Ask phase of Data Analytics

First phase of the data analytics

The ask phase involves; understanding the problem, asking SMART questions , structural thinking and understanding the context of the data. This phase is essential for every data analysis project because it explains the problem and allows the analyst to have a clearer picture of the project.

Understanding the problem.

Understanding the problem in a project helps the analyst to focus on the main objective of the analysis.

There are different types of ways to understand a problem, some are;

• Making predictions: Prediction is the ability to forecast future events. When trying to understand a project, it is essential to use information from data to predict problems.
• Categorizing things: This is arranging various data into different stages and groups based on their common features.
• Spotting something unusual: When working with various data, there will be different anomalies and identifying them aids in understanding the problem.
• Identifying themes: It involves taking and grouping data into bigger categories.
• Discovering connection: It is finding the connection between the various data that are related or various entities.
• Finding patterns: It is solving problems using historical data or data from the past(history matching).

Types of question

The question to be asked must be SMART, which mean Specific, Measurable, Action oriented, Relevant, Time bound. When asking questions the SMART concept has to be followed.

The question should be Specific in the sense that it should define the purpose and context of the project.

Measurable and action-oriented means that the questions should be able to provide a range of significant importance and also provide a possibility for change if any.

Relevant means that the answers to the questions should be able to solve the problem at hand.

Time-bound means that the questions asked should have a specific period that it covers.

For example, a company manufacturing coloured lights trying to survey their customer preference and include the SMART question concept. They’ll have to include questions like;

‘What are the preferred types of lights”

“ how well do you use the coloured lights”

“what changes do you think could be made to the coloured lights to make them effective”

“ has coloured lights become rampant in the last two years”

When asking questions, some questions have to be avoided they are;

1. Leading questions: these are questions that have only a particular response and may not need the opinion of the person asked. For example;

“ coloured lights are better than plain lights, aren’t they?”

The above question is leading in the sense that it doesn’t allow the user to state their preference. The above questions should be ;

“What type of lights do you prefer”

2. Close-ended questions: These questions usually ask for a short answer, especially yes/no answers. For example;

“Are blue lights good for reading”

This question is closed-ended because the user will tend to give a short answer which will not be informative to the analyst. The above question should be;

“What colour of light is most favourable during reading”.

The various types of Data

Data are information gathered to be analysed by professionals and used to make decisions. The type of data are ;

Quantitative data: they are measures of character of numerical facts. They answer the question; what, how many, how often.

Qualitative data: These are explanation measures of quality. It involves the use of charts, graphs and can be gotten through polls, surveys. They answer the question; why, how can.

Data can be organised as reports and dash boards.

Reports are collections of data given to stakeholders periodically.

Dashboards are live incoming data that can be used to pull key information from data in a quick way.

Structural thinking

These are the processes of recognising the current problem, organising the information pertaining the problems by understanding the problem domain and the scope of work.

Problem domain: it is the specific area of analysis that influences every step in solving the problem. It is the first step of structural thinking.

Scope of work: It is an outline of how to work on a particular problem or problem.

A well-defined scope of work keeps every member of the team working on a problem to be on the same page.

It involves;

• Deliverables
• Timelines
• Milestones
• Reports

Another important fact when asking questions is the contextuality of the data, understanding the context of data can affect the information in the data. The context of data can be affected by conscious and subconscious biases, data contexts must be able to answer who collected the data, where was the data collected, when was the data collected, how the data was collected, and why the data was collected.

After obtaining the data, the various data specialist and analysts working on the project can evaluate the various data generated and communicate to the stakeholders before moving on to the next phase.