SPSS 13.
For more information about SPSS® software products, please visit our Web site at http://www.spss.com or contact SPSS Inc. 233 South Wacker Drive, 11th Floor Chicago, IL 60606-6412 Tel: (312) 651-3000 Fax: (312) 651-3668 SPSS is a registered trademark and the other product names are the trademarks of SPSS Inc. for its proprietary computer software.
Preface The SPSS® 13.0 Brief Guide provides a set of tutorials designed to acquaint you with the various components of the SPSS system. You can work through the tutorials in sequence or turn to the topics for which you need additional information. You can use this guide as a supplement to the online tutorial that is included with the SPSS Base 13.0 system or ignore the online tutorial and start with the tutorials found here. SPSS 13.0 SPSS 13.0 is a comprehensive system for analyzing data.
You can also subscribe to an e-mail message list that is gatewayed to the USENET group. To subscribe, send an e-mail message to listserv@listserv.uga.edu. The text of the e-mail message should be: subscribe SPSSX-L firstname lastname. You can then post messages to the list by sending an e-mail message to listserv@listserv.uga.edu. Additional Publications For additional information about the features and operations of SPSS Base 13.0, you can consult the SPSS Base 13.
SPSS Advanced Models™ focuses on techniques often used in sophisticated experimental and biomedical research. It includes procedures for general linear models (GLM), linear mixed models, variance components analysis, loglinear analysis, ordinal regression, actuarial life tables, Kaplan-Meier survival analysis, and basic and extended Cox regression. SPSS Tables™ creates a variety of presentation-quality tabular reports, including complex stub-and-banner tables and displays of multiple response data.
Training Seminars SPSS Inc. provides both public and onsite training seminars for SPSS. All seminars feature hands-on workshops. SPSS seminars will be offered in major U.S. and European cities on a regular basis. For more information on these seminars, contact your local office, listed on the SPSS Web site at http://www.spss.com/worldwide. Technical Support The services of SPSS Technical Support are available to registered customers of SPSS.
Statistical procedures, including t tests, analysis of variance, and crosstabulations. Interactive graphics that allow you to change or add chart elements and variables dynamically; the changes appear as soon as they are specified. Standard high-resolution graphics for an extensive array of analytical and presentation charts and tables. Limitations Created for classroom instruction, the Student Version is limited to use by students and instructors for educational purposes only.
A clear description of what happened and what you were doing when the problem occurred. If possible, please try to reproduce the problem with one of the sample data files provided with the program. The exact wording of any error or warning messages that appeared on your screen. How you tried to solve the problem on your own. Technical Support for Instructors Instructors using the Student Version for classroom instruction may contact SPSS Technical Support for assistance.
Contents 1 1 Introduction Sample Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Starting SPSS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Variable Display in Dialog Boxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Opening a Data File. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Running an Analysis . . . . . . . . . . . . . . . . . .
Reading Data from a Text File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Saving Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 4 Using the Data Editor 51 Entering Numeric Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Entering String Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Defining Data . . . . . . . . . . . . . . . . . . . .
6 Working with Output 89 Using the Viewer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Using the Pivot Table Editor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Accessing Output Definitions. . . . Pivoting Tables . . . . . . . . . . . . . . Creating and Displaying Layers . . Editing Tables . . . . . . . . . . . . . . . Hiding Rows and Columns . . . . . . Changing Data Display Formats . . TableLooks . . . . . . . . . . . . . . . .
Editing Text . . . . . . . . . . . . . . . . . . . . . . . Displaying and Editing Data Value Labels . Using Templates . . . . . . . . . . . . . . . . . . . Defining Chart Options . . . . . . . . . . . . . . . Other Examples . . . . . . . . . . . . . . . . . . . . . . . ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 137 138 141 145 149 Pie Chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10 Sorting and Selecting Data 205 Sorting Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 Split-File Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Sorting Cases for Split-File Processing . . . . . . . . . . . . . . . . . . . . . . . 209 Turning Split-File Processing On and Off . . . . . . . . . . . . . . . . . . . . . . 209 Selecting Subsets of Cases. . . . . . . . . . . . . . . . . . . . . . . . . .
Nonparametric Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Chi-Square . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Time Series Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Exponential Smoothing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Chapter 1 Introduction This guide provides a set of tutorials designed to acquaint you with the various components of the SPSS system. You can work through the tutorials in sequence or turn to the topics for which you need additional information. The goal is to enable you to perform useful analyses on your data using SPSS. This chapter will introduce you to the basic environment of SPSS and demonstrate a typical session.
2 Chapter 1 Starting SPSS To start SPSS: E From the Windows Start menu choose: Programs SPSS for Windows SPSS for Windows To start SPSS for Windows Student Version: E From the Windows Start menu choose: Programs SPSS for Windows Student Version When you start a session, you see the Data Editor window.
3 Introduction Variable Display in Dialog Boxes Either variable names or longer variable labels will appear in list boxes in dialog boxes. Additionally, variables in list boxes can be ordered alphabetically or by their position in the file. In this guide, we will display variable labels in alphabetical order within list boxes. For a new user of SPSS, this provides a more complete description of variables in an easy-to-follow order.
4 Chapter 1 Alternatively, you can use the Open File button on the toolbar. Figure 1-2 Open File toolbar button This opens the Open File dialog box. Figure 1-3 Open File dialog box By default, SPSS-format data files (.sav extension) are displayed. You can display other file formats using the Files of Type drop-down list. By default, data files in the folder (directory) in which SPSS is installed are displayed.
5 Introduction Figure 1-4 Sample_files folder displayed in Open File dialog box E Click Open to open the SPSS data file. Figure 1-5 Variable labels The data file is displayed in the Data Editor. If you put the mouse cursor on a variable name (the column headings), a more descriptive variable label is displayed if one has been defined for that variable.
6 Chapter 1 By default, the actual data values are displayed. To display labels: E From the menus choose: View Value Labels Alternatively, you can use the Label tool on the toolbar. Figure 1-6 Value Labels tool Descriptive value labels are now displayed. This makes it easier to interpret the responses.
7 Introduction Running an Analysis The Analyze menu contains a list of general reporting and statistical analysis categories. Most of the categories are followed by an arrow, which indicates that there are several analysis procedures available within the category; they will appear on a submenu when the category is selected. We’ll start with a simple frequency table (table of counts). E From the menus choose: Analyze Descriptive Statistics Frequencies... This opens the Frequencies dialog box.
8 Chapter 1 E Select (click) the variable Income category. Figure 1-9 Variable labels and names in the Frequencies dialog box A more complete description of each variable pops up when the cursor is over it. The variable name (in square brackets) is inccat, and it has the variable label Income category. If there were no variable label, only the variable name would appear in the list box.
9 Introduction Figure 1-10 Defined labels for income variable All of the defined value labels for the variable are displayed. E Click Gender [gender] in the source variable list, and then click the right-arrow button to move the variable into the target Variable(s) list. E Click Income category [inccat] in the source list, and then click the right arrow button again.
10 Chapter 1 A pound sign (#) icon next to the variable name indicates that the variable is numeric. An icon with the letter “A” indicates that the variable is a string (alphanumeric) variable, which may contain both letters and numbers. A less-than sign (left angle bracket) indicates that the variable is a short string, containing eight or fewer characters. E Click OK to run the procedure. Viewing Results Figure 1-12 Viewer window Results are displayed in the Viewer window.
11 Introduction Figure 1-13 Frequency table of income categories This takes you directly to the frequency table for income categories. This frequency table shows the number and percentage of people in each income category. Creating Charts Although some statistical procedures can create high-resolution charts, you can also use the Graphs menu to create charts. For example, you could create a chart that shows the relationship between wireless telephone service and PDA (personal digital assistant) ownership.
12 Chapter 1 Figure 1-14 Define Clustered Bar dialog box E Scroll down the source variable list and select Wireless service [wireless] as the Category Axis variable. E Select Owns PDA [ownpda] as the Define Clusters By variable. E Click OK to create the chart.
13 Introduction Figure 1-15 Bar chart displayed in Viewer window The bar chart is displayed in the Viewer. It shows that people with wireless phone service are far more likely to have PDAs than people without wireless service. You can edit charts and tables by double-clicking on them in the contents pane of the Viewer window, and you can copy and paste your results into other applications. Those topics will be covered later.
Chapter Using the Help System 2 Help is available in a number of different ways, including: Help menu. Every window has a Help menu on the menu bar. The Topics menu item provides access to the Help system, where you can use the Contents and Index tabs to find topics. The Tutorial menu item provides access to the introductory tutorial. Dialog box Help buttons. Most dialog boxes have a Help button that takes you directly to a Help topic for that dialog box.
16 Chapter 2 Help Contents Tab The Topics item on the Help menu opens a Help window. E From the menus choose: Help Topics Figure 2-1 Help Contents tab The Contents tab in the left pane of the Help window is an expandable and collapsible table of contents. It is most useful if you’re looking for general information or are unsure of what index term to use to find what you’re looking for.
17 Using the Help System Help Index Tab Figure 2-2 Help Index tab E Click the Index tab in the left pane of the Help window. The Index tab provides a searchable index that makes it easy to find specific topics. The Index tab is organized in alphabetical order, just like a book index. It uses incremental search to find what you’re looking for.
18 Chapter 2 For example, you can: E Type med. Figure 2-3 Incremental index search The index scrolls to and highlights the first index entry that starts with these letters, which is median. Dialog Box Help Most dialog boxes have a Help button that displays a Help topic about what the dialog box does and how to do it. E From the menus choose: Analyze Descriptive Statistics Frequencies...
19 Using the Help System E Click Help. Figure 2-4 Dialog box Help topic In this example, the Help topic describes the purpose of the Frequencies procedure and provides an example. Statistics Coach The Statistics Coach can help to guide you through the process of finding the procedure that you want to use.
20 Chapter 2 Figure 2-5 Statistics Coach, first step The Statistics Coach presents a series of questions designed to find the appropriate procedure. The first question is simply “What do you want to do?” For example, if you want to summarize data: E Select Summarize, describe, or present data. E Then click Next.
21 Using the Help System Figure 2-6 Data type selection The next question asks about the type of data you want to summarize. If you’re unsure, each choice displays different examples. E Select Continuous, numeric data (interval, ratio).
22 Chapter 2 Figure 2-7 Selecting a different data type The example changes to reflect your choice. If you’re still unsure, you can: E Click More Examples. A new example of the same data type is displayed. If the examples don’t provide enough information, you can: E Click Help.
23 Using the Help System Figure 2-8 Statistics Coach Help topic In this example, the Help topic defines the different data types. E Close the Help window. E Select Data in categories (nominal, ordinal) and click Next.
24 Chapter 2 E Select Tables and numbers and click Next.
25 Using the Help System E Select Individual case listings within categories. Figure 2-10 Statistics Coach, final step When the Statistics Coach has enough information, the Next button changes to Finish.
26 Chapter 2 When you click Finish, the dialog box for the selected procedure opens automatically, and a Help topic for the procedure is also displayed. Figure 2-11 Dialog box and Help topic This is a custom Help topic, based on your selections in the Statistics Coach. Since some dialog boxes perform numerous functions, more than one path in the Statistics Coach may lead to the same dialog box, but the instructions in the Help topic may be different.
27 Using the Help System Figure 2-12 “Tell me more” Help topic This Help topic provides detailed information on the data type(s) appropriate for the selected procedure. Case Studies Case studies provide comprehensive overviews of each procedure. Data files used in the examples are installed with SPSS, so you can follow along, performing the same analysis—from opening the data source and selecting variables for analysis to interpreting the results.
28 Chapter 2 Figure 2-13 Accessing the case studies E Select Case Studies on the pop-up context menu. Case studies are not available for all procedures. The Case Studies choice on the context menu will appear only if the feature is available for the procedure that created the selected pivot table.
Chapter 3 Reading Data Data can be entered directly into SPSS, or it can be imported from a number of different sources. The processes for reading data stored in SPSS data files, spreadsheet applications, such as Microsoft Excel, database applications, such as Microsoft Access, and text files are all discussed in this chapter. Basic Structure of an SPSS Data File Figure 3-1 Data Editor SPSS data files are organized by cases (rows) and variables (columns).
30 Chapter 3 Reading an SPSS Data File SPSS data files, which have a .sav file extension, contain your saved data. To open demo.sav, an example file that is installed with the product: E From the menus choose: File Open Data... E Make sure that SPSS (*.sav) is selected in the Files of Type drop-down list. Figure 3-2 Open File dialog box E Navigate to the sample_files folder. E Select demo.sav and click Open.
31 Reading Data The data are now displayed in the Data Editor. Figure 3-3 Opened data file Reading Data from Spreadsheets Rather than typing all of your data directly into the Data Editor, you can read data from applications such as Microsoft Excel. You can also read column headings as variable names. E From the menus choose: File Open Data...
32 Chapter 3 E Select Excel (*.xls) from the Files of Type drop-down list. Figure 3-4 Open File dialog box E Select demo.xls and click Open to read this spreadsheet. The Opening Excel Data Source dialog box is displayed, allowing you to specify whether variable names are to be included in the spreadsheet, as well as the cells that you want to import. In Excel 5 or later, you can also specify which worksheets you want to import.
33 Reading Data E Make sure that Read variable names from first row of data is selected. This option reads column headings as variable names. If the column headings do not conform to the SPSS variable-naming rules, they are converted into valid variable names and the original column headings are saved as variable labels. If you want to import only a portion of the spreadsheet, specify the range of cells to be imported in the Range text box. E Click OK to read the Excel file.
34 Chapter 3 Reading Data from a Database Data from database sources are easily imported using the Database Wizard. Any database that uses ODBC (Open Database Connectivity) drivers can be read directly by SPSS after the drivers are installed. ODBC drivers for many database formats are supplied on the installation CD. Additional drivers can be obtained from third-party vendors. One of the most common database applications, Microsoft Access, is discussed in this example.
35 Reading Data If MS Access Database is not listed here, you need to run Microsoft Data Access Pack.exe, which can be found in the Microsoft Data Access Pack folder on the CD. Figure 3-8 ODBC Driver Login dialog box E Click Browse to navigate to the Access database file that you want to open. Figure 3-9 Open File dialog box E Select demo.mdb and click Open to continue. E Click OK in the login dialog box.
36 Chapter 3 In the next step, you can specify the tables and variables that you want to import. Figure 3-10 Select Data step E Drag the entire demo table to the Retrieve Fields in This Order list. E Click Next.
37 Reading Data In the next step, you select which records (cases) to import. Figure 3-11 Limit Retrieved Cases step If you do not want to import all cases, you can import a subset of cases (for example, males older than 30), or you can import a random sample of cases from the data source. For large data sources, you may want to limit the number of cases to a small, representative sample to reduce the processing time. E Click Next to continue.
38 Chapter 3 If you are in distributed mode, connected to a remote server (available with SPSS Server), the next step allows you to aggregate the data before reading it into SPSS. Figure 3-12 Aggregate Data step You can also aggregate data after reading it into SPSS, but pre-aggregating may save time for large data sources. E For this example, you don’t need to aggregate the data. If you see this step in the Database Wizard, click Next.
39 Reading Data Field names are used to create variable names. If necessary, the names are converted to valid variable names. The original field names are preserved as variable labels. You can also change the variable names before importing the database. Figure 3-13 Define Variables step E Click the Recode to Numeric cell in the Gender field. This option converts string variables to integer variables and retains the original value as the value label for the new variable. E Click Next to continue.
40 Chapter 3 You can also sort data after reading it into SPSS, but presorting may save time for large data sources. E For this example, you don’t need to sort the data. If you see this step in the Database Wizard, click Next. The SQL statement created from your selections in the Database Wizard appears in the Results step. This statement can be executed now or saved to a file for later use. Figure 3-14 Results step E Click Finish to import the data.
41 Reading Data All of the data in the Access database that you selected to import are now available in the SPSS Data Editor. Figure 3-15 Data imported from an Access database Reading Data from a Text File Text files are another common source of data. Many spreadsheet programs and databases can save their contents in one of many text file formats. Comma or tab-delimited files refer to rows of data that use commas or tabs to indicate each variable. In this example, the data are tab delimited.
42 Chapter 3 E From the menus choose: File Read Text Data... E Choose Text (*.txt) from the Files of Type list. Figure 3-16 Open File dialog box E Select demo.txt and click Open to read the selected file.
43 Reading Data The Text Import Wizard guides you through the process of defining how the specified text file should be interpreted. Figure 3-17 Text Import Wizard - Step 1 of 6 E In Step 1, you can choose a predefined format or create a new format in the wizard. Select No to indicate that a new format should be created. E Click Next to continue.
44 Chapter 3 As stated earlier, this file uses tab-delimited formatting. Also, the variable names are defined on the top line of this file. Figure 3-18 Text Import Wizard - Step 2 of 6 E Select Delimited to indicate that the data use a delimited formatting structure. E Select Yes to indicate that variable names should be read from the top of the file. E Click Next to continue.
45 Reading Data E Type 2 in the top section of next dialog box to indicate that the first row of data starts on the second line of the text file. Figure 3-19 Text Import Wizard - Step 3 of 6 E Keep the default values for the remainder of this dialog box, and click Next to continue.
46 Chapter 3 The Data preview in Step 4 provides you with a quick way to ensure that your data are being properly read by SPSS. Figure 3-20 Text Import Wizard - Step 4 of 6 E Select Tab and deselect the other options. E Click Next to continue.
47 Reading Data Because the variable names may have been truncated to fit SPSS formatting requirements, this dialog box gives you the opportunity to edit any undesirable names. Figure 3-21 Text Import Wizard - Step 5 of 6 Data types can be defined here as well. For example, it’s safe to assume that the income variable is meant to contain a certain dollar amount. To change a data type: E Under Data preview, select the variable you want to change, which is Income in this case.
48 Chapter 3 E Select Dollar from the Data format drop-down list. Figure 3-22 Change the data type E Click Next to continue.
49 Reading Data Figure 3-23 Text Import Wizard - Step 6 of 6 E Leave the default selections in this dialog box, and click Finish to import the data. Saving Data To save an SPSS data file, the Data Editor window must be the active window. E From the menus choose: File Save E Browse to the desired directory.
50 Chapter 3 E Type a name for the file in the File name text box. The Variables button can be used to select which variables in the Data Editor are saved to the SPSS data file. By default, all variables in the Data Editor are retained. E Click Save. The name in the title bar of the Data Editor will change to the filename you specified. This confirms that the file has been successfully saved as an SPSS data file.
Chapter Using the Data Editor 4 The Data Editor displays the contents of the active data file. The information in the Data Editor consists of variables and cases. In Data View, columns represent variables and rows represent cases (observations). In Variable View, each row is a variable, and each column is an attribute associated with that variable. Variables are used to represent the different types of data that you have compiled. A common analogy is that of a survey.
52 Chapter 4 Figure 4-1 Variable names in Variable View E In the first row of the first column, type age. E In the second row, type marital. E In the third row, type income. New variables are automatically given a numeric data type. If you don’t enter variable names, unique names are automatically created. However, these names are not descriptive and are not recommended for large data files. E Click the Data View tab to continue entering the data.
53 Using the Data Editor Begin entering data in the first row, starting at the first column. Figure 4-2 Values entered in Data View E In the age column, type 55. E In the marital column, type 1. E In the income column, type 72000. E Move the cursor to the first column of the second row to add the next subject’s data. E In the age column, type 53. E In the marital column, type 0. E In the income column, type 153000.
54 Chapter 4 Currently, the age and marital columns display decimal points, even though their values are intended to be integers. To hide the decimal points in these variables: E Click the Variable View tab at the bottom of the Data Editor window. E Select the Decimals column in the age row and type 0 to hide the decimal. E Select the Decimals column in the marital row and type 0 to hide the decimal.
55 Using the Data Editor E Click the Type cell. E Click the button in the Type cell to open the Variable Type dialog box. Figure 4-4 Button shown in Type cell for sex E Select String to specify the variable type.
56 Chapter 4 E Click OK to save your changes and return to the Data Editor. Figure 4-5 Variable Type dialog box Defining Data In addition to defining data types, you can also define descriptive variable and value labels for variable names and data values. These descriptive labels are used in statistical reports and charts. Adding a Variable Label Labels are meant to provide descriptions of variables. These descriptions are often longer versions of variable names. Labels can be up to 256 characters long.
57 Using the Data Editor E In the Label column of the sex row, type Gender. Figure 4-6 Variable labels entered in Variable View Changing Variable Type and Format The Type column displays the current data type for each variable. The most common are numeric and string, but many other formats are supported. In the current data file, the income variable is defined as a numeric type. E Click the Type cell for the income row, and then click the button to open the Variable Type dialog box.
58 Chapter 4 E Select Dollar in the Variable Type dialog box. Figure 4-7 Variable Type dialog box The formatting options for the currently selected data type are displayed. E Select the format of this currency. For this example, select $###,###,###. E Click OK to save your changes. Adding Value Labels for Numeric Variables Value labels provide a method for mapping your variable values to a string label. In the case of this example, there are two acceptable values for the marital variable.
59 Using the Data Editor E Click Add to add this label to the list. Figure 4-8 Value Labels dialog box E Repeat the process, this time typing 1 in the Value field and Married in the Value Label field. E Click Add, and then click OK to save your changes and return to the Data Editor. These labels can also be displayed in Data View, which can help to make your data more readable. E Click the Data View tab at the bottom of the Data Editor window.
60 Chapter 4 Figure 4-9 Value labels displayed in Data View Adding Value Labels for String Variables String variables may require value labels as well. For example, your data may use single letters, M or F, to identify the sex of the subject. Value labels can be used to specify that M stands for Male and F stands for Female. E Click the Variable View tab at the bottom of the Data Editor window. E Click the Values cell in the sex row, and then click the button to open the Value Labels dialog box.
61 Using the Data Editor E Click Add to add this label to your data file. Figure 4-10 Value Labels dialog box E Repeat the process, this time typing M in the Value field and Male in the Value Label field. E Click Add, and then click OK to save your changes and return to the Data Editor. Because string values are case sensitive, you should make sure that you are consistent. A lowercase m is not the same as an uppercase M.
62 Chapter 4 E In the second row, select the cell for sex and select Female from the drop-down list. Figure 4-11 Using variable labels to select values Only defined values are listed, which helps to ensure that the data entered are in a format that you expect. Handling Missing Data Missing or invalid data are generally too common to ignore. Survey respondents may refuse to answer certain questions, may not know the answer, or may answer in an unexpected format.
63 Using the Data Editor Figure 4-12 Missing values displayed as periods The reason a value is missing may be important to your analysis. For example, you may find it useful to distinguish between those who refused to answer a question and those who didn’t answer a question because it was not applicable. Missing Values for a Numeric Variable E Click the Variable View tab at the bottom of the Data Editor window.
64 Chapter 4 In this dialog box, you can specify up to three distinct missing values, or a range of values plus one additional discrete value. Figure 4-13 Missing Values dialog box E Select Discrete missing values. E Type 999 in the first text box and leave the other two empty. E Click OK to save your changes and return to the Data Editor. Now that the missing data value has been added, a label can be applied to that value.
65 Using the Data Editor E Click Add to add this label to your data file. E Click OK to save your changes and return to the Data Editor. Missing Values for a String Variable Missing values for string variables are handled similarly to those for numeric values. Unlike numeric values, empty fields in string variables are not designated as system missing. Rather, they are interpreted as an empty string. E Click the Variable View tab at the bottom of the Data Editor window.
66 Chapter 4 E Type No Response in the Value Label field. Figure 4-15 Value Labels dialog box E Click Add to add this label to your project. E Click OK to save your changes and return to the Data Editor. Copying and Pasting Value Attributes Once you’ve defined variable attributes for a variable, you can copy these attributes and apply them to other variables.
67 Using the Data Editor E In Variable View, type agewed in the first cell of the first empty row. Figure 4-16 agewed variable in Variable View E In the Label column, type Age Married. E Click the Values cell in the age row. E From the menus choose: Edit Copy E Click the Values cell in the agewed row.
68 Chapter 4 The defined values from the age variable are now applied to the agewed variable.
69 Using the Data Editor To apply the attribute to multiple variables, simply select multiple target cells (click and drag down the column).
70 Chapter 4 When you paste the attribute, it is applied to all of the selected cells. Figure 4-19 Values pasted into multiple cells New variables are automatically created if you paste the values into empty rows. You can also copy all of the attributes from one variable to another.
71 Using the Data Editor E Click the row number in the marital row. Figure 4-20 Selected row E From the menus choose: Edit Copy E Click the row number of the first empty row.
72 Chapter 4 All of the attributes of the marital variable are applied to the new variable. Figure 4-21 Values pasted into row Defining Variable Properties for Categorical Variables For categorical (nominal, ordinal) data, Define Variable Properties can help you define value labels and other variable properties. Define Variable Properties: Scans the actual data values and lists all unique data values for each selected variable. Identifies unlabeled values and provides an “auto-label” feature.
73 Using the Data Editor This example uses the data file demo.sav. This data file already has defined value labels; so before we start, let’s enter a value for which there is no defined value label: E In Data View of the Data Editor, click the first data cell for the variable ownpc (you may have to scroll to the right) and enter the value 99. E From the menus choose: Data Define Variable Properties...
74 Chapter 4 necessary for our sample data file. Even though it contains over 6,000 cases, it doesn’t take very long to scan that many cases. E Drag and drop Owns computer [ownpc] through Owns VCR [ownvcr] into the Variables to Scan list. You might notice that the measurement level icons for all of the selected variables indicate that they are scale variables, not categorical variables.
75 Using the Data Editor The current level of measurement for the selected variable is scale. You can change the measurement level by selecting one from the drop-down list or you can let Define Variable Properties suggest a measurement level. E Click Suggest. Figure 4-24 Suggest Measurement Level dialog box Since the variable doesn’t have very many different values and all of the scanned cases contain integer values, the proper measurement level is probably ordinal or nominal.
76 Chapter 4 The value that we entered, 99, is displayed in the grid. The count is only 1 because we changed the value for only one case, and the Label column is empty because we haven’t defined a value label for 99 yet. An X in the first column of the Scanned Variable List also indicates that the selected variable has at least one observed value without a defined value label. E In the Label column for the value of 99, enter No answer. E Then click (check) the box in the Missing column.
77 Using the Data Editor Figure 4-26 Apply Labels and Level dialog box In the Apply Labels and Level dialog box, select all of the variables in the list, and then click Copy.
78 Chapter 4 If you select any other variable in the list in the Define Variable Properties main dialog box now, you’ll see that they are all now ordinal variables, with a value of 99 defined as user missing and a value label of No answer. Figure 4-27 New variable properties defined for ownfax E Click OK to save all of the variable properties that you have defined.
Chapter Examining Summary Statistics for Individual Variables 5 This chapter discusses simple summary measures and how the level of measurement of a variable influences the types of statistics that should be used. We will use the data file demo.sav. Level of Measurement Different summary measures are appropriate for different types of data, depending on the level of measurement: Categorical. Data with a limited number of distinct values or categories (for example, gender or marital status).
80 Chapter 5 Scale. Data measured on an interval or ratio scale, where the data values indicate both the order of values and the distance between values. For example, a salary of $72,195 is higher than a salary of $52,398, and the distance between the two values is $19,797. Also referred to as quantitative or continuous data. Summary Measures for Categorical Data For categorical data, the most typical summary measure is the number or percentage of cases in each category.
81 Examining Summary Statistics for Individual Variables Figure 5-2 Frequency tables The frequency tables are displayed in the Viewer window. The frequency tables reveal that only about 21% of the people own PDAs, but almost everybody owns a TV (99.2%). This might not be an interesting revelation, although it might be interesting to find out more about the small group of people who do not own televisions.
82 Chapter 5 You can use the Dialog Recall button on the toolbar to quickly return to recently used procedures. Figure 5-3 Dialog Recall tool E Click Charts. E Select Bar charts and then click Continue. Figure 5-4 Frequencies Charts dialog box E Click OK in the main dialog box to run the procedure.
83 Examining Summary Statistics for Individual Variables Figure 5-5 Bar chart In addition to the frequency tables, the same information is now displayed in the form of bar charts, making it easy to see that most people do not own PDAs but almost everyone owns a TV. Summary Measures for Scale Variables There are many summary measures available for scale variables, including: Measures of central tendency.
84 Chapter 5 E Click Reset to clear any previous settings. E Select Household income in thousands (income) and move it into the Variable(s) list. Figure 5-6 Scale variable selected for analysis E Click Statistics. E Select Mean, Median, Std. deviation, Minimum, and Maximum. Figure 5-7 Frequencies Statistics dialog box E Click Continue.
85 Examining Summary Statistics for Individual Variables E Deselect Display frequency tables in the main dialog box. (Frequency tables are usually not useful for scale variables since there may be almost as many distinct values as there are cases in the data file.) E Click OK to run the procedure. The Frequencies Statistics table is displayed in the Viewer window.
86 Chapter 5 Histograms for Scale Variables E Open the Frequencies dialog box again. E Click Charts. E Select Histograms and With normal curve. Figure 5-9 Frequencies Charts dialog box E Click Continue and click OK in the main dialog box to run the procedure.
87 Examining Summary Statistics for Individual Variables Figure 5-10 Histogram The majority of cases are clustered at the lower end of the scale, with most falling below 100,000. There are, however, a few cases in the 500,000 range and beyond (too few to even be visible without modifying the histogram). These high values for only a few cases have a significant effect on the mean but little or no effect on the median, making the median a better indicator of central tendency in this example.
Chapter Working with Output 6 The results from running a statistical procedure are displayed in the Viewer. The output produced can be statistical tables, charts or graphs, or text, depending on the choices you make when you run the procedure. This chapter uses the files viewertut.spo and demo.sav.
90 Chapter 6 The Viewer window is divided into two panes. The outline pane contains an outline of all of the information stored in the Viewer. The contents pane contains statistical tables, charts, and text output. Use the scrollbars to navigate through the window’s contents, both vertically and horizontally. For easier navigation, click an item in the outline pane to display it in the contents pane.
91 Working with Output E Click the box with the minus sign ( ) to the left of the procedure whose results you want to hide, or click the box next to the topmost item in the outline pane to hide all of the output. Figure 6-2 Hidden output in the Viewer The outline collapses, visually indicating that these results are hidden. You can also change the order in which the output is displayed. E In the outline pane, click on the items you want to move.
92 Chapter 6 E Drag the selected items to a new location in the outline and release the mouse button. Figure 6-3 Reordered output in the Viewer You can also move output items by clicking and dragging them in the contents pane. Using the Pivot Table Editor The results from most statistical procedures are displayed in pivot tables. Accessing Output Definitions Many statistical terms are displayed in the output. Definitions of these terms can be accessed directly in the Viewer.
93 Working with Output E Right-click Expected Count and select What’s This? from the pop-up context menu. The definition is displayed in a pop-up window. Figure 6-4 Pop-up definition Pivoting Tables The default tables produced may not display information as neatly or as clearly as you would like. With pivot tables, you can transpose rows and columns (“flip” the table), adjust the order of data in a table, and modify the table in many other ways.
94 Chapter 6 E If the Pivoting Trays window is not visible, from the menus choose: Pivot Pivoting Trays Pivoting Trays provide a way to move data between columns, rows, and layers.
95 Working with Output E Click one of the pivot icons to see what it represents. The shaded area in the table indicates what will be moved when you move the pivot icon. A pop-up label also indicates what the icon represents in the table. Figure 6-6 Pivot icon labels E Drag the Statistics pivot icon from the Row dimension to the bottom of the Column dimension. The table is immediately reconfigured to reflect your changes.
96 Chapter 6 The order of the pivot icons in a dimension reflects the order of the elements in the table. Figure 6-7 Pivoting tray icon associations E Drag the Owns PDA icon to the left of the Internet icon and release the mouse button to reverse the order of these two rows Creating and Displaying Layers Layers can be useful for large tables with nested categories of information. By creating layers, you simplify the look of the table, making it easier to read.
97 Working with Output E Drag the Gender pivot icon to the Layer dimension. Figure 6-8 Gender pivot icon in the Layer dimension To view the different layers, you can either click the arrows on the layer pivot icon, or you can select a layer from the drop-down list in the table.
98 Chapter 6 Figure 6-9 Choosing a layer Editing Tables Unless you’ve taken the time to create a custom TableLook, pivot tables are created with standard formatting. You can change the formatting of any text within a table. Formats that you can change include font name, font size, font style (bold or italic), and color. E Double-click the Level of education table. E If the Formatting toolbar is not visible, from the menus choose: View Toolbar E Click the title text, Level of education.
99 Working with Output E To change the color of the title text, click the Text Color tool and select a color. Figure 6-10 Reformatted title text in the pivot table You can also edit the contents of tables and labels. For example, you can change the title of this table. E Double-click the title. E Type Education Level for the new label. Note that if you change the values in a table, totals and other statistics are not recalculated.
100 Chapter 6 Hiding Rows and Columns Some of the data displayed in a table may not be useful or it may unnecessarily complicate the table. Fortunately, you can hide entire rows or columns without losing any data. E Double-click the Education Level table. E Ctrl-Alt-click on the Valid Percent column label to select all of the cells in that column. E Right-click the highlighted column and select Hide Category from the pop-up context menu. The column is now hidden but not deleted.
101 Working with Output E From the menus choose: Format Cell Properties E Type 0 in the Decimals field to hide all decimal points in this column. Figure 6-11 Cell Properties dialog box You can also change the data type and format in this dialog box. E Select the type that you want from the Category list, and then select the format for that type in the Format list.
102 Chapter 6 E Click OK to apply your changes and return to the Viewer. Figure 6-12 Decimals hidden in Percent column The decimals are now hidden in the Percent column. TableLooks The look and feel of your tables are a critical part of providing clear, concise, and meaningful results. For example, if your table is difficult to read, the information contained within that table may not be easily understood. Using Predefined Formats E Double-click the Marital status table.
103 Working with Output The TableLooks dialog box lists a variety of predefined styles. Select a style from the list to preview it in the Sample window to the right. Figure 6-13 TableLooks dialog box You can use a style as is, or you can edit an existing style to better suit your needs. E To use an existing style, select one and click OK. Customizing TableLook Styles You can customize a format to fit your specific needs.
104 Chapter 6 E Select the style that is closest to your desired format and click Edit Look. E Click the Cell Formats tab to view the formatting options. Figure 6-14 Table Properties dialog box The formatting options include font name, font size, style, and color. Additional options include alignment, shading, foreground and background colors, and margin sizes. The Sample window provides a preview of how the formatting changes affect your table. Each area of the table can have different formatting styles.
105 Working with Output E Click OK to return to the TableLooks dialog box. You can save your new style, which allows you to apply it to future tables easily. E Click Save As. E Navigate to the desired target directory and enter a name for your new style in the File Name text box. E Click Save. E Click OK to apply your changes and return to the Viewer. The table now contains the custom formatting that you specified.
106 Chapter 6 To change the default TableLook style for your pivot tables, from the menus choose: Edit Options... E Click the Pivot Tables tab in the Options dialog box. Figure 6-16 Options dialog box E Select the TableLook style that you want to use for all new tables. Use the Sample window to see a preview of each TableLook. E Click OK to save your settings and close the dialog box. All tables that you create after changing the default TableLook automatically conform to the new formatting rules.
107 Working with Output Customizing the Initial Display Settings The initial display settings include the alignment of objects in the Viewer, whether objects are shown or hidden by default, and the width of the Viewer window. To change these settings: E From the menus choose: Edit Options... E Click the Viewer tab.
108 Chapter 6 The settings are applied on an object-by-object basis. For example, you can customize the way charts are displayed without making any changes to the way tables are displayed. Simply select the object that you want to customize, and make the desired changes. E Click the Title icon to display its settings. E Click Center to display all titles in the (horizontal) center of the Viewer. You can also hide elements, such as the log and warning messages, that tend to clutter your output.
109 Working with Output Displaying Variable and Value Labels In most cases, displaying the labels for variables and values is more effective than displaying the variable name or the actual data value. There may be cases, however, when you want to display both the names and the labels. E From the menus choose: Edit Options... E Click the Output Labels tab. Figure 6-19 Output Labels options You can specify different settings for the outline and contents panes.
110 Chapter 6 E Then, select Values from the Variable Values in Labels drop-down list to show data values instead of labels.
111 Working with Output The new settings are applied the next time you run a statistical procedure. Figure 6-21 Variable names and values displayed Using Results in Other Applications Your results can be used in many applications. For example, you may want to include a chart or graph in a presentation or report. Applications such as Microsoft’s PowerPoint or Word can display your results as plain text, rich text, or as a metafile, which is a graphical representation of the output.
112 Chapter 6 Pasting Results as Rich Text You can paste pivot tables into Word as native Word tables. Text formatting, such as font size and color, is not retained, but columns and rows are properly aligned. Because the table is in a text format, the data can be edited after you paste it into your document. E Click the Marital status table in the Viewer. E From the menus choose: Edit Copy E Open your word processing application.
113 Working with Output Figure 6-23 Pivot table displayed in Word The table is now displayed in your document. You can apply custom formatting, edit the data, and resize the table to fit your needs. Pasting Results as Metafiles Pasting your results as metafiles maintains the formatting of your output. Your pasted output becomes an image in the target document. E Click the Marital status table in the Viewer. E From the menus choose: Edit Copy E Open your word processing application.
114 Chapter 6 E Select Picture in the Paste Special dialog box. (In some applications, the choice may be “metafile” instead of “Picture.”) Figure 6-24 Paste Special dialog box E Click OK to paste your results into the current document.
115 Working with Output The metafile is now embedded in your document. This image is a snapshot of the Marital status table. Only the visible portions of the table are copied. Information in hidden categories or layers is not included in the metafile. Pasting Results as Text Pivot tables can be copied to other applications as plain text. Formatting styles are not retained in this method, but you gain the ability to edit the table data after you paste it into the target application.
116 Chapter 6 E Click OK to paste your results into the current document. Figure 6-27 Pivot table displayed in Word Each column of the table is separated by tabs. You can change the column widths by adjusting the tab stops in your word processing application. Exporting Results to Microsoft Word, PowerPoint, and Excel Files SPSS allows you to export results to a single Microsoft Word, PowerPoint, or Excel file. You can export selected items or all items in the Viewer.
117 Working with Output There are several options for exporting the results. Figure 6-28 Export Output dialog box First, you can select which type of output file you want to create. For Word and PowerPoint, you can create a file that contains charts (Output Document) or one that does not contain charts (Output Document (No Charts)). Charts are embedded in Word documents as Windows metafiles. For Excel, you can create only documents that do not contain charts (Output Document (No Charts)).
118 Chapter 6 E Select All Objects in the Export What group. Finally, you select the file format. E Select Word/RTF file (*.doc) from the File Type drop-down list. E Click OK to generate the Word file. When you open the resulting file in Word, you can see how the results are exported. Notes, which are not visible objects, appear in Word because you chose to export all objects. Figure 6-29 Output.
119 Working with Output Pivot tables become Word tables, with all of the formatting of the original pivot table, including fonts, colors, borders, and so on.
120 Chapter 6 Charts become embedded Windows metafiles. Figure 6-31 Charts in Word Text output is displayed in a fixed-pitch font.
121 Working with Output If you export to a PowerPoint file, each exported item is placed on a separate slide. Pivot tables exported to PowerPoint become Word tables, with all of the formatting of the original pivot table, including fonts, colors, borders, and so on. Figure 6-33 Pivot tables in PowerPoint Charts selected for export to PowerPoint are embedded in the PowerPoint file.
122 Chapter 6 If you export to an Excel file, results are exported differently. Figure 6-35 Output.xls in Excel Pivot table rows, columns, and cells become Excel rows, columns, and cells.
123 Working with Output Each line in the text output is a row in the Excel file, with the entire contents of the line contained in a single cell. Charts are not exported at all. Figure 6-37 Text output in Excel Exporting Results to HTML and Text Formats You can also export results to HTML (hypertext markup language) and text formats and a number of graphic formats. When saving as HTML and text, all non-graphic output from SPSS can be exported into a single file and read by other programs.
124 Chapter 6 When you export HTML and text output, charts can be exported as well, but not to a single file.
125 Working with Output Each chart will be saved as a file in a format that you specify, and references to these graphics files will be placed in the HTML or text document created by SPSS. There is also an option to export all charts, or selected charts, in separate graphics files.
Chapter Creating and Editing Charts 7 You can create and edit a wide variety of chart types in SPSS. In these examples, we will create and edit three commonly used types of charts: Simple bar chart Pie chart Scatterplot with groups Chart Creation Basics To demonstrate the basics of chart creation, we will create a bar chart of mean income for different levels of job satisfaction. This example uses the data file demo.sav. E From the menus choose: Graphs Bar...
128 Chapter 7 E Click Summaries for groups of cases. Figure 7-1 Bar Charts dialog box E Click Define. In the Define Simple Bar dialog box, specify the variable or variables to use for creating the chart. You want to create a chart that shows bars for the mean income of each job satisfaction category. E Select Job satisfaction as the category axis variable. E In the Bars Represent group, select Other statistic (e.g., mean). E Select Household income in thousands as the variable to represent the bars.
129 Creating and Editing Charts Figure 7-2 Define Simple Bar dialog box E Click OK to create the bar chart.
130 Chapter 7 Figure 7-3 Bar chart The bar chart reveals that respondents who are more satisfied with their jobs also tend to have higher household incomes. Chart Editing Basics You can edit charts in a variety of ways. For the sample bar chart that you created, you will: Change colors. Format numbers in tick labels. Edit text. Display data value labels. Use chart templates.
131 Creating and Editing Charts To edit the chart, open it in the Chart Editor. E Double-click the bar chart to open it in the Chart Editor. Figure 7-4 Bar chart in the Chart Editor Selecting Chart Elements To edit a chart element, you first select it. E Click on any one of the bars. The blue rectangles around the bars indicate that they are selected.
132 Chapter 7 There are general rules for selecting elements in simple charts: When no data elements are selected, click any data element to select all data elements. When all data elements are selected, click a data element to select only that data element. You can select a different data element by clicking it. To select multiple data elements, click each element while pressing the Ctrl key. Note: The behavior is slightly different for grouped charts.
133 Creating and Editing Charts This opens the Properties window, showing the tabs that apply to the bars you selected. These tabs change depending on what chart element you select in the Chart Editor. For example, if you had selected a text frame instead of bars, different tabs would appear. You will use these tabs to do most chart editing. Figure 7-5 Properties window Changing Bar Colors First, you will change the color of the bars.
134 Chapter 7 E Click the Fill & Border tab. E Click the swatch next to Fill to indicate that you want to change the fill color of the bars. The numbers below the swatch specify the red, green, and blue settings for the current color. E Click the dark blue color, which is second from the left in the second row. Figure 7-6 Fill & Border tab E Click Apply.
135 Creating and Editing Charts The bars in the chart are now dark blue. Figure 7-7 Edited bar chart showing blue bars Formatting Numbers in Tick Labels Notice that the numbers on the y axis are scaled in thousands. The numbers also include two unnecessary decimal places. To make the chart more attractive and easier to interpret, we will change the number format in the tick labels and then edit the axis title appropriately. E Select the y axis tick labels by clicking any one of them.
136 Chapter 7 E Click the Number Format tab. E You do not want the tick labels to display decimal places, so type 0 in the Decimal Places text box. E Type 0.001 in the Scaling Factor text box. The scaling factor is the number by which the Chart Editor divides the displayed number. Because 0.001 is a fraction, dividing by it will increase the numbers in the tick labels by 1,000. Thus, the numbers will no longer be in thousands; they will be unscaled. E Select Display Digit Grouping.
137 Creating and Editing Charts The tick labels reflect the new number formatting: there are no decimal places, the numbers are no longer scaled, and each thousandth place is specified with a character. Figure 7-9 Edited bar chart showing new number format Editing Text Now that you have changed the number format of the tick labels, the axis title is no longer accurate. Next, you will change the axis title to reflect the new number format. Note: You do not need to open the Properties window to edit text.
138 Chapter 7 E Delete the following text: in thousands E Press Enter to exit edit mode and update the axis title. The axis title now accurately describes the contents of the tick labels. Figure 7-10 Bar chart showing edited y axis title Displaying and Editing Data Value Labels Another common task is to show the exact values associated with the data elements (which are bars in this example). These values are displayed in data labels.
139 Creating and Editing Charts Figure 7-11 Bar chart showing data value labels Each bar in the chart now displays the exact mean household income. Notice that the units are in thousands, so you could use the Number Format tab again to change the scaling factor. Instead, we will change the statistic shown in the data values. E With the data value labels selected, click the Data Value Labels tab. The Data Value Labels tab allows you to change various label properties.
140 Chapter 7 E Move Percent from the Not Displayed list to the Displayed list. Figure 7-12 Data Value Labels tab E Click Apply to update the data value labels.
141 Creating and Editing Charts As a result, the bars display the percentage of cases in each Job satisfaction category. Figure 7-13 Bar chart showing percentages Using Templates If you make a number of routine changes to your charts, you can use a chart template to reduce the time needed to create and edit charts. A chart template saves the attributes of a specific chart. You can then apply the template when creating or editing a chart.
142 Chapter 7 If you expand any of the items in the tree view, you can see which specific attributes can be saved with the chart. For example, if you expand the Scale axes portion of the tree, you can see all of the attributes of data value labels that the template will include. You can select any attribute to include it in the template. E Select All settings to include all of the available chart attributes in the template. You can also enter a description of the template.
143 Creating and Editing Charts E When you are finished, click Save. You can apply the template when you create a chart or in the Chart Editor. In the following example, we will apply it while creating a chart. E Close the Chart Editor. The updated bar chart is shown in the Viewer. Figure 7-15 Updated bar chart in Viewer E From the Viewer menus choose: Graphs Bar... E Click Define to open the Define Simple Bar dialog box.
144 Chapter 7 E Remove Job satisfaction from the Category Axis text box and replace it with Level of education. E Select Use chart specifications from. E Click File. E In the Use Template from File dialog box, locate the template file that you previously saved using the Save Chart Template dialog box. E Select that file and click Open. The Define Simple Bar dialog box displays the file path of the template you selected.
145 Creating and Editing Charts (Our example shows the path C:\... \SPSS\Looks\My Template.sgt.) E Click OK to create the chart and apply the template. The formatting in the new chart matches the formatting in the chart that you previously created and edited. Although the variables on the x axis are different, the charts otherwise resemble each other. If you want to apply templates after you’ve created a chart, you can do it in the Chart Editor (choose Apply Chart Template from the File menu).
146 Chapter 7 E From the Data Editor or Viewer menus choose: Edit Options... The Options dialog box contains many settings for configuring SPSS. Click the Charts tab to see the available options. Figure 7-18 Charts tab in Options dialog box The options control how a chart is created.
147 Creating and Editing Charts Style cycles allow you to specify the style of data elements in new charts. In this example, we’ll look at the details for the color style cycle. E Click Colors to open the Data Element Colors dialog box. For a simple chart, the Chart Editor uses one style that you specify. For grouped charts, the Chart Editor uses a set of styles that it cycles through for each group (category) in the chart. E Select Simple Charts.
148 Chapter 7 E From the Data Editor or Viewer menus choose: Graphs Bar... E Click Define to open the Define Simple Bar dialog box. E Deselect Use chart specifications from. If you don’t deselect this option, the chart will use styles from the template rather than from the SPSS options. E Click OK to create the chart. The bars in the new chart are burgundy.
149 Creating and Editing Charts Other Examples Now we will create and edit a pie chart and a grouped scatterplot to explore other editing capabilities, including: Hiding categories Moving text Converting a chart to another chart type Adding a fit line to a scatterplot Identifying points in a scatterplot Pie Chart First, we will create a simple pie chart that shows how many respondents have Internet service at home. This example uses the data file demo.sav.
150 Chapter 7 E Select Internet as the variable that defines slices (Define Slices by). Figure 7-22 Define Pie dialog box When you create charts, they do not show the missing category by default. You want to display this category to make sure that the number of cases with missing values is not excessive. E Click Options.
151 Creating and Editing Charts E Select Display groups defined by missing values, and then click Continue. Figure 7-23 Options dialog box E Click OK in the Define Pie dialog box to create the pie chart.
152 Chapter 7 Figure 7-24 Pie chart in Viewer The pie chart reveals that most respondents do not have Internet service at home. From the chart, it appears that only about a quarter of the respondents have home Internet service. For this pie chart, you will: Add a title. Remove the small category of missing data. Display percentages for the two remaining categories. Move the data labels and connect them to the slices with lines. Convert the pie chart to a bar chart.
153 Creating and Editing Charts E Double-click the pie chart to open it in the Chart Editor. Figure 7-25 Pie chart in the Chart Editor Adding a Chart Title First, you will add a title to the chart. Because the title applies to the entire chart and not to a specific chart element, you don’t need to select anything to add the title.
154 Chapter 7 The Chart Editor adds the word “Title” above the chart and enlarges the chart to accommodate the title. Like footnotes, the title is in a text frame that you can move to another position in the chart. Figure 7-26 Pie chart showing default title E Type Home Internet Service over the highlighted default title.
155 Creating and Editing Charts Note: If you clicked elsewhere in the Chart Editor, the default title may no longer be highlighted. The text frame around the title may be selected instead. If this is the case, click the title again to start edit mode. You can then double-click the title text and begin typing. Figure 7-27 Pie chart showing edited title Modifying Chart Categories Next, you will remove the small category of missing data. E In the Chart Editor, select the pie chart.
156 Chapter 7 Figure 7-28 Categories tab E Click Apply.
157 Creating and Editing Charts The pie chart now displays only the No and Yes categories for home Internet service. Figure 7-29 Pie chart with excluded missing category The pie chart clearly shows that most respondents do not have Internet service at home; it appears that about three-quarters of the respondents are in the No category. However, it might be useful to see the exact percentages. Changing Data Value Label Content and Location E In the Chart Editor, select the pie chart.
158 Chapter 7 The pie chart now displays labels of counts. We need to change the counts to percentages. E Click the Data Value Labels tab in the Properties window. E Move Count from the Displayed list to the Not Displayed list. E Move Percent from the Not Displayed list to the Displayed list. Figure 7-30 Data Value Labels tab E Click Apply.
159 Creating and Editing Charts Percentages are now displayed in the pie slices. Figure 7-31 Pie chart showing percentages The percentages are based on the two categories displayed (73.4 + 26.6 = 100). If you put the category containing missing values back into the pie, the percentages will change. You may not want the data value labels to appear in the pie slices. You can move them and add connecting lines to their respective slices. E Return to the Data Value Labels tab.
160 Chapter 7 Data value labels now appear outside the slices with lines connecting the labels to their respective slices. Figure 7-32 Labels outside of pie chart Converting a Chart Finally, you will convert the pie chart into a bar chart.
161 Creating and Editing Charts The Chart Editor converts the chart and displays the result. The bar chart retains the title and data labels. However, the bars are the same color rather than the two different colors shown in the pie chart. Because the tick labels identify the categories, two colors are unnecessary. Figure 7-33 Bar chart converted from pie chart Grouped Scatterplot In this example, we will create a scatterplot using the data file car_sales.sav.
162 Chapter 7 E Click Simple Scatter. Figure 7-34 Scatter/Dot dialog box E Click Define. For a scatterplot, you need to define a scale variable for each axis. Put the dependent variable on the y axis and the independent variable on the x axis. E Select Fuel efficiency as the y axis variable. E Select Curb weight as the x axis variable. E Select Vehicle type as the grouping variable (Set Markers by).
163 Creating and Editing Charts Figure 7-35 Simple Scatterplot dialog box E Click OK.
164 Chapter 7 E In the Viewer, double-click the resulting scatterplot to display it in the Chart Editor. Figure 7-36 Scatterplot in the Chart Editor Selecting Elements in Grouped Charts Selecting an element in grouped charts (that is, charts with categories) differs from selecting an element in simple charts. You can tell that a chart is grouped because it includes, by default, a legend for the groups. You can hide the legend if necessary (choose Hide Legend from the Options menu).
165 Creating and Editing Charts There are general rules for selecting elements in grouped charts: When no data elements are selected, click any data element to select all data elements. When all data elements are selected, click a data element in a specific group to select all data elements in the group. You can then click a data element in another group to select its associated group. You can also click the legend entry for a group to select only that group.
166 Chapter 7 Adding a Fit Line With scatterplots, you often want to add a fit line to the chart. For elements like fit lines that apply to the whole chart, it doesn’t matter what is selected in the chart. You can make the change or add the item to the whole chart. If you were creating a fit line for only one or more specific groups, you would need to make specific selections.
167 Creating and Editing Charts Displaying Data Value Labels for Specific Points Previously, in the bar and pie chart examples, we displayed the data value labels for all data elements. With scatterplots, displaying data value labels for all markers is not usually useful. There are too many data elements, making viewing of the associated data value labels difficult. E For example, from the menus choose: Elements Show Data Labels The results are cluttered and of limited use.
168 Chapter 7 unreadable. A better option would be to display data value labels for only those points that you want to highlight. First, hide the current data value labels. E From the menus choose: Elements Hide Data Labels Next, turn on data label mode, which allows you to click any data element and automatically display its data value label. E From the menus choose: Elements Data Label Mode The cursor changes to indicate that you are in data label mode.
169 Creating and Editing Charts The Chart Editor now displays the data value label for only that marker. You can repeat this process for each marker that needs a data value label. You can also hide a data value label by clicking it while in data label mode. The data value label contains the case number. You can change the contents of the data value label on the Data Value Labels tab. E When you are finished labeling the data elements, exit data label mode.
Chapter Working with Syntax 8 SPSS syntax provides a method for you to control the product without navigating through dialog boxes, viewers, or data editors. Instead, you control the application through syntax-based commands. Nearly every action you can achieve through the user interface can be achieved through syntax. Using syntax also allows you to save the exact specifications used during a session. Syntax is not available with the Student Version. The examples in this chapter use the data file demo.
172 Chapter 8 This opens the Frequencies dialog box. Figure 8-1 Frequencies dialog box E Select Marital status (marital) in the source list. E Click the arrow button to move the variable to the Variables list. E Click Charts. E In the Charts dialog box, select Bar charts. E In the Chart Values group, select Percentages. E Click Continue. E Click Paste to copy the syntax created as a result of the dialog box selections to the Syntax Editor.
173 Working with Syntax Figure 8-2 Frequencies syntax You can use this syntax alone, add it to a larger syntax file, or refer to it in a Production Facility job.
174 Chapter 8 Editing Syntax In the syntax window, you can edit the syntax. For example, you could change the subcommand /BARCHART to display frequencies instead of percentages. (A subcommand is indicated by a slash.) Figure 8-3 Modified syntax To find out what subcommands and keywords are available for the current command, click the Syntax Help button. The example shows the complete syntax for the FREQUENCIES command.
175 Working with Syntax Figure 8-4 FREQUENCIES syntax help If the cursor is not in a command, clicking the Syntax Help button displays an alphabetical list of commands. You can select the one you want. Typing Syntax You can type syntax into a syntax window that is already open, or you can open a new syntax window by choosing: File New Syntax...
176 Chapter 8 Saving Syntax To save a syntax file, from the menus choose: File Save or File Save As... Either action opens the standard dialog box for saving files. Opening and Running a Syntax File E To open a saved syntax file, from the menus choose: File Open Syntax... E Select a syntax file. If no syntax files are displayed, make sure Syntax (*.sps) is selected in the Files of type drop-down list. E Click Open.
Chapter Modifying Data Values 9 The data you start with may not always be organized in the most useful manner for your analysis or reporting needs. For example, you may want to: Create a categorical variable from a scale variable. Combine several response categories into a single category. Create a new variable that is the computed difference between two existing variables. Calculate the length of time between two dates. This chapter uses the data file demo.sav.
178 Chapter 9 Figure 9-1 Initial Visual Bander dialog box In the initial Visual Bander dialog box, you select the scale and/or ordinal variables for which you want to create new, banded variables. Banding means taking two or more contiguous values and grouping them into the same category. Since the Visual Bander relies on actual values in the data file to help you make good banding choices, it needs to read the data file first.
179 Modifying Data Values Figure 9-2 Main Visual Bander dialog box E In the main Visual Bander dialog box, select Household income in thousands [income] in the Scanned Variable List. A histogram displays the distribution of the selected variable (which in this case is highly skewed). E Enter inccat2 for the new banded variable name and Income category (in thousands) for the variable label. E Click Make Cutpoints.
180 Chapter 9 Figure 9-3 Visual Bander Cutpoints dialog box E Select Equal Width Intervals. E Enter 25 for the first cutpoint location, 3 for the number of cutpoints, and 25 for the width. The number of banded categories is one greater than the number of cutpoints. So, in this example, the new banded variable will have four categories, with the first three categories each containing ranges of 25 (thousand) and the last one containing all values above the highest cutpoint value of 75 (thousand).
181 Modifying Data Values Figure 9-4 Main Visual Bander dialog box with defined cutpoints The values now displayed in the grid represent the defined cutpoints, which are the upper endpoints of each category. Vertical lines in the histogram also indicate the locations of the cutpoints. By default, these cutpoint values are included in the corresponding categories. For example, the first value of 25 would include all values less than or equal to 25.
182 Chapter 9 Figure 9-5 Automatically generated value labels This automatically generates descriptive value labels for each category. Since the actual values assigned to the new banded variable are simply sequential integers starting with 1, the value labels can be very useful. You can also manually enter or change cutpoints and labels in the grid, change cutpoint locations by dragging and dropping the cutpoint lines in the histogram, and delete cutpoints by dragging cutpoint lines off of the histogram.
183 Modifying Data Values The new variable is displayed in the Data Editor. Since the variable is added to the end of the file, it is displayed in the far right column in Data View and in the last row in Variable View. Figure 9-6 New variable displayed in Data Editor Computing New Variables Using a wide variety of mathematical functions, you can compute new variables based on highly complex equations.
184 Chapter 9 E From the menus in the Data Editor window choose: Transform Compute... E For Target Variable, enter jobstart. E Select Age in years (age) in the source variable list and click the arrow button to copy it to the Numeric Expression text box. E Click the minus (–) button on the calculator pad in the dialog box (or press the minus key on the keyboard). E Select Years with current employer (employ) and click the arrow button to copy it to the expression.
185 Modifying Data Values Note: Be careful to select the correct employment variable. There is also a recoded categorical version of the variable, which is not what you want. The numeric expression should be age-employ, not age-empcat. E Click OK to compute the new variable. The new variable is displayed in the Data Editor. Since the variable is added to the end of the file, it is displayed in the far right column in Data View and in the last row in Variable View.
186 Chapter 9 Date and time aggregation and extraction functions Missing-value functions Cross-case functions String functions Figure 9-9 Compute Variable dialog box displaying function grouping Functions are organized into logically distinct groups, such as a group for arithmetic operations and another for computing statistical metrics. For convenience, a number of commonly used system variables, such as $TIME (current date and time), are also included in appropriate function groups.
187 Modifying Data Values Pasting a function into an expression. To paste a function into an expression: E Position the cursor in the expression at the point where you want the function to appear. E Select the appropriate group from the Function group list. The group labeled All provides a listing of all available functions and system variables. E Double-click the function in the Functions and Special Variables list (or select the function and click the arrow adjacent to the Function group list).
188 Chapter 9 Figure 9-10 If Cases dialog box E Select Include if case satisfies condition. E Enter the conditional expression. Most conditional expressions contain at least one relational operator, as in: age>=21 or income*3<100 In the first example, only cases with a value of 21 or greater for Age (age) are selected. In the second example, Household income in thousands (income) multiplied by 3 must be less than 100 for a case to be selected.
189 Modifying Data Values You can also link two or more conditional expressions using logical operators, as in: age>=21 | ed>=4 or income*3<100 & ed=5 In the first example, cases that meet either the Age (age) condition or the Level of education (ed) condition are selected. In the second example, both the Household income in thousands (income) and Level of education (ed) conditions must be met for a case to be selected.
190 Chapter 9 E Type (resale + price) / 2 in the Numeric Expression text box. Figure 9-11 Find the average This equation finds the average of the original price of a car and its four-year resale price. E Click OK.
191 Modifying Data Values A new variable has been added to the end of the data file. This variable, which we named avgsale, is in the far right column in Data View. Any cases that have missing values in either of the two variables that we computed also have a missing value for the avgsale variable. Figure 9-12 Data Editor with avgsale added The second way to compute the mean of a set of variables is to use the mean function. E Recall the Compute Variable dialog box.
192 Chapter 9 E Type mean(resale, price) in the Numeric Expression text box. Figure 9-13 Find the mean Like the previous example, this equation finds the average, or mean, of the car’s original price and its four-year resale price. E Click OK.
193 Modifying Data Values The meansale variable is added to the far right column in Data View. Unlike the previous example, the mean function removes missing values from the equation and computes the mean of the remaining values. You can modify the behavior of this function to limit the number of values that can be removed from the equation.
194 Chapter 9 If you want to find the mean of four variables and only want a result if three or more of the variables have valid values, you could use the following syntax: newvar = mean.3(var1, var2, var3, var4) Figure 9-15 Specifying number of valid values for mean calculation To raise or lower the number of required valid values, simply change the number following mean (which is 3 in this example).
195 Modifying Data Values Construct a date/time variable by merging variables containing different parts of the date or time. Add or subtract values from date/time variables, including adding or subtracting two date/time variables. Extract a part of a date or time variable; for example, the day of month from a date/time variable which has the form mm/dd/yyyy. The examples in this section use the data file upgrade.sav. To use the Date and Time Wizard: E From the menus choose: Transform Date/Time.
196 Chapter 9 The introduction screen of the Date and Time Wizard presents you with a set of general tasks. Tasks that do not apply to the current data are disabled. For instance, the data file upgrade.sav doesn’t contain any string variables so the task to create a date variable from a string is disabled. If you’re new to dates and times in SPSS you can select Learn how dates and times are represented in SPSS and click Next.
197 Modifying Data Values Figure 9-17 Calculating the length of time between two dates Step 1 E Select Calculate the number of time units between two dates and click Next.
198 Chapter 9 Figure 9-18 Calculating the length of time between two dates Step 2 E Select Date of next release for Date1. E Select Date of last upgrade for Date2. E Leave the Unit drop-down list at the default of Years. E Click Next.
199 Modifying Data Values Figure 9-19 Calculating the length of time between two dates Step 3 E Enter YearsLastUp for the name of the result variable. Result variables cannot have the same name as an existing variable. E Enter Years since last upgrade as the label for the result variable. Variable labels for result variables are optional. E Leave the default selection of Create the variable now and click Finish to create the new variable.
200 Chapter 9 Figure 9-20 New variable displayed in Data Editor Adding a Duration to a Date You can add or subtract durations, such as 10 days or 12 months, to a date. Continuing with the example of the software company from the previous section, consider determining the date on which each customer’s initial tech support contract ends. The data file upgrade.sav contains a variable for the number of years of contracted support and a variable for the initial purchase date.
201 Modifying Data Values Figure 9-21 Adding a duration to a date Step 2 E Select Date of initial product license for Date. E Select Years of tech support for the Duration Variable. Since Years of tech support is simply a numeric variable, you need to indicate the units to use when adding this variable as a duration. E Select Years from the Units drop-down list. E Click Next.
202 Chapter 9 Figure 9-22 Adding a duration to a date Step 3 E Enter SupEndDate for the name of the result variable. Result variables cannot have the same name as an existing variable. E Enter End date for support as the label for the result variable. Variable labels for result variables are optional. E Click Finish to create the new variable.
203 Modifying Data Values The new variable is displayed in the Data Editor.
Chapter Sorting and Selecting Data 10 Data files are not always organized in the ideal form for your specific needs. To prepare data for analysis, you can select from a wide range of file transformations, including the ability to: Sort data. You can sort cases based on the value of one or more variables. Select subsets of cases. You can restrict your analysis to a subset of cases or perform simultaneous analyses on different subsets. The examples in this chapter use the data file demo.sav.
206 Chapter 10 This opens the Sort Cases dialog box. Figure 10-1 Sort Cases dialog box E Add the Age in years (age) and Household income in thousands (income) variables to the Sort By list. If you select multiple sort variables, the order in which they appear on the Sort By list determines the order in which cases are sorted. In this example, based on the entries in the Sort By list, cases will be sorted by the value of Household income in thousands (income) within categories of Age in years (age).
207 Sorting and Selecting Data This opens the Split File dialog box. Figure 10-2 Split File dialog box E Select Compare groups or Organize output by groups. The examples following these steps show the differences between these two options. E Select Gender (gender) to split the file into separate groups for these variables. You can use numeric, short string, and long string variables as grouping variables. A separate analysis is performed for each subgroup defined by the grouping variables.
208 Chapter 10 If you select Compare groups and run the Frequencies procedure, a single pivot table is created. Figure 10-3 Split File output with single pivot table If you select Organize output by groups and run the Frequencies procedure, two pivot tables are created: one for females and one for males.
209 Sorting and Selecting Data Figure 10-5 Split File output with pivot table for males Sorting Cases for Split-File Processing The Split File procedure creates a new subgroup each time it encounters a different value for one of the grouping variables. Therefore, it is important to sort cases based on the values of the grouping variables before invoking split-file processing. By default, Split File automatically sorts the data file based on the values of the grouping variables.
210 Chapter 10 Selecting Subsets of Cases You can restrict your analysis to a specific subgroup based on criteria that include variables and complex expressions. You can also select a random sample of cases. The criteria used to define a subgroup can include: Variable values and ranges Date and time ranges Case (row) numbers Arithmetic expressions Logical expressions Functions To select a subset of cases for analysis: E From the menus choose: Data Select Cases...
211 Sorting and Selecting Data Selecting Cases Based on Conditional Expressions To select cases based on a conditional expression: E Select If condition is satisfied and click If in the Select Cases dialog box. This opens the Select Cases If dialog box. Figure 10-7 Select Cases If dialog box The conditional expression can use existing variable names, constants, arithmetic operators, logical operators, relational operators, and functions.
212 Chapter 10 Figure 10-8 Select Cases Random Sample dialog box You can select one of the following alternatives for sample size: Approximately. A user-specified percentage. This option generates a random sample of approximately the specified percentage of cases. Exactly. A user-specified number of cases. You must also specify the number of cases from which to generate the sample. This second number should be less than or equal to the total number of cases in the data file.
213 Sorting and Selecting Data Figure 10-9 Select Cases Range dialog box First Case. Enter the starting date and/or time values for the range. If no date variables are defined, enter the starting observation number (row number in the Data Editor, unless Split File is on). If you do not specify a Last Case value, all cases from the starting date/time to the end of the time series are selected. Last Case. Enter the ending date and/or time values for the range.
214 Chapter 10 To generate date variables for time series data: E From the menus choose: Data Define Dates... Unselected Cases You can choose one of the following alternatives for the treatment of unselected cases: Filtered. Unselected cases are not included in the analysis but remain in the data file. You can use the unselected cases later in the session if you turn filtering off.
215 Sorting and Selecting Data Figure 10-11 Case selection status
Chapter 11 Additional Statistical Procedures This chapter contains brief examples for selected statistical procedures. The procedures are grouped according to the order in which they appear on the Analyze menu. The examples are designed to illustrate sample specifications required to run a statistical procedure. The examples in this chapter use the data file demo.sav, except for the following: The paired-samples t test example uses the data file dietstudy.
218 Chapter 11 Frequencies In the chapter Examining Summary Statistics for Individual Variables, there is an example showing a frequency table and a bar chart. In that example, the Frequencies procedure was used to analyze the variables Owns PDA (ownpda) and Owns TV (owntv), both of which are categorical variables having only two values. If the variable that you want to analyze is a scale (interval, ratio) variable, you can use the Frequencies procedure to generate summary statistics and a histogram.
219 Additional Statistical Procedures E Deselect the Display frequency tables check box. (If you leave this item selected and display a frequency table for current salary, the output shows an entry for every distinct value of salary, making a very long table.) E Click Charts to open the Frequencies Charts dialog box. Figure 11-2 Frequencies Charts dialog box E Select Histograms and With normal curve, and then click Continue. E To select summary statistics, click Statistics in the Frequencies dialog box.
220 Chapter 11 Figure 11-3 Frequencies Statistics dialog box E Select Mean, Std. deviation, and Maximum, and then click Continue. E Click OK in the Frequencies dialog box to run the procedure. The Viewer shows the requested statistics and a histogram in standard graphics format. Each bar in the histogram represents the number of employees within a range of five years, and the year values displayed are the range midpoints. As requested, a normal curve is superimposed on the chart.
221 Additional Statistical Procedures E From the menus choose: Analyze Descriptive Statistics Explore... This opens the Explore dialog box. Figure 11-4 Explore dialog box E Select Years with current employer (employ) and move it to the Dependent List. E Select Income category in thousands (inccat) and move it to the Factor List. E Click OK to run the Explore procedure. In the output, descriptive statistics and a stem-and-leaf plot are displayed for the years with current employer in each income category.
222 Chapter 11 More about Summarizing Data There are many ways to summarize data. For example, to calculate medians or percentiles, use the Frequencies procedure or the Explore procedure. Here are some additional methods: Descriptives. For income, you can calculate standard scores, sometimes called z scores. Use the Descriptives procedure and select Save standardized values as variables. Crosstabs.
223 Additional Statistical Procedures Means In the demo.sav file, several variables are available for dividing people into groups. You can then calculate various statistics in order to compare the groups. For example, you can compute the average (mean) household income for males and females. To calculate the means, use the following steps: E From the menus choose: Analyze Compare Means Means... This opens the Means dialog box.
224 Chapter 11 Figure 11-6 Means dialog box (layer 2) E Select Owns PDA (ownpda) and move it to the Independent List in layer 2. E Click OK to run the procedure. Paired-Samples T Test When the data are structured in such a way that there are two observations on the same individual or observations that are matched by another variable on two individuals (twins, for example), the samples are paired. In the data file dietstudy.
225 Additional Statistical Procedures E From the menus choose: Analyze Compare Means Paired-Samples T Test... This opens the Paired-Samples T Test dialog box. Figure 11-7 Paired-Samples T Test dialog box E Click Weight (wgt0). The variable is displayed in the Current Selections group (below the variable list). E Click Final weight (wgt4). The variable is displayed in the Current Selections group. E Click the arrow button to move the pair to the Paired Variables list. E Click OK to run the procedure.
226 Chapter 11 More about Comparing Means The following examples suggest some ways in which you can use other procedures to compare means. Independent-Samples T Test. When you use a t test to compare means of one variable across independent groups, the samples are independent. Males and females in the demo.sav file can be divided into independent groups by the variable Gender (gender). You can use a t test to determine if the mean household incomes of males and females are the same.
227 Additional Statistical Procedures E From the menus choose: Analyze General Linear Model Univariate... This opens the Univariate dialog box. Figure 11-8 Univariate dialog box E Select Years with current employer (employ) as the dependent variable. E Select Income category in thousands (inccat) and Job satisfaction (jobsat) as fixed factors. E Click OK to run the procedure.
228 Chapter 11 Correlating Variables The Correlate submenu on the Analyze menu provides measures of association for two or more numeric variables. The examples in this topic use the data file Employee data.sav. Bivariate Correlations The Bivariate Correlations procedure computes statistics such as Pearson’s correlation coefficient. Correlations measure how variables or rank orders are related.
229 Additional Statistical Procedures E From the menus choose: Analyze Correlate Partial... This opens the Partial Correlations dialog box. Figure 11-9 Partial Correlations dialog box E Select Current Salary (salary) and Beginning Salary (salbegin) and move them to the Variables list. E Select Months since Hire (jobtime) and Previous Experience (prevexp) and move them to the Controlling For list. E Click OK to run the procedure.
230 Chapter 11 Regression Analysis The Regression submenu on the Analyze menu provides regression techniques. Linear Regression The Linear Regression procedure examines the relationship between a dependent variable and a set of independent variables. You can use it to predict a person’s household income (the dependent variable) from independent variables such as age, number in household, and years with employer. E From the menus choose: Analyze Regression Linear...
231 Additional Statistical Procedures E Select Age in years (age), Number of people in household (reside), and Years with current employer (employ) and move them to the Independent(s) list. E Click OK to run the procedure. The output contains goodness-of-fit statistics and the partial regression coefficients for the variables. Examining fit. To see how well the regression model fits your data, you can examine the residuals and other types of diagnostics that this procedure provides.
232 Chapter 11 E From the menus choose: Transform Automatic Recode... This opens the Automatic Recode dialog box. Figure 11-11 Automatic Recode dialog box E Select the variable Gender (gender) and move it to the Variable -> New Name list. E Type gender2 in the New Name text box, and then click the New Name button. E Click OK to run the procedure. This creates a new numeric variable called gender2, which has a value of 1 for females and a value of 2 for males.
233 Additional Statistical Procedures Figure 11-12 Chi-Square Test dialog box E Select Gender (gender2) as the test variable. E Select All categories equal, since, in the general population of working age, the number of males and females is approximately equal. E Click OK to run the procedure. The output shows a table of the expected and residual values for the categories. The significance of the chi-square test is 0.6.
234 Chapter 11 Exponential Smoothing The Exponential Smoothing procedure performs exponential smoothing of time series data. It creates new series containing predicted values and residuals. This topic uses the Inventor.sav file. For example, you can fit a model for inventory data and use it to predict the next week’s inventory. Suppose that for 70 days you have kept track of the inventory of power supplies and that you want to construct a model and then use it to forecast power supplies for the next week.
235 Additional Statistical Procedures E Click Parameters to specify the procedure. This opens the Exponential Smoothing Parameters dialog box. Figure 11-14 Exponential Smoothing Parameters dialog box E To search for the best general parameter, select Grid Search and then click Continue. E To create a new variable that contains predicted values, click Save. This opens the Exponential Smoothing Save dialog box.
236 Chapter 11 E Select Predict through and type 77 in the Observation text box. This adds 7 days to the original 70. E Click Continue and then in the main dialog box, click OK. This runs the procedure and adds the new variables fit_1 and err_1. The variable fit_1 contains the fitted values and the seven new predicted values. The variable err_1 contains residual values for the original 70 cases; you can use the residuals for further analysis.
237 Additional Statistical Procedures The resulting chart shows both the actual number of power supplies and the fit line plotted on the same axes. The predicted values are plotted for the next week on the right side of the chart.
Index Access (Microsoft), 34 bar charts, 81 cases selecting, 210 sorting, 205, 209 case studies, 27 categorical data, 79 summary measures, 80 charts bar, 81, 127 chart options, 145 creating charts, 127 editing charts, 130 histograms, 86 pie, 149 scatterplot, 161 templates, 141 computing new variables, 183 conditional expressions, 187 continuous data, 79 copy value attributes, 66 counts tables of counts, 80 create variable labels, 56 database files reading, 34 Database Wizard, 34 Data Editor entering non
240 Index Help case studies, 27 Statistics Coach, 19 Help windows contents, 16 index, 17 hiding rows and columns in pivot tables, 100 histograms, 86 HTML exporting results to, 123 index Help topic search, 17 interval data, 79 keyword Help topic search, 17 layers creating in pivot tables, 96 level of measurement, 79 measurement level, 79 missing values for non-numeric variables, 65 for numeric variables, 63 system missing, 62 moving elements in pivot tables, 93 items in the Viewer, 89 nominal data, 79
241 Index string data entering data, 54 subsets of cases based on dates and times, 212 conditional expressions, 211 deleting unselected cases, 214 filtering unselected cases, 214 if condition is satisfied, 211 random sample, 211 selecting, 210 summary measures categorical data, 80 scale variables, 83 syntax, 171 syntax files opening, 176 saving, 176 Syntax Help tool, 174 syntax windows editing commands, 174 opening a new window, 175 pasting commands, 171 running commands, 173, 176 system-missing values , 6
