Correlation Research Question

1.Potential correlation research question

Is the number of quizzes correlated to the performance of students in Secondary School X in English language studies?

The above question can be classified as Correlation Research, as it is a study on the relationship between 2 variables:

1. number of quizzes done by students; and
2. the performance of students in English language studies (as measured by the score in year-end final examination).

2. How could data for the correlated variables be collected?

Step 1: Measurement: Assume the score of final examination can measure a student's performance in English language studies.

Step 2: Collect as many samples as possible, to plot a graph with number of quizzes done by students, against the score of final examination.

Step 3: Computation of correlation coefficients: The 2 correlated variables [number of
Quizzes taken, score of final examination] are used to compute the correlation coefficient for each grade. The closer the correlation coefficient to +1, the higher the strength of the relationship between 2 variables.

3. Provide argument how answer to that question can contribute to improvements in some aspect of education/teaching and learning?

If positive correlation (>0.5) is found, school may consider to employ the use of a large number of quizzes to improve the performance of students in English language studies.

posted by Cherry

Correlational Research

Correlation Research

This video introduces measures of relationship as used in correlational research.

Correlation Research (by Ronda)

Correlation Research:
The purpose of Correlation research is to discover relationships between two or more variables.

Correlation is defined as the relationship between two or more paired variables. The most common way to pair variables is to administer the measures of these variables to the same group of people. The correlation coefficient indicates the strength of correlation. The coefficient can range from 1.00 (perfect positive correlation) to -1.00 (perfect negative correlation). A coefficient of 0.00 indicates no correlation. (Ravid, 2000).

Some correlation questions we can investigate are
What is the relationship between...
1. school attendance and grades in school?
2. hours spend each week doing homework and school grades?
3. number of children in a family and the number of bedrooms in the house?

Data can be collected by surveys, scores on various tests or rating scales, demographic information and judges or expert ratings. Correlations can be calculated by Excel and SPSS.

Correlation research process:
 1.Variables to be study are identified.
 2.Questions and or hypotheses are stated.
 3.A sample is selected. (a minimum of 30 is needed)
 4.Data are collected.
 5.Correlations are calculated.
 6.Results are reported.

Correlation between two measures obtained from the same group of people can be shown graphically by the use of a Scattergram. The scattergram is used to show the correlation between two variables. The direction of the points and the degree they cluster indicate the strength of the correlation and whether the correlation is positive or negative.

Positive correlation: an increase in one variable is associated with an increase in the other variable.
Negative correlation: an increase in one variable is associated with a decrease in the other variable.

Please try the scatterplot demo showing the directions and strength of various correlation coefficients (Created by Juha Puranen from Finland).

Correlation is used in several ways in some aspects of education: (1). Describing relationships between variables. (2). Prediction. (3). Assessment of reliability and validity.

1. Describing Relationships
Correlation provides useful information about the relationships among variables. For example, what is the relationship between certain conductor behaviors and the attitudes of ensemble members toward rehearsal?

2. Prediction
If a strong correlation exists between two variables, performance on one variable may be used to predict performance on the other.

3. Assessing Consistency
For example, the correlation between music festival judges' ratings would be a good indication of their consistency.

Furthermore, correlation research contributes to improvements in some aspect of education/teaching and learning:

1. To help us understand related events, conditions, and behaviors.
2. To make predictions of how one variable might predict another.
3. To examine the possible existence of causation.

Reference:
Ravid, R. (2000). Practical statistics for educators. (2nd ed). New York, NY.: University Press of America, Inc.

----------------------------------------------------------------------------------------------

Plan a Small Scale Correlational Study:
1. Objective of the Study: The impact of parent-school relations on academic achievement of primary five students
2. Participants: 150 primary five students
3. Hypothesis: The good parent-school relations exert a positive influence on childen's school outcomes.
4. Methodology: Questionnaire and test for students.

Correlation research: This research is to study whether there is a postitive relationship between parent-school relations and academic achievement of primary five students. The correlated variables are “Parent-school relations” and “Academic Achievement”.

Data Collection: 150 primary students will be assessed using a written test. Results of the academic achievement will be converted into scores by statistical analysis and correlation coefficient will be calculated. A positive correlation coefficient will mean that there is a positive relationship between parent-school relations and student's academic result. On the contrary, if the correlation coefficient is negative, the relationship of the two variables does not exist.

Significance: Public and professional belief that parent involvement is the important means of increasing positive educational outcomes for children.

Parents can involve five types of activities:
1.Help with homework, 2. educational activities, 3. classroom volunteering, 4. conference participation and 5. telephone calls with teachers.

Other reflections:
Predication Studies:
If two variables are highly related, scores on one variable can be used to predict scores on the other variable. Predication studies are used to facilitate decision making about individuals. If several predictor variables each correlate well with a criterion, then a prediction based on a combination of the variables will be more accurate than a prediction based on any one of them. (Gay, Mills, & Airasian, 2006)

Reference:
Gay, L. R., Mills, G. E., & Airasian, P. (2006). Educational Research: Competencies for Analysis and Applications. Upper Saddle River, N.J. : Pearson/Merrill Prentice Hall.
Ravid, R. (2000). Practical statistics for educators. (2nd ed). New York, NY.: University Press of America, Inc.

Posted by Ronda