PDA

View Full Version : Is statistics intentionally overly-complex or do psychologists just like big words?


Rude Louis
04-08-2009, 07:21 PM
"The sum of squared deviations for rows is the sum of each score's row's mean's squared deviation from the grand mean."

Yes, three possessives in a row.

"The between groups population variance estimate is the sum of squared deviations of each score from its group's mean divided by the degrees of freedom for the within groups population variance estimate."

Is there some simplified way at learning statistics that I am just not seeing? Or is the field always going to be a bitch?

rabbitweed
04-09-2009, 08:19 AM
Science is riddled with Inkhorn terms. Chemistry is also a pretty bad offender.

Mantikore
04-09-2009, 10:54 AM
using numbers is better than saying that something is "close to" something else or is "likely to happen"

its more objective. but yes, sometimes, it may be overcomplicated

superspeedz
04-09-2009, 01:26 PM
"The sum of squared deviations for rows is the sum of each score's row's mean's squared deviation from the grand mean."

Yes, three possessives in a row.

"The between groups population variance estimate is the sum of squared deviations of each score from its group's mean divided by the degrees of freedom for the within groups population variance estimate."

Is there some simplified way at learning statistics that I am just not seeing? Or is the field always going to be a bitch?

lol you read the text?!?!?!

I just read the equations and formulae

a334jv2df
04-09-2009, 01:39 PM
lol you read the text?!?!?!

I just read the equations and formulae

Likewise. Don't bother with the textbook. Have your equations on hand, and a sample problem done, and just switched the values to match your current problem.

Mr.Happy
04-09-2009, 01:47 PM
I used to bitch about the stats I do in my Psychology degree too. Then I looked at the stats my maths-studying housemates take. Be thankful.

If you can't just think in formulae, a quick way to make it easier to comprehend is to mentally group each word-string term together as one.

So, "the between groups population variance estimate is the sum of squared deviations of each score from its group's mean divided by the degrees of freedom for the within groups population variance estimate" would become "value a is the sum of term b divided by the d.f. for term c". It stops you reading all the words out in your head and getting caught up in processing semantic meaning when you've got no reason to.

Agent 008
04-09-2009, 01:50 PM
Likewise. Don't bother with the textbook. Have your equations on hand, and a sample problem done, and just switched the values to match your current problem.

It's a lot easier to actually understand what's going on, than to memorize every one of the bazillion formulae out there.

Zip
04-16-2009, 10:43 PM
That's just ANOVA in statistical terms. It's basically a technique for explaining the variation in some response (dependent) variable in terms of explanatory (independent) variables. It's a very powerful method because by "partitioning" the variance in the response variable, you can determine how much the explanatory variables, plus some error term, contribute to the overall variance. For example, suppose you're making plastic parts with an injection molding process. You have several controllable parameters on the machine, for instance melt temperature, mold temperature, clamp time and cooling time. By methodically varying one or more of these parameters at a time and observing the resulting output, by a process called design of experiments, you can determine exactly which of the parameters are causing the variation in your output or if there is any interaction between the parameters with only a few data points. In psychology, the immediate application is determining the effect of treatments on a patient's outcome.

It isn't really helpful trying to understand this through text, you have to look at the equations and practice with some real data. Work through some examples and you'll have no problem, in my first statistics course I had to do 3-way ANOVA by hand.

These are some of the most important techniques you will learn if you do research, though in psychology many statistical techniques are misapplied and the assumptions violated to come up with bullshit conclusions. It tends to happen when you apply scientific techniques to something that is intrinsically not a science. In terms of the amount of "science" in the basic tenets of psychology, I would rank it somewhere between creationism and the performing arts, but I digress...