Blog Post 4: Sampling Strategies

Individuals collected for each species:

Random/systematic sampling:

Eastern Hemlock = 537

Sweet Birch =120

Yellow Birch = 100

Chestnut Oak = 104

Red Maple =162

Striped Maple = 50

White Pine = 21

For Haphazard sampling:

Eastern Hemlock = 521

Sweet Birch = 117

Yellow Birch = 100

Chestnut Oak = 100

Red Maple = 196

Striped Maple = 29

White Pine = 13

Overall more individuals were counted using random/systematic sampling methods

Time Needed for Each Sampling Type: 

Random/systematic sampling methods had a sample time of 12 hours and 35 minutes whereas haphazard sampling time of 12 hours and 50 minutes. Not a huge difference in terms of efficiency

Percentage Error:

For Random/Systematic:

Most common species: Eastern Hemlock, Data = 537, Actual = 470

(537-470)/470*100 = 14.3%

Least common species: White Pine, Data = 21, Actual = 8

(21-8)/8*100 = 163%

For Haphazard:

Most common species: Eastern Hemlock, Data = 521, Actual = 470

(521-470)/470*100 = 10.9%

Least common species: White Pine, Data = 13, Actual = 8

(13-8)/8*100 = 63%

The abundance of Eastern Hemlock was fairly similar between the two sampling methods and the percentage error was 14% for random/systematic and 10% for haphazard. The Haphazard therefore represents a more accurate population estimate as its sampling percentage error is smaller. However, as the sampling error is still greater than 5% the data would not be considered good. The abundance of White Pine is significantly different because the population is smaller therefore the difference between 21 and 13 is almost 50% of the population. This is further represented by the percentage error of 163% for the random/systematic sampling and 63% for the haphazard. Overall haphazard sampling seems to give a more accurate representation of the true population size possible because there are reduced bia’s due to sample sites being randomly chosen without knowledge of past sites sampled tho this could have problems too if the same site is chosen multiple times. More data would need to be collected to fully determine the better method for this project and maybe random/systematic works better for some types of data versus others.

 

Leave a Reply

Your email address will not be published. Required fields are marked *