Data analysis is a process of improving, examining, modifying and molding the data with highlighting important information, supporting decision making and suggesting conclusions. Data analysis helps to describe facts, develop explanations, test hypothesis and detect patterns. Data analysis has various aspects and approaches, including various techniques in science, business, administration, policy and social science domain. How to conduct data analysis is discussed in this article.
It involves checking the data accuracy, data entry, transforming the data, processing and documenting database structure that incorporates various measures. You should set up a process for checking the data and keeping track until it is ready for data analysis. Data preparation includes:
- logging the data – that is checking the data through different sources like observational data, coded interview data etc.
- data must be screened for its accuracy then the database structure should be developed to store the data for the data analysis
- enter the data. For high data accuracy, a process called double entry is followed.
- Once the data is entered, it is necessary to transform the raw data into variables for easy analysis.
It is used in describing the fundamental features of the data. Descriptive statistics simply describes what the data shows. It is used to present quantitative descriptions. Descriptive statistics includes charts, numbers, tables, graphs, bar diagrams, pie charts which are used to describe, summarize, and present the raw data. Descriptive statistics is used to examine:
- location of the data that is central tendency of the data
- variability of the data
- symmetry of the data; and
- how the data is concentrated around a single value.
Descriptive statistics are useful when you do not need to extend your result to any larger group. Descriptive statistics are organized and summarized for clear presentation. Descriptive data allows to present the data in a meaningful manner which allows easy inference of the data.
Inferential statistics concerned in making inferences from the observation and analysis of a sample. It is useful in experimental research design or quasi experimental analysis or process outcome evaluation. Most of the inferential statistics have taken the origin from General Linear Model of general statistics. It is used in making conclusions and predictions based on the numerical data analysis. Inferential statistics include ANOVA, linear regression analysis, structural equation modeling, logistic regression analysis, survival analysis etc.
The main difference between descriptive statistics and inferential statistics is – descriptive statistics remains local to the sample whereas inferential statistics concentrates on making statements about the population.