey test

PART II Running a Tukey test

Go back to SPSS and to the One-way ANOVA

This time, you’re going to click on the Post Hoc button

PART II Running a Tukey test

Find the “Tukey” box and select it

Click continue, then click OK back on the main menu to re-run your ANOVA but this time with a Tukey test after it finishes

Part II Interpreting a Tukey Test

In our output, we will now see a new box – Post Hoc Tests

In this box, we can see our groups compared to one another

For example, in the first box we can see Chocolate (1) compared to Vanilla (2) and Strawberry (3)

Part II Interpreting a Tukey Test

In this comparison, we can see there is a significant difference for both Chocolate to Vanilla and Chocolate to Strawberry

What do you think the other comparisons mean?

Part II Putting it all together

These example show how you can go beyond comparing two groups to make bigger comparisons and get to the bottom of what differences in larger groups are.

Next week, we’ll cover how to look at multiple independent variables with one or more groups!

For now, please make sure you read these slides carefully, then do your best to complete the Lab Activity: Working with SPSS Data posted under Assignments on Canvas

As always email me with your questions or to schedule a meeting