Study questions methods for understanding consumer preference

By Nathan Gray

- Last updated on GMT

Related tags Statistics

Commonly used statistical methods for separating and understanding consumers preferences may not be working as well as believed, according to a new study.

Writing in the Journal of Sensory Studies​, researchers from Kansas State University, USA, question whether statistical package clustering (SPC) – a common method for determining consumer groups – is the best way to separate consumers to find their most or least liked products.

“The findings from this research show that the assumption that cluster analysis will produce clusters containing consumers who have the same most or least liked products is false,”​ wrote the researchers, led by Edgar Chambers IV.

The authors added that the finding is important “because it shows that clustering consumers using typical SPC may not produce the homogeneous segments that researchers would like to obtain.”

Consumer preference

Chambers IV and colleagues said that ensuring new products satisfy specific groups of consumers is of great importance to the success of new product development.

They noted that in the majority of sensory studies, cluster analysis has been used to segment consumers:

“Researchers often analyse mean values of products for consumer segments, presuming that the segmented consumers like or dislike similar products,”​ said Chambers IV and colleagues.

The new study questioned whether this presumption is correct by investigating how well most and least liked products match for individual members of the same segments.

Study details

Four statistical package clustering (SPC) methods were used to assess data from a sensory study on food preference.

The products most frequently rated highest in each cluster were examined against four manually extracted clustered groups, to compare the results.

Chambers IV and co-workers reported that the standard SPC “was not found to separate consumers appropriately to understand their ranking/rating of most/least liked products.”

“For these data, additional manual clustering was necessary to produce consumer cluster segments where the consumers within each group had the same highest or lowest scoring products,”​ they explained.

The researchers concluded that, new SPC methods “should be developed that generate cluster segments that are more homogeneous”​ in order to optimise analysis of data for consumer preference mapping.

Source: Journal of Sensory Studies
Published online ahead of print, doi: 10.1111/j.1745-459X.2011.00337.x
“Statistical package clustering may not be best for grouping consumers to understand their most liked products”
Authors: R. Yenket, E. Chambers IV, D.E. Johnson

Related topics Science

Related news

Follow us


View more