In the data processing method, the first thing to do is to know the type of research used in thesis research. One of the simplest thesis research and can be used in educational research, for example, is to examine the application of learning models, namely the approach or method of processing qualitative descriptive data.
Research using this data research method aims to describe or describe the actual conditions of a situation.
The following are the characteristics of data processing methods in qualitative research:
- Researchers conducted intensive participation in the field.
- Make careful notes on events in the field and collect documentary evidence.
- Make detailed reflections in addition to documents in the field.
- Make a report in the form of a detailed description.
The data processing method with a qualitative approach in this study was used to get a clear and real picture of the events that appeared during the learning process. The event in question is the process of implementing the learning steps applied in the classroom and student learning outcomes as a result of the application of the learning model.
Therefore, the data processing method with a qualitative descriptive approach is used in this study because it aims to describe events in the field. Meanwhile, the design used will be discussed below.
For experimental research, the design used was One Group Pre-Test and Post-Test Group Design (Campbell and Stanley in Arikunto, 2002:78).
This design is presented below:
1. One Group Pre-Test and Post-Test Group Design
Pretest Subjects Posttest Treatment
R O1 X O2
Description:
R = Research Subject
X = Learning Model applied
O1 = Pre-test
O2 = Post-test
Before conducting educational research in the field , we must first determine the research instrument to be used and then we determine the data processing method that can be used.
2. Research Instruments
In research methods, research instruments are everything that is used to collect data. One of the research instruments that is often used in educational research is the comprehension test. The test was conducted to measure student learning outcomes, namely the cognitive aspect.
The learning outcomes test is carried out after the teacher applies the learning model. The form of the test used is a written test which is an objective type. Arikunto (2002:164) suggests that "objective tests are tests that can be examined objectively". The objective test used in this study is a multiple choice test.
After we determine the type of instrument to be used and compile the instrument, the next step is to validate the instrument before it is used in educational research.
3. Instrument Validation
In research methods, research instruments are tools to collect research data. A good instrument is an instrument that has a good level of difficulty, discriminating power, validity, and item reliability. The data obtained were then analyzed.
The item analysis in the study includes four things, namely the level of difficulty, discriminating power, validity and reliability of the items. Because according to Arikunto (1998) the analysis of this item is considered to be able to represent the making of a good instrument.
After the preparation of the instrument was completed, the instrument was tested first on a non-sample class. Furthermore, to obtain a really good instrument, an analysis of the test results of the instrument was carried out as follows:
a. Item Difficulty
Level The difficulty level aims to determine the level of difficulty of the item. The formula used is as follows:
Description:
P = difficulty index
B = number of students who answered the question correctly
Js = total number of students
The difficulty index is classified as follows:
If P is between 0.00 – 0.30 including difficult questions
If P is between 0 .31 – 0.70 including moderate questions
If P is between 0.71 – 1.00 including easy questions
b. Differentiating Power of Question Items
The calculation of the power of item difference is to measure the ability of the questions to distinguish smart children and children who are less intelligent with certain criteria.
The formula used to calculate the discriminatory power of the items is:
Description:
D = the differentiating power of the items
JA = the number of students in the upper group
JB = the number of students in the lower group
BA = the number of students in the upper group who answered the question correctly
BB = the number of students in the lower group who answered questions correctly
Classification of differentiating power is as follows:
Description
Distinctive power
0.00-0.20 Poor
0.20-0.40 Moderate
0.40 -0.70 Good
0.70-.00 Very good
(Source: Arikunto, 2003)
c. Item
Validity Validity is a measure used to indicate the level of validity or validity of an instrument (Arikunto, 2003:64). In this study, item validity was calculated based on the product moment correlation formula proposed by Pearson. Researchers use the help of SPSS 16 for Windows program.
Moment Product Correlation
Criteria Correlation number Meaning
0.800 < r 1.00 Very High
0.600 < r 0.800 High
0.400 < r 0.600 Enough
0.200 < r 0.400 Low
0.000 < r 0.200 Very Low
For the interpretation of the validity of the items, the following criteria are used:
Items with calculations > r the table at a significant level of 0.05 is valid. Items with r count < r table at a significant level of 0.05 are invalid.
d. Test Question Reliability
In the research method, test reliability shows consistency or determination in measuring ability. A good learning test when carried out repeatedly is expected to obtain the same learning results. If these requirements have been met, it can be said that the test has met the requirements for reliability and can be trusted.
The form of the objective test of the reliability test technique uses the Kuder Richardson 20 formula with the help of the SPSS 16 For Windows program.
Description:
r11 = test reliability
p = proportion of subjects who answered the item correctly
q = proportion of subjects who answered the item incorrectly
n = number of items
St2 = total test reliability criteria
Correlation Figures,variance of Meaning 0.8000,6000,4000,2000,000 ( Source: Arikunto, 2003)
After the instrument validation process is complete, the next step will enter the final stage, namely the method of processing educational research data.
4. Data Processing Methods
Before implementing the data processing method to test the hypothesis, the analysis requirements test is first carried out which includes normality test and linearity test.
a. Normality Test
In the research method, the normality test aims to determine the distribution of the data to be analyzed, which is normally distributed or not. Because the research conducted includes a large sample of more than 20 students, the normality test used is the X2 (chi-squared) test with the formula. X2 (Arikunto, 2003:286)
Information:
X2= chi-square
fo= frequency of observation
fh= frequency of count
Data normally distributed if X2 Count < X2 Table for significance level 0.05, and not normally distributed if X2 Count > X2 Table for significance level 0.05 (Triton, 2006).
b. Homogeneity
Test Linearity test is used to determine whether several population variants are the same or not. Testing on SPSS using the Test for Homogeneity with a significance level of 0.05. Two variables are said to be equal if the significance level value is more than 0.05.
Thus the description of how we do data processing methods in educational research. Hopefully the explanation of data processing methods can be useful in conducting research.