Stratified Random Sampling

This method involved identifying plant species within selected 1metre squares (or quadrats) in each of the six areas. Each of the six distinct areas was a strata which was sampled separately. A zero co-ordinate point ((0,0) or origin) was selected for each area. (by convention, this is usually the bottom left-hand corner, this was the point chosen for the Wet Woodland Survey). By taking pairs of random numbers from a random number table, selected one metre squares could be obtained. If pairs of random numbers gave a point lying outside of the area, they were ignored and the next pair of numbers were selected from the number table.

samplegd.jpg (32696 bytes)

A 10x10 metre area is split into an imaginary grid of 100x1m². The origin is chosen, by convention this is usually the bottom left-hand corner. The origin in the actual area could be marked by placing a cane in the ground. Then, a pair of two-digit random numbers are taken from a random number table, in this case, 0507. Using a transect line (a long rope marked at 1 metre intervals), 5 metres is measured out horizontally from the origin. From this new point 7 metres is measured vertically. Now a 1m² quadrat (square wooden frame of side 1metre) can be placed on the ground at this point and the species within it identified (shown as the black square above).

The reason that 1 m² squares were selected was to allow for the size of the largest ground cover plants growing in any of the six distinct areas. If a smaller quadrat square had been used i.e. 0.25m², it would not have been suitable for areas with larger plants. Once selected, 1m² quadrats had to be used throughout the survey so as to maintain consistency in the results.