Central tendency error

Central tendency error belongs to rating errors or to distortions of appraisal. According to Michael G. Aamodt central tendency error is "a type of rating error in which a rater consistently rates all employees in the middle of the scale, regardless of their actual levels of performance".

Categories of central tendency error

This type of error can be divided into three main categories:

• Regression towards the mean: This type of central tendency error occurs when an experimenter or surveyor is likely to select an average value, rather than a random value. This is often seen when researchers assign a score of 5 or 6 to an answer that was not selected.
• Central Tendency Bias: This type of central tendency error occurs when an experimenter or surveyor is likely to select a response that is close to the middle range, rather than a random response. For example, if a surveyor is asked to rate a product on a scale of 1-10, they may be more likely to select a score of 5 or 6, rather than a score of 1 or 10.
• Availability Bias: This type of central tendency error occurs when an experimenter or surveyor is likely to select an answer based on what is easily available to them, rather than selecting a truly random response. For example, if an experimenter is asked to choose a color for a product, they may be more likely to select a color that is more easily accessible, such as blue or red, rather than a more obscure color.

Example of Central tendency error

Central tendency error can be seen in a number of different scenarios, but one of the most common is when a surveyor is asked to rate a product on a scale of 1-10. If the surveyor is likely to select a response in the middle range, such as a 5 or 6, rather than a truly random response, then this could be an example of central tendency error. This type of error is also seen when an experimenter is asked to choose a color for a product, and is more likely to select a color that is easily accessible, such as blue or red, rather than a more obscure color.

Formula of Central tendency error

The formula for central tendency error is:

$CE={\frac {\sum _{i=1}^{n}|x_{i}-{\bar {x}}|}{n}}$

where CE is the central tendency error, x is the value of each response, and n is the number of responses. This formula calculates the mean absolute deviation of the responses from the mean value of the responses.

Reasons of occurrence

This error occurs when the evaluators are reluctant to use the extreme ends of the rating scale. Thus limiting the scope of evaluations and the possibility of objective and honest employee evaluation. Particularly susceptible to this error are five-degree scales, with exemplary extremes such as "outstanding" and "unacceptable".

The effect of using such a scale is to limit the possibility of evaluation and increasing the chance that all evaluated people will be in the mid-range of the rating scale. This makes it difficult to accurately describe the effectiveness at work when determining promotions and pay increases.

Some evaluators, instead of a lenient or strictness assessment, give an average score to all rated people, despite their actual performance. Raters think that giving an average rating is the best way to avoid any wrong judgements. Some assessors believe that performance evaluation is a waste of time, which is why the average rating for them is the simplest solution. The evaluators do not like being too strict towards anyone giving low marks. At the same time, they believe that nobody deserves the highest possible assessment.

Some assessment systems encourage evaluators to commit a central tendency error, require them to provide written justification when choosing an extreme assessment.

Steps of managing Central tendency error

The steps to correcting central tendency error are:

• Define the Error: The first step is to clearly define the error and what it looks like in the data. This will help to identify any potential sources of the error and provide a starting point for correcting it.
• Identify the Causes: The next step is to identify the potential causes of the error, such as question wording, survey design, or sampling methods. This will help to pinpoint the areas that need to be addressed.
• Address the Causes: Once the potential causes of the error have been identified, it is important to address them in order to reduce or eliminate the error. This may involve changing the question wording, the survey design, or the sampling methods.
• Check for Accuracy: Finally, it is important to check for accuracy to make sure that the error has been corrected. This may involve running additional tests or surveys to ensure that the data is accurate and that the error has been eliminated.

Effects of occurrence central tendency error

Assessment errors indicate a situation where the evaluator does not distinguish good and poor quality of work. These errors cause problems when evaluating by several people, if one evaluator is strict and the other will be assessed in the average scale of the assessment, finally it will be taken into account strict evaluation.

The average scores due to central tendecy error discriminate against employees who achieve high scores and protect those who have poor performance. As a result, the assessments become useless as a decision aid for promotions, training or feedback to the management.

Raters training

Some performance rating distortions may be corrected by rater trainings. Frank J. Landy and Jeffrey M. Conte distinguishes three types of training for raters:

• Administrative training
• Psychometric training
• Frame-of-reference training

Other rating errors

To the most common rating errors in the assessment employees belongs:

• Strictness error
• Leniency Error
• Halo Effect
• Recency of events error
• Similarity error
• Low appraiser motivation
• Inflationary pressures

Disadvantages of ratings errors

Central tendency error and other distortions have influence on company efficiency. It usually reduces efficiency by:

• Company cannot identify strengths and weaknesses of employees.
• Demotivating effect, for example when two employees doing same work with a different efficiency and they receive the same rating.
• Worse relationship between the manager and his subordinates.
• Excess or lack of the bonuses for effective work.
• Increase in renumeration costs for the company.
• Lack of consistency with low efficiency

Other approaches related to Central tendency error

There are several other approaches which can help to minimize the effects of central tendency error. These approaches include:

• Developing clear and consistent criteria for scoring responses: Developing clear and consistent criteria for scoring responses can help to reduce the chances of central tendency error. This can be done by providing clear instructions to the surveyors or experimenters, and by providing a detailed rubric that outlines what score should be given for each response.
• Using randomized response techniques: Randomized response techniques can be used to reduce the chances of central tendency error. This technique involves randomly selecting a response from a predetermined set of options. This technique can help to reduce the chances of central tendency error, as it eliminates the possibility of a surveyor or experimenter selecting a response based on their own biases.
• Using expanded response scales: Expanded response scales can also help to reduce the chances of central tendency error. This can be done by expanding the range of possible responses, or by adding additional response categories. This will help to provide a more accurate representation of the responses that are being collected.

In summary, there are several approaches which can help to minimize the effects of central tendency error. These approaches include developing clear and consistent criteria for scoring responses, using randomized response techniques, and using expanded response scales. These approaches can help to reduce the chances of central tendency error, and provide a more accurate representation of the responses that are being collected.

Footnotes

Central tendency error — recommended articles

Pareto chart — Leniency error — Quantitative risk analysis — Control chart — Quality inspection — P chart — Assignable cause — Experimental error — RATER model

References

• Aamodt M. G., (2015), Industrial/Organizational Psychology: An Applied Approach, Cengage Learning, Boston.
• Birkenbach C. X., (1984) Halo, Central Tendency, and Leniency in performance appraisel: A comparison between a graphic rating scale and a behaviourally based measure, "Perspectives on Industrial Psychology", South African Journal of Industrial Psychology.
• Catano V. M., Wiesner W. H., Hackett R. D., (2009), Recruitment and Selection in Canada, Cengage Learning, Ontario.
• DeCenzo D., Robbins S. P., Verhulst S. L., (2016),Fundamentals of Human Resource Management, Binder Ready Version, John Wiley & Sons.
• Jones J. W., Steffy B. D., Bray D. W., (1991), Applying Psychology in Business: The Handbook for Managers and Human Resource Professionals, Lexington Books, Douglas Weston.
• Knoff H. M., (2002), The Assessment of Child and Adolescent Personality, Guilford Press, New York.
• Kumar D. Bhattacharyya, (2011), Performance Management Systems and Strategies, Pearson Education India, New Delhi.
• Landy F. J., Conte J. M., (2010), Work in the 21st Century: An Introduction to Industrial and Organizational Psychology, John Wiley & Sons, Malden.
• Lunenburg F. C., (2012), Performance Appraisal: Methods and Rating Errors, "International Journal of Scholarly Academic Intellectual Diversity", Volume 14 Number 1, Sam Houston State University.
• Levi-Jakšić M., (2012), Proceedings of the XIII International Symposium SymOrg 2012: Innovative Management and Business Performance, University of Belgrade, Belgrade.
• Murphy K. R., Cleveland J. N., (1995), Understanding Performance Appraisal: Social, Organizational, and Goal-Based Perspectives, SAGE, Thousand Oaks.
• Thyer B., (2009), The Handbook of Social Work Research Methods, SAGE, Thousand Oaks.

Author: Fryderyk Olchawa