SPSS

with AI, so a lot of times it was done in small areas.

I remember when I first started working with it, the streaming technologies like SPSS, where we were doing predictive analytics of data that was in motion.

That was one of those starting points.

Now, I'm talking about just 20 years ago, but reality, this started back in the 50s.

I'm eager to see the spreadsheet.

I'm eager to see some SPSS.

Analysis.

Yeah.

Collecting data, uh, depending on your field, right? Like collecting data, reading primary sources, secondary sources, doing library research.

Um, what else? You know, I'm, I'm talking about everybody here, so like some of you are running SPSS stats things.

Like, some people are growing plants.

Some people are like, right, all of that research and progress.

[laughs] SP, SPS.

[laughs] Na no be only SPSS.

[laughs]

That guy na guy man.

I was running all the data.

I'm in SPSS running the ... data, and it was so profound when I ran demographics.

And we had a lot of demographics.

I just kept seeing question after question, and there were 83 ...

You're trying to understand beyond just the click and the purchase, which were, albeit still... challenges with methodologies back then but like you're trying to also understand brand lift and things like that and so was that where like yeah a dynamo logic would come in you would sort of like do uh maybe it was it like a a whole like some people would be exposed to an ad some wouldn't and then you hit them with a survey and say are you familiar with network solutions is that kind of what was going on that's it that's it

yeah we also did our like hack together our own media mix modeling marketing mix modeling you know in like spss software and you know, Excel and, you know, it was probably not very good or accurate,

but, uh, yeah, we make it, I think is probably the headline for the entire series to date.

And, you know, none of us really knew what we were doing.

Yeah, no, I'm with you.

And I think I only try the kind of examples, well, many years ago, when the first SPSS and SaaS and whatever kind of, let's say, softwares were coming to the statistical market, let's say, I think it was quite simple to run statistical analysis, right? Exactly, you had to press a button, then you had a lot of different outputs with different kind of functions and different kind of assumptions behind.

But if you haven't had the kind of statistical background, you will not be able to really recommend any of those results because you didn't know the assumptions.

Hence, it was very difficult to interpret.

And interestingly enough, I had asked Claude previously if it could analyze CSV files, and it said, no, sorry, I can't do that.

But my co-author on this project said, yeah, but we got an SPSS data file, and it says it can access SPSS data files.

And it did a pretty good job of pointing out some findings that I don't think I would have found, even though I have a lot of experience at survey research analysis.

And I was generally impressed by it.