Data Science in CX Index

The Data Science team in CX Index is second to none.

The platform lets you communicate the outcomes with stakeholders across your brand network.

And do so in easy to understand terms.

Our statistic experts can identify the appropriate statistical modelling for a given scenario.

Once that's done we then write the programme to automate this analysis.

Regression

In statistical modelling, regression analysis is a set of statistical processes.

These processes estimate the relationships between variables.

They estimate relationships between

• a dependent variable and
• one or more independent variables

Dependent variables are often called the 'outcome variable'.

Independent variables are often called 'predictors', 'covariates', or 'features'.

Contact Center Conditions

CX Index can present contact center conditions in a visual format.

We do this through predictions, statistics, forecasts and graphs based on selected options.

Options you can select include:

• contact center agents
• clients
• interactions and
• CX Index platform data

We use Linear Regression for calculations.

We base them on the Ordinary Least Squares data model.

P-value

What is the P-value or probability value?

It's the probability of obtaining test results at least as extreme as the results actually observed during the test. Doing this you assume that the null hypothesis is correct.

R-squared

R-squared (R2) is a statistical measure. You can use it in a regression model.

It shows the percentage of the variance in the dependent variable that the independent variables explain.

Adding more independent variables / predictors to a regression model increases the R-squared value.

This can tempt people to add even more independent variables to the regression model.

(Called overfitting this can return an unwarranted high R-squared value).

How do you determine how reliable the correlation is?

Use an Adjusted R-squared to check the reliability. It can also show you the influence of the addition of independent variables.

These include:

• CSAT
• NPS
• CES

Agent regressors:

These include:

• Work from home (hours per month)
• Work in office (hours per month)
• Years of experience
• Level of experience
• Age (of the agent)
• Age (of the customer)