With the consistently advancing and broadening business challenges, the approach with the management of human resources has additionally experienced a paradigm shift. The upper hand accomplished through innovation, technology, new items and data is transitory and immensely volatile. The main distinctive attribute among competitors which stays, are the abilities and commitment from the employees. The administrative headship accordingly has an essential impact as they straightforwardly impact the performance and the members of the company. A finely characterized and uniform competency system is the initial move towards a sortedout way to deal with the human asset management of the association.
There is a consistent need to build effectiveness and create and deliver value in every deal. Along these lines it is basic that a more scientific approach like competencies be utilized to characterize and comprehend the learning, aptitudes and state of mind required to play out a job efficaciously. Assessing an employee's performance in view of precharacterized competencies and their behavioral patterns, thusly is called competency based assessment (CBA).

Primary Data Collection
A questionnaire containing 50 situationbased questions was designed & was methodically surveyed across all the branches of the institute in 69 cities, 21 states.
We effectively characterized the competencies, rating scales and created customized evaluation simulations. The HR group of the institute alongside our team at 3EA successfully planned the venue and schedules. Three to four panels assisted & carried out the assessments, standardizations & consolidations of the evaluations. About 200 employees of the institute appeared for the competency based structured interview. Based on their answers, they were evaluated for six competencies viz. Emotional Intelligence Quotient, Ability to work under pressure, Entrepreneurship Quotient, Adaptability Quotient, Leadership Quotient & last but not the least Loyalty Quotient.

Tools
MS Excel was used for Data Understanding, Cleaning, Preparing & making Dashboards.
IBM SPSS Statistics 20 was used for Data Analysis
SQL Server Reporting Services (SSRS) was used to automate the reports of each employee.

Data Analysis
With the help of SPSS, I carried out the method of Correlation & Regression.
Correlation
Pearson's r & Scatterplot was computed to assess the relationship among the 6 variables. In this case, the two hypotheses were
H0: There is no correlation between the pair of parameters
H1: There is correlation between the pair of parameters

Linear Regression
Firstly, to check whether the assumptions of Linear Regression are met or not I ran a test for Multicollinearity & Heteroscedasticity
a)Test for Multicollinearity was done by checking Variance Inflation Factor (1/Tolerance).
b) Test Heteroskedasticity Glejser to check Heteroscedasticity
H0: There is no problem of Heteroscedasticity
H1: There is problem of Heteroscedasticity
After that I proceeded with Linear Regression

Independent t Test
a)Levene's Test for Equality of Variances between Males & Females
b)tTest for Equality of Means between Males & Females

ChiSquare Test of Association
a)Gender*EIQ
b)Gender * Ability to work under pressure
c)Gender * Entrepreneurship

Results from Data Analysis

Correlation
1.There was a strong correlation between EIQ & AWUP [r= 0.534, p=0.0001]
2.There was a strong correlation between AQ & AWUP [r= 0.523, p=0.0001]
3.There was a moderate correlation between AQ & Loyalty [r= 0.484, p=0.0001]
4.There was a weak correlation between EIQ & Entrepreneurship [r= 0.317, p=0.0001]
5.There was a weak to moderate correlation between EIQ & AQ [r= 0.367, p=0.0001]
The correlation between EIQ & AWUP was more than that between AQ & AWUP.

Linear Regression
a)Test for Multicollinearity
Since, the VIF<10 for all predictor or independent variables There was no multicollinearity.
b)Heteroskedasticity Glejser for Heteroscedasticity –
Since pValue for all independent variable > 0.05, the null hypothesis, H0 was accepted There was no problem of Heteroscedasticity.
Thus, I proceeded with the Linear regression
Linear Regression 1:
Dependent Variable Ability to Work Under Pressure
Predictor Variables/ Independent Variables Emotional Intelligence Quotient, Loyalty Quotient, Leadership Quotient, Emotional Intelligence Quotient, Entrepreneurship, Adaptability Quotient
Emotional Intelligence Quotient (EIQ) & Adaptability Quotient (AQ) are positively correlated with 'Ability to Work Under Pressure' (AWUP). Rest of the parameters, were not associated with 'Ability to work under pressure'.
H0: Predictors don't predict AWUP
H1: Predictors predict AWUP
Results
i.We rejected the null hypotheses H0 for EIQ & AQ as p Value 0.05 EIQ & AQ predict AWUP
ii.Unadjusted R Squared= 0.40 40% Variance in AWUP is explained by EIQ & AQ (overall measure of the strength of association)
iii.Regression Equation for AWUP:
AWUP=4.70+(0.41XEIQ)+(0.48XAQ)

For every unit increase in EIQ, a 0.41unit increase in AWUP is predicted, holding all other variables constant.
For every unit increase in AQ, a 0.49unit increase in AWUP is predicted, holding all other variables constant.
i.As Beta= 0.415 in EIQ Coefficients Table & Beta= 0.435 in AQ Coefficients, i. e. Beta weight of AQ> Beta weight of EIQ, relative importance of AQ is slightly higher than that of EIQ.
ii.Residuals approximate a normal curve
Linear Regression 2:
Dependent Variable Adaptability Quotient
Predictor Variables/ Independent Variables Emotional Intelligence Quotient, Loyalty Quotient, Leadership Quotient, Emotional Intelligence Quotient, Entrepreneurship, Ability to Work Under Pressure
'Ability to work under pressure' (AWUP) & Loyalty Quotient (LOQ) were positively correlated with 'Adaptability Quotient' (AQ). Rest of the parameters, were not associated with AQ
H0: Predictors don't predict AWUP
H1: Predictors predict AWUP
Results
i.We rejected the null hypotheses H0 for AWUP & LOQ as pValue 0.05 AWUP & LOQ predict AQ
ii.Unadjusted R Squared= 0.42 => 42% Variance in AQ is explained by AWUP & LOQ (overall measure of the strength of association)
iii.Regression Equation for AWUP:
AWUP = 2.85+(0.40XAWUP)+(0.32XLOq)

For every unit increase in AWUP, a 0.40unit increase in AQ is predicted, holding all other variables constant.
For every unit increase in LOQ, a 0.32unit increase in AQ is predicted, holding all other variables constant.
iv.As Beta= 0.421 in AWUP Coefficients Table & Beta= 0.378 in LOQ Coefficients, i. e. Beta weight of AWUP> Beta weight of LOQ, relative importance of AWUP is slightly higher than that of LOQ.
v.Residuals approximate a normal curve

Independent t Test
a) Levene's Test for Equality of Variances between Males & Females
H0: The variances of Male & Female are equal
H1: The variances of Male & Female are equal are not equal
As the p Values for all the parameters >0.05, Male & Female don't have equal variances.
b) tTest for Equality of Means between Males & Females
H0: There is no significance difference between Male & Female for Emotional IQ, Ability to Work Under Pressure, Entrepreneurship, Adaptability Quotient, Leadership Quotient & Loyalty Quotient
H1: There is significance difference between Male & Female for Emotional IQ, Ability to Work Under Pressure, Entrepreneurship, Adaptability Quotient, Leadership Quotient & Loyalty Quotient
As, pValues for all the parameters are > 0.05, there was no significant mean difference between Male & Female for all the 6 parameters.

ChiSquare Test of Association
Gender*EIQ
H0: There is no association between gender i.e. being male or female and EIQ
H1: There is an association between being male or female and EIQ
Since, pvalue= 0.925> 0.05 (d.f.= 8), the null hypothesis was accepted => There was no association between gender i.e. being male or female and EIQ.
Gender * Ability to work under pressure
H0: There is no association between gender i.e. being male or female and AWUP
H1: There is an association between gender i.e. being male or female and AWUP
Since, pvalue= 0.997 > 0.05 (d.f.= 9), the null hypothesis was accepted => There was no association between gender i.e. being male or female and AWUP.
Gender * Entrepreneurship
H0: There is no association between gender i.e. being male or female and EQ
H1: There is an association between being male or female and EQ
Since, pvalue= 0.885 > 0.05 (d.f.= 10), the null hypothesis was accepted => There was no association between gender i.e. being male or female and EQ.
Similarly, it was for Gender * Adaptability Quotient, Gender * Leadership Quotient & Gender * Loyalty Quotient