As a model we choose a multiple categorical regression with optimal scaling

However, taking into account all these similarities and differences between the two regions, a comparative research could be valuable in order to understand the specificities of the relationship between life satisfaction and risk behaviors. As general satisfaction with life, perception of one’s own life compared with parents and the wish to stay in the same place are theoretically intertwined variables,we decided to build a composite indicator separately before making any further analysis.The items were grouped into a single variable through dimension reduction.The Categorical Principal Components Analysis was used. This I simplemented in the Categories module of SPSS. CATPCA is a multivariate technique developed to analyze categorical variables .This is an implementation of the optimal scaling approach to Nonlinear Principal Components Analysis .CATPCA is similar to traditional PCA, but it is designated for variables of mixed measurement level, possibly nonlinearly related to each other.

Two CATPCA analyses were carried out with two components using differents caling levels: ordinal and spline ordinal scaling levels. In terms of PVAF and Cronbach’s alpha, the spline ordinal solution turned out to be a more appropriate approach for this type of data. Table 2 shows the results of the CATPCA analysis for Frosinone province. It contains the percentage of variance accounted for by the component , the eigenvalues and the Cronbach’s alpha after carrying out the CATPCA with two components with spline ordinal scaling level. The one-dimension component fits well the data having an eigenvalue higher than 1 and Cronbach’s alpha by 0.822, explaining about 73% of the variance.Thus, only the first principal component was retained in the analysis.The object scores were saved and used as a new dependent variable for further analyses. The results of CATPCA analysis for Timis are shown in Table 3. Theone-dimension component fits well the data having an eigenvalue higher than 1and Cronbach’s alpha by 0.825, explaining about 74% of the variance. As in the case mentioned above, only the first principal component was retained in the analysis.

The object scores were saved and used as a new dependent variable for further analyses. First step consisted of computing frequency distributions. They were calculated to examine the distribution of the variables in both samples. Table 5 displays the frequencies of each risk-taking behavior. It can be noted that the use of alcohol is higher in Frosinone sample, with about 20-percent difference between Timisoara’s and Frosinone’s samples. Another important difference between Frosinone and Timis was found regarding the use of tobacco.On the other hand, data were similar with regard to the use of marijuana.In the next step, a regression model was used to determine whether a significant relationship existed between life satisfaction and the use of alcohol, tobacco and marijuana. The analysis was carried out separately for each sample. The model was employed using SPSS Statistics software package. CATREG is anonparametric method which can perform multiple regression with categorical or a mix of numerical and categorical variables. Due to this, it has recently been drawing the attention of researchers from social and behavioral sciences.

CATREG is designed to allow nonlinear transformations of the variables,including the response variable. Moreover, according to Van Der Kooij et al. it “can be used with numerical data to explore the existence of nonlinear relationships”.Regression results suggest that exposure to alcohol lead to the decrease of life satisfaction in both communities. On the other hand, there was no found any connection between life satisfaction and tobacco or marijuana use. The next section presents in more detail the results of regression analysis. The results of this study suggest that the amount of variance in life satisfaction scores in youth that can be attributed to the use alcohol is relatively small, albeit statistically significant. This research seems to confirm other international studies that proved that when two variables are related, the true relationship can be characterized by more than one possible association. For example, low levels of life satisfaction could be the cause for the use of substances, or the use of substances could affect the level of life satisfaction. Hemp is an annual herbaceous plant belonging to Cannabinaceae family known to have played a historically important role in food, fiber and medicine production. For centuries, it has been considered as one of the most important agricultural crops by providing necessities such as cordage, cloth, food, lighting oil and medicines. In Europe, it was mainly utilized as a source of fiber and seed. In 1999, the EU produced about27,000 t of hemp fiber and 6200 t of hemp seed, mostly in France, and 90% of this was used as animal feed. In particular, the seeds have traditionally been employed as feed for bird and poultry . In other parts of the world,it was primarily used as a source of psychoactive drug .