Statistics Categories

Statistics Categories makes it easy to visualise and consult data relationships and predict outcomes based on the information captured. Using advanced techniques such as predictive analysis, statistical understanding, perceptual mapping and preference scaling, you can understand which characteristics your customers most associate with your product or brand, so you can discover their opinions in relation to your competitors.

Statistics Categories includes advanced analytical techniques for:

 

  • Analysing and interpreting multivariate data and their relationships in an easy and comprehensive way.

  • Transforming qualitative variables into quantitative variables by performing additional statistical operations on categorical data.

  • Graphically displaying basic relationships in any type of category studied, including market segments, medical diagnoses, political parties or biological species.

 

Analysing and easily interpreting multivariate data

 

  • Using categorical regression procedures to predict the values of a nominal, ordinal or numeric variable from a combination of predictor variables

  • Quantifying variables to maximise the Multiple R technique with optimal scaling techniques.

  • Clearly visualising data relationships using dimension reduction techniques such as biplots and perceptual maps.

  • Obtaining detailed information of relationships between more than two variables with summary charts displaying similar categories or variables.

 

Transforming qualitative variables into quantitative ones

 

  • Predicting the values of a nominal, ordinal or numerical outcome variable from a combination of categorical predictor variables.

  • Analysing two-way tables that contain some measurements of row and column correspondences, as well as display rows and columns as points on a map. Also analysing multivariate categorical data, allowing more than two variables to be used in the analyses.

  • Using optimal scaling techniques to generalise the principal component analysis procedure to handle variables of mixed levels of measurement.

  • Comparing multiple sets of variables in the same graph, once the correlation between the sets has been removed, and visually examining the relationships between two sets of objects, e.g. between consumers and products.

  • Performing multidimensional scaling that handles one or more matrices with similarity or dissimilarity (proximity).

 

Graphical display of basic relationships

 

  • Placing the relationships between the variables in a larger reference frame with optical scaling.

  • Creating perceptual maps that graphically display similar variables or categories to provide exclusive information of relationships between more than two categorical variables.

  • Using biplots and triplots to visualise relationships between cases, variables and categories; for example, you can define relationships between products, customers and demographic characteristics.

  • Visualizing further relationships between objects using preference scaling, which simplifies non-metric analyses for ordinal data and allows for more consistent results.

  • Analysing the similarities between objects and integrating the characteristics of objects in the same analysis.

 

 

 Tecnical sheet 
Statistics Categories