CORRELATION
• Correlations can be used to study relationship between two or more variables.
• It is a measure of association between two or more variables and this relationship is determined not only in terms of
direction, whether negative or positive but also in terms of its magnitude, whether high or low. However, it will not provide
information about any causal relationship between the variables.
• Sir Francis Galton’s contribution to development of correlation is noteworthy. He carried out studies on individual
differences and also studies on the influence of heredity. He studied the association between the height of parents and
that of their children with the help of bivariate distribution (that studies relationship between two variables) and found
that the parents who are tall have children who are also tall (Veeraraghavan and Shetgovekar, 2016). Further, in 1986, Karl
Pearson put forth mathematical procedure for correlation.
• Correlation can be categorized in to linear and nonlinear correlation.
• Linear Correlation: Linear correlation is denoted by a single straight line in a graph that denotes linear relationship
between given two variables. Such a graph indicates whether increase in one variable leads to increase in another
variable and vice versa, or decrease in one variable leads to increase in another variable and vice versa. For example,
if the scores on emotional intelligence increase or decrease, the scores on self esteem also increase or decrease
• Nonlinear correlation: As opposed to linear relationship, in nonlinear relationship. The relationship between two
given variables is not denoted by a straight line.
Direction and Magnitude of Correlation
• Correlation can be either positive or negative (though there could also be no relationship between the given two
variables). This can be described as direction of correlation
• Positive correlation: Positive correlation denotes that increase in one variable leads to increase in another variable
and decrease in one variable leads to decrease in another variable. For example, if the scores on emotional
intelligence obtained by adolescents increase, then the scores obtained by them on achievement motivation will
also increase or if the scores on emotional intelligence obtained by adolescents decrease then the scores obtained
by them on achievement motivation will also decrease. Positive correlation indicates that both the variables are
moving in same direction
• Negative Correlation: Negative correlation denotes that increase in one variable leads to decrease in another
variable or decrease in one variable leads to increase in another variable. For example, if the scores on
occupational stress obtained by employees increase, then the scores obtained by them on work motivation will
decrease or if the scores on occupational stress obtained by employees decrease, then the scores obtained by
them on work motivation will increase. In this case, the two variables are not moving in same direction.
• No Correlation or Zero Correlation: It may so happen that there is no relationship between the two variables. In
such a case the correlation will be zero (this will be further clear as we discuss the magnitude of correlation). Thus,
in this case the relationship is neither positive or negative. There are such variables where there might be no
relationship, for example, there may be no correlation between height of persons and years of their work
experience or there may exist no relationship between weight of persons and attitude towards environment.
Besides the direction of the correlation, it is also significant to understand the magnitude or strength of the
correlation. Magnitude is denoted by the degree of linearity of the correlation. Correlation between any two
variables is Coefficient of Correlation that is quantitatively represented. And the range for a coefficient of
correlation is between -1 to +1. Thus, coefficient of correlation can be obtained as 0.28 or -0.09 or 0.75 and so on.
The number will lie between -1 to +1 and the + and - signs denote the direction of correlation, whether it is
positive or negative
Coefficient of Correlation Range Interpretation
+ 1 or - 1 This can be interpreted as a
correlation that is perfect, though the
direction could be positive or negative
+ 0.91 to 0.99 Correlation is very high
+ 0.71 to 0.90 Correlation is high
+ 0.41 to 0.70 Correlation is moderate
+ 0.21 to 0.40 Correlation is low
0 to + 0.20 Correlation is negligible
0 No correlation
Properties of Correlation
1) The variables used in correlation are quantitative in nature.
2) The value of coefficient of correlation will range from -1 to +1 and correlation can be positive or negative or there can
also be zero correlation.
3) As we have discussed before, correlation will provide information about relationship between the variables, but it will
not denote whether a cause and effect relationship exists between the variables. For example, we may obtain a positive
correlation between organizational culture and job satisfaction of employees, but correlation will not denote if there is a
cause and effect relationship between these two variables.
4) It is not possible to make any predictions based on correlation. For example, there may be positive correlation
between rise in temperature and sale of ice cream or cold drinks but the sale of ice cream or cold drinks cannot be
predicted based on the temperature with the help of correlation.
5) Even if variables randomly vary, correlation can be used.
6) Sampling errors can have an effect on correlation
USES OF CORRELATION
1) Validity and reliability: Validity and reliability are important aspects of psychological testing and correlation can be used to
obtain validity and reliability of a psychological test. Validity is whether a test is measuring what it is supposed to measure
and reliability provides information about consistency of a test.
2) Verification of theory: Correlation can also be used to verify or test certain theories by denoting whether relationship exists
between the variables. For example, if a theory states that there is a relationship between parenting style and resilience, the
same can be tested by computing correlation for the two variables.
3) Putting variables in groups: Variables that show positive correlation with each other can be grouped together and variables
that show negative correlation can be grouped separately based on the coefficient of correlation obtained.
4) Computation of further statistical analysis: Based on the results obtained after computing correlation, various statistical
techniques can be used like regression. Further, correlation is also used for multivariate statistical analysis, especially for
techniques like Multivariate Analysis of Variance (MANOVA), Multivariate Analysis of Covariance (MANCOVA), Discriminant
Analysis, Factor analysis and so on (Mohanty and Misra, 2016).
5) Based on correlation, one can decide whether or not to determine prediction: By computing correlation, it is not possible to
predict one variable based on another variable, but based on the information that two or more variables are significantly
related to each other, further statistical techniques can be used to make predictions. For example, if we obtain a positive
correlation between family environment and adjustment of children, then further statistical techniques can be employed to
find if adjustment of children can be predicted based on family environment.
LIMITATIONS OF CORRELATION
1) As was stated earlier, correlation will not provide any information about cause and effect relationship or causation.
2) The coefficient of correlation, mainly, Pearson’s product moment correlation and Spearman’s rank order correlation
are suitable, when there is a linear relationship between the variables.
3) With regard to distributions that are discontinuous, the coefficient of correlation obtained may be overestimated or
higher.
4) Sample variations can have an effect on correlation (as is also true with other statistical techniques). 5) In case of
pooled sample, the correlation will be determined by relative position of the scores in X and Y dimensions or variables
CORRELATION COEFFICIENT
• The degree of association or the degree of relationship between two variables is measured quantitatively in the form of
an index which is termed as co-efficient of correlation. Co-efficient of correlation is a single number that tells us to what
extent the two variables are related and to what extent the variations in one variable changes with the variations in the
other. This coefficient of correlation is determined to find the relationships because it is simple to understand and
convenient to express. It is a constant.
• The measurement of correlation between two variables results in a maximum value that ranges from –1 to +1, through
zero. The ±1 values denote perfect coefficient of correlation.
• In case of ungrouped data of bivariate distribution, the following methods are used to compute the value of co-efficient
of correlation. 1. Rank Difference Co-efficient of Correlation or Spearman’s Rank Order Co-efficient of Correlation.
• Rank Difference Coefficient of Correlation is used only when the data is in ordinal scale and it is calculated with the
help of ranks. It may be defined as the correlation between the ranks assigned to the individuals in two characters.
This is the most widely used because of its easiness to compute. This is denoted by the symbol rho
• Pearson’s Product Moment Co-efficient of Correlation is determined when the data of variables consists of scores
paired in some meaningful way. There are several ways of calculating ‘r’ in such cases.
** for formula and numerical refer to your class notes.**
CHI SQUARE TEST
The Chi-square test is used for comparing experimentally obtained results with those to be expected. The important
application of Chi-square is as given below:
1) Chi-Square goodness of fit test
• It is used to find out how the observed value of a given phenomena is significantly different from the expected value.
• In ChiSquare goodness of fit test, the term goodness of fit is used in order to compare the observed sample distribution
with the expected probability distribution.
• Chi-Square goodness of fit test determines how well theoretical distribution (such as normal, binomial, or Poisson) fits
the empirical distribution.
• In Chi-Square goodness of fit test, sample data is divided into intervals. Then the numbers of points that fall into the
interval are compared, with the expected numbers of points in each interval.
• In regard to the procedure, set up the hypothesis for Chi-Square goodness of fit test, that is set upo both null and
alternative hypothesis. a) Null hypothesis: In Chi-Square goodness of fit test, the null hypothesis assumes that there is no
significant difference between the observed and the expected value. b) Alternative hypothesis: In Chi-Square goodness
of fit test, the alternative hypothesis assumes that there is a significant difference between the observed and the
expected value. Calculate chi-square using the formula and find out for the given degrees of freedom if chi-square value
is significant at .05 or .01 levels. If so reject the null hypothesis. If not significant accept the null hypothesis.
2) Testing Hypothesis of Equal Probability
• The Chi-square test is a useful method of comparing experimentally obtained results with those to be expected
theoretically on some hypothesis.
• The formula for calculating χ2 is
χ2 =∑[(fo-fe)2 ]/fe Where; fo = observed frequency of a phenomenon or even which the experimenter is
studying fe = expected frequency of the same phenomenon based on “no difference” or “null” hypotheses.
STEPS FOR CHI-SQUARE TESTING
1) First set a null of hypotheses.
2) Collect the data and find out observed frequency.
3) Find out the expected frequencies by adding all the observed frequencies divided by number of Categories (In
Ist example 240/3=80, in II example 100/5=20).
4) Find out the difference between observed frequencies and expected frequencies.(fo-fe)
5) Find out the square of the difference between observed and expected Frequency.
6) Divide it by expected frequency. Square of (fo-fe) /fe . You will get a quotient
7) Find out the sum of these quotients.
8) Determine the degree of freedom and find out the critical value of χ2 from table.
9) Compare the calculated and table value of χ2 and use the following decision rule.
Chi-square as a Test of Independence In addition to testing the agreement between observed frequencies and those
expected from some hypothesis, that is, equal probability and normal probability, chi-square may also be applied to test the
relationship between variables. In this we test whether two variables are dependent or independent to each other. The
following table is known as Contingency table. Contingency table is a double entry or two way table in which the possession
by a group of varying degree of the characteristics is represented. The same person for example may be classified as on the
basis of sex that is boys and girls or can be categoriesed in terms of the colour of their eyes i.e. brown or black etc. These
attributes are noted in terms of the gender factor.
** Refer to formula and table from your class notes