Data Analytics 101 Part 5 of 5: Presenting findings and recommendations

All that hard work and we are almost at the end. Business (and many parts of life) require effective communication:  your analysis can come to live with the right narratives and visualisations. It is key to be able to  effectively communicate with data, present findings and make recommendations to decision makers.

Data within a business context

If you want to know more about any of the topics explored in the module, you can find further resources here:

• Access the Storytelling with Data resources in the form of a blog and a podcast to find out more about the techniques for data storytelling and for practical examples of the techniques discussed in this module.
• ICAEW presents webinars on various topics including one on data driven decision making.
• TowardsDataScience has an article containing further information on data storytelling.

Data visualisation for presentations

If you want to know more about any of the topics explored in the module, you can find further resources here:

• The Office for National Statistics discusses their findings on their use of visualisations to communicate to the wider public.
• ICAEW has a webinar called ‘The 7 Hats of Data Visualisation Design’ that looks at the skills needed to be a ‘superhero’ in data visualisation.
• This article discusses exactly when a visualisation is a good choice for communicating findings.
• For inspiration when creating visualisations: this Tableau article shares some beautiful examples, and this article on Analytics Vidhya also demonstrates what good visualisation looks like.

And this concludes the five-part series on Data Analytics, hope this has been useful for you, and thanks for reading!

Data Analytics 101 Part 4 of 5: Decision Making

What do you do with all the insights developed in the previous stages of data analytics?
Yes, it is time to decide. Data-driven decision making is a key skill set to acquire and hone in this big data era.

Time series modelling

If you want to know more, you can find further resources here:

• TowardsDataScience demonstrates how to forecast data, including discussing how to identify the best model.
• The statsmodels library is used frequently in this module. Find out more about the time series analysis (tsa) package including a list of commands.
• Learn more from Machine Learning Mastery on autocorrelation and partial autocorrelation.

Anomaly detection

If you want to know more, you can find further resources here.

• DataCamp provides useful cheat sheets and tutorials that detail the various uses and commands in packages. These include a cheat sheet for the sklearn package that is used frequently within this module.
• DataCamp also has tutorials examining the sklearn package to perform KMeans clustering and Principal Component Analysis.
• TowardsDataScience provides an article explaining how to use techniques such as Principal Component Analysis and clustering to detect anomalies.
• Medium has an article explaining unsupervised learning and how it can be used in detail.This includes examining different types of clustering, dimension reduction techniques such as PCA and methods.

Ethical considerations

If you want to know more, you can find further resources here:

• Access this link for the UK Government’s Framework for incorporating Data Ethics into public sector data projects. This includes practical advice to ensure ethical issues are considered throughout.
• Royal Society have examined Data Ethics in depth through a series of seminars, roundtables and in-depth research with professionals in various fields, summarised in this report.
• The Alan Turing Institute is dedicated to using analytics and Artificial Intelligence ethically. To ensure this, they have published an online guide about data ethics.

Any other resources that are relevant to decision making? Please share in the comments below.

Data Analytics 101 Part 3 of 5: Analysing the Data

This is the third part of the series (you may refer to the first and second articles if interested) To analyse data meaningfully, it is necessary to understand the different techniques available (and how/when to use them),  and produce data visualisations to communicate the messages and insights.

a) Data manipulation techniques

If you want to know more about any of the topics explored in the module, you can find further resources here:

• The pandas website provides detailed information on the various uses of functions within the pandas package for Python. Make a start by looking at the user guides under this link.
• Information on how pandas can be used to select data from a dataset and the drop command is available under this link..
• DataCamp provides useful cheat sheets including one covering data wrangling techniques examined in the previous unit and the manipulation techniques examined in this module.
• A useful summary of the different way to manipulate rows and columns is available under this link and covers selecting, adding and deleting.
• Access this link to find out more about aggregating data using the groupby function, including a list of the aggregation functions.

b) Summary statistics in exploratory data analysis

If you want to know more about any of the topics explored in the module, you can find further resources here:

• A deeper look at summary statistics within Python can be found in this article from Real Python.
• TowardsDataScience provides a useful article that explains hypothesis testing in further detail under this link.
• Further documentation on the t-test commands in the SciPy package: scipy.stats.ttest_1samp and scipy.stats.ttest_ind.

c) Basic visualisations in exploratory data analysis

If you want to know more about any of the topics explored in the module, you can find further resources here:

• DataCamp provides a useful cheat sheet that summarises the main visualisation commands in the Seaborn Package.
• TowardsDataScience examines the different data visualisations with the corresponding code and a breakdown of what visualisation to use when in the article under this link.
• Access this TowardDataScience article for a walkthrough of performing exploratory data analysis including summary statistics and visualisations.

Any thoughts on data analysis? Leave your insights in the comments below!