Data, Metrics, Reporting and Analytics: An analysis of Organizational performance

This paper analyzes the relationship between key performance measures using predictive analysis in Disability services organization. It explores the difference in employee performance across the four locational branches and the two divisional branches. The paper provides recommendations to enhance organizational performance based on the results of predictive analysis. The data used in this project contains quantitative information on 138 employees working in its 2 business divisions (Community outings and Home cares) from the four branches. The key measures used in this analysis are efficiency measures, effectiveness measures, and business outcome measures.

Data, Metrics, Reporting and Analytics
An analysis of Organizational
performance
Author:

Introduction
Background information
Performance of an organization is often thought to be connected to the kind of workers it
employs. To reinforce such as supposition, is the current need of the CEO to determine whether
there exists a difference in employee performance across the four locational branches and the
two divisional branches, which is part of an effort to improve organizational performance. With
the increasing orientation of the labor market toward knowledge and information which has
pushed the management of organizations to need of quality workers. Despite the high
productivity of star workers, “…they cannot constitute a sustained competitive advantage if their
skills are mobile and transferable across firms.”1 In his paper on the myth of talent and
performance portability2 notes that the mantra of “the people make the place”, has been prevalent
in many an organization hence acting as a drive force for choice of human resources.
Generally, organizational performance refers to the measure extent to which an
organization performs in terms of its underlying mission, vision as well as goals3. Therefore,
organizational performance can be split into performance measures and performance referent
where performance measures refer to the metrics employed in gauging organizations while
performance referents are used in assessment of how well the organization is doing.
Usage of a range of performance metrics and referents are key due to the value imported
from the depth and information offered on the organization performance.
1 Groysberg, Boris, Lee, Linda-Eling and Nanda Ashish. “Can They Take It with Them? The Portability of Star
Knowledge Workers' Performance.” Management science 54. No. 7 (2007): 217.
2 Viswesvaran, Vish. “Chasing Stars: The Myth of Talent and the Portability of Performance.” Human resource
management 50, no.3 (2010): 68.
3 Janice, Edwards, Organizational Performance: A Complex Concept (Canada: Press books, 2018), 19.

Project objectives
i. To determine the relationship between key performance measures using predictive
analysis in Disability services organization.
ii. Provide recommendations to the CEO on the issue of staffing policy in order to enhance
organizational performance given the results of predictive analysis
Description of data and key measures
Data
The data used in this project contains quantitative information on 138 employees working in
its 2 business divisions (Community outings and Home cares) from the four branches that is:
Community outings from Brighton, Denver, Eaton as well as Victoria and Home cares from
Brighton, Denver, Eaton, Victoria.
In addition, there are 12 descriptive variables which include: employee code, last name, first
name, Location, Division description for each employee, Gender, Employee status code,
Employee position, Year that the employee begun working, date of birth, Work experience, Year
of education.
Key measures
In an article on “Measuring Your Organization’s Performance”, the author reinstates that, there
is importance of performance measurement as a means of keeping track of the organizational
performance4. He further argues that performance measurement includes gauging of the actual
performance outcomes. Profit, productivity, sales and market share, customer services,
4 Hookana, Heli. “Measurement of Effectiveness, Efficiency and Quality
in Public Sector Services - Interventionist Empirical Investigations.” Managing sustainability 4, no.7. (2011): 491.

subjective estimates of financial performance are some of the measures of how an organization is
performing5. Such factors are then classified to performance measurements that is: efficiency,
efficiency and business outcomes.
Therefore, from the organization’s data, there are three key sets of HR process measures
that is, Efficiency measures, Effectiveness measures, and Business outcome measures.
Efficiency measures
Efficiency measures “…focus on cost and report the financial efficiency of human resources
operations”6
In broad terms, efficiency measures are the metrics used to examine the relationship between
production inputs and outputs, it is also viewed as the success rate of the conversion of inputs
into outputs. It is therefore the ability of the organization to implement its plans with minimal
resource expenditure. According to Porter’s Total productivity, an organization should seek to
remove the six losses which comprise:
 Reduced yields
 Process defects
 Reduced speed
 Idling and minor stoppages
 Set-up and adjustment
 Equipment failure
5 Moss, Simon. “Measures of organizational performance.” Academy of Management Journal 39, No.57. (2016): 19.
6 Fitz-enz. and Mattox John. Predictive Analytics for Human Resources. (Wiley, 2014), 214

Thus in measuring organizational efficiency, exploration of how well the inputs are
optimized is key. In analysis of organizational efficiency, factors such as the staffing process,
and focus on time to fill in, hiring cost and salary associated with positions will be analyzed.
Effectiveness measures
Efficiency measures are inclined towards successful input conversion outputs, while
effectiveness examines interaction of outputs with economic and social environment. An
organization’s effectiveness is therefore an examination of how the organization is performing in
both long term and short term targets. As such, analysis of Focus on target groups, beneficiaries,
clients that is, sponsor satisfaction score in the organization.
Therefore, effectiveness has an orientation towards output, sales, profits, cost reduction,
innovativeness etcetera. As a result, in analysis of the organization’s effectiveness factors such as
sponsor satisfaction, the staffing process, focus on speed to competency, and performance rating
are analyzed.
Business outcome measures
“…business performance measures are a set of quantifiable metrics taken from various
sources.”7 Consequently, business performance measures enable the executive to keep track of a
given business process that is being examined. Hence, in measuring the performance of the
business, profitability and worker engagement of the organization are explored.
7 Bartuševičienė, Ilona and Šakalytė, Evelina. “Organizational Assessment: Effectiveness vs. Efficiency.” Social
Transformations in Contemporary Society 1, no 12. (2013): 45.

Analysis of the relationship between key measures
Predictive analysis
The initial objective of the project is to determine if there is a difference in employee
performance a factor which is highly correlated with the performance of the organization
In order to explore the relationship between effectiveness, efficiency and business
outcome measures, the method of predictive analytics is used. To achieve successful analysis,
exploration of the two business divisions is done separately i.e. for their independent
organizational performance. Initially, effectiveness measure is measured through which
efficiency can then be measured. Now, regression and correlation analysis will be used to
determine the relationship between the following factors which are drawn from efficiency,
effectiveness and business performance:
i. Outcomes staffing process
ii. Focus on time to fill in
iii. Hiring cost and salary
iv. Sponsor satisfaction
v. Focus on speed to competency
vi. Performance rating
vii. Profitability and worker engagement
Correlation analysis
Used as preparatory for predictive linear regression models8 correlation analysis explores
the association between quantitative variables. For instance, in determination of the relationship
8 Michael, Stanleigh. “Measuring Your Organization’s Performance.” Business improvement architects, 12th June ,
2016, accessed December 1st 2018, https://bia.ca/measuring-your-organizations-performance/

between organizational performance, the executive might be interested in determining whether
there is a relationship between education level and engagement score.
Results of Correlation analysis
Time To Fill Salary Hiring CostPerformance90SpeedToCompetencySponsor Satisfaction90profitabilityProductivityEngagement
Time To Fill 1
Salary 0.128585 1
Hireing Cos0.135356 0.99997 1
Performan0.785156 -0.0413 -0.03625 1
SpeedToCo-0.75209 0.17241 0.168321 -0.80134 1
Sponsor Sa 0.40903 -0.20793 -0.20624 0.765889 -0.84484 1
profitabilit 0.488555 -0.09929 -0.09667 0.485126 -0.59442 0.607211 1
Productivit0.837613 -0.00287 0.002489 0.930204 -0.86532 0.780128 0.726481 1
Table 1:Correlation analysis
0 1 2 3 4 5 6 7 8 9 10
-1.5
-1
-0.5
0
0.5
1
1.5
Relationship between key measures
Time To Fill Salary Hiring Cost
Performance90 SpeedToCompetency Sponsor Satisfaction90
profitability Productivity Engagement
Figure 1: Scatter plot of correlation analysis

Relationship between efficiency, effectiveness and business outcome
Assumptions
From the previous section, organizational efficiency is assumed to be measured by time
to fill in, hiring cost and salary whereas effectiveness is assumed to be measured by sponsor
satisfaction, focus on speed to competency, and performance rating. Business outcome
performance is measured by profitability and worker engagement.
Interpretation
From table 2 and figure 1 above, there is a positive correlation between performance and
time taken to fill the position a worker is holding i.e. with correlation coefficient of 0.7851.
Other factors that indicate a strong positive correlation are:
 Hiring cost and salary- 0.9999
 Sponsor satisfaction and performance rating-0.7658
 Profitability and sponsor satisfaction- 0.6272
 Productivity and performance rating- 0.8376
 Worker engagement and profitability- 0.6705
 Productivity and profitability- 0.7264
 Productivity and sponsor satisfaction- 0.7801
 Worker engagement and performance rating- 0.9101
 Engagement and sponsor satisfaction- 0.8779
 Worker engagement and productivity- 0.9501

However, speed to competency has got a strong negative correlation with performance
rating, Sponsor satisfaction, productivity and engagement with a Pearson correlation of -0.8013,
-0.8448, -0.8653, and -9405 respectively.
Linear Regression
When exploring how a response variable related with predictor variables, linear
regression models are used as one of the methods of predictive analysis. In regression analysis,
examination of which combination of factors lead to optimum productivity among workers is
examined.
Predictive analytics using linear regression
Linear regression is used to examine the factors that influence a worker’s productivity.
Linear regression model:
Yi= β0 +β1X1+β2X2 +…+ βnXn + £I Where: Yi is the response variable, βi are the
coefficients of the explanatory variables Xi
and £i is the error term
Worker’s productivity model:
Productivity= β0 + β1 (Age)+ β2 (years of service) + β3 (Work experience) + β4 (Years of
education) + β4 (Salary) eqn 1
Hypotheses
At a confidence level of 95% the following two hypotheses are formulated:

Null
There is sufficient statistical evidence to indicate a relationship between productivity and
age, years of service, work experience, years of education and salary.
Alternative
There is no sufficient statistical evidence to indicate a relationship between productivity
and age, years of service, work experience, years of education and salary.
Regression Results
The r-squared statistic which is used to measure how good the model is has a value of
0.906047 when the salary variable is not included (table 3)
SUMMARY OUTPUT
Regression Statistics
Multiple R 0.951865
R Square 0.906047
Adjusted R0.780777
Standard E0.677023
Observatio 8
Table 2
While it is 0.92268 when the salary variable is included hence increases which indicates
that salary is relevant in predicting productivity.

SUMMARY OUTPUT
Regression Statistics
Multiple R 0.960562
R Square 0.92268
Adjusted R 0.72938
Standard E0.752213
Observatio 8
Table 3
SUMMARY OUTPUT
Regression Statistics
Multiple R 0.960562
R Square 0.92268
Table 4
SUMMARY OUTPUT
Regression Statistics
Multiple R 0.960562
R Square 0.92268
Adjusted R 0.72938
Standard E0.752213
Table 5
When taking the explanatory variables together, the P-value for the Fisher’s statistic is
0.182236 which is greater than 0.05 indicating that the model cannot be used in predicting
productivity. However, when using years of experience and education as explanatory variables,
the following regression output is obtained:

SUMMARY OUTPUT
Regression Statistics
Multiple R 0.943418
R Square 0.890037
Adjusted R0.846052
Standard E0.567346
Observatio 8
ANOVA
df SS MS F Significance F
Regression 2 13.02649 6.513246 20.23492 0.00401
Residual 5 1.609408 0.321882
Total 7 14.6359
CoefficientsStandard Error t Stat P-value Lower 95%Upper 95%Lower 99.0%Upper 99.0%
Intercept 11.57801 4.78536 2.419465 0.060157 -0.72315 23.87917 -7.71724 30.87327
Work Exper0.305069 0.098026 3.112128 0.026487 0.053086 0.557053 -0.09019 0.700324
Table 6
4.00
5.00
6.00
7.00
8.00
9.00
10.00
11.00
12.00
13.00
14.00
-1
-0.5
0
0.5
1
Work Experience (years)
Residual Plot
Work Experience (years)
Residuals
Figure 2

13.00 13.50 14.00 14.50 15.00 15.50
-1
-0.5
0
0.5
1
Years OF education Residual
Plot
Years OF education
Residuals
Figure 3
The p-value of the F-statistic is 0.00401 indicating that the two explanatory variables are
suitable for use in the model. The p-value of Work experience is 0.026487< 0.05 at 95%
confidence level indicating that work experience is significant when measuring a worker’s
productivity. Elsewhere, the p-value of Years of education is 0.023799< 0.05 at 95% confidence
level, implying that years of education is significant in predicting productivity.
From table 7, the regression coefficient of the regression model is 11.57801 while that of
work experience and years of education is 0.305069 and 1.174513 respectively which gives the
final regression model as:
Productivity= β0 + β1 (Work experience) + β2 (Years of education)
Since the other variables are not significant in predicting productivity hence:
Productivity= 11.57801+ 0.305069(Work experience) + 1.174513(Years of education)
That is, for every unit increase in years of education there is an increase of 12.7525 in worker
productivity. While for every unit increase in work experience there is an increase of 11.8830
worker productivity.

Lastly, it can be inferred that work experience and years of education are the only
variables that are statistically significant in predicting productivity. Hence, after modification of
the null and alternative hypothesis, we fail to reject the null hypothesis and conclude that there is
sufficient statistical evidence to conclude that there is a relationship between productivity, work
experience and years of education. Recommendations on using the significant variable in staffing
policy are given in the succeeding section.

Recommendations on staffing policy
From the predictive analytics on the factors that influence worker productivity in the previous
section, the following recommendations on staffing policy regarding employee recruiting,
promotion, salary increment considerations are made:
Employment
From the preceding section of analysis, years of education are directly correlated with
employee productivity that is, the higher the level of education the higher the productivity of the
worker in their respective fields. For instance, if the executive employs a person with an
educational level of 14 years, their productivity is projected to reach up to 28.02. In contrast, a
person with educational level of 17 years will have a productivity rate of 31.544731.
The challenge is therefore on how to attract skilled workers given that the assumption of high
education reflects skill is made. To ensure the company gets quality talent workers, it should
embrace measures such as:
i. Value added benefits for employees, i.e. offering competitive salaries to their
employees, both new and existing.
ii. Put forward well-defined criteria for worker promotion. A move that will offer
promise of personal growth to employees and hence motivate new employees to
consider the organization as well as the existing to be more productive.
Attraction and retention of high skilled workers will ultimately ensure improved long run
organizational performance.

Worker promotion
Years of work experience are also significant in measuring productivity. Years of
experience often are accumulative of how long the employee has been on the given work post.
Therefore, in case the executive assumes that years of experience is the only parameter set to
determine whether a worker is promoted or not, it should first consider persons with relatively
higher level of experience. For instance, an employee with work experience of 15 years has an
estimated productivity of 16.15 while an employee with 25 years of experience has a 19.20
productivity rate.
Consequently, persons with more years of experience in any given field should be given a
priority when the human resource department is recruiting new employees as well as when
considering giving the current employees a promotion.
Worker motivation
Assuming that worker engagement is linearly related to the worker’s productivity, incentives
such as health benefits, insurance, etcetera, should be adopted by the executive in a move to
ensure worker motivation. Since there is a strong positive correlation between salary and worker
engagement, where a salary increment positively affects the engagement level of a worker, the
executive should consider revising employee salaries. Such a move, will ensure an increase in
worker motivation which will in turn increase engagement score and hence productivity which
has a positive correlation with engagement. That is an increase in worker engagement positively
affects productivity which ultimately improves the overall organizational performance.
In order to ensure that there is reflection of salary increment in the productivity of workers,
an algorithm to determining the rate of salary increment should be implemented for instance,
hours put into work, level of education, level of experience, etc.

Equality and production uniformity
Often, workers who feel excluded offer little input in the production process. Measures
such as equal involvement in the production chain and decision-making process are prone to
reinforce the feeling of belonging and equality among workers and hence foster loyalty, and
worker engagement. Hence, the executive should consider setting up structure to ensure total
worker involvement such as the aforementioned involvement in decision-making.
Moreover, in order to ensure uniformity in performance across the two business divisions
of the organization, the executive should lay emphasis on uniform distribution of human
resource. For instance, it should employ equally skilled employees whose pay is relatively the
same given similar job fields so as to promote equality in worker productivity. Moreover, the
executive should come up with worker motivation programs such as overtime payments, a move
that will promote engagement hence productivity.
Conclusion
Organizational performance is a complex concept. From effectiveness through efficiency
to business performance, an organization which is keen to improve its sustainability index should
be able to put in place measures with which to increase its effectiveness, effiency as well as
business performance. However, when examining the organization’s performance there are a
number of factors which need to be considered given that various metrics can fall into either
groups that is either in efficiency key measure or effectiveness key measure since there is a “thin
line” between the two. For instance, it is crucial to measure effectiveness of an organization’s
performance before its efficiency since an ineffective organization cannot be efficient whereas an
effective organization can be inefficient that is it uses lots of inputs with little output.

Therefore, from the project results, it can be concluded that different factors are
determinant of employee performance and hence confirm the CEO’s concerns that there is a
difference in the worker performance across the two divisional branches in the 4 locational
branches regions.
Furthermore, it can be inferred that there exists a relationship between the various key
measures of organizational performance. Hence, the overall organizational performance can only
be improved through simultaneous improvement of the three key measures i.e. effectiveness,
efficiency and business outcome performance.

References
Bartuševičienė, Ilona and Šakalytė, Evelina. “Organizational Assessment: Effectiveness vs.
Efficiency.” Social Transformations in Contemporary Society 1, no 12. (2013): 45.
Accessed December 1st, 2018. http://stics.mruni.eu/wp-content/uploads/2013/06/45-
53.pdf
Edwards, Janice. Organizational Performance: A Complex Concept. Canada: Press books, 2018.
Fitz-enz. and Mattox John. Predictive Analytics for Human Resources. Wiley, 2014.
Groysberg, Boris, Lee, Linda-Eling and Nanda Ashish. “Can They Take It with Them? The
Portability of Star Knowledge Workers' Performance.” Management science 54. No. 7
(2007): 213-219. Accessed December 01, 2018. December 01, 2018. DOI:
10.1287/mnsc.1070.0809
Hookana, Heli. “Measurement of Effectiveness, Efficiency and Quality
in Public Sector Services - Interventionist Empirical Investigations.” Managing
sustainability 4, no.7. (2011): 491. Accessed December 1st 2018.
www.pdfs.semanticscholar.org/05f9/4755aa45b71dffcd686fe6a48173b2ba28ce.pdf
Moss, Simon. “Measures of organizational performance.” Academy of Management Journal 39,
No.57. (2016): 19. Accessed December 1st, 2018. DOI: 10.11070.0342
Michael, Stanleigh. “Measuring Your Organization’s Performance.” Business improvement
architects, 12th June , 2016, accessed December 1st 2018, https://bia.ca/measuring-your-
organizations-performance/

Viswesvaran, Vish. “Chasing Stars: The Myth of Talent and the Portability of Performance.”
Human resource management 50, no.3 (2010): 67-74. Accessed December 01, 2018. DOI:
10.1002/hrm.20427
Appendices
Data variables:
i. Time to Fill- Days spent to fill the employee's position after advertisement
ii. Salary- An employee’s annual salary
iii. Hiring Cost- Annual salary plus administrative costs of hire
iv. Performance 90- Performance appraisal scores 90 days after joining the organization. Ranked from 1 to 9.
Where, a high score indicates better performance.
v. Speed to Competency- Number of days before reaching competency after joining the organization
vi. Sponsor Satisfaction 90- Sponsor satisfaction score ranges from 1 to 5. Where, a high score indicates a
higher level of sponsor satisfaction.
vii. High Potential- A dichotomous variable. 1 = 8 & 9 for Performance 90, and zero otherwise. It is a measure
of an employee’s potential to improve performance.
viii. Assessment Score 90- Test score 90 days after joining the organization. Assessment score ranges from 1 to
5. With a higher score indicating better assessment result.
ix. Employee engagement- Engagement score ranges from 1 to 5. Whereby, a higher score indicates a higher
level of employee engagement.
x. Productivity- employee productivity is assumed to be directly related with billable hours, hence it is an
employee's average number of billable hours per week
xi. Bradford Factor- the Bradford factor is a methodology with which to relatively measure an employee’s
unplanned absenteeism. It is obtained by using the formula
B=S * S * D; Where, B=Bradford factor score,

S= total number of occasions when the employee
has been absent
D= total number of days the employee has been
absent
xii. Profitability- Average revenue minus the employee's costs including salary costs (hypothesized)
From the data tabulation (Table 1), there are a total of 45 male employees and 93 female employees in the
organization.
Location Division Description
Brighton Community Outings 30 12 18
Denver Community Outings 23 9 14
Eaton Community Outings 19 5 14
Victoria Community Outings 33 13 20
Brighton Home Cares 5 0 5
Denver Home Cares 8 3 5
Eaton Home Cares 14 2 12
Victoria Home Cares 6 1 5
138 45 93
Total no of
Employees
Total no of
Male
Employees
Total no of
female
Employees
Table 7
eqn 1 Assumes that the average of the explanatory variables is a representative of the other
variable groups i.e. average age is representative of the other age groups in the dataset.