Using SPSS and PASW/Print version

From Wikibooks, open books for an open world
Jump to: navigation, search

Note: current version of this book can be found at http://en.wikibooks.org/wiki/Using_SPSS_and_PASW

Installing SPSS[edit]

SPSS is a software package designed for statistical analysis. This chapter will illustrate how to install SPSS 16.0 on Windows 7.

Obtain SPSS[edit]

The first step, of course, is to obtain a copy of SPSS. This can be done online via the SPSS store [1] or, if you are in an educational setting, your computer store or bookstore may have physical copies available for purchase.

Insert Disc Into Disc Drive[edit]

In Windows, once you insert the SPSS installation disc you will likely receive some sort of prompt as the disc will "autoplay." The initial prompt in Windows 7 looks like this:

01installingSPSS01.png

Select "Run setup.exe".

Follow Prompts to Install[edit]

Once you select "Run setup.exe", you'll get an autoplay window like this one:

01installingSPSS02.png

Select the very first option, "Install SPSS 16.0". You'll then have to click through a series of windows asking you to set various options. The default options are typically fine. The following screenshots illustrate these windows:

01installingSPSS03.png

Unless you are installing a site license or network license, the single user license is the option you want.

01installingSPSS04.png

You have to accept the License Agreement in order to continue installing SPSS.


01installingSPSS05.png

This window contains a ReadMe with installation instructions and a list of changes to SPSS. It is generally a good idea to read documents like this, but not necessary. Click "Next".

01installingSPSS06.png

This window asks for your User Name and Organization name. Fill them in. It also asks for your Serial Number. You do not typically have to complete this at this point as you can complete it in the License Authorization stage. But you can enter it in if you'd like. Click "Next".

01installingSPSS07.png

This window asks you where you want to install SPSS. Unless you have a reason to change the default directory, you can simply click "Next."

01installingSPSS08.png

This is the last prompt before the software will be installed. Simply click "Install."

01installingSPSS09.png

You should then see a progress bar as the software is installed.

Once the software is installed, it may or may not automatically start the License Authorization Wizard. If it does not, you can select it from the Start Menu:

01installingSPSS13.png

Authorize the License[edit]

Once the software is installed, you need to Authorize your copy of SPSS using the License Authorization Wizard. This may start automatically. If it does not, you can select it in the Start Menu. The License Authorization Wizard shows you a series of windows that allow you to authorize your license. This is the first window:

01installingSPSS10.png

The easiest option is simply to "Register with spss.com", but you can also register via email or over the phone. These options are presented to you if you unselect "Register with spss.com, which is selected be default.

Following the prompts in these windows, you will arrive at a window where you can enter your license and serial number into the window. That information is checked and then your copy of SPSS is licensed. A screenshot of that window is not included as it includes the serial number of the author of this book. But upon completing the licensing, you will see this window:

01installingSPSS12.png

Once you see this screen, your installation of SPSS is complete. You can select "Finish" then start SPSS. How to start the software is covered in the next chapter.

Installing GNU PSPP[edit]

Ubuntu Linux[edit]

If you are using Ubuntu Linux, you can open the software center and install SPSS.

Opening SPSS[edit]

Opening SPSS[edit]

There are two ways to open SPSS. The first way is quite simple if the SPSS icon is already on your desktop all you have to do is double click the icon:

02openingSPSS01.png

Double-clicking the SPSS icon should open the program, as illustrated below:

02openingSPSS02.png

Alternatively you can open SPSS through the start button on your computer (if you're running Windows). If the SPSS icon is not on your desktop you must go through your computer's Start button. Click “Start” → “All programs” → “SPSS Inc.” From this point if you have more then one version of SPSS on your computer you have to pick the version you are going to open. I am going to use version 16.0. From here click SPSS 16.0 , and finally click the SPSS 16.0 icon. All the steps are shown below in a screen shot:

02openingSPSS04.png

If you followed the steps above when you click SPSS 16.0 the program should open on your computer.

Chapter contributed by Joseph Ranalli

Opening a dataset[edit]

Opening an Existing SPSS Data File[edit]

This chapter will explain how to open new and already saved data sets. Once you have SPSS open (see Chapter 2), click "File" → "Open" → “Data”:

02openingSPSS04.png

After clicking the Data icon, SPSS with pull up a screen called “Open Data.” Open Data will give you option to find your data set. This can be seen in the screen shot below:

02openingSPSS05.png

Browse through your directories until you find your data set, then select it and click “Open.” Your data set should now be open:

02openingSPSS06.png

Opening a New Data Window[edit]

If you'd rather enter data into a blank data file, you can create an empty SPSS data file. To do so, once you've opened the program, click "File" → "New" → "Data"

03openingdatafileSPSS06.png

The result will be a blank or empty data file. You can enter data in the Data View (Chapter 4) and set variable characteristics in Variable View (Chapter 5).

03openingdatafileSPSS07.png

Chapter contributed by Jennifer Gadarowski

Understanding the Data View[edit]

The Data Editor window of SPSS has two tabs that provide slightly different information. This chapter explores the Data View tab.

The tabs are shown in the bottom left corner of the Data Editor window, as illustrated in this screenshot:

04dataview01.png

The Data View tab shows the raw data in your data set. The rows represent individual cases. If your level of analysis is individuals, each row will represent a single person.

04dataview02.png

The columns represent variables. Each column contains an individual case's data on that particular variable.

04dataview03.png

A cell, at the intersection of a row and a column represents an observation.

Depending on the size of your data set, you may need to scroll up or down to see more cases or left or right to see more variables.

The Data View tab in SPSS also allows you to modify your raw data. You can do so by simply clicking on any cell in the Data View tab:

04dataview04.png

Once you click on the cell, if you begin typing, you will replace all of the information in the cell. If you double-click on a cell, you can then modify the contents without replacing them:

04dataview05.png

While a nice feature of SPSS, the ability to modify your raw data is also a little bit scary as it makes it very easy to change your raw data, which is always something that should be done with caution as that will change your analyses. Recent versions of SPSS include an "undo" feature, but not all changes in SPSS can be undone. Thus, whenever you are in the Data View tab of the Data Editor window, you should be very careful that you do not unintentionally modify the raw data. It is good practice to work on copies of original data and to keep back-ups so that you can restore your data to good condition in the case of serious mishaps.

There are several additional things you can do in this tab. By selecting a column or row header, then clicking on the selected column or row header, you can move columns and rows to rearrange them:

04dataview06.png

Moving through the data can be accoomplished using the arrow keys, the page up and down keys, the home and end keys as you expect from using other Windows programs. One useful habit is to use the tab key to move through data. The tab key wraps around the data when it encounters an empty, undefined column, moving on to the next case.

Additionally, you can sort by variable values, which is illustrated in a separate chapter on sorting data.

Understanding the Variable View[edit]

The Variable View tab is another tab in the Data Editor window in addition to the Data View tab, which was discussed in the last chapter. Again, you can select between the tabs at the bottom left corner of the Data Editor Window:

05variableview01.png

This tab does not show raw data but rather shows information about the variables included in the data set. In fact, after examining the Data View, it may seem a little counter-intuitive to look at the Variable View window because the rows now show variables, not cases.

05variableview02.png

The columns provide information about the various characteristics of variables.

05variableview03.png

There are 10 columns total. Each column and its significance for variables is discussed in the table below:

Column What it Means
Name This column provides the name of the variable. Older versions of SPSS were limited to 8 character names, which is why you often find rather intriguing names for variables in data sets. New versions of SPSS are not limited to 8 characters, but lengthy descriptions should not be included in the Name. They go in the Label column.
Type This column indicates the type of variable that is reflected in this particular row. There are 8 options to choose from: Numeric, Comma, Dot, Scientific notation, Date, Dollar, Custom currency, and String. Most variables beginning users will encounter are either Numeric or String variables. Numeric variables are numbers that either represent a value (e.g., 1=Catholic) or are the value of interest (height=73 inches). String numbers are text and can only be treated as such. As a result, very few manipulations can be performed on them in SPSS.
Width This column indicates the number of spaces available for the variable values.
Decimals This column allows you to control the number of characters after the decimal place.
Label This column allows you to provide a more extensive description of the variable.
Values This column allows you to provide a key for what the numbers of a numeric variable may represent (e.g., 1=Catholic, 2=Protestant).
Missing This column allows you to indicate whether there are any missing values in a variable. Values marked as missing are excluded from analyses in SPSS.
Columns This column indicates the total number of columns a variable's values may have.
Align This column indicates the alignment of the variable in the Data View.
Measure This last column indicates the level of measurement of the variable. There are three from which you can choose: Nominal, Ordinal, and Scale.

Every variable in your data set should have all of the columns filled out such that it is clear exactly what the variable's characteristics are. However, not every column may be relevant for a variable. For instance, if you have a variable "ID" that simply provides a random indicator of a case in your data set, there is no reason to create "Values" for that variable as each case value should be unique.

Understanding the Output Window[edit]

The Output Window is a window in SPSS that will open when you start manipulating or analyzing data. Whenever you run an analysis in SPSS, the results are reported in the Output Window. If you change your preferences to include syntax in the log (see corresponding chapter), that will be included in the Output Window as well. Often when the Output Window first opens, you will see that SPSS is processing your requests. Once the analysis is complete, the Output Window will display your results, like this:

06understandingoutputwindow01.png

The above screenshot provides an example of an Output Window after descriptive statistics were run on the variable “age” in the example data set of Genetic Counselors. The window includes toolbars and menus that are similar to those in the other SPSS windows, including: the Menu bar, the Tool bar, and the Status Bar. These items are included in the Output Window to help the user navigate his/her analyses as well as to edit, save, and print results.

You can see there are two parts to this window: the left panel and the right panel. Each part shows a different way of looking at the results of the analysis. The left panel shows a hierarchical tree of the output. You can use this tree to rapidly navigate through your output. Clicking on any one of the items in the tree moves the right panel to that part of the analysis. The right panel shows the results of the analyses.

The image below shows the left panel of the Output Window:

06understandingoutputwindow02.png

As you can see, the list starts with “Output”, emphasizing that this is the beginning of the output window. “Log” is the next item in the tree. The Log includes your syntax or the computer programming language that runs commands in SPSS (see corresponding chapters). The label “Frequencies” indicates that a frequencies analysis was conducted. The actual analyses that are shown in the right panel are next in the outline. The right panel shows the actual results of the analyses:

06understandingoutputwindow03.png

Above the output you can see the Log, which includes the syntax. This indicates what was requested in SPSS. The Frequencies title indicates that the results are to follow. The table shows the statistics that were requested, including the mean, median, and mode of the variable “age.”

Below are two additional screenshots illustrating the types of tables and graphs available in the Output Window. The first screenshot is a frequencies table that includes the ages of participants in the Genetic Counselor study along with how many reported those ages:

06understandingoutputwindow04.png

This last screenshot is of a histogram of the variable “age” with a normal curve:

06understandingoutputwindow05.png

Thus, the Output Window serves two purposes. It presents the results of your analyses and keeps a record of everything you've done (if you turn on the syntax; see corresponding chapter). Learning how to understand the Output Window is very important in working with SPSS.

Chapter contributed by Kimberly Duggan.

Changing Preferences to Include Syntax in Output[edit]

The previous chapter mentioned syntax and including it in your Log in the Output Window. Syntax is computer programming language and can be used to tell SPSS to run analyses and to manipulate data (see Chapter 24 for more information). Sometimes SPSS is set to include syntax in the Output Window by default; sometimes it is not. This chapter will show you how to change the Preferences in SPSS to include the syntax in the Log and Output Window. Doing so provides you with a history of everything you've done in each session with SPSS.

To begin with, you can see what a basic analysis looks like without syntax:

07syntaxinoutput01.png

To tell SPSS to include syntax in your Output Window, go to “Edit” → “Options.”

07syntaxinoutput02.png

Once you click options, this window will pop up. Click the viewer tab.

07syntaxinoutput03.png

Select the “Viewer” tab, then check the box at the bottom left hand corner that says “Display commands in the log.”

07syntaxinoutput04.png

Click “OK.”

To see the syntax in the output window you can now run any test and the Log will include the syntax. This is illustrated below with a Frequencies analysis:

07syntaxinoutput05.png

You can now track everything you've done in SPSS and slowly teach yourself SPSS syntax. The utility of doing so is explained in detail in the chapter on syntax.

Chapter contributed by Christine Fernandez.

Changing Preferences to Order Lists Alphabetically[edit]

This chapter will illustrate how to change the settings to order the variable list in windows alphabetically. By default SPSS arranges the variable list in the same order as found in the data editor, however it is also possible to arrange the list alphabetically. This is very useful when you have many variables as you can then search for variables simply by typing the name of the variable in the variable list box and SPSS will jump to that variable.

Setting SPSS to display variable names instead of labels was covered in Chapter 7. Setting SPSS to display the variables alphabetically is very similar. To begin, select “Edit” → “Options”:

08variablesalphabetical01.png

After having selected the “options” tab, you will see this window:

08variablesalphabetical02.png

Select the “Alphabetical” option in the “Variable Lists” box on the top left hand side:

08variablesalphabetical03.png

Once you've selected the “Alphabetical” option, click on the “OK” button as shown above.

Once you have done this, another small window will pop up. This window notifies you that “Changing any option in the Variable List group will reset all dialog box settings to their defaults, and all open dialogs will be closed.” Basically this just means you won't see any changes in any dialogs you currently have open. Click “OK” once again:

08variablesalphabetical04.png

To see the ordered variable lists you can click on any option that includes a variable list. This is illustrated below using “Descriptives.” Click on “Analyze” → “Descriptive Statistics” → “Descriptives”:

08variablesalphabetical05.png

You should end up with a window similar to the below one. The variables in the left hand side box should be ordered alphabetically:

08variablesalphabetical06.png

You can now easily search through a list of variables by typing in the name and SPSS will jump directly to that variable.

Chapter contributed by Clarine Ovando-LaCroux.

Recoding Variables[edit]

Recoding variables is useful tool in SPSS for when you want to change the codes for categories of a variable or when you have too many variable options (there are other times you may want to use it as well). Recoding variables allows you to modify the existing numbers assigned to variable values. There aren't really right or wrong ways to recode variables. However, recoding variables is often done for one of two reason: (1) to facilitate statistical calculations; (2) to combine or “collapse” categories when the number of responses in one category are too small for statistical analysis. A time when this is often helpful is when you are analyzing age. Instead of having ages range from 1-100 and having 100 options, you could recode the variable into categories such as 29 and under, 30-60, and 61+. This transforms age from a ratio variable into an ordinal variable, which may be useful in certain situations. When recoding variables, it is generally a good idea to first write down all of the categories of your variable on a piece of paper. Then, next to each category in the existing variable, write the code to which you want to recode it, as illustrated in the table below using a variable measuring frequency of religious attendance:

Old Codes New Codes
0=never 1=never to infrequently
1=less than once a year 1=never to infrequently
2=about once or twice a year 1=never to infrequently
3= several times a year 1=never to infrequently
4=about once a month 2=relatively frequently
5=two to three times a month 2=relatively frequently
6=nearly every week 2=relatively frequently
7=every week 2=relatively frequently
8= several times a week 2=relatively frequently
9=no answer 9=no answer

Writing this out is not a necessary step but does make recoding easier.

To actually perform the recode, click on the “Transform” menu at the top of data editor window:

09recodingvariables01.png

Click on “Recode into Different Variables.” Most of the time you should recode into different variables as doing so does not destroy the values of the existing variable. About the only time you would use the “Recode into Same Variable” command is when you are fixing the labels of an existing variable that were somehow messed up.

09recodingvariables02.png

You will now see the “Recode into Different Variables” window. You can choose variables from the list on the left and insert them into the box into the “Numeric Variable → Output Variable” box. Do this by selecting the variable and then clicking on the arrow button:

In the example above, the variable "attend" is selected and moved to the recode window.

Now type the new variable name label in the “Output Variable” “Name” and “Label” box. People have different preferences for how recoded variables are indicated. One common one is to add an “x” or “z” to the end of the original variable (e.g., attendx). Once you've added the name and label, click Change. This last part is very important as you cannot recode the variable until you select “Change.”

09recodingvariables04.png

To enter the old and new values you laid out in the table above, click “Old and New Values.” You'll get the following window:

09recodingvariables05.png

Using the handmade list we previously made, convert the old values into new values. Old Values are entered on the left, New Values are entered on the right. Click “Add” after you enter the old and new values for each variable to add them to the “Old → New” list. As you can see for variables 0 through 3 they have been recoded as “1.” Values 4 through 8 have been recoded as “2.” This can also be done by selecting the radio button next to “Range” and entering a range of values simultaneously.

When you are finished, click “continue.” This will bring you back to the variable window, where you can click “OK.”

You're not quite finished. You should return to the Data Editor window, Variable View. At the very bottom of your variable list you'll find your newly recoded variable. It's not a bad idea to drag the new variable next to the old one. You should also immediately edit the Value Labels to reflect the newly recoded values lest you forget what they are:

09recodingvariables06.png

Click “OK” and you’re done.

Chapter contributed by Megan Hauf.

Computing Variables[edit]

Computing variables in SPSS basically refers to modifying existing variables mathematically to create new variables. Possible computations include adding, subtracting, dividing, or multiplying existing variables, along with many more options. The goal, of course, is to manipulate one or more variables to form a new variable with that data. One of the most common reasons for using the “compute variable” command in SPSS is to create a scale measure that combines several existing variables into a single variable. Scale measures often do a better job capturing the phenomenon of interest.

In this chapter an abortion attitudes scale is created by combining a set of questions that people were asked about abortion (e.g., Do you think a woman should be allowed to obtain an abortion if: she has been raped, she doesn't want anymore kids, etc.?). While the individual questions are useful for gaining a sense of people's views on whether or not abortion should be legal in specific cases, by combining the variables it is possible to get a better sense of peoples' overall views toward abortion.

To compute variables, select: “Transform” → “Compute Variable”:

10computingvariables01.png

The resulting window allows you to run computations on existing variables to create a new variable.

10computingvariables02.png

On the left, under “Target Variable,” you can provide a name for your new variable. Under “Numeric Expression,” you can build a formula using mathematical operators and the variables in your data set. To move variables into the “Numeric Expression” box, highlight the variable you want, and then use the arrow key to move it over to the box. Use the keypad and list to the right of the keypad to combine the variables how you like.

For the example, seven variables are being combined to create a new scale: abscale. To do this all the abortion variables are added to the “Numeric Expression” box and added together, as shown above. Once you're done, hit “OK.”

After hitting “OK,” if you look in the Data Editor under “Variable View,” your new variable “abscale,” will appear at the very bottom.

10computingvariables03.png

You'll notice that there is no “Label” and your “Values” are listed as “None.” Whenever you compute a variable, you should immediately add a descriptive label (e.g., views on abortion; should range from 0 to 7).

In order to ensure that you created the scale accurately, you can check the frequencies and/or descriptives. Select “Analyze” → “Descriptive Statistics” → “Frequencies”:

10computingvariables04.png

You will get a window like this:

10computingvariables05.png

Select your newly created variable, use the arrow to put it into the “Variable” box, and click okay.

In your Output Window you will see a frequency distribution table for your new variable.

10computingvariables06.png

Following these steps, you can create new variables using a myriad of mathematical computations and check to make sure it was accurately computed.

Chapter contributed by Gen Guzman.

Creating New Variables[edit]

This chapter will cover how to create new variables in SPSS. Creating a new variable is a foundational skill for users of SPSS statistical software, and a necessary step before performing the statistical tests and procedures covered in later chapters. This chapter will provide examples with screenshots to illustrate how you can create variables with ease.

Creating a new variable in SPSS can be completed with just a few clicks, and some simple data entry. First, switch the Data Editor window to “Variable View” (see Chapter 5 for more information on Variable View):

11creatingnewvariables01.png

Once in the “Variable View” tab click on an empty cell in an empty row in which you wish to create the new variable under the column labeled “name.” (If working from a blank SPSS data set this will be in the first row, if working from a preexisting data set, you will need to scroll down to the next open row). Once you have created the new variable, it will appear as, “VAR0001”:

11creatingnewvariables02.png

You can rename the variable by clicking into the cell and typing the name you desire to change it to. In this example the variable has been named “Age.”

11creatingnewvariables03.png

You can specify the type of the variable by clicking in the cell under the column labeled “Type.” An ellipsis will appear:

11creatingnewvariables04.png

Click on the ellipsis to open the Variable Type Dialog box:

11creatingnewvariables05.png

Using this option you may specify the type of the variable you are preparing to analyze, as well as the width and decimal places. Most variables will be “Numeric,” meaning there will be a number used, either because the number itself is meaningful (interval/ratio variables) or because it represents distinct categories. String variables are text variables. You can also have Date and Currency variables.

Closing the variable type window, we continue to move to the right, examining the various characteristics of variables. You can input a label for your variable, which is a description of the variable:

11creatingnewvariables06.png

You can also assign values for the different categories of the variable. This is particularly useful for nominal and ordinal variables for which you assign a unique number to each category (e.g., for race you may have: 1=white, 2=black, 3=Native American, etc.) To access the values option, click on the cell under the “Value” column and an ellipsis will appear. Click on the ellipsis and enter your values and value labels in the “Value Label” dialog box:

11creatingnewvariables07.png

Next you will need to specify the values of any missing values you may have. It is generally a good idea when using any statistical program to include a Value for cases where people did not respond. This Values should be included in the Value Labels, but should also be included in the “Missing” column as it will prevent SPSS from attempting to use those values in calculations. To add these values to the list of Missing Values, click on the cell in the “Missing” column. Click on the ellipsis that appears in order to access the “Missing Values” dialog box:

11creatingnewvariables08.png

While there are no right or wrong values to assign for missing data, it is common to uses iterations of 9. For instance, if a variable has 4 categories (1, 2, 3, and 4) and missing values, you can assign a “9” to indicate that the value is missing for that person. If there are 20 categories, you could use “99.” If there are 200 categories, you could use “999,” and so on. This is a widespread practice, but not mandatory. You could choose a different system to indicate missing values.

Finally, you must specify your level of measurement. This may be done by clicking on the cell under the column labeled “Measure” and choosing from Nominal, Ordinal, and Scale. Nominal and ordinal variables should be labeled under their corresponding names, while interval and ratio variables should be labeled as “Scale.”

11creatingnewvariables09.png

Repeat this process for as many variables as you need to create and/or define.

Chapter contributed by Damian Patrinostro.

Sorting Variables[edit]

One of the functions you can perform in the Data View of the Data Editor is to sort by variable values. This can be a very useful tool when you are exploring the raw data in your dataset. There are two ways to do this. One is quite simple, the other allows for more complex sorting.

The simple way to sort variable values is to make sure you are looking at the Data View tab. Then scroll to the variable by which you want to sort. Right click on the column heading and a context menu will appear:

12sort01.png

At the bottom of the context menu are two sorting options: "Sort Ascending" and "Sort Descending." Choosing the first will move the lowest values to the top of the data set. Choosing the second will move the highest values to the top of the data set. Keep in mind that SPSS automatically (unlike Excel) moves all the other columns to retain their correspondence as a single case, so you don't need to worry about cases becoming misaligned when you do this.

The more complicated way to sort also allows for greater complexity in sorting. Choose "Data" -> "Sort Cases":

12sort02.png

You'll get this window:

12sort03.png

You'll see below the "Sort by:" box the option to sort by either Ascending or Descending order again. To the left of the window is a list of all the variables in the data set. You can select multiple variables and move them to the "Sort by:" box using the blue arrow:

12sort04.png

Sorting using multiple variables tells SPSS to sort the data set by the values of the first variable, then to sort by the values of the second variable, and so on. This, too, can be useful at times when examining your data.

Sorting data does change the data set, which means that you must save it if you want to retain the changes resulting from sorting.

Creating Charts and Graphs[edit]

Charts and graphs are a useful way of organizing data so it can be read and interpreted more easily. In order to choose which type of chart or graph to use you must first decide the level of measurement, whether the variable is of nominal/ordinal or interval/ratio measurement. Next, you must consider the objective behind creating a chart. Finally, you must decide the target audience. For example if your targeted audience is the general public, you want your graph to be colorful, uncluttered and have an overview of the statistics you want to present. If your audience is a more technical audience, you may want to use tables instead.

One rule of thumb to follow is a “lost in the parking lot” test. The basic idea of this test is that if a relative stranger were to stumble upon your graph or chart it would include enough information in a simple enough format that this person would be able to interpret it.

As stated before, choosing which type of graph to create requires that you first determine the level of measurement. In statistics, the basic rules are as follows:

  • For nominal/ordinal variables, use pie charts and bar charts
  • For interval ratio/variables, use histograms.

Using the Genetic Counselors data set, we can create and interpret different types of graphs.

Pie Charts[edit]

This type of chart is particularly useful for conveying a sense of fairness, relative size, or inequality among categories.

To get started, click on the “Analyze” button and a drop-down menu will appear with various options.

13creatingcharts01.png

After this you want to move your mouse to “Descriptive Statistics” and click on “Frequencies”. A separate window will pop-up.

13creatingcharts02.png

Select “Frequencies.”

This is the box where you will determine which variable you want to use as well as the type of graph you want to use to illustrate the information.

To create a pie chart you must first decide which variable you want to use. (Remember that pie charts are used with nominal/ordinal variables.)

For this example we will use the nominal/ordinal variable of religion to determine the religious make-up of the people surveyed. To do this, you must scroll down to “relig”, highlight it by clicking on the variable name, and then move it over to the variable box using the blue arrow.

13creatingcharts04.png

Next you want to click on the charts button and a separate window will pop-up, where you will be able to choose which type of graph to use. Click on pie chart. You can also determine if you would like frequencies or percentages to appear with your pie chart. For this example, clicking on percentages will be most useful. Do this, and then click continue.

This window, entitled “Frequencies: Charts” is where you will create not only pie charts, but also bar charts, and histograms.

This will return you to the “Frequencies” window. Click okay again. A separate window (the output window) will pop-up displaying the pie chart. This is called the “Output Window”. From this you will be able to interpret the pie chart.

From this pie chart we are able to interpret many different things. For example if someone were to ask what religion most people surveyed affiliate themselves with, you can look at the chart and see that “Protestant” takes up most of the chart and therefore, most people consider themselves to be Protestant. By using the box above the pie chart you can also determine exact percentages and frequencies. For example, if someone were to ask you what percentage of Genetic Counselors surveyed consider themselves to be Jewish, you can say 9.8%.

Bar Charts[edit]

Bar charts are useful for showing a sense of competition among categories. Like pie charts, bar charts are also used with variables of a nominal/ordinal level of measurement.

To create a bar chart you use the same steps to create a pie chart except when the window, entitled “Frequencies: Charts” pops-up, you want to click on Bar Charts instead of pie charts.

13creatingcharts07.png

For this example we are going to ask the question of how often do the genetic counselors in this survey attend religious services and see the competition among the different categories.

To do this we go to “Analyze” → “Descriptive Statistics” → “Frequencies”

Next we choose our variable (which in this case is titled “attend”), move it over using the blue arrow, then click on charts, click on the chart type “bar charts”, and click continue and then okay. Your bar chart will be displayed in the Output window:

Here, just as in the pie chart output, you can see the frequencies and percentages in the box located above the bar chart. As you can also see, the most frequent answer when asked “How often do you attend religious services” is “several times a year”. You can also interpret that the most common answers are “several times a year”, “never” and “less than once a year”. With this we can conclude that the majority of the genetic counselors surveyed do not attend religious services very often.

Histograms[edit]

Histograms, also known as frequency histograms, are similar to bar charts except that the columns of a histogram touch to account for real limits and the principal of inclusiveness. As stated before, interval/ratio variables are used when creating a histogram.

For this example we can use the interval/ratio variable of age to determine the general distribution of ages among the genetic counselors surveyed. To do this we go to “Analyze” → “Descriptive Statistics” → “Frequencies”

Next, just like before, you want to move your chosen variable (age) over using the blue arrow. Next click on charts, choose Histograms, click continue, and then okay. The output window will pop-up displaying the frequency histogram:

Within frequency histograms, the frequency and percentage table is often large. This is because it accounts for every score, and not just a range of scores.
13creatingcharts10.png

Not only does this frequency histogram give you a chart, it also gives you the mean, standard deviation and sample size, located in the upper right hand corner of the chart.

Through this chart we can see that the most frequent answer when asked each person’s age was just under 30 years old. If you were to look under the frequency chart you would see that 57 people answered 28 years old and 56 people answer 29 years old. This explains why the bar directly before 30 is the highest one.

Remember that whenever you present a chart or graph you should provide a clear interpretation geared toward your audience.

Chapter contributed by Caitlin McGrath.

Frequencies[edit]

Frequencies, in statistics, refers to counts of categories or responses. It's a basic statistical tool that provides a sense of how often specific response options occur in a population.

Using the sample data set, let's say we want to know the geographic distribution of genetic counselors. In the sample data set there is a variable that codes for geographic region, "region." Using the frequencies function in SPSS, we can get a sense of the geographic distribution of genetic counselors.

To do so, begin by selecting "Analyze" -> "Descriptive Statistics" -> "Frequencies":

14frequencies01.png

The "Frequencies" window will appear:

14frequencies02.png

Scroll through the variables until you find the variable you want. In our example, we're using "region." Use the arrow to move it into the "Variable(s):" box:

14frequencies03.png

Once you've done that, you don't actually need to do anything else except make sure that the "Display frequency tables" box is checked. Then hit "OK". In the Output Window you should see something like the following:

14frequencies04.png

There are two tables in this output. The first gives a summary of the cases included. The "Valid" N tells you how many of your cases had a response to this variable that is not marked as missing. The "Missing" N indicates the number that are missing.

The second table is called a frequency table of frequency distribution. It includes four columns. The first, labeled "Frequency," simply reports the total number of cases that fall into each category of the variable of interest, "region." For instance, 100 genetic counselors live in the "pacific" region. The second column, labeled "Percent," provides a percentage of the total cases that fall into each region. The percentage of genetic counselors that work in the "pacific" region is 15.3%. The third column, labeled "Valid Percent," is a variation of the "Percent" column; it recalculates the percent without including the missing cases. The fourth column, labeled "Cumulative Percent," adds the percentages of each region from the top of the table to the bottom, culminating in 100%. This last column is more useful when you have ranked or ordinal variables you are analyzing as it makes it easy to get a sense of what percentage of cases fall below a specific rank.

The above example should help illustrate the utility of calculating frequencies in SPSS.

Descriptives[edit]

Descriptive statistics are used to describe variables. Examples of descriptive statistics include: mean, median, mode, standard deviation, and range. In this chapter you'll learn how to have SPSS calculate three descriptive statistics for you: the mean, median and mode.

In order to generate descriptives in SPSS, you first need to open up a data set. In this case we are going to use the Genetic Counseling example data set. The variable we'll use to illustrate descriptive statistical analysis is “age.”

To calculate descriptives in SPSS, click “Analyze” → “Descriptive Statistics” → “Frequencies.”

15descriptives01.png

Once you click on Frequencies, a Frequencies Window will appear on the screen. Find the variable labeled “age” and click the blue arrow to move it over into the Variable Box.

15descriptives02.png

Once you have done this, click on the tab that is labeled Statistics. A new window will appear. In this window, select mean, median, and mode then click “Continue.”

15descriptives03.png

Once the box disappears, click “OK” in the Frequencies Box and the mean, median, and mode, for the age of participants should appear in the Output window.

15descriptives04.png

In the example above, the mean (average) age of participants is 35.93. The median, or middle number in the data set is 33, and the mode or the most commonly occurring age of participants in this study is 28.

Distribution Curves in SPSS[edit]

If you want to see where the mean, median, and mode fall on a frequency distribution curve, SPSS can do this for you. This is useful when you are trying to determine if the curve is normal or skewed. Skewness statistics are used to determine whether sample data are so skewed that they suggest that the population scores are skewed (a detailed discussion of skewness statistics is beyond the scope of this book). We are going to continue with the above example with using the variable of “age.”

First, repeat all of the above steps when you were finding the mean, median, and mode. When you have the “Frequencies” window open, click the tab that is labeled “Charts.”

15descriptives05.png

Now select the radio button that says “Histograms” and choose “With Normal Curve.” Click “Continue” then “OK,” and your chart should pop up in the Output box:

15descriptives06.png

The frequency curve in this example shows that the mean (35.9), median (33), and mode (28), create a positively skewed distribution (meaning some very large values in a positive or higher direction).

Chapter 15 contributed by Kristin Mraz.

Confidence Intervals[edit]

Confidence Intervals help us estimate parameters of a population from sample statistics. Confidence Intervals are a range of possible values of a parameter. We express this interval with a specific degree of confidence. The degree of confidence tells a reader how confident we are that the population parameter falls within our stated interval.

For this example we will use the variable “How Spiritual do you consider yourself?” in our example data set made up of genetic counselors. The name of this variable in the data set is “sprscale.” This variable allowed the sample of genetic counselors to rank how spiritual they are on a scale of 1 to 10.

To compute a confidence interval in SPSS, you begin by selecting “Analyze” → “Descriptive Statistics” → “Explore.” Once the “Explore” window pops up, scroll down on the left list until you find “sprscale,” then click on the arrow that will send it to the dependent list:

16confidenceintervals01.png

Now, click on “Statistics”:

16confidenceintervals02.png

Check the box “Descriptives” and type in the level of confidence that you would like to use. For this example we'll use 95%. Click “Continue” then “OK.” The Output Window will now pop up with a variety of statistics, including the confidence interval:

16confidenceintervals03.png

The confidence interval we found for how spiritual Genetic Counselors are on a scale of 1 to 10 is 5.99 to 6.36. We interpret this in plain language by saying “We are 95% confident that the true mean spirituality on a scale of 1 to 10 for the population of genetic counselors is between 5.99 and 6.36.”

Chapter contributed by Lauren Takemoto.

Single Sample Means Tests[edit]

A single sample means test is a statistical test that can determine whether or not a population estimate (i.e., sample statistic) is significantly different from a known value.

A hypothetical example may help illustrate this. Let's say a national survey of the adult population asks how religious people are on a 10-point scale, ranging from 1 being "not at all religious" and 10 being "very religious." And let's say that the mean on this scale in the U.S. is 6.01. With our sample of genetic counselors, we can ask whether or not genetic counselors are similar to the general population in their religiosity. To do this, we use a single sample means test, which takes into consideration sampling error. Similar tests are possible for proportions (i.e., with nominal/ordinal variables), but SPSS cannot do those. How to do a single sample means test in SPSS is illustrated below.

To begin with, go to "Analyze" -> "Compare Means" -> "One-Sample T Test":

17singlesamplemeans01.png

The "One-Sample T Test" window will open:

17singlesamplemeans02.png

Select the variable you want to compare to a target value, then move it into the "Test Variable(s):" box:

17singlesamplemeans03.png

The next part is absolutely essential and easy to overlook the first time you run this. In the "Test Value" box below the "Test Variable(s):" box you need to put the target value. In our case, we'll put 6.01:

17singlesamplemeans04.png

(Note: It's not uncommon for beginners to leave the "Test Value" box blank. And, for some reason, SPSS puts a default value in the box of zero, which means the test will run if you don't put your target value in the box as SPSS will assume your target value is zero. Just make sure you put the correct target value in the "Test Value" box.)

The results will open in the Output Window:

17singlesamplemeans05.png

There are two tables. The first simply describes the sample and variable included in the analysis. The N, Mean, Standard Deviation, and Standard Error of the Mean are all included. Of particular interest for the example, the mean religiosity score of genetic counselors is 4.65, which is lower than our hypothetical religiosity of the U.S. population generally. However, the question is whether or not this is due to random error or an actual affect of genetic counselors.

The second table tells us whether or not the difference between the two is likely due to chance. The first column is the t-statistic. Using a t-table, one could use this value to determine whether or not the two means are significantly different. SPSS does the calculations for you. The next column, "df," provides your degrees of freedom, which is n-1 for single sample means tests. The third column, Sig. (2-tailed), provides the p-value for this particular test, which is the probability of finding the difference between the genetic counselors' mean and the general U.S. adult population mean we found (again, this is hypothetical) simply due to random error (assuming they are actually the same). In our example, the p-value is .000, which is smaller than the standard p-value used in statistics of .05, indicating the odds of us finding this difference if the population means are the same simply due to chance is less than 1 in 1000. In this case, we would reject the null hypothesis that there is no difference and accept the alternative hypothesis that there is a difference between genetic counselors and the general population in their average religiosity.

Independent Samples t-test[edit]

An independent samples t-test is used for comparing the means on an interval/ratio variable between two categories on a nominal/ordinal variable. It answers the question of whether the difference between means is statistically significant in the population of interest (assuming good sampling) or whether the difference is due to sampling error. To do this test, you need two variables from one population and sample. The independent variable is nominal/ordinal and the dependent is interval/ratio.

To illustrate this concept, this chapter tests to see if there is a statistically significant difference between whether or not Genetic Counselors believe in life after death and how spiritual they consider themselves.

To run the t-test you will click “Analyze” → “Compare Means” → “Independent Samples T-Test”:

18independentttest01.png

Since the spirituality scale is the interval/ratio variable it will act as the dependent variable. That is, how spiritual one considers himself depends on his belief in life after death. Find the spiritual scale variable (labeled “sprscale”) and click the arrow to move it over to the test variable section. Next, find the views on life after death variable (labeled “postlife”) and click the arrow to move that over to the grouping variable section.

You will see question marks next to the grouping variable. This is because SPSS allows you to choose which categories of the independent variable you want to include (the variable could include more than just two categories). You will need to tell SPSS which groups to compare by clicking the “define groups” button. To find these labels, go back to the Data Editor in Variable View and click on the “Values” button for the “postlife” variable. You will see this screen:

18independentttest02.png

The above screenshot shows that not believing in an afterlife is labeled “0” and believing is labeled “1.” Once you know what your values are for believing in an afterlife, return to the t-test dialog. In the define groups window, add zero for “No” and one for “Yes”:

18independentttest03.png

Once you've defined your groups, click “Continue.” You'll return to the primary “Independent-Samples t-test” window:

18independentttest04.png

Click OK, and SPSS will run the T-Test. Two tables will appear in the output:

18independentttest05.png

A lot of information appears in these tables. The first table is simply the “Group Statistics” table, which includes: sample sizes, means, standard deviations and standard errors of the means.

The second table, labeled the “Independent Samples Test,” is what you use for determining whether or not you can reject the null hypothesis. Because you are comparing two means, two different variances are obtained. There is a long equation used to determine which variance to use, but SPSS does this for you by running the Levene’s Test for Equality of Variances. If the variances are relatively equal, that is one sample variance is no larger than twice the size of the other, then you can assume equal variances. By looking at the output of the Levene’s test you decide which row to use. If the significance is .05 or below, use the bottom row, or “equal variances not assumed.” If the significance is above .05 use the top row. In this example, since the significance is .000, we'll examine the bottom row.

Both the top and bottom row provide the same information, they just use different tests to calculate the test statistic, which results in slightly different calculations. This table provides the test statistic or t value, the degrees of freedom and other values helpful for determining confidence intervals. A key statistic provided is the p-value, listed in the “Sig (2-tailed)” column. If your p-value is greater than .05 you fail to reject the null (meaning the difference in means is likely due to chance or sampling error). If your p-value is less than .05 you can reject the null (meaning there is in fact a statistically significant difference in the means and it is not due to sampling error). In this case, you can reject the null hypothesis (because the significance is .000, which is substantially less than .05).

You should always interpret your analysis, “How spiritual someone considers him/herself is affected by someone's belief in an afterlife.”

Chapter contributed by Alison Moser.

Paired Samples t-test[edit]

A paired samples t-test is a test that is useful when you have two interval/ratio variables from the same people in a sample that are measured exactly the same way. You can use a paired samples t-test to compare the scores on the two variables. The most common use of this test is for pre- and post-test scores for a sample when they are exposed to some intervention in between the pre- and post-tests. The reason a paired samples t-test is used instead of an independent samples t-test is because the scores are for the same people, which suggests there is an underlying relationship between the scores.

Another time paired samples t-tests are useful is when you have two variables with the same units of measure from the same subjects from the same time and you want to see if the subjects score differently on one test compared to the other. This is the example we will use in this chapter to illustrate the test. In the sample data set of genetic counselors, there are two interval variables that have the same unit of measure - a 10 point scale. However, one variable asked genetic counselors how religious they were and the other asked them how spiritual they were. Since it is the same subjects being asked different questions that use the same metric, we can compare their scores on these two variables using a paired samples t-test.

Here's how this is done in SPSS. To begin with, select "Analyze" -> "Compare Means" -> "Paired-Samples T Test":

19pairedsamples01.png

The "Paired-Samples T Test" window will open:

19pairedsamples02.png

To select the variables to compare, you have to actually select two variables at the same time. To do this, select the first variable, then hold down "Control" and select the second, then move them over to the "Paired Variables" box with the blue arrow:

19pairedsamples03.png

Then hit "OK." You'll get the following tables in the Output Window:

19pairedsamples04.png

The first table provides basic sample and variable statistics for the two variables, including the Mean, the sample size, the Standard Deviation, and the Standard Error of the Mean. Of interest in our comparison are the two means. On the spirituality scale, genetic counselors report a mean of 6.17. On the religiosity scale, genetic counselors report a mean of 4.64. Those appear to be quite different scores, but the question of interest is whether those scores are different due to chance or due to an actual affect in the population. The paired-samples t test allows us to determine that.

The second table, a correlation table, is discussed in the chapter on correlation, so it will not be discussed here.

The third table is the "Paired Samples Test" table. The first column is labeled the "Mean" and it is the average difference between the two variable means. The "Std. Deviation" and "Std. Error of the Mean" are calculated on the average mean in the first column. Columns four and five are lower and upper confidence intervals for the difference. For more information on confidence intervals, see the chapter on confidence intervals. The sixth column, "t", is the t-score. The seventh is the degrees of freedom. The eight column provides the p-value for the difference of means. In our example, the p-value is .000, which is less than a standard alpha of .05, indicating that the odds of these two scores being different due to random chance assuming they are actually the same is less than 1 in 1000. In this case, we would reject the null hypothesis that they are the same and accept the alternative. There is a significant difference between the religiosity and spirituality of genetic counselors; genetic counselors are significantly more spiritual than they are religious.

Chi-Square[edit]

The Chi-Square test is used when trying to find a relationship between two nominal or ordinal variables. To reiterate, a nominal variable is one that is only measured by naming categories such as class, quality or kind. An ordinal variable is similar to a nominal variable, but the categories can be put in an order (e.g., ranked highest to lowest).

To calculate Chi-Square, we use a cross-tabulation, crosstab for short, which shows the frequencies of joint occurrences between two variables.

The hypothesis we will test in this chapter is whether or not there is a relationship between religious affiliation and belief in the afterlife. We will test this using an alpha of .05. To do a Chi-Square test in SPSS, complete the following steps:

You must have two nominal variables from a single sample that you can use to see if there is a relationship between them. For this example we will use religion (relig) and belief in the afterlife (postlife). For this example religion is the independent variable and belief in the afterlife is the dependent variable. This will be important later.

To run the test, select: Analyze → Descriptives → Crosstabs.

Once you click Crosstabs, a window will pop up where you will enter your chosen variables to be tested. Select the independent variable (relig) to go in the column and the dependent variable (postlife) to go in the row.

20chisquare01.png

Next, click the statistics button and check the Chi-Square box.

20chisquare02.png

Hit continue and then click on the cells button and check observed, expected and column percentages.

20chisquare03.png

Then hit OK and an output window with the resulting cross-tabulation will pop up.

NOTE: The expected frequency of each cell must be at least five in order to rely on the Chi-Square test. If it is not five, a different statistical test must be completed.

Next you must interpret the data. At the bottom of each column and end of every row there is a total for that specific group. The bottom right cell is the total number of cases in the sample. In each cell, it shows the observed joint occurrence between the two variables and the expected occurrence based on the totals that were observed. If there is NO relationship between the variables, the observed and expected frequencies will be the same. For our example, this is clearly not the case. They are different but we must look at the last box from our output window to see if our relationship is legitimate or if it was due to sampling error.

20chisquare05.png

As we stated in the beginning, our alpha is .05. If the “Asymp. Sig. (2-sided)” for the Pearson Chi-Square statistic is less than .05, there is a relationship between the variables based on the level of confidence we stated in the beginning. As seen in the table below, the Chi-Square significance value is .000 which is less than our value of .05 which shows that there is a relationship between one’s religion and their belief in the afterlife.

Finally, we interpret the test in everyday terms, which also means we look more closely at the crosstab table as well. In the crosstab table it is clear that Protestants and Catholics are much more likely to report a belief in the afterlife than are Jews or Nones. Thus, we can say, “Jews and the non-religious are significantly less likely to believe in an afterlife than are Protestants and Catholics.” Do note that care must be taken in interpreting Chi-Square crosstabs as it is not always perfectly clear where the significant differences between scores lie.

Chapter contributed by Sarah Friswell.

ANOVA[edit]

ANOVA is an extension of the two group difference of means test (t-test). The t-test is used to compare two group means, but ANOVA allows for comparing three or more group means, which is easier than conducting numerous t-tests.

To conduct a One-Way Analysis of Variance (ANOVA) test in SPSS, you must first begin by choosing two variables. The variables we will use in this example are religious affiliation and frequency of prayer. Basically we are asking, “Does religion affiliation have an influence on how often people pray?”

Once you know what you want to compare, you can tell SPSS to run the analysis by clicking on “Analyze” → “Compare Means” → “One-Way ANOVA”:

21ANOVA01.png

The One-Way ANOVA dialog box will appear:

21ANOVA02.png

In the list displayed on the left, click on the variable that corresponds to your dependent variable (should be an interval/ratio variable). In our example this is frequency of prayer. Move it into the Dependent List by clicking on the upper arrow button. In this example, we are asking if religious preference has any effect on how often people pray.

Now select the (quasi) independent variable from the list on the left and click on it. Move it into the Factor box by clicking on the lower arrow button. In our example this is religious affiliation.

21ANOVA03.png

Click on the Options button in the One-Way ANOVA dialog box. The One-Way ANOVA dialog box appears:

21ANOVA04.png

Click in the check box to the left of Descriptives (to get descriptive statistics), Homogeneity of Variance (to get a test of the assumption of homogeneity of variance) and Means plot (to get a graph of the means of the conditions).

Click on the Continue button to return to the One-Way ANOVA dialog box. In the One Way ANOVA dialog box, click on the “OK” button to perform the analysis of variance. The SPSS output window will appear. The output consists of six major sections. The first is the descriptives section:

21ANOVA05.png

The Descriptives table provides various descriptives for the groups being compared, including the group sample size, mean, standard deviation, minimum, maximum, standard error, and confidence interval for the mean. In this example, there were 64 Jewish individuals, the mean frequency of prayer was 4.70 (on a six point scale; technically it is an interval-like ordinal variable), and the standard deviation was 1.217. There were 132 Catholic individuals with a mean frequency of prayer at 3.38, and a standard deviation of 1.438.

The ANOVA output gives us the analysis of variance summary table. There are six columns in the output:

Column Description
Unlabeled (source of variance) This column describes each row of the ANOVA summary table. It tells us that the first row corresponds to the between-groups estimate of variance. The between-groups estimate of variance forms the numerator of the F ratio. The second row corresponds to the within-groups estimate of variance. The within-groups estimate of variance forms the denominator of the F ratio. The final row describes the total variability in the data.
Sum of Squares The sum of squares column gives the sum of squares for each of the estimates of variance.
Df The third column gives the degrees of freedom for each estimate of variance. The degrees of freedom for the between-groups estimate of variance are given by the number of levels of the IV-1. In this example there are five levels of the independent variable.
Mean Square The fourth column gives the estimates of variance (the mean squares). Each mean square is calculated by dividing the sum of square by its degrees of freedom.
F The fifth column gives the F ratio. It is calculated by dividing mean square variance between-groups by the mean square variance within groups.
Sig. The final column gives the significance of the F-ratio. This is the p-value. If the p-value is less than or equal to your alpha level, then you can reject null hypothesis that all means are equal. In our example, the p-value is .000.

Here is the actual ANOVA table from the example:

21ANOVA06.png

Based on this analysis, we can conclude that religious affiliation does significantly affect frequency of prayer for genetic counselors.

Chapter contributed by Sheena Wright.

Correlation[edit]

Two variables can be considered to correlate when there is a systematic change in their scores. The purpose of correlation is to improve estimates and/or make predictions about a population. Simple linear correlation is particularly useful for improving best estimates of a dependent variable by accounting for its relationship with the independent variable using the straight-line formula.

In SPSS we are able to run correlations between two interval/ratio variables. To do so, click on “Analyze” → “Correlate”. You will be shown three options. If you have two interval/ratio variables, choose “Bivariate….”

22correlation01.png

The pop up window will ask you to select two variables to move into the “variable” box to the right of the window. Select the variables of interest. In this example, we will use a religiosity scale and a spirituality scale as our two variables. Make sure that the box next to Pearson’s r is checked.

22correlation02.png

Then Click “OK.”

You'll get the following table:

22correlation03.png

This table is called a correlation table. On the diagonal each variable is correlated with itself, thus they are perfect correlations of 1.0. The table is also symmetrical, meaning the cells above the diagonal are identical to those below the diagonal, thus you really only need to examine one set of cells – either those above or below the diagonal.

As far as what the cells tell us... The first value in the cell correlating religiosity with spirituality reports a correlation coefficient of .585. Correlation coefficients range from -1.0 to +1.0. An absolute value of 1.0 indicates a perfect correlation, which is rare in the social sciences. A negative correlation means that as one variable goes up, the other goes down. A positive correlations mean that as one variable goes up, the other also goes up. A correlation of .585 is positive, meaning as religiosity increases, so, too, does spirituality. But the correlation is not perfect, meaning some of the variation in spirituality is not explained by the variation in religiosity.

The asterisks at the end of the correlation indicate that the correlation is significant. The p-value of the significance is indicated in two places – at the very bottom of the table where the asterisks are defined and just below the correlation coefficient, where the p-value is provided. In this case, the p-value is .000, which means the odds of finding this relationship between these two variables just due to chance is less than .001, or less than 1 in a thousand. This indicates there is a significant relationship between religiosity and spirituality.

The last value in the cell is the sample size for the correlation – 646. If you were to correlate more than two variables simultaneously, which you can do in SPSS by simply adding more than two variables in the correlation dialogue, the sample sizes being compared can vary due to things like missing values or non-response.

Technically, before conducting a correlation, one should always run a scatterplot to insure that the relationship between the two variables is linear. This is done in SPSS by clicking on “Graphs” → “Chart Builder”.

22correlation04.png

Once the Chart Builder window is open, click on Scatter Plot. Select the first display example of a scatter plot graph and drag the display into the large box located towards the right of the pop up window. Select your independent variable from the side scroll bar and drag it to the X-axis on the box to the right. Select your dependent variable from the side scroll bar and drag it to the Y-axis on the box to the right.

22correlation05.png

Once this is done, click “OK.”

You will get something similar to the graph below:

22correlation06.png

This scatter plot illustrates that there is something like a linear relationship between the two variables. Basically, no one who scores high on religiosity scores low on spirituality. However, there are some people who score low on religiosity but who score high on spirituality. Even so, the general trend is a linear relationship, which means these two variables are suitable for correlation and regression analysis.

Chapter contributed by Brittany Harder.

Ordinary Least Squares Regression[edit]

Ordinary Least Squares (OLS) regression (or simply "regression") is a useful tool for examining the relationship between two or more interval/ratio variables. OLS regression assumes that there is a linear relationship between the two variables. If the relationship is not linear, OLS regression may not be the ideal tool for the analysis, or modifications to the variables/analysis may be required. The basic idea of linear regression is that, if there is a linear relationship between two variables, you can then use one variable to predict values on the other variable. For example, because there is a linear relationship between height and weight, if you know someone's height, you can better estimate their weight. Using a basic line formula, you can calculate predicted values of your dependent variable using your independent variable, allowing you to make better predictions.

To illustrate how to do regression analysis in SPSS, we will use two interval variables from the sample data set. These same variables were used in some of the other chapters. Genetic counselors were asked to rate how religious and spiritual they consider themselves on a 10 point scale - higher values indicate more religious or more spiritual. In the analysis below, we are going to see how well religiosity predicts spirituality.

Before we calculate the regression line for religiosity and spirituality for genetic counselors, the first thing we should do is examine a scatterplot for the two variables. A scatterplot will help us determine if the relationship between the two variables is linear or non-linear, which is a key assumption of regression analysis. This is done in SPSS by going to "Graphs" -> "Chart Builder":

23regression01.png

Once you select on "Chart Builder," you'll get the "Chart Builder" window, which looks like this:

23regression02.png

In the Chart Builder window, toward the middle of the screen make sure you've selected the "Gallery" tab, then select "Scatter/Dot" from the list of options. To the right of the options you'll see 8 boxes. If you hover over those, they will identify the type of scatterplot they will generate. Choose the one at the upper left of the choices, which is called "Simple Scatter." To choose it, you'll need to drag it up to the box above that says "Chart preview uses example data." You'll then be presented with two axes - a Y axis and an X axis. In our example, since we are using religiosity to predict spirituality, we drag relscale to the X axis and sprscale to the Y axis. We then select "OK" and get the following in our Output Window:

23regression03.png

Scatterplots with lots of values are often hard to interpret. SPSS tries to make this a little easier by making the dots with lots of occurrences darker. In this case, what we can see in the scatterplot is that there appears to be a dark line run from the bottom left to the upper right, suggesting a positive relationship between religiosity and spirituality - as one increases, so does the other. The relationship also appears to be linear, which is good for regression analysis. Having checked the scatterplot, we can now proceed with the regression analysis.

To run a regression analysis in SPSS, select "Analyze" -> "Regression" -> "Linear":

23regression04.png

The "Linear Regression" window will open:

23regression05.png

On the left is the list of variables. Find your dependent variable. In our example it is "sprscale." We move that over to the "Dependent" box with the arrow. Then find your independent variable. In our example it is "relscale." We move that over to the "Independent(s):" box:

23regression06.png

While there are a number of additional options that can be selected, the basic options are sufficient for example. Thus, choose "OK" and you'll get the following in the Output Window:

23regression07.png

The first table simply tells you which variables are included in the analysis and how they are included (i.e., which is the independent and which is the dependent variable).

The second table provides a "Model Summary," which we'll return to in a moment. The third table is an ANOVA, which is useful for a variety of statistics, but we are going to skip it in this chapter at the present.

The fourth table provides the regression statistics of most interest to our present efforts. The first column, "B", in the second row (not the first row labeled "(Constant)") provides the slope coefficient for the independent variable. What this means is that, for every 1 unit change in our independent variable, there is an XX unit change in the dependent variable. In our example, every 1 point increase in the 10 point religiosity scale results in a .506 point increase in the spirituality scale. This tells us that the relationship between the two variables we noticed in the scatterplot was accurate - the relationship is positive.

The second column, labeled "Std. Error," provides a standard error for the slope coefficient. The third column, "Beta," provides a standardized version of the slope coefficient (in a bivariate regression, this is also the correlation coefficient or "r"). What this means is that for every 1 standard deviation unit change in the independent variable there is a corresponding XX standard deviation unit change in the dependent variable. This is less intuitive than the slope coefficient for most variables. The fourth column, "t," is the t statistic. The fifth column, "Sig.", provides the p-value for the slope coefficient of the independent variable. In our example, the p-value is .000, which is less than a standard alpha of .05, suggesting the odds of finding the linear relationship we did between religiosity and spirituality by chance, assuming there is not in fact a relationship, is less than 1 in 1,000. In other words, we can reject the null hypothesis that there is no relationship between the two variables and accept the alternative hypothesis that there is a significant relationship between religiosity and spirituality. In practical terms, more religious genetic counselors tend to be more spiritual as well.

The first row in the fourth table provides statistics for the constant, or y-intercept. Of greatest interest to us in this chapter is the value in column "B". That value is the y-intercept, or the point at which the regression line crosses the y-axis. In our example, it is 3.822. What that means, then, is that when religiosity is zero for genetic counselors, spirituality is predicted to be 3.822.

Returning to the second table, the astute reader will notice that the first column, "R", is identical to the Beta column in the fourth table. As noted, the standardized slope coefficient in a bi-variate regression is the equivalent of the correlation coefficient or "r". The second column is the R-square statistic, which is the proportion of variability in a data set that is accounted for by the statistical model. Basically, the R-square statistic can be interpreted as saying the following: Religiosity explains 34.2% of the variation in spirituality.

Finally, to illustrate the regression line as an actual line of best fit for the many cases in our dataset, we have included another scatterplot with the regression line:

23regression08.png

This graph illustrates that the regression line tries to minimize the variation between all of the points in the scatterplot, providing a best estimate of the dependent variable (spirituality) for each value of the independent variable (religiosity). It also shows the regression line crossing the y-axis at the value noted above - 3.822.

The above explanation should provide individuals with sufficient information to run a regression analysis and interpret it in SPSS.

SPSS syntax[edit]

What Is Syntax?[edit]

Syntax is computer programming language. SPSS use a specific form of syntax that is unique to SPSS. Although the nature of SPSS does not require that you utilize syntax, it can be your friend. Knowledge of syntax is particularly beneficial if you are repeating the same task in SPSS with different variables; it can save you from tedious, repetitive clicks.

Starting Simple: Frequency Tables[edit]

Using the example data set on genetic counselors, suppose you want a frequency table for the variable “attend” because you want an idea of how often genetic counselors are attending religious services. In SPSS click “Analyze” → “Descriptive Statistics” → “Frequencies” (this is covered in more depth in Chapter 14). Select the variable “attend.” After you click “continue,” this should show up in your output window:

24syntax01.png

Now, suppose you want a frequency table of the variable “relig” in order to see what religions many of the genetic counselors identify with. You could repeat the above procedure but choose “relig” instead of “attend.” Or you can use syntax.

To begin using syntax, go to “File” → “New” → “Syntax,” like this:

24syntax02.png

A new white window will pop up that looks like this:

24syntax03.png

This window is the Syntax Editor window and it allows you to control the many statistical tests available in SPSS (and many other functions) using programming language rather than point and click mouse movements.

One of the easiest ways to get started using syntax is simply to change your Preferences in SPSS to include syntax in the Output Window (see Chapter 7). You can then use that syntax to run the same commands or similar commands in the Syntax Editor window.

Returning to the example... Since we want to run the frequencies on the variable “relig,” we can start by returning to the original output window where you can copy the syntax from your original command just above the frequency table. The syntax is located in the red box below:

24syntax04.png

To copy the syntax, you will have to double-click on it, then select it and hit CTRL+C (the copy command in most operating systems). Now return to the Syntax Editor window and paste the syntax into the blank syntax window you have already opened, it should look like this:

24syntax05.png

While it is beyond the scope of this chapter to explain all of the syntax commands available in SPSS, the above commands are relatively straightforward. Let's begin with the first word:

FREQUENCIES

This is a command that will tell SPSS which statistical test to perform. In our case, we simply want Frequencies. The second word is paired with an equals sign and includes our variable of interest:

VARIABLES=attend

This is a modifier for the first word, FREQUENCIES, that tells SPSS which variables will be analyzed. The last part of the syntax simply tells SPSS how to organize the analysis and output:

/ORDER=ANALYSIS.

The period “.” at the end of ANALYSIS tells SPSS that is the end of a series of commands, so the software knows where this small program ends.

Now, erase the word “attend” and replace it with the word “relig,” which is the name of the religious affiliation variable in our example dataset, and click the blue arrow which tells SPSS to run the syntax. The blue arrow is circled below:

24syntax06.png

The new frequency table for the variable “relig” should pop up in your output box. Voila! You have successfully run a small computer program using SPSS syntax. Your output window will now have frequencies for both variables and should look like this:

24syntax07.png

Another Example: ANOVA[edit]

This same idea of copying the syntax and using it in the Syntax Editor to quickly run repetitive tasks can be illustrated using several of the variables in the example data set. Let's pretend you want to know if the religiosity of genetic counselors varies by their attitudes towards women obtaining abortions. In our sample data set there are seven variables examining attitudes toward abortion, each for a different reason. You could run the ANOVA seven times, pointing and clicking through the menus each time for each of the seven abortion attitudes variables. Or you can run it once, copy the syntax, paste it six times into the Syntax Editor, and simply replace the independent variable with the other six abortion attitude variables.

To run the initial ANOVA, click “Analyze” → “Compare Means” → “One-Way Anova” (covered in more detail in Chapter 21). For the dependent variable, you would select “relscale” and in the “Factor” box (which is where the independent variable goes), put the first abortion attitudes variable, “abpoor.” Run the test and you'll see this:

24syntax08.png

Now, if we wanted to see if the religiosity of genetic counselors varies by their attitudes towards women getting abortions because they are single and do not wish to marry the father of their unborn baby (“absingle”), we begin by copying and pasting the syntax from the ANOVA statistical test in the Output Window into the Syntax Editor window. Replace the variable “abpoor” with the variable “absingle.” The syntax window should look like this:

24syntax09.png

If you wanted to, you could also copy this syntax and paste it five more times, then replace the abortion attitudes variable in the syntax each time. When you're ready, run the test, click “Run.” (Note: You can run all six tests at one time by highlighting them all and selecting “Run”.) In your Output Window you should see this:

24syntax10.png

Once again, genetic counselors' religiosity varies by their abortion attitudes. However, regardless of whether or not the results show there is no significant relationship, what is important is that you can save time and energy by using syntax, particularly when you have a repetitive task you need to run many times. This chapter has only scratched the surface of syntax in SPSS; it can be used to run the entire program, avoiding the Data Editor window almost entirely. It should also be noted that other statistical software, like SAS and R, rely much more heavily on syntax. SPSS is known for being more user-friendly as it allows people to point and click to run tests.

Chapter contributed by Victoria Blyde.