Afterwards, a multinomial regression analysis was used (Pettersson, Linden-Bostrom and Eriksson, 2009). The qualitative analysis of obvious barriers was completed by using content analysis on the answers to the opened-ended questions that were in the questionnaires. There were 176 responses all that considered meaning units. The significance units were then reduced into shorter sentences without altering the basis of the statements. The reduced meaning units were labeled with one or several codes. Sub-categories were shaped based on those codes and classified into eight main categories. The main lessons were separated into two main themes. The assessment was carried out by the first author, but before any concluding decision was made the co-authors read through all the steps in the analysis in order to verify the results (Pettersson, Linden-Bostrom and Eriksson, 2009).

Strengths and limitations of the mixed-methods approach

By joining quantitative and qualitative data, a better understanding of the predicament addressed in the study can be given and a more inclusive picture can surface. The mixture of quantitative and qualitative data gives a more detailed knowledge of the participants' viewpoints. The combination of the parents' quantitative answers from the three surveys along with their written answers about reasons for not participating offered additional information about non-participation. Two positions of non-participation were clarified in this study. These were limited non-participation and non-participation....

Parents who were selected as partial non-participants had put in some measures into the program but they did not believe themselves to be participants. This is a significant distinction, because the parents were allowed to describe themselves as non-participants even if they took part in some of the activities. Additional analysis will demonstrate if this distinction is important for the results of the program (Pettersson, Linden-Bostrom and Eriksson, 2009).
The HBM has been broadly used in order to look at participation in diverse health interventions, but the use of this model in order to comprehend the lack of participation in parental support programs designed to prevent underage drinking is incomplete. This study has shown that the model is useful in these kinds of studies, but not necessary in the type of study that the current one was. It is important to stress that the HBM was used as a tool in the analysis of the current study but yet the chief purpose was to understand non-participation in a parental program and not to evaluate the application of the HBM to parental involvements. The HBM model was not used as a guide in planning the study and thus was maybe not the best one to use. Nevertheless, many items in the questionnaire fit the major elements in the model theoretically and empirically (Pettersson, Linden-Bostrom and Eriksson, 2009).