# Spearman Rank + Quadratic Regression (spps) HELP! x

#### Nic0le.onStats

##### New Member
Hi,
I'm trying to find out if there is a significant correlation between salinity and no. of Bivalves found on a shore. I did a Spearman Rank and this was my result:
"There was a statistically significant correlation between Salinity and the number of Bivalvia individuals (Spearman Rank Correlation Coefficient, rs = 0.63, n = 40, P < 0.05)."
But then I did quadratic regression and it's saying it's not significant, with a P value of 0.267 !!
How can this be when my rs value is pretty close to +1?
Am I using the wrong tests?? If so, what do I use??
Many thanks

#### Karabiner

##### TS Contributor
Do you mean, you have a model of this type: y = b0 + b1*x + b2*x² + e ? And the p value for R² (the F test) was 0.267?

With kind regards

Karabiner

#### Nic0le.onStats

##### New Member
Hello,
Sorry but I'm not very skilled in this area, I am using SPSS.
My equation was: No. of Bivalvia = -2.78 + (29.01 x Salinity) + (0.07 x Salinity)^2
Hope this clarifies

#### Karabiner

##### TS Contributor
I suppose you mean (0.07 x Salinity^2) ?
Could you please report the p-values of the coefficients, and R² and its p-value ?

With kind regards

Karabiner

#### Nic0le.onStats

##### New Member
I suppose you mean (0.07 x Salinity^2) ?
Could you please report the regression coefficients, their standard errors, their p-values, and R² and its p-values ?

With kind regards

Karabiner
These?

#### Attachments

• 23.6 KB Views: 2

#### Karabiner

##### TS Contributor
Now I am not sure what your problem is. The regression model significantly predicts the dependent
variable, as indicated by the ANOVA table. The quadratic term is not significantly related to the
dependent variable, but this is regardless of whether rs was large or small.
Within the model, salinity has not a statistically significant coefficient, but this could be result of
multicollinearity (high correlation between salinity and salinity²).

Have you ever plotted y against x (scatterplot)?

With kind regards

Karabiner

#### Nic0le.onStats

##### New Member
What is multicollinearity?

Attached is my scatter plot.

Many thanks

#### Attachments

• 13.8 KB Views: 1
Last edited:

#### noetsi

##### No cake for spunky
Multicolinearity is when you can not separate out the effects of two or more predictors on the Y variable. If your model is significant (reflected by say an F test) and none of your predictors are then it is likely.

Run a VIF test and see what you get.