I don't know such a measure but I think it should consider the number of portfolio investments, and the individual weights of each investment in such a portfolio. The close to the optimal (or even optimal) value of that measure in my opinion should be a state of naive diversification. But how to put in it a number of portfolio investments I don't know. It also will be a good idea to take into account overdiversification in the construction of that measure, but this can be hard challenge.
To measure the portfolio diversification variable for primary data collection, I suggest you use critical factor analysis based on themes of similarities and differences at high, medium and low importance levels explored in the interview. Good luck.
By dividing the covariance of the two assets by the sum of their respective standard deviations, the correlation coefficient is determined. A statistical measure of diversification, correlation is basically.