Can be specifically interesting to your research:

Bagheri, A., Habshah, M., & Imon, R. H. M. R. (2012). A novel collinearity-influential observation diagnostic measure based on a group deletion approach. Communications in Statistics-Simulation and Computation, 41(8), 1379-1396.

Bagheri, A., & Midi, H. (2012). On the Performance of the Measure for Diagnosing Multiple High Leverage Collinearity-Reducing Observations. Mathematical Problems in Engineering, 2012.

Curto, J. D., & Pinto, J. C. (2011). The corrected VIF (CVIF). Journal of Applied Statistics, 38(7), 1499-1507.

Victoria-Feser, M. P. (2011). Robust VIF Regression DJ Dupuis.

Most diagnostics for multicollinearity do not provide a way for any statistical validity. A statistical test of multicollinearity is available from Haitovsky. His approach standardizes the non-constant terms in the X'X-matrix to bound the determinant from 0 to 1. The Haitovsky test statistic is calculated in part from the bounded matrix and is distributed chi-square.

Multicollinearity in Regression Analysis: Comment

Yoel Haitovsky

The Review of Economics and Statistics

Vol. 51, No. 4 (Nov., 1969) , pp. 486-489

I recommend you article:

Friendlya M., Kwana E.

Where's Waldo? Visualizing Collinearity Diagnostics, The American Statistician,

Volume 63, Issue 1, 2009,

http://www.datavis.ca/papers/viscollin-web.pdf

you can use this URL to explain that clearly.

http://www.m-hikari.com/ijcms-2011/9-12-2011/rashwanIJCMS9-12-2011.pdf

Keep using diagnostic measures of multicollinearity, such as, eigenvalues; R-square, along with small t- test values and

pairwise correlation. these are more than enough for diagnostic tests for detection of multicollinearity.