Introduction

Over the years, data quality has become a major concern for large companies
especially in the areas of customer relationship management (CRM), data
integration and regulation requirements.
Aside from the fact that poor data quality generates costs, it also affects the
customer satisfaction, company’s reputation and even the strategic decisions
of the management.

Define Data Quality

Data quality is the overall utility of a dataset(s) as a function of its ability to be

processed easily and analyzed for a database, data warehouse, or data
analytics system.

It is a perception of the data's appropriateness to serve its purpose in a given

context. Having quality data means that the data is useful and consistent.
Data cleansing can be done to raise the quality of available data (Rouse,
2005).

Define Data Quality

1.1.1. Data Accuracy using Lot Quality Assessment Sampling (LQAS)

The Lot Quality Assessment (LQAS) is a tool that allows the use of small
random samples to distinguish between different groups of data elements (or
Lots) with high and low data quality.
For health managers and supervisors, using small samples makes conducting
surveys or supervision more efficient. This technique has a wide application in
the industry for decades and has primarily been used for quality assurance of
products.
The concept and application of LQAS technique has been adopted in the
context of (District Health Information System) DHIS data quality assurance.
The adaptation was comprised of designating health facilities, DHIS monthly
reports, sections of monthly reports and group of data elements as ‘Lots’ to
provide representative samples for data quality assurance of DHIS.
Source: Dr. Sohail Amjad, Pakistan, LQAS—A useful cost-effective tool for
M&E and data quality assurance-Possible use for RBM1.1.2. Routine Data
Quality Assessment (RDQA)
The Routine Data Quality Assessment Tool (RDQA) is a simplified version of
the Data Quality Audit (DQA) which allows programs and projects to verify
and assess the quality of their reported data. It also aims to strengthen their
data management and reporting systems.
The objectives of RDQA are as follows:
Source: Page 8, RDQA User Manual, 2015

Source: Page 9, RDQA User Manual, 20151.1.3. Development

Implementation Plan
An Implementation Plan is a project management tool that shows how a
project will evolve at a high level. An implementation plan helps ensure that a
development team is working to deliver and complete tasks on time (Visual
Paradigm, 2009).
The development of an Implementation Plan is important to ensure that the
communication between those who are involved in the project will not
encounter any issues and work will also be delivered on time. The plan
validates the estimation and schedule of the project plan.
An Implementation Plan has the following key components:
Define Goals/Objectives:
Answers the question “What do you want to accomplish?”
Schedule Milestones:
Outline the high level schedule in the implementation phase.
Allocate Resources:
Determine whether you have sufficient resources, and decide how you will
procure what’s missing.
Designate Team Member Responsibilities:
Create a general team plan with overall roles that each team member will
play.
Define Metrics for Success:
How will you determine if you have achieved your goal? (Smartsheet, 2017).

Data Quality Tools

A Data quality tool analyzes information and identifies incomplete or incorrect

data. Cleansing such data follows after the completion of the profiling of data
concerns, which could range anywhere from removing abnormalities to
merging repeated information.
By maintaining data integrity, the process enhances the reliability of the
information being used by a business. Usually these data quality software
products can share features with master date management, data integration,
or big data solutions.
Gartner (2017) defines these Data quality tools used to address the data
quality problem:
As data quality becomes increasingly all-encompassing, currently, data
integration tools usually include data quality management functionality.

Application/Scope of Data Quality tools

The first generation of data quality tools was characterized by dedicated data
cleansing tools designed to address normalization and de-duplication.
But during the last 10 years, it was observed that there is generalization of
Extract, Transform, Load (ETL) tools which allowed the optimization of the
alimentation process.
Recently, these tools started to focus on Data Quality Management (DQM),
which generally integrate profiling, parsing, standardization, cleansing and
matching processes (Goasdue, Nugier, Duquennoy, and Laboisse, 2007).

Techniques in Root Cause Analysis

4.1.1. What is a Root Cause Analysis?
A root cause analysis is a class of problem solving methods aimed at
identifying the root causes of the problems or events instead of simply
addressing the obvious symptoms.
The aim is to improve the quality of the products by using systematic ways in
order to be effective (Bowen, 2011).
4.1.2 Techniques in Root Cause Analysis
Root cause analysis is among the core building blocks in the continuous
improvement efforts of the organization.
Keep in mind that root cause analysis alone will not produce any results. The
organization should embrace a relentless pursuit of improvement at every
level and in every department for this to work. The analysis will help develop
protocols and strategies to address underlying issues and reduce future
errors.
http://asq.org/learn-about-quality/root-cause-analysis/overview/roots-of-root-
cause.html
There are many techniques involved in a root cause analysis, and Bowen
(2011) describes them accordingly.
Ask Why 5 Times
This might sound like the technique of a five-year-old wanting to get
out of going to bed, but the five whys analysis can be quite useful
for getting to the underlying causes of a problem. By identifying the
problem, and then asking "why" five times - getting progressively
deeper into the problem, the root cause can be strategically
identified and tackled.
Failure Mode and Effects Analysis (FMEA)
The Failure Mode and Effects Analysis (FMEA) is a technique which
is aimed to find various modes for failure within a system. FMEA
requires several steps for execution:
All failure modes (the way in which an observed failure occurs) must
be determined.
How many times does a cause of failure occur?
What actions are implemented to prevent this cause from occurring
again?
Are the actions effective and efficient?
 FMEA is often performed and updated any time a new product or
process is generated, when changes are made to current
conditions, or to the design, when new regulations occur, or when
there is a problem determined through customer feedback.

Techniques in Root Cause Analysis

Pareto Analysis

Source: http://statisticalconcepts.blogspot.com/2010/03/pareto-analysis.html
The Pareto analysis operates using Pareto principle (20% of the
work creates 80% of the results. You will want to run Pareto analysis any time
when there are multiple potential causes to a problem. In order to perform a
Pareto analysis, you will want to create a Pareto chart using Excel or some
other program. To create a Pareto chart, you will list potential causes in a bar
graph across the bottom - from the most important cause on the left to the
least important cause on the right. Then, you will track the cumulative
percentage in a line graph to the top of the table. The causes reflected on the
table should account for at least eighty percent of those involved in the
problem.

Techniques in Root Cause Analysis

Fault Tree Analysis

Source:
http://www.govisualize.com/Visualizations/Forms/DispForm.aspx?ID=7
Fault Tree Analysis uses boolean logic to determine the root causes of an
undesirable event. This technique is usually used in risk analysis and safety
analysis. At the top of the fault tree, the undesirable result is listed. From this
event, all potential causes tree down from it. Each potential cause is listed on
the diagram in the shape of an upside down tree.
Current Reality Tree (CRT)
The current reality tree analyzes a system at once. It would be used when
many problems exist and you want to get to the root causes of all the
problems. The first step in creating a current reality tree is listing all of the
undesirables or, problems. Then begin a chart starting with each of those
problems using causal language (if...and...then). The tree will depict each
potential cause for a problem. Eventually, the tree will show one cause that is
linked to all four problems.
Fishbone or Ishikawa or Cause-and-Effect Diagrams

Source:http://www.conceptdraw.com/examples/cause-and-effect-analysis-
fishbone-diagram-management-pdf
No matter what term you use for the fishbone diagram, the truth is, that it is a
useful technique that will help you in your root cause analysis. A fishbone
diagram will group causes into categories including:

 People
 Measurements
 Methods
 Materials
 Environment
 Machines

Depending on the industry you are in, you may use different categories such
as The 4 M's (manufacturing), The 4 S's (service) or the 8 P's (also service).
The diagram gets its name due to the fact that it looks like a fishbone, with
categorized causes and their sub-causes visualized.
Kepner-Tregoe Technique
The Kepner-Tregoe technique, also known as rational process is intended to
break a problem down to its root cause. This process begins with an appraisal
of the situation - what are the priorities and orders for concerns for specific
issues? Next, the problem analysis is undertaken to get to the cause of
undesired events. Then, a decision analysis is tackled, outlining various
decisions that must be made. Finally, a potential problem analysis is made to
ensure that the actions decided upon in step three are sustainable.
RPR Problem Diagnosis
One final technique used in root cause analyses is the RPR Problem
diagnosis. RPR stands for "Rapid Problem Resolution" and it deals with
diagnosing the causes of recurrent problems.
This process has three phases:

1. Discover - team members gather data and analyze their findings

2. Investigate - a diagnostic plan is created and the root cause is identified
through careful analysis of the diagnostic data
3. Fix - the problem is fixed and monitored to ensure that the proper root
cause was identified.

Sustaining culture of information use

Choo, Bergeron, Detlor, and Heaton (2008) stress that information culture
affects the information use outcomes. The information culture is
determined by the following variables: mission, history, leadership,
employee traits, industry, national culture. It can also be shaped by
cognitive and epistemic expectations which is influenced by the way that
tasks are performed and decisions are made.
The result suggests that in order to have a sense of the information
attitudes and values, the managers should consider taking the information
pulse of their own organizations. The sets of identified behaviors and
values were able to account for significant proportions of the variance in
information use.
Thus, the management plays an important part in sustaining the culture of
information and should continuously work on maintaining and improving
the quality of data and information used in their daily operations.
 KEY POINTS TO REMEMBER

 Data quality is the overall utility of a dataset(s) as a function of its ability

to be processed easily and analyzed for a database, data warehouse, or
data analytics system.
 The Lot Quality Assessment (LQAS) is a tool that allows the use of
small random samples to distinguish between different groups of data
elements (or Lots) with high and low data quality.
 The Routine Data Quality Assessment Tool (RDQA) is a simplified
version of the Data Quality Audit (DQA) which allows programs and
projects to verify and assess the quality of their reported data.
 The development of an Implementation Plan is important to ensure that
the communication between those who are involved in the project will
not encounter any issues and work will also be delivered on time.
 A root cause analysis is a class of problem solving methods aimed at
identifying the root causes of the problems or events instead of simply
addressing the obvious symptoms.
 Techniques in Root cause analysis include: Ask why 5 times, Failure
Mode and Effects Analysis (FEMA), Pareto Analysis, Fault Tree
Analysis, Current Reality Tree (CRT), Fishbone or Ishikawa or Cause-
and-Effect Diagrams, Kepner-Tregoe Technique and RPR Problem
Diagnosis.