Data Analysis

Data analysis summarised in ONE FLASHCARD!!!

Data vs. Information

Data: facts and figures: useless on themselves.

Information: produced when data is manipulated into meaningful/useful forms.

Qualitative v. Quantitative Data

Quantitative data is numbers-based, countable, or measurable; collected through observations and experiments.

Qualitative (why, what, how) data is interpretation/opinion-based, descriptive, and relating to language; collected through interviews, surveys, etc.

Primary v. Secondary Data

• Primary Data: data that are collected from the field under the control /supervision of the investigator.

• Collected from: interviews, questionnaires, surveys, etc.

• Secondary Data: data gathered/recorded by second party, i.e. someone else.

• Collected from: data from a book, etc.

Open-ended Questions

Questions that don’t limit the answers that can be given by the respondent, generates qualitative data. E.g., What do you want to do when you grow up?

Closed Questions

Questions that limit responses by the answerer generate quantitative data. E.g., What is your name?

Referencing Primary Sources

Includes interviews, questionnaires, and observations.

Name of respondent, date/time, place of interview, etc., will need to be carefully recorded for appropriate referencing.

Examples of Referencing

Footnotes—referencing listed at the bottom of page (where citation is made).

APA (American Phychological Association) - a style guide with academic writing such as essays and books.

Permission

MUST BE SOUGHT for collecting data due to laws. E.g., Australian Privacy Principles.

One method is consent forms.

Consent Forms: Components

Title, name of researchers

Purpose, reason of project to be undertaken

Requirements from participants e.g., time, cost, resources

Rights and interests of participants

Statement of whether the participant’s identity will be preserved.

Quality of Data/Information

Data Integrity - refers to quality of data

Factors that affects the quality of data:

1. Reference - how well is data/info connected to/appropriate to the matter in question.

2. Accuracy - how well do the data represent the objective truth.

3. Bias - may be intentional/not. Mostly based on factors such as culture, religion, etc. Note that people tend to support things they prefer.

4. Reliability - if all conditions above are ok - it is reliable.

Data types and structure

Data type: a set of data with predefined characteristics.

Common data types:

• Text (String)

• Numeric (Integer, Floating Point)

• Date, Characters

• Boolean - true (1) or false (0) evaluation

Common data structures:

• Array

• Record

• File

Physical and software security controls

• Encryption, Backups

• Username, passwords

• Firewall, malware protection

• Whitelisting (allow) /blacklisting (block)

Australian Private Principles

• Privacy Act 1988: dealt with handling data to the federal government/government agencies.

• Australian Privacy Principles (APPs) - 13 privacy principles.

Ethical Dilemmas

Ethical - accepted moral standards that guide behaviour.

Ethical Dilemma - situations which only allow unethical/unacceptable outcomes.

Data Visualisation

The process by which large amounts of data are taken and processed → effective graphical representations that meet the needs of clients, such as graphs, charts, etc.

Purpose: educating, informing, persuading, etc. Making reading data more interesting, understandable, and less time-consuming (allowing data to be compared when glanced at) and effective (precise).

Types of Data Visualisation

1. Charts/Graphs;

2. Flowcharts/Diagrams;

3. Images;

4. Hierarchy charts;

5. Animations;

6. Maps/Timelines/Infographics.

Infographics

Shows data/information using different elements, which include:

• Graphs, pictures, diagrams, narratives, timelines.

• Complete story of data—more engaging than traditional ways of data.

Steps to create a data visualisation

1. Investigation;

2. Interpret Primary Data;

3. Familiarise yourself with ways to visualise data (refer to previous flashcards).

4. Mockup design;

5. Create (include citations and references!);

6. Formats/Conventions;

7. Validations.

Design Principles of Data Visualisation

Functional requirements: What the visualisation should do.

→ Usability (easiness to use) & accessibility (navigation, error tolerance).

Non-Functional Requirements: how the visualisation appears rather than what it does.

This includes:

• Alignments of objects/text

• Repetition of visual elements

• Contrast of colour/tone

• Space around/between objects/texts/images

• Balance of visual appeal, harmonions.

Constraints/Scope of a solution

Constraints: Factors that limit/restrict the visualisation, such as economic, technological, social, legal, and usability constraints.

Scope of a solution: Work that must be done to complete solution.

Design tools

• Functionality: IPO (Input-Process-Output)*/Flowcharts.

*a way of how people process information.

• Annotated Diagrams

  • Mockups: hand-drawn representations of page/screen with detailed formatting information, i.e., prototype/demo model.

  • Designing an interface for a solution. Includes object, colours, and annotations when required (sometimes name of object).

Formats and Conventions

Formats: related to font type, size, background colour, etc.

Conventions: general rules followed when using a format. E.g., titles, text styles, shapes, etc.

Software Functions

Extracting data: using built-in functions, e.g., web query (Excel) and/or query (Access).

Validation of data:

1. Manual - data entered are checked for accuracy by person rather than machine, such as proofreading.

2. Electronic (machine) - ensure accuracy of data, built-in software functions e.g., range/existence/data type checking.

Evaluating data visualisations

Does the the solution meet the information needs of the client?

Relates to the initial solution, quantifiable/measurable criteria, effectiveness/efficiency, e.g., time/cost saved.

Evaluation Report - involves stating if the solution meets the information need(s) of the user.

Database and its components

A database is a structured collection of data, stored manually (e.g., telephone book) or electronically (e.g., library records).

A DBMS (Database Management System) is software that manages databases.

Key Components:

• File: A group of related data stored together.

• Record: A single entry containing multiple fields (e.g., a student's details).

• Field: A specific data point within a record (e.g., "Name," "Age").

• Character: The smallest unit of data, including letters, numbers, and symbols.

Data Dictionary

A comprehensive description of each field in a database used to plan storage of software elements, e.g., variables, data structures, objects, etc.

Should list all variable names and data type/structure. May include data’s purpose, size, description, formatting, and validation.

Table/Form

Table: displays data using columns (fields) and rows (records);

Forms: displays data for every field of a single record (basically a specific SINGLE ROW OF DATA).

Queries

The ability to select a set of specific data based on a series of criteria.

Querying - selecting specific data based on criterias.

Reports

Information produced at the end of the conversation process of manipulating the list produced from the query into useful information.

SQL

Structured Query Language—a declarative language used for querying and updating tables in a relational database. Also can be used to create tables.

Basic Syntax for Extracting Data

1. Select - list of fields to be displayed;

2. From - list the table(s) data will come from;

3. Where - list of search criteria;

4. Order By - list the fields that the results are to be sorted on. By default, ascending order;

5. And - filter records based on more than one condition being true;

6. Or - filter records when any conditions are true.

Note: AND and OR functions are to be used ONLY in WHERE syntax exists.

Spreadsheets

Software tools that essentially perform calculations and create basic charts.

Best suited when working with smaller amounts of data, performing quick calculations, and simple sorting and analysis.

Their true power is in the ability to perform complex mathematical functions. These are:

• Arithmetic operations (+, -, /, *)

• Statistical or other mathematical functions (max, min, average, median, SD)

• If (), Lookup(), etc.

Spreadsheet Terminology

• Cell: A single unit in a spreadsheet to store data.

• Columns and Rows: Vertical and horizontal arrangements forming a spreadsheet grid structure.

• Formulas: Mathematical expressions for calculations within spreadsheet cells.

• Functions (table 2.23): Predefined formulas for specific tasks, like SUM or AVERAGE.

• Relative Referencing: Cell references adjust relative to their new position.

• Absolute Referencing: Fixed cell references remain constant when copied or moved

Functions that can be used in spreadsheets

Hungarian Style Naming Convention

Labels - lbl

Textbox - txt

Buttons - btn

Option Buttons - opt

Check Box - chk

Combo Boxes - combo

Relationships in MS Access

A relationship connects two table using a common field (key).

Ensures data integrity and eliminates redundancy.

Types of relationships includes one-to-one, one-to-many (common), many-to-many (junction table).

Preparing for Relationships

• Identify common fields (e.g., StudentID in both Students and Enrollments).

• Ensure tables have primary keys.

• A foreign key is a primary key used in another table to establish a relationship.

• Close all open tables to prevent errors.

Steps to create relationship

• Go to Database Tools > Relationships.

• In Show Table, select and Add tables.

• Drag the common field between tables.

• In Edit Relationships, verify relationship type.

• Check Enforce Referential Integrity to prevent orphan records.

• Click Create to establish the relationship.

Referential Integrity

• Maintaining valid relationships – Preventing orphan records.

• Restricting deletions – A parent record cannot be deleted if child records exist.

• Ensuring data consistency – Keeping data accurate across related tables.

Relationships Types

• One-to-One:

• Each record in Table A has one matching record in Table B.
  E.g. Vehicle to Registration

• One-to-Many:

• A record in Table A can have multiple related records in Table B.
  E.g. Customer to Orders

• Many-to-Many:

• Requires a junction table to link records from both tables.

Edit/Deleting Relationships

To Edit a Relationship:

1. Open the "Relationships" window.

2. Double-click on the relationship line.

3. Modify settings as needed.

To Delete a Relationship:

1. Select the relationship line.

2. Press "Delete" to remove it.

SQL JOIN Example - Relational Database

• Definition: Relationships allow querying multiple tables using JOIN in SQL to retrieve related data.

• Example: Matching customers with their orders using CustomerID.

• SQL Query:
  SELECT Customers.CustomerID, Customers.Name, Orders.OrderID

  FROM Customers

  INNER JOIN Orders ON Customers.CustomerID = Orders.CustomerID;