Data Analysis SPSS: Statistics with SPSS can be complicated, but here’s how to get started with data analysis using SPSS. After data importing into SPSS or typing follows the verification of the data, ask help in the essay writers. Are they complete? Are they valid? How is the data distributed? The evaluation of your data can be roughly divided into three areas: the descriptive (descriptive), the exploratory (exploratory) and the closing (inferential statistical) analysis. In a practical example we demonstrate the general procedure. Furthermore, the result of a hypothesis test is explained in more detail.
Data analysis SPSS: General for descriptive statistics
The descriptive statistics are suitable for the first steps of your analysis. Because it can help you find errors in the data. In SPSS you can also add the descriptive statistics with matching diagrams. Thus, for example, you get an idea of your SPSS data distribution. It also gives you a first idea of whether certain conditions are met for the implementation of the following statistical procedures.
Loading the sample file
SPSS provides some sample data sets with the installation. Open the file demo.sav in the welcome screen. It contains data of a telephone provider.
You can switch between the data view and the variable view in SPSS. The data view usually contains the cases and the data points. Whereas the variable view contains the detailed description of the variable. For example, here you can insert the name of the variable and a more detailed description, taking into account the SPSS syntax.
After you have checked the record, you finish the descriptive statistics.
Calculate mean, mode, median, standard deviation, and variance with SPSS
Key figures of the descriptive statistics are mainly median, mode and the arithmetic mean. But also parameters paper writers of dispersion such as the standard deviation and variance. For example, you use the variables “old”, “marriage”, “income” and “internet”.
First, you switch to the data view. You realize that the age was queried in whole numbers and not as a category. Here the measurement level for “age” is stored as metric.
The variable “marriage” is given as a numeric data type. In addition, the data value is entered under the “Values” column. 0 = means “not married” and 1 = “married”. The variable “einkomm” contains the “Household income in thousands” according to the registered inscription. If you have difficulties with the assignment here, the statistics advice is recommended.
Descriptive Statistics – Data Analysis with SPSS
Now we’ll explain how to use descriptive statistics analysis with SPSS. To create the descriptive statistic, in SPSS, choose Analyze, then Descriptive Statistics and Frequencies. Then you drag the variables into the right box. Make the selection as shown in Figure 3. Finally, confirm the entry with the “Next” button.
The evaluation window now appears in the result window of the SPSS software. The columns represent the evaluated variables. However, the calculation of the median or the arithmetic mean does not make sense for all variables. You use the arithmetic mean for metric scaled variables. Whereas the median presupposes at least ordinal scales.
The median divides the data of the variable in half. Because he is almost the value in the middle. The arithmetic mean of the variable “Household income in thousands” is M = 69.47. However, the median for the same variable is MD = 45. The standard deviation is the mean deviation of the values from their mean. Furthermore, the variance forms the mean square deviation.
The mode is the most common value for a variable. 39 years is in our example in the variable “age in years” so the most common name. The arithmetic mean can not actually be used for nominally scaled data. However, you recognize in our case that about the same number of interviewees are married (M = 0.5). More precisely here is the output of the table of frequencies of the variable “Married”.
Key figures by groups in the descriptive statistics
Now you have calculated some key figures of individual variables. Furthermore, we want to form the descriptive statistics by groups.
Select “Data” and “Split File”. Then you set a categorical variable as a partition variable. Now you can perform the analysis again, just as previously described. Alternatively, of course, you can relax using the data analysis service. After you’ve done the analysis, you’ll get the metrics per group. The median and arithmetic mean are almost equal in the sample of the unmarried and the married sample. In addition, it can be seen that in both groups the distribution is approximately centered because of the almost identical median and mean.
This leads us to the next step: statistical tests. First of all, the hypothesis test. Do not forget to delete the split command in SPSS for the following analysis.
Hypothesis test principle for data analysis with SPSS
In statistical testing of a hypothesis, you give a pair of hypotheses. This you then check by calculation for validity. The null hypothesis is often formulated so that there is no effect and, for example, a correlation in SPPS does not result. The null hypothesis for a non-directional t-test for dependent samples is: “The mean values of both groups are equal”. The alternative hypothesis contains the effect. At the t-test she says: “The two groups are different”. Now you compare the calculated value with a threshold value. You can discard the null hypothesis if the calculated value is below the comparison value. Then the alternative hypothesis applies.
The t-Test – Example of Data Analysis with SPSS
In a simple example, let’s look at the principal in a t-test for data analysis in SPSS. For example, you want to test whether married and unmarried persons differ in their age.
Select “Analyze” then “Compare averages” and “T-test for independent samples”.
For the success of the t-test, the dependent variable should be metric. The distributions of the two groups should be approximately normally distributed in the dependent variable. In addition, you should check whether the variances in the groups are approximately equal. You can test these with the Levene test. Finally, you can read the result directly in the t-test’s output (for more information, check the website of a professor at the University of Sussex).
Levene test with SPSS
The Levene test examines the null hypothesis: “The variances of the groups are the same”. The significance test yields the result p = 0.888. First, you compare the p-value with a reference value. Usually you can use 0.05 for reference. If the calculated value is above this, you can not discard the null hypothesis. In this case, the variances can be assumed to be the same.
Interpretation of the t-test
Now you stay in the first line of the SPSS output. The null hypothesis of the t-test is: “The two mean values of the samples are the same”. In the column “Sig (2-sided)” you now read the p-value of the t-test. Here too, the value is above the 0.05 threshold. Again, the null hypothesis can not be rejected. If in doubt, the statistics service can provide more clarity.
If the p-value were below the 0.05 limit, you could reject the null hypothesis. Then the alternative hypothesis applies. In that case, there would be a difference between the two samples in terms of age.
In this example, the interlocking of descriptive and closing statistics becomes clear. So use the advanced data analysis capabilities of SPSS to increase the quality of your analysis. Also, be sure to consider SPSS Python programming. The descriptive statistics help you to check prerequisites for further analysis in SPSS. In most cases, these include methods for testing hypotheses. With their help you can answer your research questions about essay writers.