CHAPTER II RESEARCH METHOD RESEARCH METHOD
2.1 Research Design
The present study employs a quantitative design to examine the patterns of relationship among the students‘ topic familiarity, writing performance and
critical thinking skills. Therefore, in terms of the analysis, this study employs a correlational design involving Path Analysis. Path Analysis examines the comparative strength of direct or indirect relationship among variables (Lleras, 2005). It is chosen as it has the ability to examine the underlying relations among many variables which can be used as a basis to argue for causal inference (Lomax
& Li, 2009).
Prior to the Path Analysis several conditions need to be made to fulfill the statistical assumption. The first assumption is that there is a one way causal flow in the system and any reciprocal causation between variables is ruled out. In addition, the relations among all the variables in the current study indicated that the patterns are linear, additive and causal. The residual is not correlated with the variables that precede in the model.
By identifying the patterns of relationship among the students‘ topic familiarity, writing performance and critical thinking skills, the study can result in a more comprehensive interpretation on the phenomenon on the dynamic of students‘ critical thinking skills as reflected in their argumentative writing. As the study investigates the students‘ critical thinking skills as reflected in their
argumentative writing and there is no effort made by the researcher to manipulate the characteristics of any variables involved, this study is also called an ex-post-facto research (Ary et al., 1995).
This study begins with the observation of the individual students‘
performance in terms of topic familiarity, writing performance and critical thinking skills. Accordingly, the unit of analysis is individual students. The description of the group of individuals in terms topic familiarity, writing performance and critical thinking skills can actually be aggregated and
manipulated from the description of the individual. The description of the unit analysis in this study is taken from the phenomenon occurring at a certain point of time, without any focus on long term effect. For that reason, based on the
dimension of time, this design of this study is a cross-sectional one (Olsen &
George, 2004). The assessments of topic familiarity, writing performance and critical thinking skills are done only at the certain point of time of the data collection when the present study is conducted. It means that during the
instrument administration there is no changes in subjects‘ characteristics would occur. Due to this assumption, this study does not intend to observe the
development of the students‘ topic familiarity, writing performance and critical
thinking skills over a period of time longer than that is needed for the data collection.
In line with the objective of the study, four different variables are
identified namely (1) topic familiarity of student initiated topic (TF-SIT); (2) topic familiarity of teacher initiated topic (TF-TIT); (3) writing performance in student initiated topic SIT); (4) writing performance in teacher initiated topic (WP-TIT); (5) critical thinking skills in student initiated topic (CTS-SIT); and (6) critical thinking skills in teacher initiated topic (CTS-TIT). Based on the theory explicated in the background, the first four variables modify the last two variables.
Therefore, the variables on topic familiarity and writing performance are classified as independent variables (X) while the variables on critical thinking skills belong to dependent variables (Y).
The relationship among variables is depicted in Figure 2.1 which
illustrates the hypothetical relations to be verified further through the data analysis in this study. The hypothetical relations are constructed based on the theoretical and empirical foundation explored in the background of the study. The path model presents the simultaneous relations between the endogenous variables (variables whose variation is explained by one or more variables within the model) and exogenous variables (variables whose variation is explained by factors outside the model and which also explains other variables within the model) so that the further analysis is done to examine the comparative strength of the direct and indirect relationships among variables (Lleras, 2005).
Furthermore, it should be noted that the path relating each variable represents the hypothetical relation which means that the variable contribute to
another one. There is not enough empirical evidence and theoretical background supporting the contributive relationship between topic familiarity in student initiated topic and topic familiarity in teacher initiated topic. Therefore, in the path model there is no link relating between the two variables. The similar reason also applies to the variable of writing performance in student initiated topic which is not related to writing performance in teacher initiated topic. The same case
happens to the variable of critical thinking skills in student initiated topic which is not related to critical thinking skills in teacher initiated topic.
Notes:
X1 : Topic familiarity of student initiated topic (exogenous variable) X2 : Topic familiarity of teacher initiated topic (exogenous variable) X3 : Writing performance in student initiated topic (endogenous variable) X4 : Writing performance in teacher initiated topic (endogenous variable) Y1 : Critical thinking skills in student initiated topic (endogenous variable) Y2 : Critical thinking skills in teacher initiated topic (endogenous variable) e : Residual or errors
Figure 2.1 Estimated Path Model
As shown in Figure 2.1, the direction of relationships is symbolized by the arrow. It means that the variable of critical thinking skills in student initiated topic (Y1) is related with topic familiarity of student initiated topic (X1); writing
performance of teacher initiated topic (X3) and the variable of critical thinking X1
skills in teacher initiated topic (Y2). While, the variable of critical thinking skills in teacher initiated topic (Y2) is related with the variable of critical thinking skills in student initiated topic (Y1); topic familiarity of teacher initiated topic (X2) and writing performance in teacher initiated topic (X4). The similar way of reading the direction of relationship also occurs for the other variables.
Based on the estimated path model in Figure 2.1, there are mathematical models to solve for the direct effect in which each endogenous variable is regressed on all the variables with direct paths leading to it. The path equations are as follow:
X3 = p3.1 X1 + e2 (1) X4 = p4.2 X2 + e3 (2) Y1 = py1.1 X1 + py1.3 X3 + e4 (3) Y2 = py2.2 X2 + py2.4 X4 + e5 (4)
Where p: the regression coefficient between the variables e: the error or the value of residual regression