How to enter data in SPSS 1 1

How to enter data in SPSS 1. 1 Introduction of SPSS 1. 2 Data Entry 1. 3 Data Cleaning using SPSS 1

Statistical Software Packages Most Commonly Cited in the NEJM and JAMA between 1998 and 2002 302 SAS 87 SPSS STATA 80 49 Epi Info 43 SUDAAN S-PLUS 33 Stat. Xact 18 BMDP 9 Stat. View 9 Statistica 8 0 100 200 300 400 Number of articles software was sited 2

Before you perform analysis in SPSS, let’s set up the following option. Go to Edit, Options, . . 3

SPSS Windows has 3 windows: Data Editor Viewer or Draft Viewer which displays the output files Syntax Editor, which displays syntax files The Data Editor has two parts: Data View window, which displays data from the active file in spreadsheet format Variable View window, which displays metadata or information about the data in the active file, such as variable names and labels, value labels, formats, and missing value indicators. 4

SPSS Data View 5

SPSS Variable View 6

1. 2 Data Entry into SPSS There are 2 ways to enter data into SPSS: 1. Directly enter in to SPSS by typing in Data View 2. Enter into other database software such as Excel then import into SPSS Let’s start with the second option, using data in Excel. 7

Figure 1. Data from Hell 8

Data from Heaven 9

How to move from Hell to Heaven (1): 1. Add a patient’ ID number 2. Delete the first row with the title of the project 3. Delete the 2 rows under the variable name. 4. Delete the 2 row between the groups. 5. Delete the row of average at the bottom. 6. Add a variable called group and code the first 10 with Drug A as 1 and the next 10 as 2. 7. Change the variable names to less than 8 or 8 characters with no spaces, (you can use numeric, but not starting with numeric, avoid symbols). 8. Insert 2 columns before BP as SYSBP and DIASBP. Delete the BP text column. 9. Change missing values, NA, unknown, ? , to blanks. 10. Change age of 6 months to 0. 5 (years). Fix errors. 11. Code males=1 and females=2. 12. Code complications as 0 for no and 1 for yes 13. Go back to the source and complete the missing information 14. If a column was entered as a string (words), you may have to select the column and format the cells for change it to numeric. 10

General guidelines for data entry 1. Give each variable a valid name (8 characters or less with no spaces or punctuation, beginning with a letter not a numeric number). Short, easy to remember word names. Avoid the following variable names: TEST, ALL, BY, EQ, GE, GT, LE, LT, NE, NOT, OR, TO, WITH. These are used in the SPSS syntax and if they were permitted, the software would not be able to distinguish between a command a variable. Each variable name must be unique; duplication is not allowed. Variable names are not case sensitive. The names NEWVAR, New. Var, and newvar are all considered identical. 2. Encode categorical variables. Convert letters and words to numbers. 3. Avoid mixing symbols with data. Convert them to numbers. 4. Give each patient a unique, sequential case number (ID). Place this ID number in the first column on the left 11

5. Each variable should be in its own column. Avoid this: Change to: Animal Control 1 Control 2 Experiment 1 Experiment 2 Animal 1 2 3 4 Group 0 0 1 1 * Do not combine variables in one column * It is recommended to use 0/1 for 2 groups with 0 as a reference group. 6. All data for a project should be in one spreadsheet. Do not include graphs or summary statistics in the spreadsheet. 12

7. Each patient should be entered on a single line or row. Do not copy a patient’s information to another row to perform subgroup analysis. 8. However when data are repeatedly collected over a patient, it’s recommended to have patient-day observation on a simple line to ease data management. SPSS has a nice feature to convert from the longitudinal format to horizontal format. When the number of repeats are few 2 or 3, horizontal format may be preferred for simplicity. Longitudinal data entry Horizontal data entry Date 1/2/2005 1/3/2005 1/4/2005 3/1/2005 3/2/2005 ID SYSBP 1 SYSBP 2 SYSBP 3 1 130 120 2 110 140 ID 1 1 1 2 2 SYSBP 130 120 110 140 13

9. For yes/no questions, enter “ 0” for no and “ 1” for yes. Do not leave blanks for no. Do not enter “? ”, “*”, or “NA” for missing data because this indicates to the statistical program than the variable is a string variable. String variables cannot be used for any arithmetic computation. 10. Put ordinal variables into one column if they are mutually exclusive. Avoid: Preferred: Pain Mild 1 0 0 Pain Moderate 0 1 0 Severe 0 0 1 1 2 3 11. Do not make columns wider then 8 characters, unless absolutely essential. 14

Entering Date in Excel. In Excel, go to: Format, Cells, select Date under Category, Choose Type for a format you like 15

Entering Time in Excel. In Excel, go to: Format, Cells, select Time under Category, Choose Type for a format you like 16

Entering Date / Time in Excel. In Excel, go to: Format, Cells, select Time under Category, Choose Data/Time format 17

Entering Date, Time in SPSS In SPSS, open Variable View, Click Type for the variable you want to Assign date format, click on Date, and select a format of your choice. 18

Importing data from Excel spreadsheet into SPSS. In SPSS, go to: File, Open, Data Select Type of file (for example, Excel) you want to open Select File name you want to open 19

Importing data from SPSS to Excel. In SPSS, go to: Data, Save as, Select Type of file (for example, Excel) you want to save into Give File name you want to save into 20

Data merging in SPSS (1) 1. Make sure that both files are sorted by Key variable in ascending order 2. In SPSS, open Data from Hell to Heaven. sav 3. Select Add Variables under Data, Merge Files 21

Data merging in SPSS (2) 4. Select the dataset you want to merge into the working file. 22

Data merging in SPSS (3) 5. Click on Match cases on key variables in sorted files, 6. Click on Both files provide cases 7. Highlight ID in the excluded variables box, then click ► near key Variables 23

Note in Data merging in SPSS (3) Cases must be sorted in the same order in both data files. If one or more key variables are used to match cases, the two data files must be sorted by ascending order of the key variable. Variable names in the second data file that duplicate variable names in the working data file are excluded by default because Add Variables assumes that these variables contain duplicate information. Thus before you merge data files, you need carefully to check two variables with the same name. If two variables contain different information, SPSS automatically delete variable from the file, which is being merged into (Birthday. sav). 24

1. 3 Data Cleaning in SPSS 1. Re-coding existing variables 2. Creating new variables 3. Creating new variable from existing variables 4. Data labeling and formatting 25

Data cleaning in SPSS (1): Recoding existing variables (1) We want to use numeric coding for group instead of A and B. Old ID 1 2 3 4 New Group A A B B Group 0 0 1 1 26

Data cleaning in SPSS (2): Recoding existing variables (2) From SPSS dialog box, go to: Transform Recode Into Same variables 27

Data cleaning in SPSS (1): Recoding existing variables (3) 1. Select Group from the variable box into String Variables box 2. Click on Old and new Values to proceed 28

Data cleaning in SPSS (1): Recoding existing variables (4) 1. Type the old value and the new value you want to convert into 2. Click on Add (To remove, or change, click on Change or Remove) 3. Type all values in the Old New box, then click Continue 4. Click OK to execute the commands. 29

Data Cleaning in SPSS (2) Creating a new variable for Diastolic blood pressure (Dias. BP): In SPSS, go to Variable View, Then type Dias. BP at the last row under Name Go back to Data View and directly type diastolic blood pressure to separate from Sys. BP. For ease of data entry, you can move Dias. BP right after Sys. BP. Now also edit sys. BP. 30

Data Cleaning in SPSS (3) Computing patient’s age from birthday and date enrolled into the study. 31

Data Cleaning in SPSS (4): Data labeling and formatting (1) Specifying Type of Variable HT 61. 00 68. 00 47. 00 66. 00 72. 00 67. 00 72. 00 66. 00 60. 00 61. 00 59. 00 73. 00 65. 00 71. 00 68. 00 69. 00 66. 00 68. 00 32

Data Cleaning in SPSS (4): Data labeling and formatting (2) Data Labeling 33

Data Cleaning in SPSS (4): Data labeling and formatting (3) Variable Formatting 34

Data Cleaning in SPSS (4): Data labeling and formatting (4) Specifying missing values 35

Data Cleaning in SPSS (4): Data labeling and formatting (5) Measurement category 36

Retrieve data property from existing files in SPSS (1) This property is extremely handy when you need to construct a similar database for expanded, or new group of patients. You can save time on creating variable label, format, etc, rather you can retrieve these information from existing files. Now let’s create a copy from “Data from heaven. sav” after you delete formats and labels you just created. Save it as “Data from hell to heaven without format. sav”. Modified Note: Before you perform this commands, make sure that Type of variables matched between the two datasets. 37

Retrieve data property from existing files in SPSS (2) 38

Retrieve data property from existing files in SPSS (3) 39

Using syntax in SPSS: SPSS has its great advantage in producing high level graphs and statistical analysis by easy point-and-click operations. However, some people may criticize SPSS for irreproducibility of analysis which were conducted before. In fact, SPSS has a high level capacity of programming syntax which can be saved and repeatedly operated. Throughout the course, I will provide “how to” box to conduct all analysis used in the class, here I will show to save your commands in syntax. I highly recommend the use of syntax for better organization on haw has been done. 40

Using syntax in SPSS (1): Creating a new syntax file 41

Using syntax in SPSS (2): Editing a syntax file 42

Using syntax in SPSS (3): Saving a syntax file 43

Using syntax in SPSS (4): Opening an existing syntax 44

Using a syntax in SPSS (5): Example Syntax I find syntax very handy especially when you get tired of clicking so many times! 45

Using syntax in SPSS (6): Recoding syntax from command dialog box You can in fact use command dialog box (point and click method) as your main tool and still save what you did with point and click into syntax. Then later you can simply execute the syntax to repeat the analysis. Step 1 46

Step 2: Saved syntax from the previous PASTE command 47

Using syntax in SPSS (7): Executing the syntax 48

Data confidentiality Data need to be stored in a secure locked place, need to be back-up daily or once a week. When you send your data to a biostatistician for further statistical analysis, delete patient name, social security numbers, medical record numbers, actual dates (birth day, admission date, etc) 49

Communication with a biostatistician: Most statisticians prefer to have data submitted as SPSS format or in the statistical software they use. An advantage of entering data directly into a statistical package, such as SPSS is that one can enter variable label and value labels in the file. When communicating with a biostatistician, also describe the research problem, study hypothesis, and the primary comparison that you are interested in. Explain any variables that need to be controlled for. Explain the code used for missing values. Also answer the following questions: What is the name of your study? What is the purpose of your study? What is the type of your study? Will all subjects be included in the analysis? Was there any matched (repeated) measures? How will outliers be defined and handled? Has the data been cleaned? What is our goal and deadline for this goal? 50