Interpretation of Results: ANOVA

ANOVA: Interpretation of Results Analysis of Variance Modern-day learning institutions make use of many course delivery methods to deliver content and assignments to students.

The choice of method, of course, depends on the tutor's and the student's preferences, as well as the availability of relevant facilities and learning equipment - but all the same, one question lingers; how effective are these methods in enhancing the learning process and instilling relevant skills in students? A college instructor conducted a study to assess the differences between three common course delivery methods -- online, face-to-face, and hybrid, and to consequently identify the most effective method of the three.

The methods were assessed on the basis of their abilities to instill statistical learning skills in students. Research Questions and Hypotheses The research question and hypotheses guiding the study have been stated below.

RQ1: How are the various course-delivery methods related in terms of being able to instill statistical learning skills in students? H0: there are no discernible differences in the different methods' abilities to instill statistical skills H1: there are significant differences in the different methods' abilities to instill statistical skills in students Variables Judging from the research question, 'course-delivery methods' is the independent variable, and 'statistical learning skills' the dependent variable. In other words, the study is interested in establishing how students' statistical skills are affected by the course-delivery technique used.

The independent variable, 'course-delivery methods' is categorized into three -- the online method, the face-to-face method, and the hybrid method. All the three sub-variables of the independent variable are discrete and nominal in nature. The dependent variable, 'statistical learning skills', on the other hand, is measured as a quantitative, interval variable through a statistical skills test (SST), a 100-item test that rates students' test scores on a scale of 1 to 100. Descriptive Statistics 24 students -- eight from each of the three programs were selected for analysis.

Upon completion of the course, the students were given a statistical skills test (SST) to assess their understanding of key statistical concepts learnt throughout the course. The SST contained 100 statistics-related questions, each worth 1 point. Students' statistical skills were measured on the basis of their scores on the SST test. The higher the average SST score for students subscribed to a particular program, the higher the program's perceived degree of effectiveness.

Guided by existing literature on course instructional techniques, the author speculated that the face-to-face program would yield higher mean scores than the other two programs. From the values of the mean, the median, and the mode, it is possible to predict the nature of the distribution for the quantitative dependent variable (statistical learning skills) in relation to each of the three independent variables.

A distribution is negatively-skewed if the median and the mode are greater than the mean; and positively-skewed if the mean is greater than the mode and median (Hippel, 2005; Stopher, 2012). The closer the three values are, the more normal (and the less skewed) the distribution is (Stopher, 2012).

Students' statistical skills score in relation to the first independent variable (online instruction) and the second independent variable (face-to-face instruction) yielded median and mean values of 83.0 and 83.13, and 85.0 and 86.38 respectively (both showing that the data is almost normally-distributed and approximately four students apiece are above-average). The -.310 value for skewness (mean = 83.25, median 84.0) in the case of the third independent.