Call us Chat now

Analysis of Employee Performance -

Analysis of Employee Performance

Project Outline

This research was carried out to study the employee performance in organizations. It considers such issues as Customer Satisfaction Scores, Age and Experience of Employees, Accuracy Scores, and Employee Productivity. The impact of experience and age of employees is identified with the use of various statistical methods. This impact is studied regarding such factors as Accuracy, Productivity, and Customer Satisfaction.

Theoretical Framework

XYZ Corporation (Illinois, USA) hired an additional consultant who studied the impact of mentioned factors on productivity and other indicators of employees’ performance. XYZ Corporation is going to rebuild its strategy according to the results of the research. After that, the organization plans to hire new employees who will be able to show better performance.

Methodology and Design

The internal consultant used the identification of different performance factors, for various types of business in XYZ Corporation. All businesses were estimated by common performance characteristics:

  1. Customer Satisfaction
  2. Accuracy
  3. Productivity

Statistical methods were used to understand the impact of experience and age on these factors.

Sampling Methods

During the sampling process, a small amount of elements is selected from a bigger group. The total amount of elements that are studied is called Population. The sample represents a subgroup of the population that is studied at the moment. In other words, ‘m’ employees are chosen from among all the employees (‘M’) of the organization. The sampling method is used in such situations as:

  • Costs and time are limited;
  • The process requires certain destructive tests, like car crashes, or taste tests;
  • The population can hardly be captured.

Sampling cannot be used when the products, as well as events, are unique, and so not replicable. Several methods of sampling are also unacceptable: Cluster sampling, Stratified random sampling. Simple random sampling, and Systematic sampling. Such methods as Quota sampling, Convenience sampling, Snowball and Judgment sampling are acceptable. These methods are distinguished by their non-probability.

Systematic sampling includes selecting every mth element from the entire population. Simple random sampling provides equal chances to be selected for every element. Cluster sampling implies sampling samples once in a certain period of time. Stratified random sampling implies creating groups which serve as a basis for randomly selected units.

Non-probability methods differ from methods described above. Quota sampling pays attention to a certain type of characteristics. Convenience sampling is based on access and convenience. Snowball sampling is based on respondent’s opinion of others. Judgment sampling is based on a belief that all respondents meet certain characteristics.

In this research, the consultant organization conducted a study using a method of Simple random sampling. 75 employees were chosen randomly. The organization collected such data as their experience, Productivity scores, Customer Satisfaction scores, age, and Accuracy Scores.

All the employees were sorted into three groups determined by age: 40-50 years, 30-40 years, and 20-30 years. In turn, these groups were classified according to their experience (0-10, 10-20, and 20-30 years).

Data Analysis

There are different options for data analysis available for the consultant. These options represent the impact of age and experience factors on accuracy, productivity, and Customer Satisfaction.

Hypothesis testing methods are used for each option above. It helps understand whether there is any significant difference between different data sets, in order to represent distribution. Hypothesis testing detects the difference in average, the difference in variation (Continuous Data), and the difference in Proportion Defective (Discrete Data).

There are four steps of Hypothesis Analysis:

  1. Determine the appropriate Hypothesis test.
  2. State the Alternate Hypothesis Ha and Null Hypothesis H0.
  3. Calculate table value of test statistic against P-value and Test Statistics.
  4. Interpret results of Hypothesis Analysis (reject or accept Ho).

Generally, Hypothesis testing implies following mechanisms:

  • Ha – Alternative Hypothesis. It means a significant difference between various groups.
  • Ho – Null Hypothesis. It means that two groups show no difference.

If we are using this type of testing, some errors may occur. These are:

  • Type I Error – P (Reject Ho when Ho is true) = α
  • Type II Error – P (Accept Ho when Ho is false) = β

P is a statistical measure that represents the probability of α errors to occur. This value is binary. Usually, alpha risk equals about 5%. If this value is less than 0.05, it means that the Null hypothesis should be rejected. In this case, we accept the alternate hypothesis. Alpha is defined before we start the test. In case P is more than 0.05, Ho is considered true, and that means that there is no difference between groups. Ho is accepted.

There are also several types of such testing:

  1. 1-Sample t-test. It’s used if there is a Discrete X, and Normal Continuous Y. A population is compared to a certain standard.
  2. 2-Sample t-test. It’s used in the same situation with X and Y, but when we have to compare two different populations.
  3. ANOVA. It’s also used in comparing the means of more than two populations.
  4. Homogeneity of Variance: this method helps compare the variance of two populations or several populations.
  5. Mood’s Median Test. This method is used in case X is Discrete. And Y is Non-Normal Continuous. It also helps compare more than two populations.
  6. Simple Linear Regression illustrates how Y changes while X changes. Y represents output, and X represents input.
  7. In case we have Discrete X and Discrete Y, we can use Chi-Square Test of Independence. It illustrates how Y counts from several sub-groups which differ by X.

This analysis gave researchers an opportunity to see how an increase in age and experience lower the Productivity indications, though increasing levels of the Accuracy and Customer Satisfaction.

Leave a Reply

Your email address will not be published. Required fields are marked *

We accept:
Follow us: