Data Processing in Research
The steps involved in data analysis cycle are:
Table of Content
Input: It involves the collection of data from various available sources. Here, it should be noted that these sources should be authentic and accurate so the data collected is of the highest possible quality. Once the data is collected, it enters the data preparation stage, wherein raw data is diligently checked for any errors and organised for easy accessibility.
The main of the data preparation step is to eliminate redundant, incomplete, or incorrect data and maintain high-quality data for the effective business intelligence. The clean data is then entered into its destination (computerised applications) and translated into a language that it can understand. Data input is the first stage in which raw data begins to take the form of usable information.
Processing: At this stage, the data entered into to the computer in the previous stage is actually processed for interpretation using machine learning algorithms.
Output: At this stage, the data is translated, readable and presented in the form of graphs, videos, images, plain text, etc. It can be now used by different users.
Storage: This is the final stage of data processing. Once the data is processed, it is then stored for future use. Although some information is used immediately, much of it serves a purpose later on. When data is properly stored, it can be quickly and easily accessed by users whenever needed.
It is important to take into account the difference between data processing and a data processing system. Data processing is the rules through which the data is converted into useful information. A data processing system is an optimised application for a certain type of data processing. For example, a time-sharing system is designed to optimally carry out time sharing processing activities. It can also be used to perform batch processing.
The data collected from various sources may lack uniformity. For example, data collected through questionnaires may have answers which may not be ticked properly or some questions are left unanswered. Apart from that, there are some instances when data collected needs reconstruction in a category designed for analysis; for example, converting monthly income in annual income.
Here, the role of data editing comes into the picture. Data editing is a process of examining the collected data to identify errors and omissions and correct them to ensure data quality. Apart from that, edited data also facilitates in the coding and tabulation of data.
The main aim of data editing is to ensure:
- Legibility: The recorded data must be legible so that it can he coded later. An illegible response may be corrected by contacting the people who recorded it or it may be inferred from other parts of the question.
- Completeness: All the items in the questionnaire must be fully completed. If some questions are not answered, the interviewers may be contacted to find out whether the respondent failed to respond to the question or refused to answer the question. If the interviewer doesn’t remember the response, the respondent may be contacted again or this particular piece of data may be treated as missing data.
- Accuracy: The inaccuracy of the survey data may be due to the interviewer biasness. One way of identifying is to look for a common pattern of responses in the instrument of a particular interviewer.
Coding of data refers back to the manner of remodelling gathered records or observations to a fixed of meaningful, cohesive categories. It is a manner of summarising and re-importing records to be able to offer a scientific account of the recorded or determined phenomenon. Data consult with a huge variety of empirical items inclusive of ancient documents, newspaper articles, TV programming, area notes, interview or attention institution transcripts, pictures, face-to-face conversations, social media messages (e.g., tweets or YouTube annotations), and so on.
Codes are principles that hyperlink records with theory. They can both be predefined via way of means of the researcher or emerge inductively from the coding manner. It must be noted that the data that is already coded is known as precoded data. The data that is coded at the time of data processing is known as postcoded data.
Data Classification is comprehensively characterised as the most common way of getting sorted out information by significant classes with the goal that it might be utilised and secured all the more proficiently.On a fundamental level, the arrangement interaction makes information simpler to find and recover. Information characterisation is of specific significance with regards to hazard the executives, consistency, and information security.
Data order includes labelling information to make it effectively accessible and identifiable. It additionally takes out various duplications of information, which can lessen capacity and reinforcement costs while accelerating the hunt interaction. However, the characterisation interaction might sound exceptionally specialised, a point ought to be perceived by your association’s authority.
Types of Data Classification
Information characterisation regularly includes a huge number of labels and marks that characterise the sort of information, its classification, and its uprightness. Accessibility may likewise be thought about in information characterisation measures. Information’s level of affectability is regularly characterised dependent on shifting degrees of significance or privacy, which at that point, relates to the safety efforts set up to ensure every order level.
Three fundamental kinds of data types are viewed as industry norms:
- Content-based classification: Investigates and deciphers records searching for touchy data
- User-based classification: It is related to an application, area, or maker among different factors as backhanded pointers of delicate data
- Content-client determination: Client put together order depends concerning client information and carefulness at creation, alter, audit, or spread to hail delicate records.
Content-setting, and client-based methodologies can be both set in stone contingent upon the business need and data type.
Data entry is defined as the method of transcribing data into an electronic medium such as a laptop or alternative device. It will either be performed manually or mechanically by employing a machine or laptop. Data entry is taken into account a non–core method for many organisations and is sometimes performed on knowledge forms such as spreadsheets, written or scanned documents, audio, or video.
To perform data entry, you do not need any special qualifications, information, and solely need accuracy and quick turnaround. Most data entry jobs are usually outsourced to people having lower skills which means the cost of data entry work is lowered. Computers also are employed in machine-driven knowledge entry, as they are extremely correct and might be programmed to fetch and transcribe information into the desired medium.
Accurately entered data is the base upon which the organisation will perform analysis and create plans. Manual entry usually needs sensible concentration and focus over a protracted length of your time, and this may prove physically and mentally difficult for data entry staff.
The process of placing categorised facts into tabular shapes is known as tabulation. A table is the asymmetric arrangement of statistical data in rows and columns. Rows are horizontal preparations, while columns are vertical preparations. It could be easy, double, or complex relying upon the form of category.
Types of Tabulation
Simple Tabulation or One-way Tabulation
The information is tabulated to at least one feature in a one-way tabulation. For example, tabulation of world population by country is an example of a simple tabulation.
The facts are tabulated in line with traits at a time, it is stated to be a double tabulation or two-way tabulation. For example, tabulation of data of world population in terms of males and females.
The data is tabulated according to the traits, and it is stated as the complicated tabulation. An example of complex tabulation is shown in Figure: