Using IBM SPSS statistics

Nowadays, most quantitative data analysts use SPSS or an equivalent statistical software package. Such tools are widely regarded as being much faster and more efficient than mental arithmetic, as they can generate huge volumes of complex statistical data within seconds. If you prepare a probability sample, SPSS can help you to produce high-quality results. If you have a very small data set, though, using SPSS would be akin to using a sledge-hammer to crack a nut.
Reference: Bryman: Social Research Methods: 5th Edition Page(s) 353

The Data Viewer is one of the two screens that comprise the Data Editor in SPSS, the other being the Variable Viewer. The Data Viewer is a spreadsheet grid into which you can enter your data for analysis. It is actually the first screen you will see when you start up the programme, and you can go to work straightaway by entering the data you have collected; questionnaire by questionnaire, interview by interview etc.
Reference: Bryman: Social Research Methods: 5th Edition Page(s) 355 (see plate 16.1)

It is conventional to represent an inferred relationship between two variables in this way because it makes tables easier to read. Typically this is done when you feel you can make a claim of causality, so that a change in the independent variable produces a change in the dependent variable. Similarly, when producing a bar chart or scatter-plot, you should assign the independent variable to the x axis (to produce columns) and the dependent variable to the y axis (to produce horizontal readings).
Reference: Bryman: Social Research Methods: 5th Edition Page(s) 367,

Recoding variables involves changing the way scores or values for a particular variable are distributed across the range. For example, “age” (an interval/ratio variable) can be re-categorized into five different “age groups” (an ordinal variable). This creates a new variable (or variables) and transforms the way in which a concept can be analyzed and represented.
Reference: Bryman: Social Research Methods: 5th Edition Page(s) 357,358

Clicking the tab on the bottom of the Data Editor screen will switch the programme to the ‘Variable View’. You are limited to eight characters for the variable name, so there is a limit on how you can express the variable for the purposes of SPSS calculations. However, you can enter a longer and more meaningful name as a variable label. SPSS will use the label for all printed output. An example within the Gym example-dataset would be reasons. A variable label provides a more detailed description of what this means, and serves as a memo to oneself: for example: reasons for visiting gym.
Reference: Bryman: Social Research Methods: 5th Edition Page(s) 355

This is a straightforward way of printing your bar chart as a piece of “output” from SPSS. If you do not specify which things you want to print from the output summary box on the left of the screen, SPSS will print all of the graphs and tables in the Output Viewer. You can also locate a printer ‘icon’ like you have seen in many other computer programmes, which will open a ‘Print dialog box’. SPSS will warn you that your output has not been saved if you try to close the Output Editor. If that should happen, save your output as a file (SPSS gives you many types to choose from) and decide later on which material you want to print (and even which programme to print from).
Reference: Bryman: Social Research Methods: 5th Edition Page(s) 370

The Chi-square test is down a number of levels in the Analyse drop-down menu. The entire sequence would look like this: click ‘Analyze’; select ‘Descriptive Statistics’; select ‘Crosstabs’; choose your dependent variable for the ‘Row(s)’ box and your independent variable for the Column(s) box); click Cells, then check ‘Observed’, ‘Column’ and ‘Round cell counts’ on the Cell Display dialog box and then ‘Continue’; back in the Crosstabs box, click ‘Statistics’, then check ‘Chi-square’ and ‘Phi and Cramér’s V’ on the Statistics dialog box and then ‘Continue’; finally, click ‘OK’ on the Crosstabs box and you will get an output like that shown in Table 16.2 on page 364.
Reference: Bryman: Social Research Methods: 5th Edition Page(s) 363-366

If you only have access to a monochrome printer, this can make it difficult to see where the different coloured “slices” of a pie chart begin and end. A practical solution is to represent groups of cases in terms of patterns rather than colours. Even if you have access to a colour printer, it is usually much more expensive to print in colour than in ‘black and white’. This might be the time to find out what facilities are available to you in your institution and how the printing credits are calculated.
Reference: Bryman: Social Research Methods: 5th Edition Page(s) 361

Following this set of steps will open the “Frequencies” dialog box, in which you can select the variables you want to analyse and then click “OK”. It is worthwhile experimenting with the various drop-down menus to discover what else SPSS can do for you. Probably the best course of action is to ‘play’ with the gym-set data until you feel you are comfortable with the programme, then input your own data.
Reference: Bryman: Social Research Methods: 5th Edition Page(s) 360

Spearman’s rho is a test of correlation, so we should expect to find the SPSS function under ‘Analyse’ – ‘Correlate’. Selecting ‘Bivariate’ opens up the “Bivariate Correlations” dialog box and allows you to generate a coefficient to show the strength of the relationship between variables you selected. Plate 16.16 on page 367 shows the dialog box featuring age, cardmins and weimins as the selected variables. Remember, Spearman is used when one or both variables are ordinal (Table 16.3, on page 366, shows the output for a Pearson test).
Reference: Bryman: Social Research Methods: 5th Edition Page(s) 363 (and see plate 16.16 on page 367)