Multilinear regression is one of the major multivariate analyses approaches, and has been applied to MLAS [12], [13].
The presence/absence data of all samples were analysed using multivariate analyses (similar in approach to [3], [13], [54], [64], [65] using PRIMER-e (version 6.0; Clarke 1993).
Recognizing that some characteristics can be interdependent, we performed multivariate analyses with two approaches whereby interactions among variables are emphasized: recursive partitioning via receiver operating characteristic (RP-ROC) [ 26] curves and (boosted) classification and regression trees (CART®) [ 24, 27, 28].
Use was significantly more common among men reporting recent unprotected anal intercourse with a sexual partner whose HIV status was positive or unknown than among men not reporting such unprotected sex (28.9% of 45 vs. 6.2% of 195), though in multivariate analyses, this association approached but was not statistically significant (OR = 2.7, 95% CI 0.9 8.0).
In order to minimise the number of individuals lost in multivariate analyses, a modified hot-deck approach was used for imputation.
For the multivariate analyses, the choice of estimation approach was informed by the nature of the dependent variable, i.e. utility decrement.
For the multivariate analyses, the choice of estimation approach was informed by the nature of the dependent variable under consideration in each model.
Significant variables with p?<?0.1 in univariate analyses were then tested by multivariate analyses with the forward stepwise approach, with the p value set at 0.05 for entry and 0.1 for removal.
All these tests can be performed under different assumptions and with slightly different approaches, and multivariate analyses are generally performed.
Univariate and multivariate analyses were performed in both approaches.
