quantitative toolbox
These are some of our commonly used quantitative tools.

• Crosstabular analysis
  At the most basic level, we always examine how data vary by important sub-groups, such as gender, income or age. The subgroups that matter depend on your objectives and the population we’ve surveyed.
  
  
• Correlation analysis/measures of association
  These statistics measure the strength and direction of relationships between two variables. For example, purchase behavior may vary by gender and age groups, but which group has a greater impact on behavior?
  
  
• Analysis of variance
  This serves the same purpose as correlation analysis, but is used when the variable we seek to explain (the dependent variable) is continuous (year of birth or income, for example) and the independent variable is categorical (such as ethnicity). Analysis of variance measures the significance of differences between three or more means.
  
  
• Factor analysis
  This technique reduces data by finding questions that are highly correlated with each other. Two highly correlated questions could be asking the same thing or could be about different aspects of the same underlying phenomenon. Factor analysis finds general patterns in data and helps reveal the big picture.
  
  
• Multiple regression/ predictive modelling
  Through multiple regression, we develop models to explain behavior or attitudes. Each model consists of a dependent variable (what we’re seeking to explain) and a set of independent variables (what we hypothesize as the causes). Multiple regression shows how much we can explain the dependent variable using those independent variables, and which independent variables matter most. This allows clients to understand behavior and attitudes better, and develop effective strategies to change them.
  
  
• Cluster analysis
  Cluster analysis is a type of segmentation analysis. It finds homogeneous groups within a sample. The members of these groups are similar to each other in their attitudes and/or behavior. Cluster analysis allows clients to visualize and prioritize their various target markets, and to develop customized product and communications strategies for each.
  
  
• Discriminant analysis
  Discriminant analysis shows which variables best predict group membership.
  
  
• Perceptual mapping
  Perceptual mapping is particularly useful in competitive analysis. It visually maps your organization relative to your competitors on a set of variables. A typical map uses brand attributes (such as reliability, cost-effectiveness, friendliness and innovation) as the set of variables. It shows the degree to which your brand and competing brands represent each of these attributes. It tells you who "owns" particular market positioning.
  
  
• Gap analysis
  Like perceptual mapping, gap analysis shows you the degree to which your brand and competing brands "own" brand attributes - for example, which brands have the value positioning or the quality positioning. Gap analysis then shows which brand attributes matter most to consumers in your category. Clients can identify priority areas for performance improvements.
  
  
• Conjoint analysis
  Conjoint analysis tells clients the optimal value proposition for a product or service. What, for instance, is the optimal mix of price and product features (such as color, size or quantity)? Or, for a particular product mix, what price is feasible?