Actuarial Outpost
Go Back   Actuarial Outpost > Actuarial Discussion Forum > Software & Technology
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions

Thread Tools Search this Thread Display Modes
Old 06-23-2019, 11:11 AM
campbell's Avatar
campbell campbell is offline
Mary Pat Campbell
Join Date: Nov 2003
Location: NY
Studying for duolingo and coursera
Favorite beer: Murphy's Irish Stout
Posts: 91,001
Blog Entries: 6
Default Human data errors: categorization

Human data errors

Data quality problems all, at their root, involve some form of human error. Whilst this is easy to say, it is perhaps harder to identify and resolve the causes of these human errors.

In this blog post, I will explore ways to categorise human data errors and propose strategies to reduce the severity of these errors.

Much of this article is based on the work of J Rasmussen (see bottom) but has been adapted for situations relating to data.

ISO 8000-61 (and -150) define approaches to data quality management and includes an activity ‘Root Cause Analysis & Solution Development’. The purpose of this activity is to prevent data errors being repeated, therefore, you need to determine the root cause(s) and then instigate actions to prevent them happening again.

Some of the different types of human data error are illustrated in the diagram below: [above]

Types of human data error
Most people come to work intending to do a good job, so we will leave consideration of deliberate errors/actions to the end of this post.

The four types of unintentional error are:

Slips – This occur when a lack of care/diligence means that data is entered in the wrong data field, for example, or the wrong button is pressed on an on-screen form – the correct action is visible, but it is not taken

Lapses – Forgetting to carry out a step in a common procedure where there are no visual clues as to what is required. For example, in a call-centre forgetting to inform a caller that they will be asked to take part in a feedback survey at the end of the call

Rule based mistakes – These tend to occur in unfamiliar situations, such as, failures in decision-making or errors in judgement. For example, selecting an incorrect job code from a long list of possible job codes, or deciding the wrong next process step for an unusual situation

Knowledge based mistakes – These occur in familiar situations where actions are taken based on incorrect knowledge. For example, an employee may have been recruited without being given formal training on a process/system and has relied on incorrect advice from a colleague

To prevent, or at least minimise, such errors, the following steps should be considered:

Consider software/ forms from a users perspective – are they clear and easy to interpret
Check that field names and pop-up prompts are clear and correct
Does the order of fields on-screen support data entry?
If two or more columns of fields are present, is it clear whether the fields are entered row-by-row or one column at a time
If a form is complex, consider the use of more formalised UX (User Experience) approaches to optimise form design

Are business processes clear and easy to follow? Are simple process diagrams available to staff?
Are on-screen prompts present to indicate the next steps in a process?
If there is a situation which does not fit the ‘standard’ process, is there a clear and easy route to find information on the correct approach to take?

Rule based mistake
Is there easy guidance on the correct process to follow, such as a process diagram?
Is there easy access to advice for situations that do not fit the ‘normal’ process?
If there are many options that could be selected to classify a job, is the list of possible codes ordered logically and with clear, unambiguous descriptions?

Knowledge based mistake
Have all staff been trained in the correct process to follow and the way that data and systems supports the process?
Is there readily available supporting documents and diagrams to act as an aide-memoire?
Are competency checks used to check staff understanding of the correct approach?
Are there checks on the quality of data and decisions made by staff with timely reminders where errors may have been made?

For deliberate errors, there is a whole range of different possible motivations and actions to consider:

‘I know better’ – staff finding short-cuts or not following all process steps in the mistaken belief that an action may not be needed
‘Pushing the boundaries’ – providing less and less data to reduce the time spent on an activity until a manager/ supervisor notices and corrects them
Malicious acts – deliberate deviations from the process, not for personal gain, but to cause problems for a manager, colleague or other team
Fraudulent acts – deliberate deviations from the process for personal gain, such as ‘creative’ completion of timesheets and expenses forms
The above examples are likely to be indications of an employee culture that is not in the best interests of the organisation. Care should be taken to investigate such situations to determine severity and extent and then to agree, with relevant individuals, the most appropriate remedial action to take.

This blog post was inspired by an article on the Maintenance and Engineering web site. That article was in turn inspired by:

Rasmussen, J. Human errors: a taxonomy for describing human malfunction in industrial situations, J. Occupational Accidents, pp. 311 – 333, Vol. 4, 1982.


LinkedIn Profile
Reply With Quote

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

All times are GMT -4. The time now is 05:14 PM.

Powered by vBulletin®
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.
*PLEASE NOTE: Posts are not checked for accuracy, and do not
represent the views of the Actuarial Outpost or its sponsors.
Page generated in 0.15216 seconds with 11 queries