SPSS stands for Statistical Package for the Social Sciences. It is designed to make statistical analysis easier for researchers and analysts. Over the years, SPSS has evolved significantly, with new versions introducing additional features and improved functionality. One of the most valuable developments was the introduction of syntax, and even more so the ability to automatically generate and save syntax while performing analyses through the menu options. This made it possible to reproduce analyses and work more efficiently. SPSS was later acquired by IBM. Its popularity largely came from the fact that users did not need to write code: they could simply follow the step‑by‑step procedures, choose the appropriate statistical methods, and interpret the output produced by SPSS.
SPSS is used for statistical analysis, primarily in the social sciences. It became popular because users did not need to write code, and the syntax that SPSS does use can hardly be considered programming. One of the lecturers in Human Geography and Spatial Planning wrote the book ‘Basishandboek SPSS’ that explained each statistical procedure step by step. Every new SPSS release had its own updated edition of the book, which was very convenient because I could simply buy the version that matched the software version I was using. Of course, a solid understanding of statistics is still necessary to use both the book and SPSS effectively. More on statistics in the blog 'What is statistics?'.
I learned SPSS during my Urban Geography Studies and later while working at E‑Quality. I took many methodology courses in the social sciences, which covered SPSS, statistical analysis, and other analytical techniques used in social research. I was even taught by the author of ‘Basishandboek SPSS’. More about my master’s program Urban Geography in the blog 'Universiteit Utrecht Master Sociale Geografie'.
During my time at E‑Quality (2005-2007), I worked on a project called ‘Genderbudgettering in de sport’, which at the time was a pilot in several Dutch municipalities. Data for this project was collected in various ways: literature review, contributions from an intern, interviews, a small‑scale survey, and data from the Sociaal Cultureel Planbureau (SCP). Through DANS, E‑Quality obtained micro‑level data on how people in The Netherlands spend their free time. I used SPSS to conduct several straightforward analyses for the advisory report on gender budgeting in sports. More about this research in the blog 'Doelgericht Begroten In De Sport'.
The SCP data on how people in The Netherlands spend their free time also included information on the frequency and types of cultural activities they attended. I was allowed to use this dataset for my final thesis. In that research, I carried out several basic analyses as well as one more advanced statistical method: logistic regression. More about my final thesis can be found in these two blog posts ‘Factsheet Cultuurdeelname Voor iedereen?’ & ‘Thesis Cultuur voor Jou & Mij’.
At STOGO Onderzoek + Advies, some research projects required more analytical power than Excel could provide. For that reason, SPSS was purchased, and several straightforward analyses were carried out in SPSS for the advisory reports. The software also made it possible to coach students who were writing their final thesis or wanted to complete an internship placement. More about my work at STOGO Onderzoek + Advies can be found in the blog post 'Stogo Onderzoek + Advies'.
While working at STOGO Onderzoek + Advies, I had the opportunity to take several advanced SPSS courses at the SPSS Center. One of these courses focused specifically on creating syntaxes in SPSS. Using syntax allowed me to run analyses repeatedly on the dataset whenever needed. Instead of manually performing all the steps for each analysis, I could simply execute the syntax. The Syntaxes saved time and reduced the likelihood of errors, provided the syntax was written correctly. If there is a mistake in the syntax, every analysis based on it will contain the same error. A reliable way to create a good syntax is to use SPSS’s Record function, which automatically generates syntax based on the steps I perform. After recording, the syntax should be checked, adjusted where necessary, and tested with several control runs to ensure that every step is included and that the output is correct. Once the syntax is ready, the analysis phase becomes much more efficient. Running analyses takes less time, and it becomes possible to perform multiple identical analyses in a short period.
Before leaving for the Pacific for a sabbatical (2013-2014), I founded my business Databeet in 2012. One of the first services I offered was SPSS support. My very first assignment at Databeet was providing SPSS assistance in a promotion trajectory for a lecturer at Saxion in Enschede. After that I assisted around three people, who needed help using SPSS for their final thesis. Most of the support took place online, using examples to guide them through the analyses, and one person received hybrid support so I could teach her SPSS more directly. She ultimately earned an excellent grade for her thesis, and in her first job after graduating she was working with SPSS.
A friend’s sister asked whether I could teach her and her colleagues SPSS at their workplace, Media Contacts. I visited their office first to discuss their existing knowledge of SPSS and what they needed the software for. This resulted in a two‑day in‑house training. On the first day, I covered the basics of SPSS and guided them through exercises using their own data. For the second day, I invited a specialist in factor analysis to teach part of the session. The training was successful, but it also taught me that I do not enjoy teaching full days.
Today, SPSS is still widely used, and it remains the option I would choose when there is sufficient budget. However, when budgets are tight, there are many alternatives available. For those who can code, or are willing to learn, there are statistical packages such as R. Another option is the Excel add‑on used in the Big Data eCourse at Laudius, which I co‑developed. More about this can be found in the blog 'How Laudius asked for a Big Data eCourse and I and another teacher made it'.
In the Big Data eCourse I co‑developed with another instructor and Laudius, launched in 2018, an Excel add‑on is used instead of SPSS. We were selected for this project based on our professional backgrounds in research. I had worked with SPSS at VUMC as a data manager, at Kennisbank Filantropie as a data-manager, briefly at Altrecht as a volunteer, and for an assignment for Rotterdam Festivals. The decision to use an Excel add‑on was made to ensure that the statistical analyses required for Big Data would be accessible to everyone. SPSS is an excellent and user‑friendly tool, but it is also expensive. Using SPSS would have made the course more costly and less accessible. Almost everyone who uses Word or Outlook already has access to Excel, and the price of Microsoft Office is far lower than the cost of the SPSS package needed for this eCourse. On my website, you can find instructional blog posts explaining how to use the Excel add‑on. More about this can be found in the blog 'Big Data analysis with Real-Statistics in Excel'.
In the Big Data eCourse I co‑developed with another instructor and Laudius, launched in 2018, an Excel add‑on is used instead of SPSS. We were selected for this project based on our professional backgrounds in research. In addition to the above mentioned SPSS work, I also have worked with SPSS at VUMC as a data manager, at Kennisbank Filantropie as a data-manager, briefly at Altrecht as a volunteer, and for an assignment for Rotterdam Festivals. The decision to use an Excel add‑on was made to ensure that the statistical analyses required for Big Data would be accessible to everyone. SPSS is an excellent and user‑friendly tool, but it is also expensive. Using SPSS would have made the course more costly and less accessible. Almost everyone who uses Word or Outlook already has access to Excel, and the price of Microsoft Office is far lower than the cost of the SPSS package needed for this eCourse. More about the Excel add-on can be found in the blog 'Big Data analysis with Real-Statistics in Excel'.
AI is everywhere and used for almost everything, but it is not always the right solution. Ai can also be used for statistical analysis. In fact, Ai can be programmed to work with most of the statistical packages I have already mentioned, SPSS, R, Python and others, to produce the results requested by the user. However, Ai does not always provide the correct answer. This means that the person giving instructions to Ai must have sufficient knowledge to guide it properly. Good understanding of SPSS, careful verification, and accuracy in checking Ai‑generated output are essential. Without this expertise, errors can easily go unnoticed.