How many digits are really significant when expressing the correlation coefficient? More is better? I beg to differ. Can someone please direct me to an article that discusses this topic? Thank you.
I agree with David, it depends on your audience and it's accepted conventions.
In general, I would report 2 or 3 digits past the decimal.
I agree with David and William, and Farooq brings up a good point to consider if it's applicable. The best way to verify a linear fit is also to plot the residual of each point (take the observed value and subtract the theoretical value from a linear fit). If the plot of residual versus the x-value of the point exhibits a pattern (such as a parabola), then a linear fit may not be appropriate. This may not apply to all data, but it's still a good test.
Why don't you present the Scatter-plot of the data?
Graphs have more data to show and they are more comprehensive.
You may want to check the SPSS help for this Q. However, The Pearson coefficient or r , indicates the strength of the association between the variables. r varies between -1 & +1. A correlation coefficient near +1 indicates there is a strong positive relationship between the two variables ,with both always increasing together. A correlation of -1 indicates there is a negative relationship between the two variables,with one always decreasing as the other increases. A correlation of 0 indicates no relationship between the two variables.
About significat digits , if you look at different papers ,r usually expressed as 2 digits after decimal and its explanation is as above. One of my papers "Subclinical hypothyroidism in Li treated ..." is an example that Pearson coefficient is used.
Good Luck .
- Badges
- Science topic
- Similar topics
- Mathematics
- Applied Mathematics