Socio-Economic Typologies of Coastal Communities

asNeeded

In creating the typology 42 variables were used (for a full list please see the methodology sub report. In order to meet the needs of the clustering technique the variables were standardised to fall within a scale of 0 - 1. Prior to standardisation natural log transformations were applied topull in the extremes in distribution. The K-means algorithm was applied using the methodology to develop the output areas. using 4 clusters produced the most equally sized groups and was the only solution where the smallest cluster was 60% of the size of the average cluster. The K-means algorithm was then run separately on each of the groups to generate 2 or 3 subgroups. After feedback 1 cluster group was further divided into 2 categories. A full methodology of how the dataset was created can be found within the methodology sub report. Known Limitations: - Dataset was developed for use internally by the MMO. - Where variables correlated highly and were closely associated with other measures, they were excluded them from the typology. Additional work carried out on dataset by MMO Data/GIS Officer on 07/05/2013. Six additional columns have been created in the attribute table NW_MPA, SW_MPA, S_MPA, SE_MPA, E_MPA and NE_MPA. A select query was ran on each Marine Plan Area (MPA), both inshore and offshore, selecting all typology areas within 10km of the MPA. A distance of 10km was used to match the report. All Typology areas within 10km were assigned a value of 1 and those not a value of 0. Caveat - The MPAs do not separate the typology areas clearly therefore it was decided the 10km selection area would be the set standard with the knowledge that some typology areas will overlap at MPA boundaries.