To deliver value with our data we need to be confident that we are saying is correct and others need to be confident that the data, insight and forecasts we share with them are accurate. Unfortunately the measures we analyse will be affected by random chance and external and internal factors that can make 100% accuracy unachievable. In this box set we will.
- Share tools and techniques to allow you to measure the impact of chance and understand the limitations of the data.
- Discuss the ways that our data can become bias and ways we can manage this
- Teach you to find the right balance between cost/time and accuracy. And make sure the MI, Insight and Forecasts we share are “good enough”
An Introduction to Accuracy and confidence
Understand the key statistical considerations that will ensure your conclusions are well founded and can drive confidence. The separate Statistical Confidence box set is an opportunity to drive this further
Descriptive Statistics Part 1
It is easy to lie with statistics it is easier to lie without them. Whether we are an analyst or a decision maker a basic knowledge of statistical knowledge is essential to;
- Understand the predictability, variability and repeatability of our data and allow us to separate the exceptions from the normal
- Ensure we are giving the right message
- Make sure we aren’t being misled and are focussing on the normal rather than the rare exceptions
In this module we the introduce the principles of statistics including, standard deviation, standard error and confidence levels. The accompanying excel workbook allows you to pause and play, and practice with real life examples at your own speed.
Descriptive Statistics Part 2
This module talks you through how to use the examples
Calculating Sample Size
Not having enough data or doing enough checks can lead to erroneous conclusions, but if we don’t know when to stop testing, checking and analysing then we drive cost into our organisations while delaying improvements. In this module we look at calculating the right sample size to be able to draw conclusions from our data that are not the result of random chance.
Our data often throws up anomalies that can be counterintuitive. Million to one chances happen far more often than we would expect. In this module we will explore Statistical Probability and Bayes Theorem. A technique that will help you explain these anomalies and prevent you drawing incorrect conclusions.
Understanding & Managing Sample Bias
- How to Identify Sample Bias & ensure you reduce the likelihood of bias
- The impact on your data and insight validity from Sample Bias and the potential impact to your business
- Common Sample Bias issues to look out for