English Deutsch Français Italiano Español Português 繁體中文 Bahasa Indonesia Tiếng Việt ภาษาไทย
All categories

I am working with SPSS right now. I collected and entered data and now I am trying to make sense of it. The problem is that I don't have that great of a grasp on statistics. I mean, I understand them, but I'm not sure how to interpret the results. Here's the deal:

I'm running correlations. The results I'm getting are super significant (yay!). So, I'm getting Pearson Correlation numbers like .444, .668, and .385 paired with Sig. (2 tailed) numbers of .000. Here is my problem: What does this all mean? I understand that it's significant, but what does .000 mean? What do the numbers like .385 and .668 mean? How do I interpret them?

2006-12-07 03:08:45 · 2 answers · asked by jr_crime_fighter 2 in Education & Reference Higher Education (University +)

2 answers

The Pearsons Corr values you are getting correspond to R-squared, a measure of correlation strength. Depending on the kind of study you are running, you have to set an arbitrary bar for what makes a strong correlation. For example, when I was a kid, learned that R-squared of 80% was a very strong correlation. however, now that I do studies of real world occurrences (i.e., biostatistics), I find that I am VERY happy to get an R-squared value over 20%.

So with R-squares of 44%, 67% and 39%, if you were running a clinical trial, you'd want to throw a victory party.

And with the two-tailed p-values of ".000", that means the p-value is sooooooo small, that it is smaller than the 3 decimal places displayed by SPSS. So maybe your p-value is 0.00001.

Anyway, good news, your results are very significant.

2006-12-07 04:49:21 · answer #1 · answered by Gumdrop Girl 7 · 1 0

I will not be the first to give you this answer because I would need to be sitting beside you to do that, lol , but when dealing with Stats, what I have found when I am struggling with what my numbers mean, I go back to the basics. No one wants to go into their old notes at school, so I find this website can help.

http://www.ats.ucla.edu/stat/seminars/

2006-12-07 03:27:45 · answer #2 · answered by Raven 1 · 0 0

fedest.com, questions and answers