After thirty-five years in the workforce, I did something I never thought I would: I went back to college.
At first, it wasn’t even about earning a degree. It was about confronting a recurring nightmare. For decades I kept dreaming about dropping out of my Sociology class the week of finals. The dream nagged at me like unfinished business. So I re-enrolled to finish that course and earn the “A” my subconscious had been demanding for half my life.
Once that psychological debt was paid, something unexpected happened. I discovered I actually wanted to keep going to learn more, to explore history, politics, and the shifting patterns of a world I’ve spent my career analyzing. I’m 52 now, disabled, and taking all my courses online. It has been rewarding, frustrating, and eye-opening in equal measure.
A Digital Culture Shock
Returning to school after three decades felt less like a reunion and more like landing on a new planet. Every class uses the Canvas learning system, which, while ambitious, often feels like an archaeological dig through mismatched folders and broken links. Each professor organizes material differently: different calendars, file types, and submission rules. Some are brilliantly engaged, while others never answer an email.
Add in constant outages, especially when Amazon’s AWS servers crash, and you start to feel like education itself is hanging by a fiber-optic thread. Because I run two companies and produce about twenty films a year, my coursework happens mostly after midnight, which is prime time for system failures. When Canvas goes dark at 12:45 a.m., you can almost hear students screaming across California’s time zones.
Enter SPSS: A Software Relic
Then came Statistics. It is a required General Education course that fulfills the mathematics category. Mine, Psychology 104: Statistics for the Behavioral Sciences, uses Gregory J. Privitera’s Essential Statistics for the Behavioral Sciences (Second Edition).
As of this writing we are up through Chapter 9, covering z-scores, t-tests, Cohen’s d, confidence intervals, correlations, and regression analysis.
The idea behind the class is sound: to teach students how to interpret data, identify bias, and make informed decisions. In practice, the course often feels trapped inside its own formulas, obsessed with precision for precision’s sake. For many students, the goal isn’t learning; it’s surviving. Pass the class, check the box, move on.
That challenge is made exponentially worse by the required software, IBM SPSS Statistics, a program that looks and behaves as if it hasn’t evolved since the Clinton administration. When I first opened it, the interface felt like stepping into a time machine with no instructions. It took me hours just to install, license, and figure out basic navigation, hours that should have been spent actually learning statistics. I even wrote to my professor to thank her for her clarity and patience while explaining that my frustrations were not with her teaching but with IBM’s outdated software. It was clear that she understood the problem and shared the same concerns.
A Short History Lesson
To be fair, SPSS wasn’t always a dinosaur. It was developed in the late 1960s at Stanford by Norman Nie, Dale Bent, and Hadlai Hull, and was originally written in Fortran to free social scientists from coding every analysis by hand. In the punch-card era it was liberation. The company SPSS Inc. was incorporated in Chicago in 1975, and IBM bought it in 2009 for $1.2 billion, rebranding it “IBM SPSS Statistics.”
But somewhere along the line, progress froze. The program that once liberated researchers from code now traps students in menus and syntax windows that feel older than the internet itself. In an age where software and hardware become obsolete yearly, SPSS seems content to stand still. I sometimes wonder if IBM even realizes the stress this causes students, or if it has simply accepted that frustration is part of its brand.
My “Military Command Center” Setup
Because I’m on a Mac, my SPSS nightmare multiplied. One version worked on one OS, another refused to launch, while Canvas behaved only on a third machine. A large part of the incompatibility stems from SPSS being designed primarily for PC systems, making it a continual hardship for Mac users who must find workarounds just to complete assignments. To complete a single lab assignment, I ended up with three computers running simultaneously plus two iPhones for backup.
If anyone had walked in, they would have thought I was operating a military command center, not a statistics lab. Meanwhile, the college requires accepting endless cookies just to log into Canvas, a small but real privacy trade-off that adds irony to the mix.
A Professor Doing Her Best
To her credit, my professor has been outstanding, patient, responsive, and openly frustrated by SPSS herself. She even offered extra credit for attending a series of discussions on The Twilight Zone, a creative, human touch that made the class feel alive again. I couldn’t attend because of my disability, which made me wonder about equal-access provisions, but I admired the intent.
I also suspect her hands are tied. Departments cling to legacy tools because they are standardized, even when everyone knows they are ineffective. The system feels grandfathered in, a bureaucratic machine that cannot see how far the world has raced ahead.
Math Without Meaning
My frustration isn’t with statistics itself but with how it is taught. A course like this should bridge the human and the numerical, not divorce them. As a social-science major, I expected to explore why data matters, how bias enters research, how numbers can be weaponized, and how interpretation shapes truth.
Instead, we memorize terms like z-scores, Cohen’s d, and t-distributions that most of us will never use again. Peers admit they just want to pass. A quick Reddit search confirms it: nationwide, thousands of students voice the same despair. “Just cheat and get it over with,” one post reads.
Is that really the message higher education wants to send? That statistical literacy means surviving a clunky program rather than learning to think critically about information?
A National Irony
IBM’s SPSS once democratized analysis; now it alienates learners. Critics cite the same problems everywhere: high cost, outdated interface, slow performance, limited flexibility, and a “black-box” design that hides what is really happening behind the screen. For beginners it is confusing; for experts, limiting.
And yet universities keep paying for it, millions in licensing fees, because tradition dictates that “this is how it’s done.” Meanwhile, private industry has already moved on to AI-driven analytics that complete in seconds what once took hours. Bias hasn’t vanished; it has merely been automated.
Technology’s False Promise
I often laugh at the absurdity of it all: a 52-year-old filmmaker juggling three laptops at midnight, wrestling with a relic of 1960s software to satisfy a 21st-century math requirement. My livelihood doesn’t depend on passing this class, but for the 19-year-olds beside me, it might. They need this credit to graduate, to keep scholarships, to build futures.
I can’t imagine how any 18- or 20-year-old could navigate SPSS and its labyrinth of lab assignments without losing heart. It’s not education; it’s endurance training.
California, Colleges, and Compassion
Still, I have enormous respect for my college and for California’s determination to keep higher education accessible. These institutions are fighting uphill against underfunding, changing technology, and the impossible expectation to serve everyone.
My professor embodies that fight. She admits SPSS is flawed, she improvises around it, and she gives students permission to think beyond the formula. My critique isn’t of her, or even of the college, it’s of the outdated systems that hold brilliant educators hostage to dying software.
Reclaiming the Human Element
Researchers will always be needed, but research itself must evolve. Real analysis requires a human component, someone willing to question assumptions instead of accepting every statistical output as fact. We already know polls and algorithms can be twisted to serve agendas.
Higher education needs to rethink its structure. Why make an antiquated statistics course mandatory for every major? Teach data literacy, yes, but teach it alongside ethics, interpretation, and human bias. Many analytic courses are struggling to redefine themselves for a new generation but lack the tools to do it.
A Final Reflection
Whether I continue this course through the end or not, what I have learned about myself, and about the collision of old systems with new technology, has already been more valuable than any equation. The lessons about adaptability, frustration, and perseverance feel far more applicable to real life than the numbers on a spreadsheet.
Going back to college after three decades has reminded me that learning should open doors, not crash browsers. Education should make knowledge more accessible, not more complicated. It should meet students where they are, not in a 1968 command window.
Until then, I’ll keep juggling my midnight laptops like a weary commander at his post, grateful for professors who still care and hopeful that someday soon, education will rediscover what it’s meant to do: teach people, not programs.
Gregory Hatanaka is a Los Angeles–based filmmaker and social commentator. He writes about the intersection of culture, technology, and education.
By Gregory Hatanaka