Analyse the data

Descriptive statistics are used to describe the basic features of a data set and provide simple summaries and measures. Together with simple visuals like graphs, charts, diagrams, flow charts and tables, they form the basis of virtually every quantitative data analysis.

Commonly used descriptive statistics include mean, median, mode, range, and standard deviation.

To calculate the mean, add all of the values and divide by the number of values. For example, if you have five values (1, 2, 3, 4, 5), the mean would be (1+2+3+4+5) ÷ 5=3.

To calculate the median, first arrange all of the values in order from smallest to largest. If there is an odd number of values, the median will be the middle value. If there is an even number of values, the median will be the average of the two middle values.

To calculate the mode, find the value that occurs most often.

The range is the difference between the highest and lowest values in a data set. To calculate the range, subtract the lowest value from the highest value.

Calculate the square of the difference between each data point and the mean (the variance) to identify the standard deviation. Once we have calculated the variance, we take its square root to get the standard deviation.

Watch the video to learn more about data summaries.

Keep in mind that summary statistics are useful for getting a general sense of what a dataset looks like, but they don't tell the whole story. For example, they don't show how much variability there is or the shape of the distribution.

A case study showing how to complete a descriptive analysis is available here.

Use descriptive data analysis to complete tasks like:

• tracking enrolments
• monitoring the number of times a product is bought
• collating results of a survey
• determining the time taken to achieve a goal
• identifying the success rate of a product as opposed to the failure of another.

Watch

Watch the video to learn how to use Excel to complete a descriptive analysis.

Diagnostic analysis uses data to determine the causes of trends and correlations between variables.

Note: Regression modelling can be used to determine the causes of trends and relationships between variables.

The main purpose is to identify and respond to anomalies within your data. For example, if your descriptive analysis shows that there was a 10% drop in sales in June, a diagnostic analysis could explain why.

To identify the cause, the data analyst will identify additional data sources to provide insight. They might drill down to find that fewer customers were progressing to check out despite heavy site traffic and several add-to-cart actions.

Data might show that checking out was abandoned by most customers at the point they were asked to complete delivery details. A problem with the form loading may be the cause.

With further investigation, the data analyst aims to find a feasible explanation for the data anomaly.

Watch

Watch the video to learn more about diagnostic analysis.

Use diagnostic data analysis to complete tasks like:

• determining why enrolments are low
• understanding why a product is popular
• determining why a company's profits are dropping or improving.
• figuring out why a website has seen an increase in traffic.
• diagnosing technical problems within a business' digital infrastructure.
• understanding the factors that contribute to staff attrition.
• evaluating the effectiveness of products and services.

Diagnostic analysis answers the questions, 'What happened and why?'

Predictive analysis statistics is a type of data analysis that predicts future events based on past events. This analysis can be used to forecast customer behaviour, financial trends, and future demand.

Inferential statistics are used to draw conclusions from data that are subject to random variation. This means that inferential statistics can be used to make predictions about a population based on a sample.

Note: Sampling is a technique used to predict future events.

Watch

Watch the video to learn about other techniques to complete a predictive analysis.

Predictive analysis answers the question, 'What might happen in the future?'

For example, a credit card company might employ a data analyst to use logistic regression to predict whether a customer will default on their payments so they can classify them according to risk categories.

Prescriptive analytics looks at what has happened, why it happened, and what might happen to determine what should happen. What steps can the business take to avoid a future problem? What can an organisation do to capitalise on an emerging trend?

Watch

Watch the video to learn more about prescriptive analysis.

Prescriptive analysis answers the question, 'What should we do next?'