Descriptive Analysis

In quantitative research, we often collect data from hundreds or even thousands of people. Based on the collected data, we try to come to general conclusions about experiences of a group of people rather than individuals. In other words, we aim to get an overview of our data by:

• describing and presenting characteristics of our sample as a whole
• summarising the outcomes we observe
• becoming aware of patterns or irregularities in the data

Developing such an overview of the data, is called descriptive analysis.

Descriptive analysis is (a) the process of describing and summarising a specific dataset and (b) an important step to prepare for further analysis which aims to make general conclusions about the population from which the sample has been taken.

Researchers can do descriptive analysis in two ways:

1. Mathematically by calculating summary values
2. Visually by generating graphs and figures.

(Intellspot, n.d.)

It is possible to find out about the distribution of the data and the relationships between variables both mathematically and visually.

Let’s use an example. A researcher investigates language skills of refugees starting group therapy. The researcher is interested in learning about the following aspects of language skills:

1. Distribution, i.e., the number of times an outcome occurs in the data (‘frequency’). For example, how many attendees have beginner, intermediate or advanced language skills?
2. Central tendency, i.e., the centre of the distribution. For example, what is the average language skill level in the group?
3. Dispersion, i.e., the extent to which a distribution is stretched or squeezed. For example, how big is the difference in language skills between the least and most advanced attendee?
4. Correlations, i.e., relationships between variables. For example, do people who have lived longer in a new country tend to have higher language skill levels?

In order to handle such a large amount of data, such descriptive analysis is usually done using computer software.

(Author: Leonie Ader)

Videos:

Books:

Introduction to epidemiological study design by Jayati Das-Munshi et al. (2020)  

This resource is a chapter of an epidemiology methods book for psychiatric epidemiology. The information about descriptive studies is most relevant to this content page.

(Academic reference: Ford, T., Das-Munshi, J., & Prince, M. (2020). Introduction to epidemiological study design. In Das-Munshi, J., Ford, T., Hotopf, M., Prince, M., & Stewart, R.  (Eds.), Practical psychiatric epidemiology (2nd ed., pp. 127-143). Oxford University Press)

Essential medical statistics by Betty Kirkwood and Jonathan Sterne (2003)

This resource is a statistical methods book for epidemiologists. Descriptive analysis techniques are primarily presented in Part A and Part B.

(Academic reference: Kirkwood, B., & Sterne, J. (2003). Essential medical statistics (2nd ed). Wiley-Blackwell. https://www.wiley.com/en-gb/Essential+Medical+Statistics,+2nd+Edition-p-9780865428713)

Websites:

Descriptive statistics by Research Connections (2022)

This website gives a structured list and definitions of methods for descriptive statistics, including some examples of how to calculate them.

(Academic reference: Child Care & Early Education Research Connections. (2022, June 28). Descriptive statistics. https://www.researchconnections.org/research-tools/descriptive-statistics)

Descriptive statistics by Curtin University (2022)

This website includes more advanced descriptions of descriptive statistics for different types of variables. It introduces interactive graphs and self-test to test your knowledge.

(Academic reference: Curtin University (2022, June 28). Descriptive statistics. https://libguides.library.curtin.edu.au/uniskills/numeracy-skills/statistics/descriptive)

Videos:

Manuals & Guides:

Websites:

Articles:

Reports:

Presentations: