The art of fetching useful insights from the data is now just not limited to the professional’s hand but modern tools like IBM SPSS are also capable of serving the best from a massive data set. Big organizations are using such tools to get valuable information from the data generated from multiple sources. Let’s see what is SPSS and how it is getting used in today’s work culture.
In this SPSS Tutorial, we will go through below topics |
SPSS is a versatile program that was developed for managing statistical procedures. SPSS is the Statistical Package for Social Science and is used for the analysis of complex statistical data by researchers and organizations. This software package was launched in 1968 by SPSS Inc. But later in 2009, it was acquired by IBM. It was designed especially for the management and analysis of social science data.
Officially SPSS was given the name - IBM SPSS Statistics. But the people mostly call it SPSS. Whenever there is a task of social data analysis then first preference is always SPSS because of its user-friendly English-like command language. Its user manual is very simple to understand as it is straightforward and easy to get.
The application areas of SPSS are health research centers, marketing companies, government projects, education, data mining, etc. It is also used in survey organizations, data mining, and many other statistical analysis operations.
Most of the research companies use SPSS for the analysis of survey data so that they can get their desired output and can fetch useful insights from the data.
SPSS Statistics data is stored in the data files which are specifically designed for being used by the IBM SPSS Statistics, In this section, we will discuss the different forms of data files and how to use these files.
You can simply import the data from various sources and also can enter it manually. Here, we will discuss how to read stored data from:
Data of IBM and SPSS statistical files are saved with the .sav extension. You can open the saved data by the following steps:
Want to enhance your skills in dealing with the worlds best IBM, enroll in our SPSS Training |
Consider the following steps to import Excel Database in SPSS:
From the database, the data can be imported using the Database Wizard which is not that complex. You simply need to install drivers for reading data directly from the database. The Open Database Connectivity (ODBC) driver is mostly used for this purpose and for different database formats. Any database with this driver can easily read the data. The installation CD is available for it. Other additional drivers are also available in the market by third-party vendors.
Example: Let us consider an example. It is based on Microsoft Windows and the driver in this is ODBC. This driver is only compatible with the 32-bit version of SPSS and IBM statistics. You can follow the same steps if you are using some other platform but you need a third-party ODBC driver to access it.
Define the Variables step
In this section, we will discuss the steps to read a text file in SPSS. The text file is stored with the extension .txt. The file which we will consider in the example is named as awards.txt. It includes the following steps:
Select File Menu => Read Text Data
Browse the file that you want to read. The dialog box of the Open Data window appears.
Click the required file awards.txt and press the open button.
The Text Import Wizard will appear on the screen to load the data. Here, you can format the data to make it as per the requirement.
Check the input data and then press the Next button.
Verify the selected file to analyze whether the file is correct or not by checking the contents of the input file. If your file is already in a defined format then you can skip some of the steps and can select your file here. And, if it is not in a predefined manner then you can follow the further steps.
Mention that the data to read is delimited.
You can specify that the data in your file is organized and can be separated using tabs, space, and commas, etc. Maybe your data is already separated by having a fixed width and cannot be divided as data items are fixed together.
Click on the Next button.
Mention and check where the data are located in the file.
SPSS interpreting the text data.
You can specify to the SPSS the file which you want to read. And also, you will have one line of text representing the one row of data in the SPSS. Or you will have the SPSS count the variables to check from where the row is starting.
It is not required to read the whole file. You can randomly select the lines from a file or can select a percentage of the total lines in the file. This selection of lines will work as a test section and it is a very useful process if you are dealing with a large data file.
Click Next => define space as delimiters => double quotes as text qualifiers.
The SPSS knows very well that how to use the spaces, commas, tabs, etc., for the separation in the file. Some other characters can also be used as a delimiter by selecting the ‘Other’ option. Then you are able to write the character into the blank. You are enabled to define that your file text is formatted within the quotes. You can format your data by using both single and double-quotes. Strings should be written within quotes if they are containing any of the characters as delimiters.
The missing data item in the text file can also be specified in the file by using two delimiters.
Press the Next button. Modify the variable name and if required change the format.
By default, SPSS will assign the names to variables like V1, V2, V3. You can change the variable name by updating the new name in the variable name field. From the data format drop-down list, you can also modify the data format.
Click the Next button. Save format and also analyze the syntax
If you want to save the file for future purpose then click the no button to get the file save.
In SPSS this file will be used if you are going to upload a similar life with the same formatting. Next time when you load the file, it will save your time of formatting and reduce answering the questions again.
Press the Finish button.
The SPSS may take some time to finish the processing and this time depends on the formatting you have done on the data. And, your data will be displayed on the Data Editor Window.
Data View.
The file name will be asked and you need to enter the file name. You can call the name Awards. As the new file will be saved with extension .sav and it represents the SPSS file.
Reading the SPSS file is quite flexible as compared to the example shown. And, the reason will be explained with the second example. The file AwardHeader.txt contains the same data as the previous file but has different formatting.
In this file, the variable names will be mentioned in the first line. The whole data is placed in one line and is separated by commas. The data reading process is the same as we have done before. This time it will not be done in a single step as SPSS cannot predict the header and the data.
Modify data types and formats. Then Save File.
You are supposed to explain the data parts otherwise the data will remain like a block of text.
In the previous steps, it was explained that the SPSS is notified to show the variable name in the first line of the text file.
Step 3 of 6, the data starts from line 2 of the file. Every case is having five data items.
In the input text file, the data can be entered in the lower lines. But the variable names are entered in the first row of the file. SPSS may not read the starting and ending lines of the file but when it has a complete case it counts the data values.
Specifying the delimiters. Put the characters in the quote.
In the previous steps, 4 of 6, commas and spaces were used as the delimiters in the file. As in this example, space was not used as a delimiter but there no problem if you want to use it. Here SPSS analyzes by itself that where space is required and applies the setting by default. And after that, it organizes the data as per the requirements defined by you. SPSS will read the variable names. Also, mention these names as the column headers.
Modify variable names and mention their types.
SPSS will automatically detect the type of variable name by itself.
After completion of Step 6 of 6, press the Finish and wait to get the data load.
The data will appear in SPSS as you have formatted it.
Check Out: SPSS Tutorials |
Data Editor is used to displaying the data in the file. In the Data editor, the information is displayed in the form of the variables and the cases.
Here the variables are used to show the different types of data which you have compiled. If we consider a survey then each question is equal to the variable. A variable can be a string, integer, date, currency, etc. It can be of any type.
The design of the Data Editor is similar to a spreadsheet. It has some differences which are mentioned below.
In the Data Editor, you are able to enter the data. This can be used for both small as well as the large file. For large files, it can be used to make small edits.
The Variable names in the Variable View
Values entered in the Data View
In addition, to define the data types, you are free to specify the descriptive variable labels and also the value labels for variable names and data values. These defined labels can be used in statistical reports and charts.
Four programs are offered by the SPSS that guides the researchers to deal with their complex data analysis requirements.
The SPSS Statistics program facilitates a plethora of basic statistical functions. These functions consist of frequencies, bivariate statistics, and cross-tabulation.
Modeler program for SPSS guides the researchers in performing their research operations like designing and implementing the predictive models with the support of advanced mathematical statistical measures.
For Survey program, SPSS Text Analytics helps the survey administrators. They represent the suggestions on the basis of the responses to the survey questions.
Visualization Designer program of SPSS enables the researchers to create a large variety of visuals easily such as radial boxplots by using their data.
The SPSS also includes solutions for data management processes which makes the complex programs simple. It makes the researchers able to perform the case selection, file reshaping and also to create derived data.
SPSS is the best solution for multiple tasks including data documentation. The researchers can easily save their confidential data as the metadata dictionary. And, this dictionary act as a centralized repository of information which is used at the runtime. It is pertaining to data like meaning, origin, format, and relationships with the other data.
For manipulation of the data, the usage of SPSS tool is one of the benefits that the researcher can take. It is best for the statistical analysis of survey data.
You can export any data in SPSS for the analysis purpose. The survey data gathered by SurveyGizmo can also be used.
This saved data file has .SAV extension and with this, you can apply the process of manipulation, data analysis, etc.
With this, the SPSS will automatically define the default variable names, types and the value labels. And also, it reduces the time and workload of the researchers. There is a number of features and opportunities for the users in SPSS.
SPSS is customizable and flexible so that you can use it as per your requirement. You can find the solutions for the complex datasets and can easily develop the predictive model.
In this section, the simple summary measures are discussed. And also, it includes how the level of measurement of a variable is influencing the types of statistics. Here, the data file names as demo.sav will be used.
There are different summary measures for different types of data and it depends on the level of measurement:
The most complex and difficult summary measure for categorical data is the number of cases for each category. The category mode is the one with the greatest number of cases. And, if the ordinal data has more categories then the median can be the beneficial summary measure for it. The median is the value where half of the cases fall above and below.
The frequency tables represent both the number and the percentage of cases for every single observed value of a variable. Frequencies procedure generates this table.
Frequency tables
The frequency tables can be viewed in the Viewer window. The frequency tables represent that only 20.4% of the person own PDAs. And, almost all the people (99%) owns a TV. This study may not be interesting but searching for those small groups of people will be interesting who do not use the TV.
In a frequency table, the information can be displayed graphically with a bar chart or with a pie chart.
The frequency tables can also display the same information in the form of bar charts. It makes it easy for people to see because most of the people have TV rather than PDAs.
Several summary measures are available for scale variables. It includes:
The table of Frequencies Statistics will be shown in the Viewer window.
Table of Frequencies Statistics
Mean and Median both are the different measures. For example, the value of mean is about 25,000 times greater than the median value. It indicates that these values are not normally distributed. The distribution can also be check by the histogram.
At the lower end of the scale, with the fall below 100,000, the majority of the cases are clustered. There are only a few cases in the range of 500,000 and beyond. These high values have more effect on the mean than the median. And it makes the median better statistic measure.
This product comes loaded with a lot many features which can be really helpful when you are working on an analysis of data sets particularly when the analysis is being performed on a similar type of data sets.
The data file in IBM SPSS contains the Metadata attributes which are linked with the data file that can be anything like variables, value labels or the missing value definitions, etc. So this avoids recreating the information every time you start working on a new data file. This can save a lot of time and you just need to copy the variable settings from a saved IBM SPSS Data file. The file properties from one file can be applied to the currently active dataset.
This method is helpful in:
This contains the dictionary of information for the different variables for the active data file.
Information could be like:
These are the steps you can refer to go to display Data Dictionary.
File > Display Data File Information > Working File...
This will display each variable with its specific properties. You can easily save this information in some output file or you can have that Information printed as a hard copy for your reference.
Dialog Recall
This button will show you the list of recently accessed dialog boxes. You can click this button to return to any listed dialog box.
This button looks like this:
This is helpful if you need to rerun some analysis or make some modifications to the existing analysis in the session and run.
This feature helps you to customize the user interface and this can really save a lot of time because it provides you easy access to the most frequently used features.
These Menu items can be easily created to run command syntax files, scripts. You can create your own menu with your choice of items.
Steps to go to Menu Editor: View > Menu Editor…
Here you can create your entire menu as per your choice or you can easily edit the existing one too.
Custom toolbars provide you a quick way to perform specific tasks. Let’s say that you want to create a graph or a chart all this can be achieved with just a simple click of a toolbar button.
Steps for Creating Custom toolbar: View > toolbars > Customize…
This will show you all the toolbars which are available and can be used in the currently active window.
This can be used for generating command syntax. You can create your own version of a dialog for built-in IBM SPSS Statistics procedure. You can also create a user interface for any set of command syntax and that can be parameterized based upon user input.
Below are the steps for building your custom dialog:
Once you are done with the above steps you can then begin using it. This can also be stored at some external source location and can be shared later with other users also.
Using Production Jobs IBM SPSS Statistics can be made to run in an automated fashion. This program runs unattended and will terminate after it executes the last command, so this helps you to perform other tasks while it is running. Production jobs can be scheduled to run automatically at scheduled times.
Let’s talk about analyzing weekly reports which we all know is a time-consuming process, So Production Jobs helps us out here.
Below are the ways in which we can make Production Jobs to run:
In this, the program runs unattended in a separate session which could be on your local computer or on a remote server. If the program is running on your local computer then that must remain ON and connected to the server until the job is complete.
In this case, the program is running on a separate session on a remote server. So, your local computer doesn’t need to be kept ON or connected to the remote server. You can also fetch the results after some time.
Do keep in Mind that running a production job on a remote server requires access to the server that is running IBM SPSS Statistics Server.
Scoring is the term used here and which is referred to as applying a predictive model to a set of data. IBM SPSS Statistics can be used to build procedures such as regression, neural network, tree, clustering. The specifications for a model can be saved in a file and that can be used to reconstruct the model and can be used to generate predictive scores in other datasets. The file can be XML File or a compressed file (.zip file).
Scoring Process consists of the below steps:
IBM SPSS provides you numerous procedures for building predictive models. The Predictive models pick up the useful data and these models are trained to work on that useful data. If that model gives you accurate predictions then that can be used to score real data.
Historical data which is pre-aggregated is made to run through predictive models in IBM SPSS Modeler. If the predictions are accurate than can be used to create a runtime predictive model. Retraining of models is done by changing the behavior patterns.
IBM SPSS Modeler 15 User’s Guide lists down the information about the IBM SPSS Modeler
Select the Scoring Wizard in the Utilities and choose the data model file which could be an XML File or a ZIP File. Select the Case from the Data Section which was created any by this the Model can be applied for the prediction work.
SPSS Syntax is the language which contains the instructions for the data. These instructions are used to predict, analyze and modify the data.
Click on Analyze > Descriptive Statistics > Frequencies > Paste.
This will open up the Syntax Editor which holds the Instructions which were given in the Frequencies dialog. Then by running the commands, you have created you can get the data in the form of charts.
You can select the commands and click the “Run” Button.
A little practice in writing SPSS syntax will help to get good and really fast results.
The following ways can be used:
Here in this section of the post, we will discuss some of the mathematical statistics technique:
Frequency Distribution: This lets us know the number of distinct values in a variable and the number of times they occurred.
Z-Scores What and Why? : These are linearly transformed data values which have a standard deviation of 1 and mean of 0. These are also referred to as Standardized scores or data values which are given a common standard.
Standard Deviation: Standard deviation is a main statistical measure which gives value as output. This value tells us to what extent the number lie apart.
Crosstabulation is the method to measure the relationship in the two categorical variables. For instance, Gender is considered as a column variable and the Age is taken as a row variable. In the age category, you can display the number of males and females by developing the two-dimensional crosstabulation.
A simple crosstabulation: The main factor is the money that people are going to spend on the product. Here, we will consider an example to show the relation between the Personal Digital Assistant (PDA) and the income level.
Note: the Statistics Base option is needed.
Crosstabs dialog box
Crosstabulation table
In this table, the cells are representing the number of the cases for value joint combination.
Here the cells of the table are representing the number of cases for each joint combination of values. For example, 455 people fall under the category of the income in thousands i.e. $25,000–$49,000 own PDAs. And, the relationship between the variables is not indicated by the numbers in the table.
Counts vs percentage in crosstabulation: In each cell, it is quite difficult to estimate the crosstabulation. There are more than two crosstabulations as the several PDA owners come under the $25,000–$49,000 income category. Also, it is difficult for the ones who are under $25,000 income category as many people fall in this range.
Crosstabulation with row percentages
In the above figure, it is clear that that the own PDA is increasing as the income increases.
Significance Testing for the Crosstabulation:
The key reason for evaluating the crosstabulation is to find the relation between the two variables. Many tests are there to find whether the relation between the cross-tabulated variables is significant or not. Chi-square is one of them and the benefit of using it is, it can be applied to various types of data.
Pearson chi-square is used for hypothesis testing. It is done to estimate the relationship between the row and column variables and analyze that they are independent or not. We need the significance value as the actual value does not contain the information which we need. More the significant value is, more the variables are independent. Suppose if the significance value is coming as .000 then it means that the two variables are, indeed, related.
Add the Layer variable:
To develop a three-way table, you are allowed to insert a layer variable. In this, the row and column variables are further divided by the layer variable categories. It can also be called are the control variable because after adding a layer variable you can analyze the change in the relation of column and row variables.
If you have a look at the crosstabulation table then you will notice that the table is large in size and difficult to analyze. And if you check the chi-square statistics table, it is easy to understand and interpret. In the education category, the apparent relationship between the two variables of PDA and income becomes negligible or disappears. This table mentions the relationship between the income column and the PDA owner.
Related Article: SPSS Interview Questions |
Multiple types of charts are available to create and use. You can use these charts to display your result of tasks. There are three most used charts which are mentioned below:
This section shows the basics of chart creation. Here, a bar chart is created which represents the mean income for different levels. In this example, a demo1.sav file is used.
Select the “Menu” option and choose: Graphs => Chart Builder
You can preview the chart while creating it and can modify it at the same time. You can view it in the Chart Builder dialog box which is an interactive window.
The result of your statistical procedure can be the tables, graphs, and charts, etc. It can be shown in the Viewer. It depends on you and your need that how you want to display the output means in the form of chart, text or table.
Using the Viewer
Viewer
The Viewer window is divided into two parts, one is the outline pane and the other is contents pane. Here the outline pane is used for an outline of the saved information and the contents pane includes the tables, statistical charts, and the text as a result.
You can explore the window’s content vertically or horizontally by using the scroll bar. Also, click on the item simply to display it in the contents pane.
You can drag and drop the result items into the contents pane as per your requirement.
Using the Pivot Table Editor
Viewer
Creating a pivot table using pivot table editor is a simple task. With pivot tables, you can represent the results of your statistical procedures as shown in above figure.
TableLooks
The table is a way to represent your results in a proper format to make the result more clear and concise. Create the table in such a way that it will be easy to read and understand as it is a critical part of the process.
Using Results in Other Applications
You can use the results of this process in various applications like using table or chart in your report, documentation or presentations, etc.
You can run several tasks by simply using the command language. It helps you to store and automate the files. It offers a number of functions which are not available in the dialog box or the menu. These commands allow you to store your jobs for further use in the syntax file. You can use these syntax files to save your time when you will work on the same analysis of data.
The command language syntax file includes the IBM and SPSS statistics syntax commands and is a text file. You can directly type a command in a syntax window. You simply need to run the command and no need to perform many actions from dialog box or menu.
Use the paste button to write the syntax. This button is available on the dialog box.
Frequencies syntax
You can edit the syntax in the syntax window. You can modify the commands to represent frequency instead of the percentage. And also, you can directly enter the keyword for the frequency. Get the keyword by placing the cursor anywhere on the subcommand name. Then press the Ctrl+Spacebar and you will get the auto-completion control for the command. From the BARCHART subcommand, delete the PERCENT keyword. Then Press Ctrl+Spacebar together.
The auto-completion control will provide you with the list of all available terms when you type.
The F1 key is used to get the complete information about the current command. It will redirect you to the command syntax reference information. The syntax window shows the colored text which enables you to distinguish between the code easily. In this only the recognized coding terms are colored. If you make some spelling mistake while coding then you can easily identify it. Suppose if you have typed PRNT instead of PRINT command. PRINT command will be shown as colored by default whereas PRNT will be uncolored.
A standard dialog box for opening files is displayed.
If the command you are going to execute is applied to a specific data file. Then before you run the command, make sure that the data file is open. Or else you can open that file using the command.
Breakpoints are used to stop the working of the code at the particular points and then you can run the code again when it will be ready. It makes you able to view the result at some point in the syntax and you can also get the current information by running a syntax command like FREQUENCIES. And, the breakpoints will be applied only at the particular level of commands. You cannot set it on any of the lines within the command.
How to Insert a Breakpoint in the Command:
By the time you execute a syntax command which has breakpoints, the running process halts prior to every single command holding the breakpoint.
The progress of the executing command can be seen through the downward arrow pointing towards the left of the command text. It spans the first to last command of execution. It is very useful while executing command syntax having breakpoints.
To resume execution following a breakpoint:
In this section, we will understand why we need to modify the data. As we know, the information or data that we have to optimize may not be as per the requirement. So first we need to arrange it in a proper manner so that we can apply the procedure and model over it. Let us consider some examples:
In the data file, most of the categorical variables are taken from the scale variables.
Several categorical variables in the data file demo1.sav is, in fact, derived from scale variables in that data file. These categorical variables include the integer values i.e. 1 to 4 which represents the income categories and this income is in thousands.
Steps to create the categorical variable - inccat:
In the Visual Binning dialog box, you can choose the ordinal variables and can create a new binned variable. Here, binning means combining two or more variables to make a group of the same category.
The visual binning depends on the value present in the data file and this will make you able to make the correct decision. For this purpose, you need to read out the complete file. It may be a time-consuming process as your file may have large cases. You can also scan the number of cases which are required to read. But it is not necessary in every case.
Main Visual Binning dialog box
The number of cutpoints is one less than the number of binned categories. In the example, the binned variable has four categories and out of the first three include the range of 25 thousand. Where the last one will include all the values having the highest cutpoint value of around 75 thousand.
Now you will get values displayed in the grid and these values indicate the defined cutpoints. These cutpoints are the upper endpoints of categories. The locations of the cutpoints in the histogram indicated by the vertical lines.
Automatically generated value labels
It will automatically create descriptive value labels for each of the categories. The cutpoints and labels can also be added manually in the grid. You can also change its location by drag and drop line in the histogram.
With this, a new variable will be created and it will be viewed in the Data Editor.
It is easy to solve and run complex equations by applying different statistical functions on the variables. Here in the example below, the difference between the two existing variables will be estimated.
In the file, there is a variable containing the respondent's current age and also the variable for years spent at current work. There is no variable consisting of time and age when the person starts his/her job. A new variable will be calculated which shows the difference between these two variables.
Compute Variable dialog box
Date and time wizard is used to compute the tasks related to date and time data. You can easily perform these jobs using the wizard. With this wizard, you are able to execute the following tasks:
Use the Date or Time Wizard:
The screen shown in the above picture is showing general tasks set. The tasks which are not in use for the current data will be disabled. If you are working on it for the first time, then you get a brief overview by clicking the “Next” option. And also, from the help button, you will get complete information about the variables.
Conclusion
This digital era is revolving across the millions of data sources throwing information at an unstoppable pace. The requirement for tools like IBM SPSS is obvious for dealing with this information. Thus, you can get useful insights from your data and solve complex business problems by making quick decisions. This tools will help you to perform predictive analysis giving out effective and efficient results.
Explore SPSS Sample Resumes Download & Edit, Get Noticed by Top Employers! |
Our work-support plans provide precise options as per your project tasks. Whether you are a newbie or an experienced professional seeking assistance in completing project tasks, we are here with the following plans to meet your custom needs:
Name | Dates | |
---|---|---|
IBM SPSS Training | Nov 26 to Dec 11 | View Details |
IBM SPSS Training | Nov 30 to Dec 15 | View Details |
IBM SPSS Training | Dec 03 to Dec 18 | View Details |
IBM SPSS Training | Dec 07 to Dec 22 | View Details |
Pooja Mishra is an enthusiastic content writer working at Mindmajix.com. She writes articles on the trending IT-related topics, including Big Data, Business Intelligence, Cloud computing, AI & Machine learning, and so on. Her way of writing is easy to understand and informative at the same time. You can reach her on LinkedIn & Twitter.