by Ted Cleaver
How dreadful. Most people who are confronted with this program hate it. Most people hate mathematics and statistics. They require abstract thought. There is no, at least in simple calculations, margin for error or interpretation. You’re either right or wrong. Its hard to pinpoint what might be the cause for the contempt most people have for studying how to manipulate numbers. But its not a sexy subject, not at all.
I’ll be using SPSS again this week, processing binomial regression analyses – although I might be able to transform my categorical into continuous data. I used to hate the subject and the program, but once you get into it it’s really okay. Statistics is all about understanding the underlying principles. I think it’s one of the subjects were a wise old man with a white beard can provide more insight with three well formulated sentences than a 600-page textbook. Honestly, the phrasing in some of my textbooks unnecessarily complicates things. Maybe consultancy firms should look into teaching statistics and provide supplementary courses where the principles are summarized on sexy slides.
What is most striking is that we perform statistical analyses every day, even if we are not aware of it. Like when estimating how long it will take for the next bus to arrive. Or, more fundamentally, when using a number of superficial indicators to estimate how much we’d like to be friendly with someone we meet. In essence that is a statistical calculation, grouping a number of predictors together and inferring a result from them. “Oh, you’re friend is from Cali? Bet she knows how to party.” See, the expected result is based on the predictor. The predictor is generalized to the point that it’s useful for making inferences.
Starting tomorrow, I’ll be trying to find a causal link between a specific attribute of employees and the strength of their social relations in the workplace. The research is interesting and relevant. Peers at the workplace are crucial to our happiness. If you don’t get along with your colleagues you’ll dread your workplace. Looking for factors that influence social relations at work is part of the growing research in the softer aspects of organizational ordering.
Use SPSS well and it’s a pretty nifty tool. Notice how the geek in me just chose the word nifty – and ran with it. The geek in me is looking forward to produce significant results. Yay.