Data Analysis

What is Data Analysis?

Data analysis is the process of examining, cleansing, transforming, and modeling data with the objective of extracting useful information for decision-making. It is often used in different domains, such as business, science, and the humanities.

Types of Data Analysis

While many types of data analysis can be used, the following are five of the most well-known methods:

Statistical analysis

Statistical analysis analyzes a set of data or a sample of data and interprets past data. It is done via the collection, analysis, interpretation, presentation, and modeling of past data. The two subcategories of statistical analysis are descriptive analysis and inferential analysis.

Descriptive analysis analyzes and summarizes the characteristics of a data set to determine what is happening.

Inferential analysis differs from descriptive analysis because it allows analysts to test a hypothesis and assess whether the data from the sample can be generalized to the general population. In other words, inferential analysis allows one to make predictions or inferences from the data through testing a smaller sample instead of the entire population.

For example, in accounting, auditors will often use inferential analysis to determine the risk of material misstatement in a client’s financial statements. It is done by auditors taking a sample of data from the client and then determining if the results from the sample apply to the entire population. In such a case, auditors will need to remove any high-value items or non-recurring items (in a process called stratification) before analyzing the sample to reduce the risk of sampling error.

Diagnostic analysis

The primary purpose of diagnostic analysis is to determine the root cause of any results found after statistical analysis. Diagnostic analysis is useful because it helps identify any patterns in the data. When new issues arrive in the business process, one can use diagnostic analysis to find similar patterns.

Text analysis (data mining)

Also known as data mining, text analysis is one of the most popular data analysis methods used to discover patterns in large data sets by utilizing databases or data mining tools. Text analysis is primarily used to transform raw data into business information and, specifically, derives patterns after examining data to put it to good use.

Predictive analysis

Predictive analysis is used to predict what is likely to happen given the previous data. It makes predictions of future outcomes; however, it is important to note that it is just an estimate. Other factors may be needed to be taken into account, such as industry trends or macroeconomic developments in the economy or society as a whole.

Prescriptive analysis

Prescriptive analysis is the final phase of business analytics and includes descriptive and predictive analysis. Its purpose is to combine the insight of the previous analysis to determine which actions should be taken to address the current problem or make a decision. It is used because predictive analysis and descriptive analysis are often not enough to improve data performance.

While prescriptive analysis can help prevent fraud, limit risk, increase efficiency, and meet business goals, it is not foolproof. It is only effective if the organizations involved know which questions to ask and how to address the answers throughout the analysis process.

Prescriptive analysis utilizes state-of-the-art technology and data practices and, as such, is a huge organizational commitment. Therefore, companies need to be sure that they are ready and able to afford the extensive human and financial resources.

Artificial intelligence (AI) is an example of prescriptive analysis, as AI systems use a large amount of data to learn continuously and use the information to make informed decisions. High-quality AI systems are capable of communicating decisions and put them into action. Using artificial intelligence allows business processes to be performed and optimized daily without needing any human action.

Additional Resources

• Data Warehousing
• Hypothesis Testing
• Descriptive Statistics
• Sampling Errors
• See all data science resources