Vegetation sample methods

One of the things that I found interesting as we analyzed the vegetation data that we collected was that the two methods of collecting data we compared were not interchangeable. We used a Bland-Altman plot to determine if they were comparable methods of collecting data and found many values that were out of the tolerance range, indicating that they are not equivalent methods of collecting data. However this plot does not tell which method is better for sampling, merely that they do not record similar data. This is essential because if the visual analysis is more accurate then using the sample point program then all of the that data we collected might not be very accurate or give us a good representation of what the area is actually like. This means that if another researcher uses this experiment as the basis of a further study they must also perform the same methods that we did and can not use the visual analysis method or their results will not be correlated to our research. One way that would could further examine the differences between these two sampling methods is we could do all 17 transect with both sampling methods. That way when further researchers come to use our data they will have two types of tests to determine whether the shoreline vegetation is actually changing or not. Doing this would lead to more accurate future data then just picking one method. However due to time constraints this was not possible and we must settle for only performing one method.

Bland-Altman graph of the visual vs the sample point methods. (Does not show several values far above the upper limit.)