Using Hierarchical Linear Modeling Methods

(Conklin p 170). This study uses a research technique that involves repeated observations of the same variables over a few decades. (ADD Health, n.d.).

| Based on explicit knowledge and this can be easy and fast to capture and analyse.Results can be generalised to larger populationsCan be repeated – therefore good test re-test reliability and validityStatistical analyses and interpretation are

Study mentioned more than once that the study used hierarchical linear regression model. However, to be a hierarchical model there should be the multilevel of observations or completely nested level. Considering the explanation on the article and looking at the analysis this is not actually a hierarchical regression model instead it is only a multiple regression model.

The multiple variables enlisted in the statistical hypothesis testing were measured at designated intervals under a layered conception to utilize prior study tools as resources and benchmarks.

Data used in this study is from a previously prepared collection for a different study and hence does not require any instruments. Use of Microsoft Excel and graphpad, an online software, helped in calculating the results and analyzing the data.

Table 2*. Correlations of the key study variables, with means (M) and standard deviations (SD).

The researchers used a multimethodological approach in order to gain the data that was needed to form the conclusion and other studies were referenced in order to add to the findings of this study.

The analysis included t-tests (Studies III & IV), Pearson correlation, multiple regression (Studies I & III), and repeated measures ANOVAs (Study IV).

Mendelson et al. (2010) reported multiple limitations, or characteristics of design or methodology that impacted the interpretation of the research findings. First, the small sample size of students did not provide enough power to determine true group differences and also prevented testing of moderation and mediation (Mendelson et al., 2010). There is also the limitation that the researchers do not know the Type I error rate, or the chance that a false positive occurred do to the rejection of a hypothesis when the results could have happened by chance (McMillan & Schumacher, 2010). Even thought the findings are consistent with an intervention effect, there can be no inference of causal effects because Mendelson et al. (2010) did not take into account for the clustered nature of the data. Mendelson et al. (2010) also reported that they did not utilize hierarchical linear models due to the fact that there were only four schools in the study. Mendelson et al. (2010) may have recruited more highly motivated students and students who have parents that are more engaged, thus biasing the sample population of students. This bias of sampling alludes to the fact that the students who participated may not truly represent the intended study population.

The relevance I found was the use of multilevel models to construct and examine the hierarchical relationship between the climates and commitment (p. 74). Task commitment on the individual level, and a task-involving climate was significantly related to commitment on the team level (p. 74). In an effort to enhance commitment, coaches and sport psychology consultants may wish to employ strategies targeted toward the facilitation of task involving and caring climates (p. 74). The article developed evidence based psychology to help the health professionals and their sponsoring organizations to analyze and structure reviews that have produced the guidelines for practice and also writing about some of the limitations of the evidence they produced

SEM allows researchers to use more than one outcome at the same time and to test the direct and indirect effects of the predictors on the outcome. It allows the same variable to be used as a predictor and as an outcome at the same time. It also gives a map for all the studied variables in order to understand the overall relationships among them. SEM comprises both path and measurement models. Measurement models represent latent variables and provide a better adjustment for error in measurement of a construct [207, 208].

Well there were the 'prom rulers,' the 'goth gathering,' the 'Mexican posse,' the "athletes," and the "performers." I arranged them, unwittingly, by the gathering they hung out with, the garments they wore and the segment they sat in.

The objective of this chapter is to describe the procedures used in the analysis of the data and present the main findings. It also presents the different tests performed to help choose the appropriate model for the study. The chapter concludes by providing thorough statistical interpretation of the findings.

Referencing McMillan and Wergin (2010) assertion on the commonality of mixed method designs having one dominate approach, determined through a series of questions. The conclusion is that Oreck (20014) has a dominate quantitative focus. Therefore, to determine whether it contributes significantly to the knowledge base, it must be judged on how well it meets the non-experimental quantitative evaluative criteria offered by McMillian and Wergin (2010). Particularly in its ability to provide connections and rationale to previous studies and illuminate gaps in the present knowledge base that this study will examine through mainly statistical means to discover associations amongst variables.

mixed linear model (Zhang et al. 2010) implemented in the GAPIT package (Lipka et al. 2012) in R. determines the trade-off between misclassifying training examples and minimizing the norm of the weights. Parameter controls the band of the insensitive zone that in turn affects the number of support vectors in building the regression function. Bigger means lesser support vectors and produces more ‘flat’ estimates. III.