Statistical methods have emerged as some of the most critical aspects of research work in a contemporary world. It is through statistics that researchers have been able to collect vital data that is subsequently analyzed before an appropriate inference was made. As a branch of mathematics, it has proven useful in communicating the findings made during studies together with supporting the hypotheses made (Dowdy, Wearden, & Chilko, 2011, p. 28). It is therefore appropriate for researchers to include suitable analyses methods to ensure that the research is holistic and tackles every angle of the problem in question. Various methods are implemented in conducting these studies, the common denominator being the design, planning execution of objectives, a collection of data, evaluation and finally making an inference. Statistical tests are a requirement if the measurement of both qualitative and quantitative variables is to be done with fidelity to a particular research method (Subong, 2005, p. 45). Understanding the appropriate statistical process of analysis is vital if parametric and non-parametric tests are to be fully appreciated as methods used in analyzing data. In this essay, I will present two research situations, their hypotheses and the best analytical approach to test the hypotheses.
Research Situation 1
The first research scenario involves a study into procrastination and the relationship of two contrasting perspectives. In carrying out an investigation, the researcher’s hypothesis seeks establish if active procrastination has any bearing on a student’s level of academic motivation and whether passive procrastination stands on its own as a distinct variable. The researcher seeks to establish the relationship that exists between active and passive procrastination, using the Self-Determination Theoretical framework as the basis of the study. In conducting the investigation, students from a local university will participate as subjects to provide the much-needed data. In addition to using a questionnaire, the researcher also decided to enlist the use of an Aitken Procrastination Inventorythat would come in handy when seeking to establish different procrastination levels. The best analytical approach in this scenario is multiple-regression as the study aims to develop the relationship that exists between two criterion variables. An advantage of using the various regression models lies in the fact that the researcher will be able to determine the relative influence of the predictor variables. By so doing, they will, therefore, succeed in pointing out anomalies that may arise or any outliers that are worth investigating (Jaccard & Turrisi, 2011, p. 78). On the flipside, the disadvantage of using this method lies in the fact that the results are dependent on the data provided, even when it is incomplete or falsified. The plethora of measures used will ultimately be the determining factor in the reliability of the final results.
Research Situation II
The second scenario involves a researcher interested in research mentoring and the role it plays in determining the productivity of doctoral graduates. Research mentoring is a relatively new strategy that includes experts guiding students on how they should conduct their case studies after completing graduate school. Of particular concern in this scenario is the fact that most of these graduates rarely do their independent research. The hypotheses in this study focus on the research training environment and whether it is a motivational factor in leading these doctoral students towards their research practice. All this would occur, even when it was evident that the students had sufficient knowledge of the Scientist-Practitioner model. The appropriate method of analysis, in this case, would be theANCOVA(analysis of covariance). It is fitting since the researcher can control all the covariates in a one-way analysis where there are independent variables. Strengths of using this method of analysis is that any of the covariates can be put to use in the event significant data is missed. In such a situation, the bias is significantly lowered allowing the researcher to heighten the sensitivity levels (Anova and Ancova, 2012, p. 70). Nonetheless, this system of analysis presents a computational labor challenge that makes it quite cumbersome. The research environment, therefore, would be a factor to consider when making a prediction of the results in this scenario.
In the second hypothesis you predict that those who receive mentoring will have a higher motivation. You then go on to suggest that the best way to analyze this example would be to use a ANCOVA. Recall that we only use ANCOVAs when we have categorical independent variables, one continuous dependent variable, and at least one covariate. In your example I was unclear what your different variables would be in your analysis. What is your independent variable and how many groups does it have? What about your dependent variable and your covariates?
- Cite all sources in APA format.
- Attach a Turnitin.com Report.