An Updated Version to TDQM-93-12

Submitted to Information Systems Research (ISR) for publication.








Data Consumers'

Perspectives Of Data Quality

May 1994 TDQM-94-01

Richard Y. Wang

Diane M. Strong

Lisa M. Guarascio

Total Data Quality Management (TDQM) Research Program

Room E53-320, Sloan School of Management

Massachusetts Institute of Technlogy

Cambridge, MA 02139 USA

Tel: 617-253-2656

Fax: 617-253-3321






© 1994 Richard Wang, Diane Strong, and Lisa Guarascio

Acknowledgments: Work reported herein has been supported, in part, by MIT's Total Data Quality Management (TDQM) Research Program, MIT's International Financial Services Research Center (IFSRC), Fujitsu Personal Systems, Inc., Bull-HN and Advanced Research Projects Agency and USAF/Rome Laboratory under USAF Contract, F30602­93­C­0160. The authors wish to thank Professors France LeClerc and Paul Berger for their input on the design and execution of this research. Thanks are also due to Professors Donald Ballou, Izak Benbasat, Yang Lee, Stuart Madnick, and Veda Storey for their careful reviews of earlier versions of this paper. We also thank Daphne Png for conducting the confirmatory experiment and summarizing its results.

Beyond Accuracy:

What Data Quality Means to Data Consumers

1. INTRODUCTION

Many databases are not error-free, and some contain a surprisingly large number of errors [3-5, 21, 28, 33]. A recent industry report, for example, notes that more than 60 percent of the surveyed firms (500 medium size corporations with annual sales of more than $20 million) have problems with data quality. Data quality problems, however, go beyond accuracy to include other aspects such completeness and accessibility. A big New York bank found that the data in its credit-risk management database was only 60% complete, necessitating double-checking by anyone using it. A major manufacturing company found that it could not access all sales data for a single customer because many different customer numbers were assigned to represent the same customer. In short, poor data quality can have substantial social and economic impacts.

To improve data quality, we need to understand what data quality means to data consumers (those who use data). The purpose of this research, therefore, is to develop a framework that captures the aspects of data quality that are important to data consumers.

1.1 Related Research

The concept of "fitness for use" is now widely adopted in the quality literature. It emphasizes the importance of taking a consumer viewpoint of quality because ultimately it is the consumer who will judge whether a product is fit for use or not [13, 15, 22, 23]. In this research, we also take the consumer viewpoint of "fitness for use" in conceptualizing the underlying aspects of data quality. Following this general quality literature, we define data quality as data that are fit for use by data consumers. In addition, we define a data quality dimension as a set of data quality attributes that represent a single aspect or construct of data quality.

Three approaches are used in the literature to study data quality: (1) intuitive, (2) theoretical, and (3) empirical. An intuitive approach is taken when the selection of data quality attributes for any particular study is based on the researchers' experience or intuitive understanding about what attributes are "important." Most data quality studies fall into this category. The cumulative effect of these studies is a small set of data quality attributes that are commonly selected. For example, many data quality studies include accuracy as either the only or one of several key dimensions [4, 28, 31] . In the accounting and auditing literature, reliability is a key attribute used in studying data quality [7, 11, 21, 24, 25, 45] .

In the information systems literature, information quality and user satisfaction are two major dimensions for evaluating the success of information systems [12] . These two dimensions generally include some data quality attributes, such as accuracy, timeliness, precision, reliability, currency, completeness, and relevancy [2, 20, 27] . Other attributes such as accessibility and interpretability are also used in the data quality literature [40, 41] .

Most of these studies identify multiple dimensions of data quality. Furthermore, although a hierarchical view of data quality is less common, it is reported in several studies [27, 33, 40] . None of these studies, however, empirically collect data quality attributes from data consumers.

A theoretical approach to data quality focuses on how data may become deficient during the data manufacturing process. Although theoretical approaches are often recommended, there are few research examples. One such study uses an ontological approach in which attributes of data quality are derived based on data deficiencies, which are defined as the inconsistencies between the view of a real-world system that can be inferred from a representing information system and the view that can be obtained by directly observing the real-world system [39] .

The advantage of using an intuitive approach is that each study can select the attributes most relevant to the particular goals of that study. The advantage of a theoretical approach is the potential to provide a comprehensive set of data quality attributes that are intrinsic to a data product. The problem with both of these approaches is that they focus on the product in terms of development characteristics instead of use characteristics. They fail to capture the voice of the consumer. Evaluations of theoretical approaches to defining product attributes as a basis for improving quality find that they are not an adequate basis for improving quality and are significantly worse than empirical approaches [38] .

To capture the data quality attributes that are important to data consumers, we take an empirical approach. An empirical approach to data quality analyzes data collected from data consumers to determine the characteristics they use to assess whether data are fit for use in their tasks. Therefore, these characteristics cannot be theoretically determined nor intuitively selected by researchers. The advantage of an empirical approach is that it captures the voice of customers. Furthermore, it may reveal characteristics that researchers have not considered as part of data quality. The disadvantage is that the correctness or completeness of the results cannot be proven based on fundamental principles.

1.2 Research Approach

We follow the methods developed in marketing research for determining the quality characteristics of products. Our approach implicitly assumes that data can be treated as a product. It is an appropriate approach because an information system can be viewed as a data manufacturing system acting on raw data input to produce output data or data products [1, 6, 16, 19, 34, 42] . While data consumers in most cases are not purchasing data, they are choosing to use or not use data in a variety of tasks.

Approaches for assessing product quality attributes that are important to consumers are well-established in the marketing discipline [8, 26] . Three tasks are suggested in identifying quality attributes of a product: (1) identifying consumer needs, (2) identifying the hierarchical structure of consumer needs, and (3) measuring the importance of each consumer need [17, 18] .

Following the marketing literature, this research identifies the attributes of data quality that are important to data consumers. We first collect data quality attributes from data consumers, and then collect importance ratings for these attributes and structure them into a hierarchical representation of data consumers' data quality needs. Our goal is to develop a comprehensive, hierarchical framework of data quality attributes that are important to data consumers.

Some researchers may doubt the validity of asking consumers about important quality attributes because of the well-known difficulties with evaluating users' satisfaction with information systems [29] . Importance ratings and user satisfaction, however, are two different constructs. Griffin & Hauser [17] , for example, demonstrate that determining attributes of importance to consumers, collecting importance ratings of these attributes, and measuring attribute values are valid characterizations of consumers' actions such as purchasing the product, but satisfaction ratings of these attributes are uncorrelated with consumer actions.

1.3 Research Method

We first developed two surveys that were used to collect data from data consumers (referred to as the two-stage survey later). The first survey produced a list of possible data quality attributes, attributes that came to mind when the data consumer thought about data quality. The second survey assessed the importance of these possible data quality attributes to data consumers. The importance ratings from the second survey were used in an exploratory factor analysis to yield an intermediate set of data quality dimensions that were important to data consumers.

Because the detailed surveys produced a comprehensive set of data quality attributes for input to factor analysis, a broad spectrum of intermediate data quality dimensions were revealed. We conducted a follow-up empirical study to group these intermediate data quality dimensions for the following reasons. First, it is probably not critical for evaluation purposes to consider so many quality dimensions [27] . Second, although these dimensions can be ranked by the importance ratings, the highest ranking dimensions may not capture the essential aspects of data quality. Third, the intermediate dimensions seem to form several families of factors. Grouping these intermediate data quality dimensions into families of factors is consistent with research in the marketing discipline. For example, Deshpande [14] grouped participation in decision making and hierarchy of authority together as a family, named centralization factors, and job codification and job specificity as a family, named formalization factors.

In grouping these intermediate dimensions into families, we used a preliminary conceptual framework developed from our experience with data consumers. This conceptual framework consisted of four "ideal" or target categories. Our intent was to evaluate the extent to which the intermediate dimensions matched these categories. Thus, our follow-up study moved beyond the purely exploratory nature of the two-stage survey to a more confirmatory study.

This follow-up study consisted of two phases (referred to as the two-phase study later). For the first phase, subjects were instructed to sort these dimensions into categories, and then label the categories. For the second phase, a different set of subjects were instructed to sort these dimensions into the categories revealed from the first phase to confirm these findings.

The key result of this research is a comprehensive framework of data quality from data consumers' perspectives. Such a framework serves as a foundation for improving the data quality dimensions that are important to data consumers. Our analysis is oriented toward the characteristics of the quality of data in use, in addition to the characteristics of the quality of data in production and storage; therefore, it extends the concept of data quality beyond the traditional development view. Our results have been used effectively in industry and government. Several Fortune 100 companies and the U.S. Navy [32] have used our framework to identify potential areas of data deficiencies, operationalize the measurements of these data deficiencies, and improve data quality along these measures.

1.4 Preliminary Conceptual Framework

Based on the limited relevant literature, the concept of fitness for use from the quality literature, and our experiences with data consumers, we propose a preliminary conceptual framework for data quality that includes the following aspects:

The data must be accessible to the data consumer. For example, the consumer knows how to retrieve the data.

The consumer must be able to interpret the data. For example, the data are not represented in a foreign language.

The data must be relevant to the consumer. For example, data are relevant and timely for use by the data consumer in the decision-making process.

The consumer must find the data accurate. For example, the data are correct, objective and come from reputable sources.

Although we hypothesize that any data quality framework that captures data consumers' perspectives of data quality will include the above aspects, we do not bias our initial data collection in the direction of our conceptualization. To be unbiased, we start with an exploratory approach that includes not only the attributes in our preliminary framework, but also the attributes mentioned in the literature. For example, our first questionnaire starts with some attributes (timeliness and availability) which are not part of this preliminary framework.

2. THE TWO-STAGE SURVEY

The purpose of this two-stage survey is to identify data quality dimensions perceived by data consumers. In the following, we summarize the method and key results of these surveys. The reader is referred to Wang, Strong, & Guarascio [43] for more detailed results.

2.1 Method

The method for the two-stage survey was as follows. For Stage one, we conducted a survey to generate a list of data quality attributes that capture data consumers' perspectives of data quality. For Stage two, we conducted a survey to collect data on the importance of each of these attributes to data consumers, and then performed an exploratory factor analysis on the importance data to develop an intermediate set of data quality dimensions.

2.1.1. The First Survey

The purpose of the first survey was to generate an extensive list of potential data quality attributes. Since the data quality dimensions resulting from factor analysis depend, to a large extent, on the list of attributes generated from the first survey, we decided that: (1) the subjects should be data consumers who have used data to make decisions in diverse contexts within organizations; and (2) we should be able to probe and question the subjects in order to fully understand their answers.

Subjects Two pools of subjects were selected. The first consisted of 25 data consumers currently working in industry. The second was MBA students at a large U.S. university. We selected 112 students who had work experience as data consumers. The average age of these students was over 30.

Survey Instrument The survey instrument (see Appendix A) included two sections for eliciting data quality attributes. The first section elicited respondents' first reaction to data quality by asking them to list those attributes that first came to mind when they thought of data quality (beyond the common attributes of timeliness, accuracy, availability, and interpretability). The second section provided further cues by listing 32 attributes beyond the four common ones to "spark" any additional attributes. These 32 attributes were obtained from data quality literature and discussions among data quality researchers.

Procedure For the selected MBA students, the survey was self-administered. For the subjects working in industry, the administration of the survey was followed by a discussion of the meanings of the attributes the subjects generated.

Results This process resulted in 179 attributes as shown in Figure 1.

2.1.2. The Second Survey

The purpose of the second survey was to collect data about the importance of quality attributes as perceived by data consumers. The results of the second survey were ratings of the importance of the data quality attributes. These importance ratings were the input for an exploratory factor analysis to consolidate these attributes into a set of data quality dimensions.

Subjects Since a sample consisting of a wide range of data consumers with different perspectives was necessary, we selected the alumni of the MBA program of a large university who reside in the U.S. These alumni consisted of individuals in a variety of industries, departments, and management levels who regularly used data to make decisions, thus satisfying the requirement for data consumers with diverse perspectives. From over 3200 alumni, we randomly selected 1500 subjects.

Survey Instrument The list of attributes shown in Figure 1 was used to develop the second survey questionnaire (see Appendix B). The questionnaire asked the respondent to rate the importance of each data quality attribute for their data on a scale from 1 to 9, where 1 was extremely important and 9 not important. The questionnaire was divided into four sections depending on the appropriate wording of the attributes, e.g., as stand-alone adjectives or as complete sentences. Since we did not include definitions with the attributes, it is possible that data consumers responding to the surveys could differently interpret the meanings of the attributes. Attributes that are not important or that are not consistently interpreted across data consumers will not show up as significant in the factor analysis.
Ability to be Joined With Ability to Download Ability to Identify Errors Ability to Upload
Acceptability Access by Competition Accessibility Accuracy
Adaptability Adequate Detail Adequate Volume Aestheticism
AgeAggregatability Alterability Amount of Data
Auditable Authority Availability Believability
Breadth of Data Brevity Certified Data Clarity
Clarity of Origin Clear Data Responsibility Compactness Compatibility
Competitive Edge Completeness Comprehensiveness Compressibility
Concise Conciseness Confidentiality Conformity
Consistency Content ContextContinuity
Convenience Correctness Corruption Cost
Cost of Accuracy Cost of Collection Creativity Critical
Current Customizability Data Hierarchy Data Improves Efficiency
Data Overload Definability Dependability Depth of Data
Detail Detailed Source Dispersed Distinguishable Updated Files
Dynamic Ease of Access Ease of Comparison Ease of Correlation
Ease of Data Exchange Ease of Maintenance Ease of Retrieval Ease of Understanding
Ease of Update Ease of Use Easy to Change Easy to Question
Efficiency Endurance Enlightening Ergonomic
Error-Free Expandability Expense Extendibility
Extensibility Extent Finalization Flawlessness
Flexibility Form of Presentation Format Integrity
Friendliness Generality Habit Historical Compatibility
Importance Inconsistencies Integration Integrity
Interactive Interesting Level of Abstraction Level of Standardization
Localized Logically Connected Manageability Manipulable
Measurable Medium Meets Requirements Minimality
Modularity Narrowly Defined No lost information Normality
Novelty Objectivity Optimality Orderliness
Origin Parsimony Partitionability Past Experience
Pedigree Personalized Pertinent Portability
Preciseness Precision Proprietary Nature Purpose
Quantity Rationality Redundancy Regularity of Format
Relevance Reliability Repetitive Reproducibility
Reputation Resolution of Graphics Responsibility Retrievability
Revealing Reviewability RigidityRobustness
Scope of Info Secrecy SecuritySelf-Correcting
Semantic Interpretation Semantics Size Source
Specificity Speed Stability Storage
Synchronization Time-independence Timeliness Traceable
Translatable Transportability Unambiguity Unbiased
Understandable Uniqueness Unorganized Up-to-Date
Usable Usefulness User Friendly Valid
Value Variability VarietyVerifiable
Volatility Well-Documented Well-Presented

Figure 1: Data Quality Attributes Generated from the First Survey

A pre-test of the questionnaire was administered to fifteen respondents: five industry executives, six professionals, two professors, and two MBA students. Minor changes were made in the format of the survey as a result of the pre-test. Based on the results from the pre-test, the final second survey questionnaire included 118 data-quality attributes (i.e., 118 items for factor analysis) to be rated for their importance, as shown in Appendix B.

Procedure This survey was mailed along with a cover letter explaining the nature of the study, the time to complete the survey (less than twenty minutes), and its criticality. Most of the alumni addresses were their home address. To assure a successful survey, we sent the survey questionnaires via first-class mail. We gave respondents a six week cut-off period to respond to the survey.

Response Rate Of the 1500 surveys mailed, sixteen were returned because they were undeliverable. Of the remaining 1484, 355 viable surveys (an effective response rate of 24 percent) were returned by the six week cut-off response date.

Missing Responses While none of the 118 attributes (items) had 355 responses, none had fewer than 329 responses. There did not appear to be any significant pattern to the missing responses.

Results Descriptive statistics of the 118 items (attributes) are presented in Appendix C. Most of the 118 items had a full range of values from 1 to 9, where 1 means extremely important and 9 not important. The exceptions were accuracy, reliability, level of detail, and easy identification of errors. Accuracy and reliability had the smallest range with values ranging from 1 to 7; level of detail and easy identification of errors ranged from 1 to 8. 99 of the 118 items (85%) had means less than or equal to 5; that is, most of the items surveyed were considered to be important data-quality attributes. Two items had means less than 2 and thus were overall the most important data-quality attributes: accuracy and correct, with means of 1.771 and 1.816 respectively. An exploratory factor analysis of the importance ratings produced twenty dimensions, as shown in Table 1. As mentioned earlier, more detailed results can be found in Wang, Strong, & Guarascio [43] .

Factor Interpretation Factor analysis is appropriate for this study because its primary application is to uncover an underlying data structure [10, 36]. An alternative method would be to ask subjects to group the attributes into common dimensions, as we did in the second phase of this research. Grouping tasks provide more assurance that the factors are interpretable. While both factor analysis and grouping tasks are used in the literature to uncover dimensions, grouping tasks become impractical when the number of items increases. Furthermore, factor analysis can uncover dimensions that are not obvious to researchers.

Table 1: Description of the Dimensions
Dim.
Name of Dimension (Attribute List)
Mean
S.D.
C.I.
Cronbach a
1
Believability
2.71
0.10
2.51-2.91
N/A
(Believable)
2
Value-added
2.83
0.09
2.65-3.01
0.70
(Data Give You a Competitive Edge, Data Add Value to Your Operations)
3
Relevancy
2.95
0.06
2.82-3.08
0.69
(Applicable, Relevant, Interesting, Usable)
4
Accuracy
3.05
0.10
2.86-3.24
0.87
(Data Are Certified Error-Free, Error Free, Accurate, Correct, Flawless, Reliable, Errors Can Be Easily Identified, The Integrity of the Data, Precise)
5
Interpretability
3.20
0.09
3.03-3.37
N/A
(Interpretable)
6
Ease of Understanding
3.22
0.07
3.07-3.37
0.79
(Easily Understood, Clear, Readable)
7
Accessibility
3.47
0.08
3.32-3.62
0.81
(Accessible, Retrievable, Speed of Access, Available, Up­To­Date)
8
Objectivity
3.58
0.09
3.40-3.76
0.73
(Unbiased, Objective)
9
Timeliness
3.64
0.11
3.43-3.85
N/A
(Age of Data)
10
Completeness
3.88
0.09
3.74-4.06
0.98
(The Breadth, Depth, and Scope of Information

Contained in the Data)

11
Traceability
3.97
0.09
3.79-4.14
0.79
(Well-Documented, Easily Traced, Verifiable)
12
Reputation
4.04
0.10
3.83-4.25
0.87
(The Reputation of the Data Source, The Reputation of the Data)
13
Representational Consistency
4.22
0.09
4.04-4.39
0.84
(Data Are Continuously Presented In Same Format, Consistently Represented, Consistently Formatted, Data Are Compatible with Previous Data)
14
Cost Effectiveness
4.25
0.10
4.05-4.44
0.85
(Cost of Data Accuracy, Cost of Data Collection, Cost Effective)
15
Ease of Operation
4.28
0.08
4.13-4.44
0.90
(Easily Joined, Easily Changed, Easily Updated, Easily Downloaded/Uploaded, Data Can be Used for Multiple Purposes, Manipulable, Easily Aggregated, Easily Reproduced, Data Can Be Easily Integrated, Easily­Customized)
16
Variety of Data & Data Sources
4.71
0.12
4.48-4.95
N/A
(You Have a Variety of Data and Data Sources)
17
Concise
4.75
0.08
4.59-4.92
0.92
(Well-Presented, Concise, Compactly Represented, Well­Organized, Aesthetically Pleasing, Form of Presentation, Well­Formatted, Format of the Data)
18
Access Security
4.92
0.11
4.70-5.14
0.84
(Data Cannot be Accessed by Competitors, Data Are of a Proprietary Nature, Access To Data Can Be Restricted, Secure)
19
Appropriate Amount of Data
5.01
0.11
4.79-5.23
N/A
(The Amount of Data)
20
Flexibility
5.34
0.09
5.17-5.51
0.88
(Adaptable, Flexible, Extendable, Expandable)

A potential disadvantage of factor analysis in this research is that attributes with nothing in common could load on the same factor because they have the same importance ratings. This would lead to problems in interpreting the factors. Importance ratings collected from a sample of 355 data consumers with diverse backgrounds, however, will be similar only if data consumers perceive these items consistently. If these items form different constructs, data consumers will rate their importance differently, resulting in different factors. Furthermore, the twenty dimensions show face validity because it was easy for us to name these factors.

Factor Stability Since the number of survey responses relative to the number of attributes is lower than recommended, it is possible that the factor structure is not sufficiently stable. To test factor stability, we re-ran the analysis using two different approaches. First, in a series of factor-analysis runs, we varied the number of factors to test whether the attributes loading on those factors changed as the number of factors changed. Second, we ran the factor analysis using as input only those attributes that actually loaded on the twenty dimensions to test whether the insignificant attributes affected the results.

In the first approach, our analysis of factor stability found that fourteen out of the twenty dimensions were stable across the series of runs. That is, the same dimensions consisting of the same attributes emerged (See Table 1). Two dimensions, 15 (ease of operation) and 20 (flexibility), were stable in terms of the attributes loading on them, but were combined into a single dimension in runs fixed at fewer dimensions. Four dimensions, 5, 9, 16, and 19, which are all single-attribute factors, have less than desirable stability. Specifically in some runs, these dimensions either were not significant or they were combined with another single-attribute dimension.

In the second approach, the factor analysis used the 71 attributes shown in Table 1, which meets responses-to-attribute ratio recommendations. The second approach produced the same results as the first approach. Thus, we conclude that these dimensions are stable with the caveat that additional research is needed to verify most of the single-attribute dimensions.

A dimension mean was computed as the average of the responses to all of the items with a loading of .5 or greater on the dimension. For example, the dimension ease of understanding consisted of the three items: easily understood, readable, and clear. The mean importance for ease of understanding was the average of the importance ratings for easily understood, readable, and clear. (See Table 1 for the means, standard deviations, and confidence intervals.)

Cronbach's alpha, a measure of construct reliability, was computed for each dimension to assess the reliability of the set of items forming that dimension. As shown in the rightmost column of Table 1, these alpha coefficients ranged from 0.69 to 0.98. As a rule of thumb, alphas of 0.70 or above represent satisfactory reliability of the set of items measuring the construct (dimension). Thus, the items measuring our dimensions are sufficiently reliable.

3. THE TWO-PHASE SORTING STUDY

Twenty dimensions were too many for practical evaluation purposes. In addition, although these dimensions were ranked by the importance ratings, the highest ranking dimensions might not capture the essential aspects of data quality. Finally, a grouping of these dimensions was consistent with research in the marketing discipline, and substantiated a hierarchical structure of data quality dimensions.

Using our preliminary conceptual framework, we conducted a two-phase sorting study. The first phase of the study was to sort these intermediate dimensions into a small set of categories. The second phase was to confirm that these dimensions indeed belonged to the categories in our preliminary conceptual framework.

3.1 Method

We first created four categories (see Column 1 of Table 2) based on our preliminary conceptual framework, following Moore & Benbasat [30] . We then grouped the 20 intermediate dimensions into these four categories (see Column 2 of Table 2). Our initial grouping was based on our understanding of these categories and dimensions. The sorting study provided the data to test this initial grouping and to make adjustments in the assignment of dimensions to target categories (see Column 3 of Table 2), which will be further discussed.

Table 2: Four target categories for the 20 dimensions
Target Category DimensionAdjustment
accuracy of databelievability none
accuracynone
objectivitynone
completenessmoved to category 2
traceabilityeliminated
reputationnone
variety of data sources eliminated
relevancy of datavalue-added none
relevancynone
timelinessnone
ease of operationeliminated
appropriate amount of data none
flexibilityeliminated
representation of datainterpretability none
ease of understanding none
representational consistency none
concise representation none
accessibility of dataaccessibility none
cost-effectivenesseliminated
access securitynone

Note: A target category is a hypothesized category based on our preliminary conceptual framework.

3.1.1. The Sorting Study: Phase One

Subjects A pool of 30 subjects from industry were selected to participate in the overall sorting procedure. These subjects were enrolled in an evening MBA class in another large university. Eighteen of these 30 subjects were randomly selected to participate in the first phase.

Design Each of the 20 dimensions, along with a description, was printed on a 3x5-inch card, as shown in Appendix D1. These cards were used by each of the subjects in the study to group the 20 dimensions into a small set of categories. In contrast to Phase two, the subjects for Phase one were not given a pre-specified set of categories each with a name and description. The study was pre-tested by two graduate-level MIS students to clarify any ambiguity in the design or instruction.

Procedure The study was run by a third party who was not aware of the goal of this research in order to avoid any bias by the authors. Before performing the actual sorting task, subjects performed a trial sort using dimensions other than these 20 dimensions to ensure that they understood the procedure. In the actual sorting task, subjects were given instructions to group the 20 cards into three to five piles. The subjects were then asked to label each of their piles.

3.1.3. The Sorting Study: Phase Two

The original assignment of dimensions to categories was adjusted based on the results from the phase-one study. For example, as shown in Column 3 of Table 2, completeness is moved from the accuracy category to the relevancy category because only four subjects assigned this dimension to the former category whereas twelve assigned it to the later. This was a reasonable adjustment because completeness could be interpreted within the context of the data consumer's task instead of our initial interpretation that completeness was part of the accuracy category.

In addition, five dimensions were eliminated: traceability, variety of data sources, ease of operation, flexibility, and cost-effectiveness. These dimensions were eliminated because of both of the following two reasons. First, subjects did not consistently assign the dimension to any category. For example, seven subjects assigned cost-effectiveness to the relevancy category, three assigned it to the other three categories, and eight assigned it to a self-defined category. Second, the dimension was not ranked highly in terms of importance. For example, cost-effectiveness was ranked 14 out of 20.

The purpose of the second phase of the sorting study was to confirm that the dimensions indeed belonged to these adjusted categories.

Subjects The remaining 12 subjects from our subject pool participated in this phase of study.

Design For each category of dimensions revealed from Phase one, the authors provided a label, as shown in Appendix D2, based on the underlying dimensions. Descriptive phrases rather than single words were used as labels to avoid confounding category labels with any of the dimension labels.

Procedure The third party that ran the phase-one study also ran the phase-two study. The procedure for Phase two was similar to that of Phase one, with the exception that subjects were instructed to place each of the dimension cards into the category that best represents that dimension.

3.2 Findings

In this section, we present the results from the two-phase study. Using the adjusted target categories, we tabulated the results from the phase-one study. As shown in Table 3, the overall placement ratio of dimensions within target categories was 70%. This indicated that these 15 dimensions were generally being placed in the appropriate categories.

Table 3: Results from the phase-one study (15 dimensions and 18 subjects)
Actual Categories
Target

Categories
Accuracy
Relevancy
Representation
Accessibility
N/A
Total
Target

(%)
Accuracy
57
10
2
1
2
72
79%
Relevancy
16
56
11
2
5
90
62%
Repres.
8
4
50
5
5
72
69%
Access.
1
2
3
26
4
36
72%
Total Item Placements: 270
Hits: 189
Overall Hits Ratio: 70%

Note: A target category is a hypothesized category based on our preliminary conceptual framework. An actual category is the category selected by the subjects for a dimension. "N/A" denotes "Not Applicable," which means that the actual category does not fit into any target category.

These results, together with the adjustment of dimensions within the target categories, led us to refine the four target categories as follows:

(1) the extent to which data values are in conformance with the actual or true values.

(2) the extent to which data are applicable (pertinent) to the task of the data user.

(3) the extent to which data are presented in an intelligible and clear manner, and

(4) the extent to which data are available or obtainable.

These four descriptions were used as the category labels for the phase-two study. The results from the phase-two study (Table 4) showed that the overall placement ratio of dimensions within target categories was 81%.

Table 4: Results from the phase-two study (15 dimensions and 12 subjects)
Actual Categories
Target

Categories
Accuracy
Relevancy
Representation
Accessibility
Total
Target

(%)
Accuracy
43
4
1
0
48
90%
Relevancy
7
44
3
6
60
73%
Repres.
2
6
40
0
48
83%
Access.
1
1
3
19
24
79%
Total Item Placements: 180
Hits: 146
Overall Hits Ratio: 81%

4. TOWARD A HIERARCHICAL FRAMEWORK OF DATA QUALITY

In our sorting study, we labeled each category based on our preliminary conceptual framework and our initial grouping of the dimensions. For example, we labeled as accuracy the category that includes accuracy, objectivity, believability, and reputation. Similarly, we labeled the three other categories as relevancy, representation, and accessibility. We used such a labeling so that we would not introduce any additional interpretations or biases into the sorting tasks.

However, such representative labels did not necessarily capture the essence of the underlying dimensions as a group. For example, as a whole, the group of dimensions labeled accuracy was richer than that conveyed by the label accuracy. Thus, we re-examined the underlying dimensions confirmed for each of the four categories, and picked a label that captured the essence of the entire category. For example, we re-labeled accuracy as intrinsic DQ because the underlying dimensions captured the intrinsic aspect of data quality.

As a result of this re-examination, we re-labeled two of the four categories. The resulting categories, therefore, are: intrinsic DQ, contextual DQ, representational DQ, and accessibility DQ (see Figure 2). Intrinsic DQ denotes that data have quality in their own right. Accuracy is merely one of the four dimensions underlying this category. Contextual DQ highlights the requirement that data quality must be considered within the context of the task at hand; i.e., data must be relevant, timely, complete, and appropriate in terms of amount so as to add value. Representational DQ and accessibility DQ emphasize the importance of the role of systems; i.e., the system must be accessible but secure, and the system must present data in such a way that they are interpretable, easy to understand, and represented concisely and consistently.



Figure 2: A conceptual framework of data quality

This hierarchical framework confirms and substantiates the preliminary framework that we proposed. Below we elaborate on these four categories, relate them to the literature, and discuss some future research directions.

4.1 Intrinsic Data Quality

Intrinsic DQ includes not only accuracy and objectivity which are evident to IS professionals, but also believability and reputation. This suggests that, contrary to the traditional development view, data consumers also view believability and reputation as an integral part of intrinsic DQ; accuracy and objectivity alone are not sufficient for data to be considered high quality. This has an analogy to some aspects of product quality. In the product quality area, dimensions of quality emphasized by consumers are broader than those emphasized by product manufacturers. Similarly, intrinsic DQ encompasses more than the accuracy and objectivity dimensions that IS professionals strive to deliver. This finding implies that IS professionals should also ensure the believability and reputation of data. Research on data source tagging [41, 44] is a step in this direction.

4.2 Contextual Data Quality

Some individual dimensions underlying contextual DQ were reported previously; for example completeness and timeliness [4]. However, contextual DQ was not explicitly recognized in the data quality literature. Our grouping of dimensions for contextual DQ revealed that data quality must be considered within the context of the task at hand. This was consistent with the literature on graphical data representation, which concluded that the quality of a graphical representation must be assessed within the context of the data consumer's task [37] .

Since tasks and their contexts vary across time and data consumers, attaining high contextual data quality is a research challenge. One approach is to parameterize contextual dimensions for each task so that a data consumer can specify what type of task is being performed and the appropriate contextual parameters for that task. Below we illustrate such a research prototype.

During Desert Storm combat operations, naval researchers recognized the need to explicitly incorporate contextual DQ into information systems in order to deliver more timely and accurate information. As a result, a prototype is being developed which will be deployed to the U.S. aircraft carriers as stand-alone image exploitation tools [32] . This prototype parameterizes contextual dimensions for each task so that a pilot or a strike planner can specify what type of task (e.g., strike plan or damage assessment) is being performed and the appropriate contextual parameters (relevant images in terms of location, currency, resolution, and target type) for that task.

4.3 Representational Data Quality

Representational DQ includes aspects related to the format of the data (concise and consistent representation) and meaning of data (interpretability and ease of understanding). These two aspects suggest that for data consumers to conclude that data are well-represented, they must not only be concise and consistently represented, but also interpretable and easy to understand.

Issues related to meaning and format arise in database systems research in which format is addressed as part of syntax, and meaning as part of semantic reconciliation. One focus of current research in that area is context interchange among heterogeneous database systems [35] . For example, currency figures in the context of a U.S. database are typically in dollars, whereas those in a Japanese database are likely to be in yens. This type of context belongs to the representational DQ, instead of contextual DQ that deals with the data consumer's task.

4.4 Accessibility Data Quality

IS professionals understand accessibility DQ well. Our research findings show that data consumers also recognize its importance. Our findings appear to differ from the literature which treats accessibility distinctly from information quality; see, for example, Culnan [9] . A closer examination reveals that accessibility is presumed (i.e., perfect accessibility DQ) in earlier information quality literature because hard-copy reports were used instead of on-line data. In contrast, data consumers in our research access computers for their information needs, and therefore, view accessibility DQ as an important data quality aspect. However, there is little difference between treating accessibility DQ as a category of overall data quality, or separating it from other categories of data quality. In either case, accessibility needs to be taken into account.

5. SUMMARY AND CONCLUSIONS

To improve data quality, we need to understand what data quality means to data consumers (those who use data). This research develops a hierarchical framework that captures the aspects of data quality that are important to data consumers. Specifically, 118 data quality attributes collected from data consumers are consolidated into twenty dimensions, which in turn are grouped into four categories. Using this framework, information systems professionals will be able to better understand and meet their data consumers' data quality needs.

In developing this framework, we conduct a two-stage survey and a two-phase sorting study. The resulting framework has four data quality (DQ) categories: (1) intrinsic DQ consists of accuracy, objectivity, believability, and reputation; (2) contextual DQ consists of value-added, relevancy, timeliness, completeness, and appropriate amount of data; (3) representational DQ consists of interpretability, ease of understanding, representational consistency, and concise representation; and (4) accessibility DQ consists of accessibility and access security.

Intrinsic DQ denotes that data have quality in their own right. Contextual DQ highlights the requirement that data quality must be considered within the context of the task at hand. Representational DQ and accessibility DQ emphasize the importance of the role of systems. These findings are consistent with our understanding that high quality data should be intrinsically good, contextually appropriate for the task, clearly represented, and accessible to the data consumer.

The salient feature of this research study is that quality attributes of data are collected from data consumers instead of defined theoretically or based on researchers' experience. Furthermore, this study provides additional evidence for a hierarchical structure of data quality dimensions. At a basic level, the justification for the framework is that a data quality framework does not exist and one is needed so that data quality can be measured, analyzed, and improved in a valid way. Information systems researchers have chosen many different dependent variables for assessing information systems in general, and data quality in particular, with little empirical or theoretical foundation for their choice. This framework provides a basis for deciding which aspects of data quality to use in any research study.

The framework is further justified by the use of well-established empirical methods to develop it. Thus, we argue that the framework is methodologically sound, and that it is complete from the perspectives of data consumers. Furthermore, this framework will be useful as a basis for measuring, analyzing, and improving data quality. While we have only anecdotal evidence to support this claim, the anecdotal evidence is strong and convincing.

This framework was used effectively in industry and government. For example, IS managers in one investment firm thought they had perfect data quality (in terms of accuracy) in their organizational databases. However, in their discussion with data consumers using this framework, they found several deficiencies including: (1) additional information about data sources was needed so that data consumers could assess the reputation and believability of data, (2) data downloaded to servers from the mainframe were not sufficiently timely for some data consumers' tasks, and (3) the currency ($, £, or ¥) and unit (thousands or millions) of financial data from different servers were implicit so data consumers could not always interpret and understand these data correctly.

Based on this hierarchical framework, several research directions can be pursued. First, a questionnaire could be developed to measure perceived data quality. The data-quality categories and their underlying dimensions in this framework would provide the constructs to be measured. Second, methods for improving the quality of data as perceived by data consumers could be developed. Such methods could include user training to change the quality of data as perceived by data consumers. Third, the framework could be useful as a checklist during data requirements analysis. That is, many of the data quality characteristics are actually system requirements or user training requirements. Finally, since a single empirical study is never sufficient to validate the completeness of a framework, further research is needed to apply this framework in specific work contexts.


6. REFERENCES

[1] Arnold, S. E., Information manufacturing: the road to database quality. Database, 15(5) 1992, pp. 32.

[2] Bailey, J. E. and S. W. Pearson, Development of a Tool for Measuring and Analyzing Computer User Satisfaction. Management Science, 29(5) 1983, pp. 530-545.

[3] Ballou, D. P. and H. L. Pazer, Designing Information Systems to Optimize the Accuracy-Timeliness Tradeoff. Information Systems Research (ISR), 6(1) 1995, pp. 51-72.

[4] Ballou, D. P. and H. L. Pazer, Modeling Data and Process Quality in Multi-input, Multi-output Information Systems. Management Science, 31(2) 1985, pp. 150-162.

[5] Ballou, D. P. and K. G. Tayi, Methodology for Allocating Resources for Data Quality Enhancement. Communications of the ACM, 32(3) 1989, pp. 320-329.

[6] Ballou, D. P., R. Y. Wang, H. Pazer and K. G. Tayi (1993). Modeling Data Manufacturing Systems to Determine Data Product Quality. (No. TDQM-93-09). Total Data Quality Management (TDQM) Research Program, MIT Sloan School of Management.

[7] Bodnar, G., Reliability Modeling of Internal Control Systems. Accounting Review, 50(4) 1975, pp. 747-757.

[8] Churchill, G. A., Marketing Research: Methodological Foundations. 5th ed. Dryden Press, 1991.

[9] Culnan, M., The Dimensions of Accessibility to Online Information: Implications for Implementing Office Information Systems. ACM Transactions on Office Information Systems, 2(2) 1984, pp. 141-150.

[10] Cureton, E. E. and R. B. D'Agostino, Factor Analysis: An Applied Approach. Lawerence Erlbaum, Hillsdale, N.J., 1983.

[11] Cushing, B. E., A Mathematical Approach to the Analysis and Design of Internal Control Systems. Accounting Review, 49(1) 1974, pp. 24-41.

[12] Delone, W. H. and E. R. McLean, Information Systems Success: The Quest for the Dependent Variable. Information Systems Research, 3(1) 1992, pp. 60-95.

[13] Deming, E. W., Out of the Crisis. Center for Advanced Engineering Study, MIT, Cambridge., 1986.

[14] Deshpande, R., The Organizational Context of Market Research Use. Journal of Marketing, 46(4) 1992, pp. 92-101.

[15] Dobyns, L. and C. Crawford-Mason, Quality or Else: The Revolution in World Business. Houghton Mifflin, Boston, MA., 1991.

[16] Emery, J. C., Organizational Planning and Control Systems: Theory and Technology. Macmillan, New York., 1969.

[17] Griffin, A. and J. R. Hauser, The Voice of the Customer. Marketing Science, 12(1) 1993, pp. 1-27.

[18] Hauser, J. R. and D. Clausing, The House of Quality. Harvard Business Review, 66(3) 1988, pp. 63-73.

[19] Huh, Y. U., F. R. Keller, T. C. Redman and A. R. Watkins, Data Quality. Information and Software Technology, 32(8) 1990, pp. 559-565.

[20] Ives, B., M. H. Olson and J. J. Baroudi, The Measurement of User Information Satisfaction. Communications of the ACM, 26(10) 1983, pp. 785-793.

[21] Johnson, J. R., R. A. Leitch and J. Neter, Characteristics of Errors in Accounts Receivable and Inventory Audits. Accounting Review, 56(2) 1981, pp. 270-293.

[22] Juran, J. M., Juran on Leadership for Quality: An Executive Handbook. The Free Press, New York., 1989.

[23] Juran, J. M. and F. M. Gryna, Quality Planning and Analysis. 2nd ed. McGraw Hill, New York., 1980.

[24] Knechel, W. R., A Simulation Model for Evaluating Accounting Systems Reliability. Auditing: A Journal of Theory and Practice, 4(2) 1985, pp. 38-62.

[25] Knechel, W. R., The Use of Quantitative Models in the Review and Evaluation of Internal Control: A Survey and Review. Journal of Accounting Literature, 2 1983, pp. 205-219.

[26] Kotler, P., Marketing Essentials. Prentice-Hall, Inc., Englewood Cliffs, NJ., 1984.

[27] Kriebel, C. H., Evaluating the Quality of Information Systems, in Design and Implementation of Computer Based Information Systems, N. Szysperski and E. Grochla, Editors. 1979, Sijthtoff & Noordhoff, Germantown., 1979.

[28] Laudon, K. C., Data Quality and Due Process in Large Interorganizational Record Systems. Communications of the ACM, 29(1) 1986, pp. 4-11.

[29] Melone, N., A Theoretical Assessment of the User-Satisfaction Construct in Information Systems Research. Management Science, 36(1) 1990, pp. 598-613.

[30] Moore, G. C. and I. Benbasat, Development of an Instrument to Measure the Perceptions of Adopting an Information Technology Innovation. Information Systems Research, 2(3) 1991, pp. 192-222.

[31] Morey, R. C., Estimating and Improving the Quality of Information in the MIS. Communications of the ACM, 25(5) 1982, pp. 337-342.

[32] Page, W. and P. Kaomea. Using Quality Attributes to Produce Optimal Tactical Information. in Proceedings ofthe Fourth Annual Workshop on Information Technologies and Systems (WITS). Vancouver, British Columbia, Canada: pp. 145-154, 1994.

[33] Redman, T. C., Data Quality: Management and Technology. Bantam Books, New York., 1992.

[34] Ronen, B. and I. Spiegler, Information as inventory: A new conceptual view. Information & Management, 21 1991, pp. 239-247.

[35] Siegel, M., E. Sciore and A. Rosenthal (1993). Using Semantic Values to Facilitate Interoperability Among Heterogeneous Information Systems. (No. 3543-93). MIT Sloan School of Management.

[36] Stevens, J., Applied Multivariate Statistics for the Social Science. Lawerence Erlbaum, Hillsdale, NJ., 1986.

[37] Tan, J. K. and I. Benbasat, Processing of Graphical Information: A Decomposition Taxonomy to Match Data Extraction Tasks and Graphical Representations. Information Systems Research, 1(4) 1990, pp. 416-439.

[38] Von Hippel, E., Successful Industrial Products from Customer Ideas. 1978, pp. 39-49.

[39] Wand, Y. and R. Y. Wang, Anchoring Data Quality Dimensions in Ontological Foundations. Forthcoming, Communications of the ACM, 1995, .

[40] Wang, R. Y., H. B. Kon and S. E. Madnick. Data Quality Requirements Analysis and Modeling. in Proceedings ofthe 9th International Conference on Data Engineering. Vienna: pp. 670-677, 1993.

[41] Wang, R. Y., M. P. Reddy and H. B. Kon, Toward quality data: An attribute-based approach. Decision Support Systems (DSS), 13(1995) 1995, pp. 349-372.

[42] Wang, R. Y., V. C. Storey and C. P. Firth, A Framework for Analysis of Data Quality Research. IEEE Transactions on Knowledge and Data Engineering, 7(4) 1995, pp. 623-640.

[43] Wang, R. Y., D. M. Strong and L. M. Guarascio (1993). An Empirical Investigation of Data Quality Dimensions: A Data Consumer's Perspective. (No. TDQM-93-12). Total Data Quality Management (TDQM) Research Program, MIT Sloan School of Management.

[44] Wang, Y. R. and S. E. Madnick. A Polygen Model for Heterogeneous Database Systems: The Source Tagging Perspective. in Proceedings ofthe 16th International Conference on Very Large Data bases (VLDB). Brisbane, Australia: pp. 519-538, 1990.

[45] Yu, S. and J. Neter, A Stochastic Model of the Internal Control System. Journal of Accounting Research, 1(3) 1973, pp. 273-295.

7. APPENDICES

7.1 Appendix A: First Data Quality Survey Questionnaire

Position Prior to Attending the university: Finance Marketing

(Circle One) Operations Personnel

IT Other______________

Industry you worked in the previous job : ______________________

When you think of data quality, what attributes other than timeliness, accuracy, availability, and interpretability come to mind? Please list as many as possible!

_________________ _________________ _________________

_________________ _________________ _________________

_________________ _________________ _________________

_________________ _________________ _________________

_________________ _________________ _________________

_________________ _________________ _________________

_________________ _________________ _________________

PLEASE FILL OUT THIS SIDE BEFORE TURNING OVER. THANK YOU!!

---------------------------------------------------------------------------------------------------------------------

---------------------------------------------------------------------------------------------------------------------

(Side Two)

The following is a list of attributes developed for data quality:

Completeness Flexibility Adaptability Reliability

Relevance Reputation Compatibility Ease of Use

Ease of Update Ease of Maintenance Format Cost

Integrity Breadth Depth Correctness

Well-documented Habit Variety Content

Dependability Manipulability Preciseness Redundancy

Ease of Access Convenience Accessibility Data Exchange

Understandable Credibility Importance Critical

After reviewing this list, do any other attributes come to mind?

_________________ _________________ _________________

_________________ _________________ _________________

_________________ _________________ _________________

_________________ _________________ _________________

THANK YOU!

7.2 Appendix B: Second Data Quality Survey Questionnaire

Thank you for participating in this study. All responses will be held in strictest confidence.

Industry: _________________________ Job Title: _____________________________

Department: Finance Marketing/Sales Operations Human Resources

Accounting Information Systems Planning Other ________________

The following is a list of adjectives and phrases which describes corporate data. When answering the questions, please think about the internal data such as sales, production, financial, and employee data that you work with or use to make decisions in your job.

We apologize for the tedious nature of the survey. Although the questions may seem repetitive, your response to each question is critical to the success of the study. Please give us the first response that comes to mind and try to use the FULL scale range available.

Section I: How important is it to you that your data are:

Extremely Not Important Important Important At All

Accurate 1 2 3 4 5 6 7 8 9

Believable 1 2 3 4 5 6 7 8 9

Complete 1 2 3 4 5 6 7 8 9

Concise 1 2 3 4 5 6 7 8 9

Verifiable 1 2 3 4 5 6 7 8 9

Well-Documented 1 2 3 4 5 6 7 8 9

Understandable 1 2 3 4 5 6 7 8 9

Well-Presented 1 2 3 4 5 6 7 8 9

Up-To-Date 1 2 3 4 5 6 7 8 9

Accessible 1 2 3 4 5 6 7 8 9

Adaptable 1 2 3 4 5 6 7 8 9

Aesthetically Pleasing 1 2 3 4 5 6 7 8 9

Compactly Represented 1 2 3 4 5 6 7 8 9

Important 1 2 3 4 5 6 7 8 9

Consistently Formatted 1 2 3 4 5 6 7 8 9

Dependable 1 2 3 4 5 6 7 8 9

Retrievable 1 2 3 4 5 6 7 8 9

Manipulable 1 2 3 4 5 6 7 8 9

Objective 1 2 3 4 5 6 7 8 9

Usable 1 2 3 4 5 6 7 8 9

Well-Organized 1 2 3 4 5 6 7 8 9

Transportable/Portable 1 2 3 4 5 6 7 8 9

Unambiguous 1 2 3 4 5 6 7 8 9

Correct 1 2 3 4 5 6 7 8 9

Please try to use the FULL scale range available.

Section I (continued): How important is it to you that your data are:

Extremely Not Important

Important Important At All

Relevant 1 2 3 4 5 6 7 8 9

Flexible 1 2 3 4 5 6 7 8 9

Flawless 1 2 3 4 5 6 7 8 9

Comprehensive 1 2 3 4 5 6 7 8 9

Consistently Represented 1 2 3 4 5 6 7 8 9

Interesting 1 2 3 4 5 6 7 8 9

Unbiased 1 2 3 4 5 6 7 8 9

Familiar 1 2 3 4 5 6 7 8 9

Interpretable 1 2 3 4 5 6 7 8 9

Applicable 1 2 3 4 5 6 7 8 9

Robust 1 2 3 4 5 6 7 8 9

Available 1 2 3 4 5 6 7 8 9

Revealing 1 2 3 4 5 6 7 8 9

Reviewable 1 2 3 4 5 6 7 8 9

Expandable 1 2 3 4 5 6 7 8 9

Time Independent 1 2 3 4 5 6 7 8 9

Error-Free 1 2 3 4 5 6 7 8 9

Efficient 1 2 3 4 5 6 7 8 9

User-Friendly 1 2 3 4 5 6 7 8 9

Specific 1 2 3 4 5 6 7 8 9

Well-Formatted 1 2 3 4 5 6 7 8 9

Reliable 1 2 3 4 5 6 7 8 9

Convenient 1 2 3 4 5 6 7 8 9

Extendable 1 2 3 4 5 6 7 8 9

Critical 1 2 3 4 5 6 7 8 9

Well-Defined 1 2 3 4 5 6 7 8 9

Reusable 1 2 3 4 5 6 7 8 9

Clear 1 2 3 4 5 6 7 8 9

Cost Effective 1 2 3 4 5 6 7 8 9

Auditable 1 2 3 4 5 6 7 8 9

Precise 1 2 3 4 5 6 7 8 9

Readable 1 2 3 4 5 6 7 8 9

Section II: How important is it to you that your data can be:

Extremely Not Important Important Important At All

Easily Aggregated 1 2 3 4 5 6 7 8 9

Easily Accessed 1 2 3 4 5 6 7 8 9

Easily Compared to Past Data 1 2 3 4 5 6 7 8 9

Easily Changed 1 2 3 4 5 6 7 8 9

Easily Questioned 1 2 3 4 5 6 7 8 9

Easily Downloaded/Uploaded 1 2 3 4 5 6 7 8 9

Easily Joined With Other Data 1 2 3 4 5 6 7 8 9

Easily Updated 1 2 3 4 5 6 7 8 9

Easily Understood 1 2 3 4 5 6 7 8 9

Easily Maintained 1 2 3 4 5 6 7 8 9

Easily Retrieved 1 2 3 4 5 6 7 8 9

Easily Customized 1 2 3 4 5 6 7 8 9

Easily Reproduced 1 2 3 4 5 6 7 8 9

Easily Traced 1 2 3 4 5 6 7 8 9

Easily Sorted 1 2 3 4 5 6 7 8 9

Section III: How important are the following to you?

Extremely Not Important Important Important At All

Data are certified error-free. 1 2 3 4 5 6 7 8 9

Data improves efficiency. 1 2 3 4 5 6 7 8 9

Data gives you a competitive edge. 1 2 3 4 5 6 7 8 9

Data cannot be accessed by competitors. 1 2 3 4 5 6 7 8 9

Data contains adequate detail 1 2 3 4 5 6 7 8 9

Data are in finalized form. 1 2 3 4 5 6 7 8 9

Data contains no redundancy. 1 2 3 4 5 6 7 8 9

Data are of proprietary nature 1 2 3 4 5 6 7 8 9

Data can be personalized. 1 2 3 4 5 6 7 8 9

Data are not easily corrupted. 1 2 3 4 5 6 7 8 9

Data meets all of your requirements. 1 2 3 4 5 6 7 8 9

Data adds value to your operations. 1 2 3 4 5 6 7 8 9

Data are continuously collected. 1 2 3 4 5 6 7 8 9

Data continuously presented in same format. 1 2 3 4 5 6 7 8 9

Data are compatible with previous data. 1 2 3 4 5 6 7 8 9

Data are not overwhelming. 1 2 3 4 5 6 7 8 9

Section III (Continued): How important are the following to you?

Extremely Not Important Important Important At All

Data can be easily integrated. 1 2 3 4 5 6 7 8 9

Data can be used for multiple purposes. 1 2 3 4 5 6 7 8 9

Data are secure. 1 2 3 4 5 6 7 8 9

Section IV: How important are the following to you?

Extremely Not Important Important Important At All

The source of the data is clear. 1 2 3 4 5 6 7 8 9

Errors can be easily identified. 1 2 3 4 5 6 7 8 9

The cost of data collection. 1 2 3 4 5 6 7 8 9

The cost of data accuracy. 1 2 3 4 5 6 7 8 9

The form of presentation. 1 2 3 4 5 6 7 8 9

The format of the data. 1 2 3 4 5 6 7 8 9

The scope of information contained in data. 1 2 3 4 5 6 7 8 9

The depth of information contained in data. 1 2 3 4 5 6 7 8 9

The breadth of information contained in data. 1 2 3 4 5 6 7 8 9

Quality of resolution. 1 2 3 4 5 6 7 8 9

The storage medium. 1 2 3 4 5 6 7 8 9

The reputation of the data source. 1 2 3 4 5 6 7 8 9

The reputation of the data. 1 2 3 4 5 6 7 8 9

The age of the data. 1 2 3 4 5 6 7 8 9

The amount of data. 1 2 3 4 5 6 7 8 9

You have used the data before. 1 2 3 4 5 6 7 8 9

Someone has clear responsibility for data. 1 2 3 4 5 6 7 8 9

The data entry process is self-correcting. 1 2 3 4 5 6 7 8 9

The speed of access to data. 1 2 3 4 5 6 7 8 9

The speed of operations performed on data. 1 2 3 4 5 6 7 8 9

The amount and type of storage required. 1 2 3 4 5 6 7 8 9

You have little extraneous data present. 1 2 3 4 5 6 7 8 9

You have a variety of data and data sources. 1 2 3 4 5 6 7 8 9

You have optimal data for your purpose. 1 2 3 4 5 6 7 8 9

The integrity of the data. 1 2 3 4 5 6 7 8 9

It is easy to tell if the data are updated. 1 2 3 4 5 6 7 8 9

Easy to exchange data with others. 1 2 3 4 5 6 7 8 9

Access to data can be restricted. 1 2 3 4 5 6 7 8 9

Thank you for your time and effort in completing this survey.

7.3 Appendix C: Descriptive Statistics for Attributes

Attribute
# of cases
mean
S.D.
Min
Max
Accurate
350
1.771
1.135
1
7
Believable
348
2.707
1.927
1
9
Complete
349
3.229
1.814
1
9
Concise
348
3.994
2.016
1
9
Verifiable
348
3.224
1.854
1
9
Well-Documented
349
4.123
2.087
1
9
Understandable
349
2.668
1.671
1
9
Well-Presented
350
3.937
2.124
1
9
Up-To-Date
350
2.963
1.732
1
9
Accessible
349
3.370
1.899
1
9
Adaptable
344
4.942
2.042
1
9
Aesthetically Pleasing
350
6.589
2.085
1
9
Compactly Represented
349
5.123
2.181
1
9
Important
335
3.824
2.138
1
9
Consistently Formatted
347
4.594
2.141
1
9
Dependable
349
2.648
1.615
1
9
Retrievable
350
3.660
1.999
1
9
Manipulable
349
4.327
2.162
1
9
Objective
345
3.551
1.963
1
9
Usable
348
2.586
1.567
1
9
Well-Organized
348
3.737
1.856
1
9
Transportable/Portable
348
5.448
2.259
1
9
Unambiguous
346
3.717
2.030
1
9
Correct
348
1.816
1.252
1
9
Relevant
347
2.937
1.573
1
9
Flexible
344
5.183
1.981
1
9
Flawless
348
4.761
2.649
1
9
Comprehensive
347
4.501
1.907
1
9
Consistently Represented
347
4.058
2.002
1
9
Interesting
349
6.095
2.117
1
9
Unbiased
350
3.623
2.006
1
9
Familiar
348
6.670
5.115
1
9
Interpretable
349
3.198
1.652
1
9
Applicable
349
3.046
1.624
1
9
Robust
338
5.059
2.290
1
9
Available
347
3.357
1.902
1
9
Revealing
349
4.493
2.293
1
9
Reviewable
346
4.042
1.916
1
9
Expandable
342
5.588
2.047
1
9
Time Independent
334
6.772
2.134
1
9
Error-Free
348
3.376
2.384
1
9
Efficient
339
4.906
1.980
1
9
User-Friendly
346
4.711
2.118
1
9
Specific
339
4.189
1.915
1
9
Well-Formatted
346
4.855
2.056
1
9
Reliable
346
2.309
1.303
1
7
Convenient
346
4.673
1.864
1
9
Extendable
342
5.678
2.041
1
9
Critical
333
4.985
2.321
1
9
Well-Defined
343
3.603
1.779
1
9
Reusable
341
4.959
2.268
1
9
Clear
345
3.151
1.634
1
9
Cost Effective
349
4.183
2.167
1
9
Auditable
346
4.197
2.378
1
9
Precise
347
3.896
2.133
1
9
Readable
348
3.247
1.791
1
9
Easily Aggregated
347
3.968
2.052
1
9
Easily Accessed
347
3.510
1.827
1
9
Easily Compared to Past Data
348
3.043
1.790
1
9

Attribute
# of cases
mean
S.D.
Min
Max
Easily Changed
347
4.818
2.134
1
9
Easily Questioned
344
4.387
2.095
1
9
Easily Downloaded/Uploaded
350
4.969
2.357
1
9
Easily Joined With Other Data
349
4.619
2.204
1
9
Easily Updated
345
3.559
1.924
1
9
Easily Understood
349
3.241
1.766
1
9
Easily Maintained
348
3.879
2.004
1
9
Easily Retrieved
348
3.667
1.791
1
9
Easily Customized
348
4.825
2.134
1
9
Easily Reproduced
347
4.617
2.130
1
9
Easily Traced
347
4.573
2.225
1
9
Easily Sorted
348
4.635
2.303
1
9
Data are certified error-free.
348
4.141
2.572
1
9
Data improves efficiency.
344
3.718
2.064
1
9
Data gives you a competitive edge.
348
3.178
2.277
1
9
Data cannot be accessed by competitors.
347
4.450
2.760
1
9
Data contains adequate detail
348
3.057
1.378
1
8
Data are in finalized form.
348
5.575
2.201
1
9
Data contains no redundancy.
344
6.279
2.026
1
9
Data are of proprietary nature
346
5.867
2.612
1
9
Data can be personalized.
345
5.759
2.390
1
9
Data are not easily corrupted.
344
3.741
2.162
1
9
Data meets all of your requirements.
348
3.664
2.123
1
9
Data adds value to your operations.
349
2.479
1.708
1
9
Data are continuously collected.
347
4.608
2.443
1
9
Data continuously presented in same format.
346
4.627
2.232
1
9
Data are compatible with previous data.
348
3.578
1.893
1
9
Data are not overwhelming.
347
4.037
2.306
1
9
Data can be easily integrated.
348
4.086
1.896
1
9
Data can be used for multiple purposes
347
4.565
2.304
1
9
Data are secure
349
4.456
2.432
1
9
The source of the data is clear
350
3.291
1.836
1
9
Errors can be easily identified.
347
3.089
1.584
1
8
The cost of data collection.
349
4.304
2.180
1
9
The cost of data accuracy.
348
4.261
2.169
1
9
The form of presentation.
349
4.794
1.994
1
9
The format of the data.
348
4.917
2.045
1
9
The scope of information contained in data.
345
3.838
1.726
1
9
The depth of information contained in data.
345
3.922
1.835
1
9
The breadth of information contained in data.
344
3.872
1.796
1
9
Quality of resolution.
329
5.024
1.995
1
9
The storage medium.
348
6.534
2.148
1
9
The reputation of the data source.
348
4.144
2.172
1
9
The reputation of the data.
347
3.925
2.133
1
9
The age of the data.
350
3.640
2.044
1
9
The amount of data.
347
5.009
2.125
1
9
You have used the data before.
345
6.107
2.228
1
9
Someone has clear responsibility for data.
347
3.744
2.271
1
9
The data entry process is self-correcting.
344
4.695
2.362
1
9
The speed of access to data.
347
3.934
1.992
1
9
The speed of operations performed on data.
348
4.687
2.194
1
9
The amount and type of storage required.
349
6.209
2.030
1
9
You have little extraneous data present.
345
5.797
2.003
1
9
You have a variety of data and data sources.
344
4.712
2.234
1
9
You have optimal data for your purpose.
345
3.554
2.126
1
9
The integrity of the data.
345
2.371
1.571
1
9
It is easy to tell if the data are updated.
348
3.609
1.926
1
9
Easy to exchange data with others.
346
4.945
2.311
1
9
Access to data can be restricted.
347
4.988
2.514
1
9

7.4 Appendix D: The Two-Phase Sorting Study

7.4.1. Appendix D1: Instruction and content for Phase one

Instruction 1

Group the 20 data quality dimensions into several categories (between 3 and 5) where the dimensions within each category, in your opinion, represent similar attributes of high-quality data. (Note: A data quality dimension may also be isolated into its own category, if you see fit to do so.)

Example 3 X 5 Card



BELIEVABILITY

The extent to which data are accepted or regarded as true, real, and credible.


(1)

Content of the remaining nineteen 3 X 5 dimension cards

(2) VALUE-ADDED: The extent to which data are beneficial and provide advantages from their use.

(3) RELEVANCY: The extent to which data are applicable and helpful for the task at hand.

(4) ACCURACY: The extent to which data are correct, reliable, and certified free of error.

(5) INTERPRETABILITY: The extent to which data are in appropriate language and units, and the data definitions are clear.

(6) EASE OF UNDERSTANDING: The extent to which data are clear, without ambiguity, and easily comprehended.

(7) ACCESSIBILITY: The extent to which data are available or easily and quickly retrievable.

(8) OBJECTIVITY: The extent to which data are unbiased (unprejudiced) and impartial.

(9) TIMELINESS: The extent to which the age of the data is appropriate for the task at hand.

(10) COMPLETENESS: The extent to which data are of sufficient breadth, depth, and scope for the task at hand.

(11) TRACEABILITY: The extent to which data are well-documented, verifiable, and easily attributed to a source.

(12) REPUTATION: The extent to which data are trusted or highly regarded in terms of their source or content.

(13) REPRESENTATIONAL CONSISTENCY: The extent to which data are always presented in the same format and are compatible with previous data.

(14) COST EFFECTIVENESS: The extent to which the cost of collecting appropriate data is reasonable.

(15) EASE OF OPERATION: The extent to which data are easily managed and manipulated (i.e., updated, moved, aggregated, reproduced, customized).

(16) VARIETY OF DATA & DATA SOURCES: The extent to which data are available from several, differing data sources.

(17) CONCISE: The extent to which data are compactly represented without being overwhelming (i.e., brief in presentation, yet complete and to the point).

(18) ACCESS SECURITY: The extent to which access to data can be restricted, and hence, kept secure.

(19) APPROPRIATE AMOUNT OF DATA: The extent to which the quantity or volume of available data is appropriate.

(20) FLEXIBILITY: The extent to which data are expandable, adaptable, and easily applied to other needs.

Instruction 2

Label the categories that you have created with an overall definition (word or two/three word phrase) that best describes/summarizes the data quality dimensions within each category

7.4.2. Appendix D2: Instruction and content for Phase two

Instruction

Group each of the data quality dimensions into one of the following four categories. In case of conflict, choose the best fitting category for the dimension. All dimensions must be categorized.

Content of the four 3 X 5 category cards

Category 1: The extent to which data values are in conformance with the actual or true values.

Category 2: The extent to which data are applicable to, or pertain to, the task of the data user.

Category 3: The extent to which data are presented in an intelligible and clear manner.

Category 4: The extent to which data are available or obtainable.