Bad data is the number one enemy for the data-driven companies. Although organisations are investing money and effort into eradicating it, it still presents a challenge. Decisions made on bad data can turn fatal for the business.
Proactively building for and managing bad data, rather than trying to pretend it doesn’t exist, means that you can not only minimise the negative impact on your business, but can potentially pave the way for important business improvements too.
Proper visibility into the data correction process helps to understand the causes of bad data, and help to reveal other systemic problems that need to be addressed. Problems with your data can indicate the need for changes elsewhere in the process.
Changes in how data is sourced or processed, or how staffs are trained for example, can often not only improve your data quality, but also improve efficiency, turnaround times and so directly impact business health.
Data errors can also provide insight into revenue leaks, showing you where you need to focus attention to fix problems that affect the bottom line.
Building an efficient error handling process into your data management brings many benefits:
- Automating error handling can help get (corrected) data back into the system as quickly as possible, keeping business processes running on schedule.
- Standardizing error correction helps provide consistency in the data process, and reporting enables complete transparency into everything that happens to each record.
Data cleaning
Data cleansing or data cleaning is the process of detecting and correcting (or removing) corrupt or inaccurate records from a record set, table, or database and refers to identifying incomplete, incorrect, inaccurate or irrelevant parts of the data and then replacing, modifying, or deleting the dirty or coarse data. Data cleansing may be performed interactively with data wrangling tools, or as batch processing through scripting or a data quality firewall.
After cleansing, a data set should be consistent with other similar data sets in the system. The inconsistencies detected or removed may have been originally caused by user entry errors, by corruption in transmission or storage, or by different data dictionary definitions of similar entities in different stores. Data cleaning differs from data validation in that validation almost invariably means data is rejected from the system at entry and is performed at the time of entry, rather than on batches of data.
The actual process of data cleansing may involve removing typographical errors or validating and correcting values against a known list of entities. The validation may be strict (such as rejecting any address that does not have a valid postal code), or with fuzzy or approximate string matching (such as correcting records that partially match existing, known records). Some data cleansing solutions will clean data by cross-checking with a validated data set. A common data cleansing practice is data enhancement, where data is made more complete by adding related information. For example, appending addresses with any phone numbers related to that address. Data cleansing may also involve harmonization (or normalization) of data, which is the process of bringing together data of “varying file formats, naming conventions, and columns”, and transforming it into one cohesive data set; a simple example is the expansion of abbreviations (“st, rd, etc.” to “street, road, etcetera”).
Data quality
High-quality data needs to pass a set of quality criteria. Those include:
Validity: The degree to which the measures conform to defined business rules or constraints (see also Validity (statistics)). When modern database technology is used to design data-capture systems, validity is fairly easy to ensure: invalid data arises mainly in legacy contexts (where constraints were not implemented in software) or where inappropriate data-capture technology was used (e.g., spreadsheets, where it is very hard to limit what a user chooses to enter into a cell, if cell validation is not used). Data constraints fall into the following categories:
Data-Type Constraints; e.g., values in a particular column must be of a particular data type, e.g., Boolean, numeric (integer or real), date, etc.
- Range Constraints: Typically, numbers or dates should fall within a certain range. That is, they have minimum and/or maximum permissible values.
- Mandatory Constraints: Certain columns cannot be empty.
- Unique Constraints: A field, or a combination of fields, must be unique across a dataset. For example, no two persons can have the same social security number.
- Set-Membership constraints: The values for a column come from a set of discrete values or codes. For example, a person’s sex may be Female, Male or Non-Binary.
- Foreign-key constraints: This is the more general case of set membership. The set of values in a column is defined in a column of another table that contains unique values. For example, in a US taxpayer database, the “state” column is required to belong to one of the US’s defined states or territories: the set of permissible states/territories is recorded in a separate State table. The term foreign key is borrowed from relational database terminology.
- Regular expression patterns: Occasionally, text fields will have to be validated this way. For example, phone numbers may be required to have the pattern (999) 999-9999.
- Cross-field validation: Certain conditions that utilize multiple fields must hold. For example, in laboratory medicine, the sum of the components of the differential white blood cell count must be equal to 100 (since they are all percentages). In a hospital database, a patient’s date of discharge from the hospital cannot be earlier than the date of admission.
Accuracy: The degree of conformity of a measure to a standard or a true value – see also Accuracy and precision. Accuracy is very hard to achieve through data-cleansing in the general case because it requires accessing an external source of data that contains the true value: such “gold standard” data is often unavailable. Accuracy has been achieved in some cleansing contexts, notably customer contact data, by using external databases that match up zip codes to geographical locations (city and state) and also help verify that street addresses within these zip codes actually exist.
Completeness: The degree to which all required measures are known. Incompleteness is almost impossible to fix with data cleansing methodology: one cannot infer facts that were not captured when the data in question was initially recorded. (In some contexts, e.g., interview data, it may be possible to fix incompleteness by going back to the original source of data, i.e. re-interviewing the subject, but even this does not guarantee success because of problems of recall – e.g., in an interview to gather data on food consumption, no one is likely to remember exactly what one ate six months ago. In the case of systems that insist certain columns should not be empty, one may work around the problem by designating a value that indicates “unknown” or “missing”, but the supplying of default values does not imply that the data has been made complete.)
Consistency: The degree to which a set of measures are equivalent in across systems (see also Consistency). Inconsistency occurs when two data items in the data set contradict each other: e.g., a customer is recorded in two different systems as having two different current addresses, and only one of them can be correct. Fixing inconsistency is not always possible: it requires a variety of strategies – e.g., deciding which data were recorded more recently, which data source is likely to be most reliable (the latter knowledge may be specific to a given organization), or simply trying to find the truth by testing both data items (e.g., calling up the customer).
Uniformity: The degree to which a set data measures are specified using the same units of measure in all systems ( see also Unit of measure). In datasets pooled from different locales, weight may be recorded either in pounds or kilos and must be converted to a single measure using an arithmetic transformation.
The term integrity encompasses accuracy, consistency and some aspects of validation (see also data integrity) but is rarely used by itself in data-cleansing contexts because it is insufficiently specific. (For example, “referential integrity” is a term used to refer to the enforcement of foreign-key constraints above.)
Process
Data auditing: The data is audited with the use of statistical and database methods to detect anomalies and contradictions: this eventually indicates the characteristics of the anomalies and their locations. Several commercial software packages will let you specify constraints of various kinds (using a grammar that conforms to that of a standard programming language, e.g., JavaScript or Visual Basic) and then generate code that checks the data for violation of these constraints. This process is referred to below in the bullets “workflow specification” and “workflow execution.” For users who lack access to high-end cleansing software, Microcomputer database packages such as Microsoft Access or File Maker Pro will also let you perform such checks, on a constraint-by-constraint basis, interactively with little or no programming required in many cases.
Workflow specification: The detection and removal of anomalies are performed by a sequence of operations on the data known as the workflow. It is specified after the process of auditing the data and is crucial in achieving the end product of high-quality data. In order to achieve a proper workflow, the causes of the anomalies and errors in the data have to be closely considered.
Workflow execution: In this stage, the workflow is executed after its specification is complete and its correctness is verified. The implementation of the workflow should be efficient, even on large sets of data, which inevitably poses a trade-off because the execution of a data-cleansing operation can be computationally expensive.
Post-processing and controlling: After executing the cleansing workflow, the results are inspected to verify correctness. Data that could not be corrected during the execution of the workflow is manually corrected, if possible. The result is a new cycle in the data-cleansing process where the data is audited again to allow the specification of an additional workflow to further cleanse the data by automatic processing.
Steps of Data Cleaning
Every organization’s data cleaning methods will vary according to their individual needs as well as the particular constraints of the dataset. However, most data cleaning steps follow a standard framework:
- Determine the critical data values you need for your analysis.
- Collect the data you need, then sort and organize it.
- Identify duplicate or irrelevant values and remove them.
- Search for missing values and fill them in, so you have a complete dataset.
- Fix any remaining structural or repetitive errors in the dataset.
- Identify outliers and remove them, so they will not interfere with your analysis.
- Validate your dataset to ensure it is ready for data transformation and analysis.
- Once the set has been validated, perform your transformation and analysis.