Glossary of Statistical Terms: Definition Cluster analysis
The cluster analysis separates elements from a sample into groups called clusters. This process is carried out after the collection of the data. Features like demographic data, attitudes and so forth can be utilized in order to divide the data with the aim that elements within each cluster have similar properties. At the same time, the aim is to find the maximum difference between the clusters within the investigation. A comparison can then be carried out between the clusters.
The cluster analysis is often used in the field of market research in order to identify different types of buyers on the basis of their consumption habits. For example: taking 15 questions into account on the subject of sport, four clusters are formed (sports enthusiasts, those with a basic sports interest, opportunistic sportsmen and those not interested in sport). These four groups will be used to determine how and where different types of athletes buy sporting goods. One result could be, for example, that sports enthusiasts prove the most eager to try and test a new product – they are willing to pay higher prices to purchase it. Those with a basic sports interest are more selective in their choice, carefully browsing the internet to make their choice and paying more attention to what they eventually end up paying. A company dedicated to sports enthusiasts therefore invests more in an attractive showroom and scraps its plans for an online shop.
Please note that the definitions in our statistics encyclopedia are simplified explanations of terms. Our goal is to make the definitions accessible for a broad audience; thus it is possible that some definitions do not adhere entirely to scientific standards.
- Central limit theoremCentral limit theorem
- Cluster analysisCluster analysis
- Cluster sampleCluster sample
- Coefficient of correlationCoefficient of correlation
- Competitor analysisCompetitor analysis
- Conditional probabilityConditional probability
- Confidence levelConfidence level
- Conjoint analysisConjoint analysis
- Cross-sectional dataCross-sectional data