Research Methods and Data Types

Study Notes

Research Studies

A research study follows a structured process, beginning with defining a research question and identifying the relevant target population.
Defining variables, obtaining a sample, employing an appropriate study design, using valid statistical methods, and reporting results clearly are integral parts of a research study.
Examples of research questions include: "What are the important factors to flatten the curve of coronavirus in Australia?"

Population vs Sample

A population refers to all possible measurements within a given context, while a sample is a subset of the population.
Samples are used to make inferences about the broader population.

Types of Data

Categorical Data: Each observation falls into one of two or more categories.
- Nominal: Categories have no natural order (e.g., gender, ethnicity).
- Ordinal: Categories have a natural order (e.g., disease severity: mild, moderate, severe).
Numerical Data: Observations take numeric values
- Discrete: Countable values (e.g., the number of patients).
- Continuous: Any value within a range (e.g., weight, height).

Categorical Data Summarisation

Frequency Tables: Show counts or proportions in each category.
Contingency Tables: Examine relationships between two categorical variables and can display counts and percentages.

Cross-Tabulation (Contingency) Tables

A cross-tabulation table helps analyze the relationship between two categorical variables.
An example is a table showing the connection between antiseptic use and patient outcome, allowing for the evaluation of antiseptic's impact.

Visual Displays for Categorical Data

Bar Charts: Used for both nominal and ordinal categorical variables, bars should be presented in order of frequency for easy comparison.
Pie Charts: Can be used for categorical data, but bar charts are generally preferred for clarity.
Clustered Bar Charts: Employed to examine the association between two categorical variables.
An example with crime data illustrates how to use a bar chart and a clustered bar chart.

Example with Crime Data

A study on prisoners in 1909 summarized crime data using a bar chart, highlighting the frequency of different crimes.
Stealing was the most common crime, while fraud was less common.

Clustered Bar Charts for Two Categorical Variables

A clustered bar chart can be used to display the relationship between drinking status and crime type.
In an example, it shows a difference in crime types between drinkers and abstainers, suggesting a connection between drinking and certain crimes.

Summary

Categorical data can be nominal or ordinal.
Frequency tables and contingency tables are used to summarize categorical data.
Bar charts for single categorical variables, and clustered bar charts for relationships between two categorical variables are used in graphical displays.
The choice of statistical methods depends on the data type, whether categorical or numerical.

Description

This quiz covers the fundamental concepts of research studies, including the structured process of research, the distinction between population and sample, and the types of data. You will learn about categorical and numerical data, as well as their classifications. Test your understanding of these essential research methodologies.