A Brief Introduction to Important Aspects of Data Science

Introduction

To comprehend the need for the study of data, it is quite imperative to first of all comprehend what exactly data are. According to the Australian Bureau of Statistics (2013), data itself is a plural form for the word: datum which is a singular value for a particular variable of information.

Data and its Types

In data science, data are mostly in numeric form and are gotten through observations or even measurements. Though, data come in two forms: the qualitative and quantitative value sets whereby the qualitative deals with non-numeric data forms and quantitative refers to the numeric value sets based on a single or multiple animals, places, persons or things (OECD, 2008).

Role of a Data Scientist

A data scientist employs the use of various data types while working on his projects. These data can be as well grouped into some main classes. The classes are majorly two: the primitive or primary data or atomic data type and the non-primitive or composite or complex or secondary data type. However, in the literature, there is also a third underlying class which is often referred to as the abstract data type (Placide, 2013).

Data Classes in Programming

In each of these classes, there are some data structures embedded in them. For instance, in the primary data class, we have the following data structures; characters, integers (long), floating-point number (such as float, double, real and double precision), fixed-point number, Boolean logical values, reference (also known as pointers or handles) (Placide, 2013).

Data Structures

On the other hand, secondary data class includes the complex data structures like record (also known as structs), arrays (both one-dimensional and multi-dimensional), strings (combination of characters), pointers, linked list (single-linked list, doubly linked list, etc), stacks, queues, trees and graphs (Bullinaria, 2019). The primary data are used to form these unique data classes.

Steps in Data Collection and Analysis

Data needs to be collected in order to be worked on. Hence, there are some vital methods through which data scientists and other experts gather data. Some of the most common ways any kind of data can be elicitated either the qualitative or quantitative data are through the field work surveys (be it direct or indirect surveys) and observations. And, they can be later analyzed through the use of experiments (like manipulative, natural or observational), modelling (through simulation model) ?after of course developing hypothesis for the research.

Datasets in Data Repositiories

For the purposes of education, research and posterity, data are now kept in data repositories especially the digital kinds. All the contents of these libraries comprises of both the numeric and non-numeric types of data. In addition, these archives have in them databases and data catalogues that are readily accessible by students, researchers and data scientists online. Datasets in these repositories are identified through the use of metadata which show important information about the data searches that users are looking and the corresponding datasets (Ma, 2019). Thus, data repositories serve as a preservation tool for not just the data itself but all other properties of the research, academics and programming that corroborate the collection, development and enhancement of the data.

Reference List

Australian Bureau of Statistics (2013). Statistical Language - What are Data? Archived from the original on 2019-04-19. Retrieved 2020-03-09.

Bullinaria, J. (2019). Lecture Notes for Data Structures and Algorithms. UK: University of Birmingham.

Ma, X. (2019). Data Repository. In Schintler, L.A., McNeely, C.L. (eds.) Encyclopedia of Big Data. Switzerland: Springer, Cham In Press. https://dx.doi.org/10.1007/978-3-319-32001-4_59-1

OECD (2008). Glossary of Statistical Terms. p. 119. ISBN 978-92-64-025561.

Placide, D. (2013). Data Types and Data Structures. Cameroon: PCHS Mankon – Bamenda.

?

?

?