SPSS
SPSS (Statistical Package for Social Sciences) is the most popular, the standard software, for statistical analysis in social and communication sciences.
We will use SPSS for the data analysis phase of our project.
To be familiar with this software will elevate the value of your profile in the professional market in the field of strategic communication.
We are living in a data driven society. To be able to generate, manage, analyze and interpret data set can open professional doors for you in the area or political and corporate communication.
Central Connecticut State University offer this software to all its students. However, you will not be able to download the program to your computer. You must access it from the applications’ page in the University’s Web-site.
In the first tutorial, you will learn how to access the application through the Citrix platform.
The URL to the application page:
If you have any problem accessing SPSS through Citrix, contact immediately the help-desk:
(860) 832 1720
Or E-mail: techsupport@ccsu.edu
1 Access to the Software through Citrix
One of the tricky aspects of working through Citrix, is that you cannot save your SPSS files in your computer.
There is a way, though, to have access to the documents. From your computer at home. You can save it into your OneDrive (the university’s cloud service).
In this link you can find information about using OneDrive with Citrix (the remote application server of the university).
2 Saving Files in OneDrive
Since files can not be saved or downloaded from the Citrix platform to our computers, I strongly recommend using the OneDrive, the university’s cloud system. Contact the help desk if you cannot access the this service.
3 Questions and Numeric Values
In the next video, you will just find the questions I am using for this tutorial and the numeric values I assigned to them. These numeric values are arbitrary.
If you click here, you should be able to download a package with the pdf file with this information and the Excel spread sheet with the data, which we will import in the next step of this tutorial.
4 SPSS – Defining Our First Variable
We introduce in the next video, the data variable interface. We learn how to define the value of the most relevant columns.
5 SPSS – Different Types of Variables
After we have learned how to define the most basic variable in SPSS, the dichotomous variable, we are ready to define the rest of the questions in our tutorial.
We will add multiple choice questions and scales. In your projects, most of your variables will fall under these categories.
6 SPSS – Importing Data from Excel
In case you have not done it yet, you can download the tutorial package clicking here. In the Excel spread sheet, you will find the data you need to replicate what we are learning in the tutorial.
7 SPSS – Descriptive Statistics: Frequencies
Once we have imported the data to the SPSS file, we are ready to start the analysis. We will introduce, first of all, descriptive statistics.
In the next video, you will learn how to generate and interpret the most basic form of this modality of statistics: the frequencies.
8 SPSS – Descriptive Statistics: Crosstabs
Cross Tabulations are useful analysis to help visualize the relationship between two variables. They cannot give you any certainty about the statistical significance of that relationship, but it gives you a clear idea of the strength and the direction of the correlation.
9 SPSS – Descriptive Statistics: Correlations
In the following tutorial you can learn how to run Pearson correlations in SPSS.
The Pearson correlation quotient tells us whether the observed relationship between two variables is statistically significant or not.
Yet, this statistical calculation is still considered descriptive.
The reason is that even if we can establish the statistical significance of the relationship between the variables, we are not allowed to stablish causal relationships. Thus, the Pearson correlation quotient would never allow us to use one particular variable as predictor of other variable or sets of variables. It just confirms that the two variables co-variate in a positive or negative way.
10 SPSS Inferential Statistics: Independent Samples t-Test
We are using the Independent Samples t-Test to measure the statistical significance between the two groups of one categorical dichotomous variable and an interval/continuous variable.
This type of test allows us to differentiate between the independent variable (our predictor) and the dependent one.
The predictor will be our categorical dichotomous variable (gender).
We can use as dependent variable any or all the scales used in the tutorial.
11 SPSS Inferential Statistics: ANOVA
First of all, ANOVA stands for Analysis of Variance.
The analysis of variance is also another statistical procedure to establish significant differences among the different groups (answers) of a categorical variable. In this regard, ANOVAS are similar to the Independent Samples, t-Test we learned in the previous tutorial.
In the tutorial I study whether the variable of ethnicity may affect how students value the quality of the instructor of the course.
Clicking on the following link, you can download an Excel spread sheet with the added variable of ethnicity, which was missing in the previous one.
12 SPSS Inferential Statistics: Regression
Regression models measure the predicting value of one or multiple variables. It is a highly sophisticated procedure used frequently in economics, for instance to assess the risks of investments.
In social sciences, regressions models are used to established causal relationships between variables that have a significant correlation.
It is important to understand that. A correlation shows that two things/phenomena are happening simultaneously. Still, we cannot conclude from the correlation analysis that one variable is the cause and the other the consequence.
The regression model helps us establish more sophisticated relationships. It expresses how the variance in one variable might explain the variance in the other variable.
There are different types of regression analyses. Depending on the type of data (categorial, ordinal, interval, ratio) and the number of predictors we can apply simple linear, multiple linear, ordinal, binary or multinominal logistic regression models.
In this course, we will focus on the simple linear regression, a statistical procedure to measure the predicting power of a single interval variable.