Scribd
Upload
20 views2 pages

The Concept of Correlation

There is a high positive correlation between training and productivity. The correlation coefficient (r) between training hours and productivity is 0.71, indicating a high positive correlation. When r is between 0.71-0.90, it is considered to be a high positive correlation. Approximately 50% of the variability in productivity can be explained by training hours based on the coefficient of determination (r^2).

Uploaded by

Sherlene Borres
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
20 views2 pages

The Concept of Correlation

There is a high positive correlation between training and productivity. The correlation coefficient (r) between training hours and productivity is 0.71, indicating a high positive correlation. When r is between 0.71-0.90, it is considered to be a high positive correlation. Approximately 50% of the variability in productivity can be explained by training hours based on the coefficient of determination (r^2).

Uploaded by

Sherlene Borres
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

CHAPTER 5 - SIMPLE CORRELATION ADVANCE STATISTICS

The Concept of Correlation The value of r most always falls between -1 and +1
inclusive. If r is close to 0, we conclude that there is no
In statistics, a relationship between two variables is
significant linear correlation between x and y, but, if r is
referred to as correlation. A correlation exists between
close to -1 and +1, we conclude that there is a significant
two variables when one of them is related to the other in
linear correlation between x and y. When there is no
some way. For example, given a sample of paired data
linear correlation between x and y, Table Vi, A lists values
such as students grades in English and in Mathematics.
that are critical in this sense. They separate the usual
Given such a sample, we want to determine whether
values of r from those that are unusual.
there is a relationship between these two variables, and
if such a relationship exists, we want to identify it with an
For example, from Table Vi, A with n = 8 pairs of sample
equation so that we can predict one variable by knowing
data and with α = 0.05, the critical value is 0.707, and
the other variable.
here’s how you will interpret this: with 8 pairs of data and
Assumptions: no linear correlation between x and y, there is 5% chance
that the absolute value of the computed r will exceed
1. The sample of paired (x, y) data is random sample. 0.707.
2. The pair of (x, y) data have a bivariate normal
distribution. Table 6.1 below is range of values of Pearson’s r.
The Pearson Product Moment Correlation Value of r Interpretation
± 1.00 Perfect positive (negative) correlation
The Pearson product moment correlation reveals the ± 0.91 to ± 0.99 Very high positive (negative) correlation
magnitude and direction of relationships. The Pearson’s r ± 0.71 to ± 0.90 High positive (negative) correlation
measures relationships in variables that are linearly ± 0.51 to ± 0.70 Moderately positive (negative)
related. Its value range from +1 through 0 to -1. The r correlation
symbolizes the coefficient’s estimate of linear association ± 0.31 to ± 0.50 Low positive (negative) correlation
based on sampling data. The formula for Pearson’s r is: ± 0.01 to ± 0.30 Negligible positive (negative) correlation
0.00 No correlation

These ranges of the correlation coefficient should be


treated only as guidelines. You will notice that the range
where:
limits overlaps. For example, a value of 0.9 is both very
x = observed data for the independent variable
high and high degree of correlation. Whether the
y = observed data for dependent variable
interpretation is very high or just high will depend on the
n = sample size
type and nature of data being studied.
r = degree of relationship between x and y
Coefficient of Determination is used to determine
how well the least square regression line fits the sample
data. It is very useful in assessing how much errors of
prediction of y can be reduced by using the information
provided by x.

How to compute the value of coefficient of


determination
1. Compute the coefficient of correlation and square its
value.
2. The value of coefficient of correlation ( r ) is from – 1
to 1, therefore the value of coefficient of determination
Interpreting Pearson’s r (r2 ) is between 0 to 1.
CHAPTER 5 - SIMPLE CORRELATION ADVANCE STATISTICS
The management oat R & B Construction Company is α = 0.05 df = n – 2 = 12 – 2 CV (0.05, n = 10) =
interested in determining the extent of the relationship 0.632 (Table VI) (Method 2)
between advertising and sales. Use the 5% level of 5. Test Statistics = t – test (Method 1) For Method 2- Test
significance & the two-tailed test. Statistics: r

Seatwork:
The production manager at XYZ Company is interested in
determining the nature of the relationship between
training and productivity. The following data were
collected over a one-quarter period on 10 employees:

Interpretation: There is a very high degree of positive


correlation between the advertising expenses and sales
of R & B Construction Company.

a) Determine the correlation coefficient between these


Interpretation: two variables and interpret the result.
Using the least square regression equation,
approximately 97.16% of the variability in y (dependent
variable) can be explained by x (independent variable). It
means that the least square regression line/equation
does a good job in predicting the sales of company given
the advertising expenses.

Ten Step Process:

1. Ho: P = 0 There is no correlation between advertising


and sales.
2. Ha: P ≠ 0 There is a correlation between advertising
and sales
3. Test Statistics: two-tailed test
4. Level of Significance: α = 0.05/2 = 0.025 df = n = 2 = 12
– 2 = 10 (Table 111)

You might also like